Eastern Mediterranean Health Journal [2020; Vol.26, Issue 11]

La Revue de Santé de la Méditerranée orientale Eastern Mediterranean Health Journal Volume 26 / No. 11 November/Novembre 11 ددع / نورشعلاو سداسلا دلجملا يناثلا نيرشت/ربمفون 2020 Nurses make up the largest group of health-care providers and are the first point of contact in health service delivery of diabetes in terms of detection, treatment and rehabilitation. This year’s World Diabetes Day coincides with the International Year of the Nurse and the Midwife, as designated by the World Health Assembly, and looks to focus on and promote the role of nurses in the prevention and management of diabetes. Eastern Mediterranean Health Journal IS the official health journal published by the Eastern Mediterranean Regional Office of the World Health Organization. It is a forum for the presentation and promotion of new policies and initiatives in public health and health services; and for the exchange of ideas, concepts, epidemiological data, research findings and other information, with special reference to the Eastern Mediterranean Region. It addresses all members of the health profession, medical and other health educational institutes, interested NGOs, WHO Collaborating Centres and individuals within and outside the Region. طسوتلما قشرل ةيحصلا ةلجلما ةماعلا ةحصلا في ةديدلجا تاردابلماو تاسايسلا ميدقتل برنم ىهو .ةيلماعلا ةحصلا ةمظنمب طسوتلما قشرل ىميلقلإا بتكلما نع ردصت ىتلا ةيمسرلا ةلجلما ىه قشر ميلقإب اهنم قلعتي ام ةصاخو ،تامولعلما نم كلذ يرغو ثاحبلأا جئاتنو ةيئابولا تايطعلماو ميهافلماو ءارلآا لدابتلو ،اله جيوترلاو ةيحصلا تامدلخاو ةمظنم عم ةنواعتلما زكارلماو ،ةينعلما ةيموكلحا يرغ تماظنلما اذكو ،ةيميلعتلا دهاعلما رئاسو ةيبطلا تايلكلاو ،ةيحصلا نهلما ءاضعأ لك لىإ ةهجوم ىهو .طسوتلما .هجراخو ميلقلإا فى ةحصلاب ينمتهلما دارفلأاو ةيلماعلا ةحصلا La Revue de Santé de la Méditerranée Orientale EST une revue de santé officielle publiée par le Bureau régional de l’Organisation mondiale de la Santé pour la Méditerranée orientale. Elle offre une tribune pour la présentation et la promotion de nouvelles politiques et initiatives dans le domaine de la santé publique et des services de santé ainsi qu’à l’échange d’idées, de concepts, de données épidémiologiques, de résultats de recherches et d’autres informa- tions, se rapportant plus particulièrement à la Région de la Méditerranée orientale. Elle s’adresse à tous les professionnels de la santé, aux membres des instituts médicaux et autres instituts de formation médico-sanitaire, aux ONG, Centres collaborateurs de l’OMS et personnes concernés au sein et hors de la Région. Correspondence Editor-in-chief Eastern Mediterranean Health Journal WHO Regional Office for the Eastern Mediterranean P.O. Box 7608 Nasr City, Cairo 11371 Egypt Tel: (+202) 2276 5000 Fax: (+202) 2670 2492/(+202) 2670 2494 Email: emrgoemhj@who.int Members of the WHO Regional Committee for the Eastern Mediterranean Afghanistan . Bahrain . Djibouti . Egypt . Islamic Republic of Iran . Iraq . Jordan . Kuwait . Lebanon Libya . Morocco . Oman . Pakistan . Palestine . Qatar . Saudi Arabia . Somalia . Sudan . Syrian Arab Republic Tunisia . United Arab Emirates . Yemen طسوتلما قشرل ةيلماعلا ةحصلا ةمظنلم ةيميلقلإا ةنجللا ءاضعأ نادلبلا ةيملاسلإا ناريإ ةيروهجم . ايبيل . سنوت . نيرحبلا . ناتسكاب . ةدحتلما ةيبرعلا تاراملإا . ناتسناغفأ . ندرلأا برغلما . صرم . نانبل . تيوكلا . رطق . ينطسلف . نماُع . قارعلا . لاموصلا . نادوسلا . تيوبيج . ةيروسلا ةيبرعلا ةيروهملجا نميلا . ةيدوعسلا ةيبرعلا ةكلملما Membres du Comité régional de l’OMS pour la Méditerranée orientale Afghanistan . Arabie saoudite . Bahreïn . Djibouti . Égypte . Émirats arabes unis . République islamique d’Iran Iraq . Libye . Jordanie . Koweït . Liban . Maroc . Oman . Pakistan . Palestine . Qatar . République arabe syrienne Somalie . Soudan . Tunisie . Yémen Subscriptions and Permissions Publications of the World Health Organization can be obtained from Knowledge Sharing and Production, World Health Organization, Regional Office for the Eastern Mediterranean, PO Box 7608, Nasr City, Cairo 11371, Egypt (tel: +202 2670 2535, fax: +202 2670 2492; email: emrgoksp@who.int). Requests for permission to reproduce, in part or in whole, or to translate publications of WHO Regional Office for the Eastern Mediterranean – whether for sale or for noncommercial distribution – should be addressed to WHO Regional Office for the Eastern Mediterranean, at the above address; email: emrgoegp@who.int. Cover photo: © WHO / Budi Chandra EMHJ is a trilingual, peer reviewed, open access journal and the full contents are freely available at its website: http://www/emro.who.int/emhj.htm EMHJ information for authors is available at its website: http://www.emro.who.int/emh-journal/authors/ EMHJ is abstracted/indexed in the Index Medicus and MEDLINE (Medical Literature Analysis and Retrieval Systems on Line), ISI Web of knowledge, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), Embase, Lexis Nexis, Scopus and the Index Medicus for the WHO Eastern Mediterranean Region (IMEMR). © World Health Organization (WHO) 2020. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO licence (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Disclaimer The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. The authors alone are responsible for the views expressed in this publication and they do not necessarily represent the views, decisions or policies of the institutions with which they are affiliated. If authors are staff members of the World Health Organization, the authors alone are responsible for the views expressed in this publication and do not necessarily represent the decisions, policy or views of the World Health Organization. ISSN 1020-3397 Cover 26-7.indd 4,6 23/07/2020 01:02 Editorial Tackling diabetes: how nurses can make the difference Hicham El Berri, F Gulin Gedik, Jamal Belkhadir, Howard Catton, Asmus Hammerich, Arwa Oweis and Slim Slama ...............1318 Commentary Tobacco industry commissioned reports on illicit tobacco trade in the Eastern Mediterranean Region: how accurate are they? Erin Sandberg, Allen WA Gallagher and Raouf Alebshehy ..................................................................................................................................1320 Research articles Health information systems in Jordan and Palestine: the need for health informatics training Hussein Jabareen, Yousef Khader and Adel Taweel ...............................................................................................................................................1323 Barriers to initiation of insulin therapy in poorly controlled type 2 diabetes based on self-determination theory Armin Rajab, Pegah Khaloo, Soghra Rabizadeh, Hamid Alemi, Salome Salehi, Reza Majdzadeh, Hossein Mirmiranpour, Assadollah Rajab, Alireza Esteghamati and Manouchehr Nakhjavani............................................................1331 Prevalence of non-reporting of hospital medical errors in the Islamic Republic of Iran Mehrdad Askarian, Seyyed M. Sherafat, Maryam Ghodsi, Zahra Shayan, Charles Palenik, Nahid Hatam and Yavor Enchev ...................................................................................................................................................................................1339 Assessment of nurses’ patient safety culture in 30 primary health-care centres in Tunisia Mohamed Ayoub Tlili, Wiem Aouicha, Mohamed Ben Dhiab and Manel Mallouli ...................................................................................1347 Travel burden and geographic access to health care among children with cancer in Saudi Arabia Abdulrahman Alsultan, Abdullah Aljefri, Mouhab Ayas, Musa Alharbi, Nawaf Alkhayat, Faisal Al-Anzi, Fawwaz Yassin, Fawaz Alkasim, Qasim Alharbi, Shaker Abdullah, Mohammed Burhan Abrar and Wasil Jastaniah ...........................................................................................................................................................................................................1355 Prevalence of refractive error and visual impairment among school-age children of Hargesia, Somaliland, Somalia Zahra Abdi Ahmed, Saif Hassan Alrasheed and Waleed Alghamdi ...................................................................................................................1362 Risk factors associated with worse outcomes in COVID-19: a retrospective study in Saudi Arabia Anas Khan, Saqer Althunayyan, Yousef Alsofayan, Raied Alotaibi, Abdullah Mubarak, Mohammed Arafat, Abdullah Assiri and Hani Jokhdar ................................................................................................................................................................................1371 Quality utilization of antenatal care and low birth weight: evidence from 18 demographic health surveys Saverio Bellizzi and Susanna Padrini ..........................................................................................................................................................................1381 Macromineral enrichment of white bread reduces postprandial glycaemia without altering sensory properties: a crossover study Rania El Khoury, Noor El Solh, Ammar Olabi, Imad Toufeili, Sani Hlais and Omar Obeid ......................................................................1388 Reviews Rationing access to total hip and total knee replacement in the Islamic Republic of Iran to reduce unnecessary costs: policy brief Mohammad Soleimani, Shoresh Barkhordari, Farhad Mardani, Nasrin Shaarbafchizadeh and Fatemeh Naghavi-Al-Hosseini .......................................................................................................................................................................................1396 Application of geographic information systems in maternal health: a scoping review Leila Ahmadian, Fatemeh Salehi and Kambiz Bahaadinbeigy ...........................................................................................................................1403 Vol. 26.11 – 2020 La Revue de Santé de la Méditerranée orientale Eastern Mediterranean Health Journal Short research communications Reproductive and behavioural risk factors of low birthweight among newborns in Al Thawra Hospital, Sana’a, Yemen Idayu Idris, Manal Sheryan, Qistina Ghazali and Azmawati Nawi ..................................................................................................................1415 Barriers to the use of dental services by children in Lebanon and association with parental perception of oral health care Ingrid Karam, Miran A. Jaffa and Joseph Ghafari ..................................................................................................................................................1420 Report Profil épidémiologique d’une intoxication au méthanol, El Hajeb (Maroc) Sanah Essayagh, Mariama Bahalou, Meriem Essayagh et Touria Essayagh .................................................................................................1425 WHO events addressing public health priorities Enteric and diarrhoeal diseases surveillance, prevention and control in the Eastern Mediterranean Region ........1430 Ahmed Al-Mandhari Editor-in-Chief Arash Rashidian Executive Editor Ahmed Mandil Deputy Executive Editor Phillip Dingwall Managing Editor Editorial Board Zulfiqar Bhutta Mahmoud Fahmy Fathalla Rita Giacaman Ahmed Mandil Ziad Memish Arash Rashidian Sameen Siddiqi Huda Zurayk International Advisory Panel Mansour M. Al-Nozha Fereidoun Azizi Rafik Boukhris Majid Ezzati Hans V. Hogerzeil Mohamed A. Ghoneim Alan Lopez Hossein Malekafzali El-Sheikh Mahgoub Hooman Momen Sania Nishtar Hikmat Shaarbaf Salman Rawaf Editorial assistants Nadia Abu-Saleh, Suhaib Al Asbahi (graphics), Diana Tawadros (graphics) Editorial support Guy Penet (French editor) Eva Abdin, Fiona Curlet, Cathel Kerr, Marie-France Roux (Technical editors) Ahmed Bahnassy, Abbas Rahimiforoushani, Manar El Sheikh Abdelrahman (Statistics editors) Administration Iman Fawzy, Marwa Madi Web publishing Nahed El Shazly, Ihab Fouad, Hazem Sakr Library and printing support Hatem Nour El Din, Metry Al Ashkar, John Badawi, Ahmed Magdy, Amin El Sayed, Gehane Al Garraya Cover and internal layout designed by Diana Tawadros and Suhaib Al Asbahi Printed by WHO Regional Office for the Eastern Mediterranean, Cairo, Egypt EMHJ – Vol. 26 No. 11 – 2020Editorial 1318 Tackling diabetes: how nurses can make the difference Hicham El Berri,1 F Gulin Gedik,2 Jamal Belkhadir,3 Howard Catton,4 Asmus Hammerich,5 Arwa Oweis 6 and Slim Slama 7 1Medical Officer for NCD Management, Department for UHC/Noncommunicable Diseases and Mental Health, World Health Organization Regional Office for the Eastern Mediterranean, Cairo, Egypt. 2Coordinator, Health Workforce Development, World Health Organization Regional Office for the Eastern Mediterranean, Cairo, Egypt. 3Regional Chair, International Diabetes Federation Middle East and North Africa, Rabat, Morocco. 4Chief Executive Officer, International Council of Nurses, Geneva, Switzerland. 5Director, Department for UHC/Noncommunicable Diseases and Mental Health, World Health Organization Regional Office for the Eastern Mediterranean, Cairo, Egypt. 6Regional Advisor, Nursing and Midwifery, World Health Organization Regional Office for the Eastern Mediterranean, Cairo, Egypt. 7Regional Advisor for Noncommunicable Diseases Prevention, Department for UHC/Noncommunicable Diseases and Mental Health, World Health Organization Regional Office for the Eastern Mediterranean, Cairo, Egypt. (Correspondence to: Hicham El Berri: elhi@who.int). Citation: El Berri H; Gedik FG; Belkhadir J; Catton H; Hammerich A; Oweis A; et al. Tackling diabetes: how nurses can make the difference. East Mediterr Health J. 2020;26(11):1318–1319. https://doi.org/10.26719/2020.26.11.1318 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). This year, World Diabetes Day on 14 November coincides with the International Year of the Nurse and the Midwife, and therefore focuses on highlighting the role of nurses in the prevention and management of diabetes. Diabetes is recognized as an important cause of premature death and disability globally and in the Eastern Mediterranean Region, where its prevalence has been steadily increasing since 1990, (1). Although the annual decline of the risk of dying from a major noncommunicable disease between the ages of 30 and 70 years is slowing globally, diabetes is showing a 5% increase in attributed premature mortality (1). In 2016, diabetes was the direct cause of 1.6 million deaths globally and 43% of all deaths before the age of 70 years occur due to high blood glucose (2). Overweight and obesity are the strongest risk factors for type 2 diabetes (3,4). In addition, diabetes increases the risk of heart disease and stroke and is a leading cause of blindness, lower limb amputation and kidney failure. A study conducted in 35 countries indicated that people living with diabetes are more likely to experience catastrophic health expenditures with an estimated increase of 4% between diabetic and non- diabetic individuals, regardless of their insurance status (5). The Eastern Mediterranean Region has the highest rates of diabetes worldwide with more than 43 million people living with the disease (1). Many countries in the Region are struggling to meet the health-care needs of people with chronic diseases. Diabetes and its complications are a great economic challenge in advancing universal health coverage. Several obstacles exist in terms of organization of health and care systems to prevent and manage diabetes, including insufficient public investment, shortage of health workers, lack of availability and accessibility of medicines, and insufficient information for decision-making (6). In addition, the COVID-19 pandemic has taught us how vulnerable our systems are in terms of preparedness and imposed challenges to the provision of care for patients with diabetes (7,8). Health workforce shortages in the Region have a significant impact on the availability and accessibility to services, which implies the need for interventions to maximize their functions and performance in addressing the 5.9 million of shortage of nurses, 17% of which is in the Region (9). Health workers, at primary care level, play an important role in prevention and management of non-communicable diseases including diabetes. Nurses, as the largest group of health-care providers and the point of first contact, should take a leading role in diabetes detection, treatment and rehabilitation, as well as supporting health promotion and prevention efforts to people living with diabetes, thus allowing them to participate in the decision-making regarding diabetes care. While recognizing that adequate numbers of well- educated nurses are key to maintaining essential health services to effectively address the rise in noncommunicable diseases including diabetes, the need for strengthening the nursing workforce in the Region remains as a challenge. Increases in numbers have not been sufficient to match population growth; moreover, there has even been a decline in nursing numbers in almost half of countries in the Region since 2010 (10,11), affecting nurses’ ability to perform in their full capacity and scope of practice. Diabetes care should also be reviewed in nursing training and updated in accordance with recent developments. Furthermore, nursing leadership in policy and advocacy is imperative to change practice and expand nursing capacity to address diabetes in the Eastern Mediterranean Region (12). It is essential that nurses take on this leadership role in a more assertive way, working with decision-makers to achieve positive health outcomes and provide an effective quality of care at a lower cost (13). The nursing contribution to the health system is illustrated by six key features to prevent and manage diabetes as follows (14): 1) Care coordination to ensure that the patient’s health needs are met over time; 2) Being part of a multidisciplinary approach to care based on an integrated relationship between health professionals, allowing different practitioners to seamlessly work as a team to improve the quality of care; 3) mobilizing and empowering the nursing workforce to specialize in 1319 EMHJ – Vol. 26 No. 11 – 2020Editorial noncommunicable diseases to improve cost-effective and sustainable treatments; 4) improving access to care; 5) empowering individuals and the community; and 6) harnessing technology to maintain access to essential health services and reduce exposure to COVID-19. Such a situation has led Member States to call for accelerated action to strengthen nursing in the Region (15). The call for action highlights the need for investment in health workforce that impacts not only Social Development Goal 3 (SDG3) but also the other SDGs on eradicating poverty, inclusive and equitable education, gender equality through the employment and empowerment of women, and promoting decent work and sustainable and inclusive economic growth. Investments can target scaling up nursing education in terms of quantity, quality and relevance to roles and scopes of practice to maximize the utilization of their capacities, including diabetes care. Investment is also critical in creating jobs that will enable nurses to work across the full scope of practice in primary care, inpatient care settings and leadership roles where they can be available and actively involved in diabetes care (7). Investing in health workforce should be part of ‘Step 0’ in health system recovery (16). Overall, governments, health-care providers, civil society and individuals have a shared responsibility in raising public awareness about the threat of diabetes, its prevention and management, as well as strengthening nursing and ensuring access to acceptable standards of health care for all people living with diabetes. References 1. World Health Organization. World Health Statistics 2020: Monitoring health for the SDGs. Geneva: World Health Organization; 2020 (https://www.who.int/gho/publications/world_health_statistics/2020/en/). 2. World Health Organization. Global report on diabetes. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bit- stream/handle/10665/204871/9789241565257_eng.pdf?sequence=1). 3. Wu Y, Ding Y, Tanaka Y, Zhang W. Risk factors contributing to type 2 diabetes and recent advances in the treatment and preven- tion. Int J Med Sci. 2014;11(11):1185–1200. doi: 10.7150/ijms.10001 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4166864/). 4. Bhupathiraju SN, Hu FB. Epidemiology of obesity and diabetes and their cardiovascular complications. Circ Res. 2016 May 27;118(11):1723–1735. doi: 10.1161/CIRCRESAHA.115.306825. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4887150/). 5. Smith-Spangler CM, Bhattacharya J, Goldhaber-Fiebert JD. Diabetes, its treatment, and catastrophic medical spending in 35 developing countries. Diabetes Care. 2012 Feb;35(2):319-26. doi: 10.2337/dc11-1770. Epub 2012 Jan 11. PMID: 22238276; PMCID: PMC3263916. 6. Beran D. The impact of health systems on diabetes care in low and lower middle income countries. Current Diabetes Reports 2015;15(4):591 doi: 10.1007/s11892-015-0591-8. (https://www.researchgate.net/publication/272835492_The_Impact_of_Health_Sys- tems_on_Diabetes_Care_in_Low_and_Lower_Middle_Income_Countries/citation/download). 7. World Health Organization Regional Office for the Eastern Mediterranean. Rapid assessment of service delivery for NCDs during COVID-19. Cairo: WHO/EMRO; 2020 (http://www.emro.who.int/noncommunicable-diseases/publications/rapid-assess- ment-of-service-delivery-for-ncds-during-covid-19.html?ver=2). 8. World Health Organization. Pulse survey on continuity of essential health services during the COVID-19 pandemic. Geneva: World Health Organization; 2020 (https://www.who.int/publications/i/item/WHO-2019-nCoV-EHS_continuity-survey-2020.1). 9. World Health Organization. State of the world’s nursing 2020: investing in education, jobs and leadership. Geneva: World Health Organization; 2020. 10. FG Gedik, Buchan J, Mirza Z, Rashidian A, Siddiqi S, Dussault G. The need for research evidence to meet health workforce chal- lenges in the Eastern Mediterranean Region. East Mediterr Health J. 2018.24.9.811–812. https://doi.org/10.26719.2018.24.9.811 11. World Health Organization Regional Office for the Eastern Mediterranean (WHO/EMRO). Strengthening the nursing work- force to advance universal health coverage in the Eastern Mediterranean Region, EM/RC66/R.3. 2019. Cairo: WHO/EMRO; 2019 (https://applications.emro.who.int/docs/RC66-R3-eng.pdf?ua=1). 12. Affara FA, Tuipulotu AA, Al Darazi FA, Aiken LH, Betker C, Buchan J, et al. Nurses: a voice to lead nursing the world to health. Ge- neva: International Council of Nurses; 2020 (https://2020.icnvoicetolead.com/wp-content/uploads/2020/03/IND_Toolkit_120320. pdf). 13. Peimano M, Tabatabaei O, Paajouhi M. Nurses’ role in diabetes care: a review. Iran J Diabetes Lipid Dis. 2010;9(4):1-9 (https:// www.researchgate.net/publication/236985435_Nurses’_Role_in_Diabetes_Care_A_review). 14. World Health Organization Regional Office for the Eastern Mediterranean (WHO/EMRO). The nurse and diabetes: Report by the International Council of Nurses for World Diabetes Day 2020. Cairo: WHO/EMRO; 2020 (https://applications.emro.who.int/docs/ RC_Technial_Papers_2019_4_en.pdf?ua=1). 15. World Health Organization Regional Office for the Eastern Mediterranean (WHO/EMRO). Strengthening the nursing workforce to advance universal health coverage in the Eastern Mediterranean Region, EM/RC66/R.3. Cairo: WHO/EMRO; 2019 (https://ap- plications.emro.who.int/docs/RC66-R3-eng.pdf?ua=1, accessed on 22 October 2020). 16. World Health Organization. UHC 2030. Living with COVID-19: Time to get our act together on health emergencies and UHC. Discussion paper, published 27 May 2020. (https://extranet.who.int/sph/sites/default/files/document-library/document/ UHC2030%20Discussion%20paper%20on%20health%20emergencies%20and%20UHC%20-%20May%202020.pdf, accessed on 27 October 2020). 1320 EMHJ – Vol. 26 No. 11 – 2020Commentary Tobacco industry commissioned reports on illicit tobacco trade in the Eastern Mediterranean Region: how accurate are they? Erin Sandberg,1 Allen WA Gallagher2 and Raouf Alebshehy2 1Vital Strategies, New York, United States of America. 2University of Bath, Bath, United Kingdom. (Correspondence to: Raouf Alebshehy: r.alebshehy@ exposetobacco.org) Citation: Sandberg E; Gallagher AWA; Alebshehy R. Tobacco industry commissioned reports on illicit tobacco trade in the Eastern Mediterranean Region: how accurate are they? East Mediterr Health J. 2020;26(11):1320–1322. https://doi.org/10.26719/emhj.20.131 Received: 28/07/20; accepted: 07/10/20 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo) Tobacco industry-commissioned reports on the illic- it tobacco trade are one of the few data sources on the practice across Europe and Asia. The tobacco industry is now funding estimates of illicit trade in a number of countries in the World Health Organization (WHO) East- ern Mediterranean Region, specifically Egypt, Jordan and Lebanon. These estimates come from a recent report by Oxford Economics, which was funded by major trans- national tobacco companies. Industry-funded studies of the illicit tobacco trade have been found to consistently fail to meet the standards of quality and transparency ex- pected of peer-reviewed research. Moreover, the scale of the problem would appear exaggerated in order to aid the industry’s efforts to oppose tobacco controls by arguing that such measures would actually increase illicit trade. A critical look at this new report suggests that this trend continues, while concerns remain over the reliability of the data and estimates claimed, stressing the need for in- dependent research of the illicit tobacco trade in the East- ern Mediterranean Region. While countries in the Eastern Mediterranean Region are seeing progress in the implementation of WHO MPOWER measures (1), WHO reports on smoking prevalence in the Region indicate it will not achieve its 30% relative prevalence reduction target (12.6%) by the year 2025 (2). Decreasing the affordability of cigarettes is recognized as the most effective means to reduce prevalence and help the Region achieve its target, but a major obstacle to this is the illicit tobacco trade, which increases accessibility and affordability of tobacco products. The practice has been a problem for a number of countries in the Region for decades, driven in large part by transnational tobacco companies having smuggled their own product into countries, including Jordan and Lebanon, from as early as the mid-1970s (3,4). However, the scale of the Region’s current illicit tobacco trade is difficult to measure due to its illegality as well as data collection and analysis complexities. Transparent public data on the topic of illicit tobacco trade is limited, and in many countries it is non-existent (5). It is within this context that the tobacco industry has become a major funder of data on illicit trade, often by commissioning reports that provide estimates of illicit trade in a geographical area. This practice is now occurring in the Eastern Mediterranean Region with a recent Oxford Economics report (6) providing estimates of illicit trade in Egypt, Jordan and Lebanon. Tobacco companies have an incentive to misrepresent the size of the illicit tobacco trade, resulting in industry- funded research having been widely criticized for its unreliability and exaggeration of the scale of the illicit market (7). Tobacco companies regularly cite industry-funded reports about the illicit trade [yet fail to acknowledge the funding link, as seen in a recent Philip Morris International (PMI) interview (8)] as part of their efforts to oppose public health policy (9). As such, we have scrutinised the new Oxford Economics report to establish the reliability of its estimates of illicit trade. In March 2020, global forecasting company Oxford Economics (10) released a report titled “Levant Illicit Tobacco 2019” (6), commissioned by British American Tobacco (BAT), Japan Tobacco International (JTI) and Philip Morris SA (a subsidiary of PMI). It examines the illicit cigarette market in Egypt, Jordan and Lebanon. The report, which is a business document and not peer- reviewed academic research, is concerning for several reasons. As per the report’s disclaimer, it was prepared “in accordance with specific terms of reference” agreed by Oxford Economics and the tobacco companies. These terms of reference are not disclosed and may have influenced how the report portrays the tobacco industry and its involvement in the illicit market. Also of concern is Oxford Economics’ existing tobacco industry ties, including a working relationship with PMI dating back to 2017 when PMI announced that Oxford Economics would receive funding from its PMI IMPACT initiative (11,12). Past Oxford Economics reports on illicit trade have been criticised by academics and nongovernmental organizations (NGOs) for their reliance on the industry for their data, for the methods of analysis used, and for the presentation of the reports appearing to mislead readers (13,14). Most recently, Oxford Economics’ “Asia Illicit Tobacco Indicator 2017” report was critiqued in a report by the Southeast Asia Tobacco Control Alliance (15). Many of the concerns raised over previous Oxford Economics reports hold true in the new report on the 1321 EMHJ – Vol. 26 No. 11 – 2020Commentary Levant region. The choice of featured markets in the report is questionable and no justification is given in the report for the selection of countries mentioned. Despite the report’s title referring to the Levant region, only three countries within this geographical area are featured in the analysis and no justification is provided for why other countries were left out. Since Egypt, Jordan and Lebanon have all experienced tax increases in recent years, these countries may have been chosen to help illustrate the industry narrative that increased taxes lead to increased illicit tobacco trade. As with previous industry-commissioned reports on illicit tobacco trade (7), the primary data input for this report was highly susceptible to industry interference. To estimate levels of illicit tobacco trade, the report relies on empty pack surveys, where discarded cigarette packs are collected and then tested by tobacco companies to identify if the product is from their own supply chain. Allowing tobacco companies to determine this opens the data up to manipulation, since tobacco companies have a vested interest in under-reporting their own product on the illicit market. The report fails to disclose the known limitations of such surveys and does not provide sufficient detail for surveys to be replicated by independent researchers to validate the findings. The report does acknowledge one flaw, albeit solely via a footnote on page 7. The authors indicate that only exports from the three featured countries to those same three countries are included in the analysis. This is problematic, as leaving out products that were legally exported from the three countries to other countries not featured in the report ultimately lowers the estimated total legal consumption of the three featured countries. This, in turn, makes the percentage of illicit cigarettes in the three countries larger than if all legal exports from those countries were captured, thus skewing the final figures. Concerns over the accuracy of the report’s estimations extend to the report’s recommendations, as well. The policy recommendations come from the Transnational Alliance to Combat Illicit Trade (TRACIT), an NGO with extensive tobacco industry ties. Among other links to the industry, TRACIT has previously listed BAT, JTI and PMI as members on its website, with its current website still citing PMI as a member (16); however, none of TRACIT’s partnerships with tobacco companies were mentioned in the report. One of the report’s recommendations is to “rationalize tax policy”, which supports the tobacco industry’s efforts to create a link between increased cigarette taxes and purported growth in illicit tobacco trade. Evidence indicates16 that this depiction of the relationship between tax and illicit trade is over-simplified as countries with low cigarette taxes and prices often have larger illicit cigarette markets than countries with higher taxes and prices (17). Therefore, the report, data and estimates would appear unreliable and highlights the urgent need for independent research on the illicit tobacco trade. With no comparable, independent alternatives, Oxford Economics reports are one of the only major sources of data on the illicit trade across Asia, and now the Eastern Mediterranean Region. More independent data are needed to provide accurate insight into the illicit tobacco trade and to verify findings in industry-funded reports. Funding: All authors acknowledge the support of Bloomberg Philanthropies Stopping Tobacco Organiza- tions and Products project funding (www.bloomberg. org). Competing interests: None declared. References 1. Heydair G, Zaatari G, Al-Lawati J, El-Awa F, Fouad H. MPOWER, needs and challenges: trends in the implementation of the WHO FCTC in the Eastern Mediterranean Region. East Mediterr Health J. 2018;24(1):63–71. https://doi.org/10.26719/2018.24.1.63 2. El-Awa F, Bettcher D, Al-Lawati JA, Alebshehy R, Gouda H, Fraser CP. The status of tobacco control in the Eastern Mediterrane- an Region: progress in the implementation of the MPOWER measures. East Mediterr Health J. 2020;26(1):102–109. https://doi. org/10.26719/2020.26.1.102 3. World Health Organization Regional Office of the Eastern Mediterranean (WHO/EMRO). The cigarette “transit” road to the Islamic Republic of Iran and Iraq: illicit tobacco trade in the Middle East. Cairo: WHO/EMRO; 2008 (https://applications.emro. who.int/dsaf/dsa908.pdf, accessed 24 July 2020). 4. Tobacco Control Research Group. Eastern Mediterranean Region. In. TobaccoTactics.org: University of Bath (https://tobaccotac- tics.org/wiki/eastern-mediterranean-region/) (last edited 30 April 2020). 5. Prasad V, Schwerdtfeger U, El-Awa F, Bettcher D, da Costa e Silva V. Closing the door on illicit tobacco trade, opens the way to better tobacco control. East Mediterr Health J. 2015;Sep 8;21(6):379-80. https://doi.org/10.26719/2015.21.6.379. 6. Oxford Economics. Levant Illicit Tobacco 2019. (https://www.oxfordeconomics.com/recentreleases/levant-illicit-tobacco-2019. Published 2020, accessed 24 July 2020). 7. Gallagher AWA, Evans-Reeves KA, Hatchard JL, Gilmore AB. Tobacco industry data on illicit tobacco trade: a systematic review of existing assessments. Tob Control. 2018: tobaccocontrol-2018-054295. 8. Philip Morris International. Philip Morris International’s Taylan Suer discusses IQOS and local market response. Executive, 8 May 2020 (https://www.executive-magazine.com/brand-voice/philipmorris-internationals-taylan-suer-discusses-iqos-and-loq- cal-market-response, accessed 24 July 2020). 1322 EMHJ – Vol. 26 No. 11 – 2020Commentary 9. Ulucanlar S, Fooks GJ, Gilmore AB. The Policy Dystopia Model: an interpretive analysis of tobacco industry political activity. PLoS Med. 2016;13(9):e1002125. 10. Tobacco Control Research Group. Oxford Economics. In. TobaccoTactics.org: University of Bath. (https://tobaccotactics.org/wiki/ oxford-economics). (Last edited 9 August 2017). 11. Tobacco Control Research Group. PMI IMPACT. In. TobaccoTactics.org: University of Bath. (https://tobaccotactics.org/wiki/ pmi-impact). (Last edited 26 October 2017). 12. Tobacco Control Research Group. List of successful PMI IMPACT Applicants. In. TobaccoTactics.org: University of Bath. (https:// tobaccotactics.org/wiki/list-of-successful-pmi-impact-applicants). (Last edited 21 March 2019). 13. South East Asia Tobacco Control Alliance (SEATCA). Failed: a critique of the ITIC/OE Asia-14 Illicit Tobacco Indicator 2013. Bang- kok: SEATCA; 2015 (https://seatca.org/dmdocuments/Asia%2014%20Critique_Final_20May2015.pdf; accessed 7 November 2017). 14. South East Asia Tobacco Control Alliance (SEATCA). Illicit tobacco indicator 2012: more myth than fact. A critique by SEATCA. Bangkok: SEATCA; 2013 (http://seatca.org/dmdocuments/ITIC%20report_More%20Myth%20than%20Fact_2%20July %202014.pdf, accessed 1 October 2014). 15. South East Asian Tobacco Control Alliance (SEATCA). Still defective: Asia illicit tobacco indicator 2017 report. Bangkok: 2020 (https://stoptobacco.sharepoint.com/sites/STOPBriefingonTrackingTracing/Shared%20Docu ments/General/Still%20Defec- tive-Asia%20Illicit%20Tobacco%20Indicator_June%202020.pdf, accessed 24 July 2020). 16. Tobacco Control Research Group. Transnational Alliance to Combat Illicit Trade (TRACIT). University of Bath. (https://tobacco- tactics.org/wiki/tracit). (Last edited 11 June 2020). 17. World Bank Group. Confronting illicit tobacco trade: a global review of country experiences. Washington DC: World Bank Group; 2019 (http://documents.worldbank.org/curated/en/677451548260528135/pdf/133959-REPLPUBLIC-6-2-2019-19-59-24-WB9- GTobaccoIllicitTradeFINALvweb.pdf, accessed 28 May 2019). 1323 EMHJ – Vol. 26 No. 11 – 2020Research article Health information systems in Jordan and Palestine: the need for health informatics training Hussein Jabareen,1 Yousef Khader2 and Adel Taweel3 1College of Nursing, Hebron University, Hebron, Palestine (Correspondence to: Hussein Jabareen: husseinj@hebron.edu). 2Department of Public Health, Jordan University of Science and Technology, Irbid, Jordan. 3Department of Computer Science, Birzeit University, Ramallah, Palestine. Abstract Background: Some Arab countries have health information systems (HIS) in place but they lack well trained IT staff. Poor management and lack of appreciation of the importance of HIS are major barriers to development and adoption of HIS in Arab hospitals. Aims: This research is part of a survey carried out to determine health informatics (HI) use and to assess the training needs of health professionals in Jordan and Palestine. Methods: A survey was conducted in 2017 among employees in all health professions at 14 hospitals in Jordan and Pales- tine to assess their use of the HI system and to assess the HI skills needed in both countries. Results: The majority of respondents reported that their hospital departments were employing computer systems to run services. More than half had received training in computer skills but also half said they needed specialized training in HI. Between 58.0% and 73.6% agreed that their hospitals provided the necessary support to operate HI systems. The vast majority (86.0%) of health professionals reported that they needed skills to monitor diagnosis and treatment, including ac- cess to clinical findings. Other skills needed included using shared hospital services (85.6%), using medical records (84.7%), managing electronic patient data (84.5%), using patient medical records to conduct clinical research (83.4%), and using tele-care services and technologies effectively (75.9%). Conclusions: Health professionals in Palestine and Jordan are in need of training in HI and therefore educational pro- grammes in the area of HI are strongly recommended. Keywords: health information systems, health informatics, health professionals, Jordan, Palestine Citation: Jabareen H; Khader Y; Taweel A. Health information systems in Jordan and Palestine: the need for health informatics training. East Mediterr Health J. 2020;26(11):1323–1330. https://doi.org/10.26719/emhj.20.036 Received: 22/05/19; accepted: 16/12/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo) Introduction Using computer and health informatics is critically important in improving the quality of the health care system, patient care and health outcomes (1). Health in- formation systems (HIS) or health informatics (HI) is a multidisciplinary domain that utilizes health informa- tion technology (HIT) to enhance health care services via any combination of higher efficiency, higher quality and new opportunities. The HIS is information design and build applied to the field of health care, essentially “the management and utilization of patient health care infor- mation” (2). Electronic health systems are being used by health professionals to implement daily care manage- ment, avoid complications, prevent medical and medica- tion errors, and carry out clinical research (3,4). Electronic health systems make the management of patient records easier and safer while HI supports health professionals, consumers, patients and other stakeholders in their deci- sion-making to achieve desired outcomes. This support is accomplished through the use of information structures, information processes and information technology (5,6). Most of the Arab countries are lagging behind in using an HIS due to lack of financial resources and staff competencies (7). There is a huge gap between the Arab and Western countries in the information systems available (and used) (8–10). However, it seems that both groups of countries are facing similar challenges in the enforcement of HISs (11). For instance, in an international comparison of factors inhibiting physicians’ use of HIS in the United States, Chile and Germany, Gewald et al. found that each country developed its own HIS but took a unique approach to fostering the implementation of IT in their health care settings and applied it differently at each management level (12). They reported that leadership and management structures had a significant role in the Western health care services in fostering the use of IT allied with the user, system, and process obstacles. Similarly, each Arab country had a unique culture and approach to health services, so HIS had to be specific and appropriate and in keeping with specific health policies (13). In many health care institutions, data are collected manually on paper rather than by digital data entry, but still the collected data are not complete, which results in a burden for patient monitoring as well as data analysis. In addition, regulations regarding the ethical use of clinical data are not adequate in many Arab countries (14). It is essential that health policy experts, researchers and other professionals create systems and policies that are comprehensive and improve health care services (15). 1324 EMHJ – Vol. 26 No. 11 – 2020Research article Some Arab countries have an HIS in place but they lack well trained IT staff, data collectors, data entry personnel and data analysts (7). Poor management and bureaucracy as well as a lack of appreciation of the importance of HIS are major barriers to the development and adoption in Arab hospitals (16). Cost was an obstacle to HIS development in many Arab countries, but it was less of a concern in Saudi Arabia compared with others (16). The shortage, however, of competent IT staff was an obstacle in Syria and Saudi Arabia. Some efforts are under way to address these shortcomings, but more financial resources need to be allocated to further develop IT human resources and an HIS that fit local needs. In many Arab countries, poor funding is the main obstacle to implementing the HIS, while in countries of the Gulf Cooperation Council, economic funds are available but the main barriers were the lack of qualified personnel, the lack of engagement of clinical staff in utilizing HIS, and ethical and privacy concerns (17,18). According to Alsadan et al., Saudi Arabia and the United Arab Emirates were the most advanced Arab countries in utilizing an HIS (7), the authors recommended that cooperation and exchange of experiences between different Arab countries is needed to overcome some of the common barriers in implementing HIS. Hayajneh and Zaghloul recommended that medical and health curricula should be revised to include and integrate HIS and that new programmes in HI should be established (16). Similar findings were reported by Shaban et al. in 2010 in their study on trauma registry in the United Arab Emirates, where the lack of funding and resources, the lack of qualified personnel in HI and the need for user friendly software for data entry were the main barriers (19). In developed countries such as the United States, despite its advancement in HI, barriers to using electronic health records as revealed by studies conducted over 10 years earlier were mainly costs and technical and communication difficulties (20). Academic institutions in the Arab world successfully provide national and international health institutions with qualified health care professionals, including medical doctors, pharmacists and nurses, while HI remains a newer field. With the recent realization and emphasis on the need for HI, some universities in the Arab countries have started providing a few courses, tracks or degrees in HI. In Egypt, Jordan and Palestine a few universities provide HI training and courses for undergraduate students in the health and IT fields. In Saudi Arabia and the United Arab Emirates, some universities offer advanced degrees in HI. The current situation in the Arab world requires further development of academic programmes that can meet the needs of health care providers (15,21). In Palestine and Jordan, HIS has recently been implemented in some areas but with various challenges and obstacles, and its application is still limited to electronic patient records (22). HiCure is an Erasmus + funded project. The project has a well-placed international consortium that combines strong academic and industrial expertise in both computing and health to develop innovative, integrated HI curricula. The HiCure project develops HI as integrated pathways within the undergraduate degrees of both the information technology and health-oriented programmes in Palestine and Jordan. The development of HI skills integrated within the existing degrees will ensure graduate students with competent skills founded on a solid grounding within their educational background. In addition to meeting their intended educational learning outcomes, HI skills will also become an essential part of students’ profession employed as mechanisms that will improve the quality of health care by graduates qualified in both health and computing, and advance the health domain towards information-driven, efficient and effective evidence-based practice (23,24). Implementation of HIS is complex and relies on organizational, structural, technological and human factors to be complementary and successful (25). Assessing the impact of a complex HIS from the perspective of users is considered one of the most efficient evaluation methods in comparison with other methods (26). Although the use of HIS has pervaded health care settings in the Arab world and worldwide, methodologies to evaluate its impact in these settings have not developed with the same momentum (4). So, our study aimed to assess the use of HIS by health professionals and assess the needed HI skills in Palestine and Jordan. Methods A descriptive cross-sectional design was conducted among employees of all health professions at 14 govern- mental, nongovernmental, and private hospitals in Jor- dan and Palestine. A sample of health professionals was selected from those who were working in these hospitals during the period March–June 2017. Within each hospi- tal, the study questionnaire was randomly administered to 20% of the total number of employees. This study is part of a survey that was carried out to determine HI usage and to assess the training needs of health professionals in Jordan and Palestine. Two methods of sampling were adopted to satisfy the declared objectives of the study. The first method was a purposeful (convenience) sampling technique which was used to access targeted hospitals. Letters, questionnaire, and information sheets about the study were sent to the ministries of health responsible for governmental hospitals and to the administrations of nongovernmental hospitals asking their permission to allow the research team to distribute the questionnaire and collect data from health professions employees. All 14 hospitals we approached agreed to participate in this study. The second sampling procedure was within each hospital, where the research team targeted the available employees who were on duty during data collection shift. The data collection process was facilitated by permissions granted from the Jordanian and Palestinian ministries of health and the administrations of the nongovernmental hospitals. 1325 EMHJ – Vol. 26 No. 11 – 2020Research article This facilitated entering these hospitals during different working shifts to reach as many diverse employees as possible and include opinions of all health professionals. Participants were recruited from 3 working shifts (day, evening, night) from each hospital during the data collection phase. Data collection started with the available health employees in the different departments in the day shift. After that, the research team revisited the hospitals during the evening and night shifts. The questionnaire sought information about demographic characteristics of participants, use of computerized health systems, organizational support for HI, training and usage of HI, and the needed HI skills. The perception of employees regarding the importance and adequacy of HI for health care practice at their institutions was explored by asking 10 questions using a 5-point Likert-type scale. The questionnaire was self- completed, filled out within the hospital setting during the break time in a place that was convenient to each participant. Completing the questionnaire took on average 10 minutes. Ethical approval was obtained from the Institutional Review Board at Jordan University of Science and Technology. Data were analysed using SPSS, version 20. Data were described using means, medians, and percentages as appropriate. Results Participants A total of 579 health employees from 14 hospitals re- sponded to the questionnaire, a response rate of 83%. About 34% of the respondents were from Jordan and 66% from Palestine. The mean age of the respondents was 32 (range 20–65) years. About 18% had a Diploma, 71% a Bachelor’s degree and 7% a post graduate degree (High Diploma, Masters, PhD, or specialization in medicine). About 75% of the respondents were employed in govern- mental hospitals, 14% in nongovernmental hospitals, and 11% in private hospitals. The survey covered all health units in hospitals. About 74% of respondents were work- ing in the major medical, surgical, allied health and high dependency departments (Table 1). However, 5% stated they were working in other departments not listed in the questionnaire. Years of experience The median years of experience in the profession was 5 (range 1–35) years. While the median experience in using computer systems was 4 (range 0–30) years. The medi- an duration of using HIS was 2 (range 0–23) years. The majority of professionals in Palestinian and Jordanian hospitals started recently to use health computerized systems, and more than one-fifth did not have any expe- rience with HI (Table 2). Computer use in institutions At the institutional level, more than 67% of respondents reported that their hospital departments were employ- ing computer systems to run their services, especially in financial management (77%), medical laboratory (74%), pharmacy (71%), electronic records (67%), and top man- agement (66%) (Table 3).The use of computer systems was reported to a lesser extent for electronic prescrip- tions (59%) and managing medical devices (50%). Use of computer by physicians, allied health professionals, and nurses When the analysis was limited to physicians, allied health professionals and nurses, 84.9% in Palestine and 68.8% in Jordan reported that they used computers to carry out their duties (Table 4). Nurses and allied health professionals in Palestine were more likely to report the use of computers to accomplish their tasks. About 75.3% of health professionals in Palestine and 62.2% in Jordan reported the use of electronic medical records. Nurses and allied health professionals in Palestine were more likely to report the use of electronic medical records than those in Jordan. Training on computer use for health professionals More than half of health professionals received training on computer skills (Table 4). Almost the same proportion reported that they needed specialized training in HI (Ta- ble 5). The majority of health professionals perceived that it is important to have HIS in their institution and about 52% of Jordanian and 60% of Palestinian health profes- sionals reported that their institutions had adequate HIS (Table 5). Organizational support for health informatics Most employees were satisfied with the support provid- ed by their organizations for the use of HI programmes (Table 6). For instance, about three-quarters (73.6%) re- ported that their institutions provided a supportive envi- ronment for HI programmes; two-thirds (66.9%) reported that their institutions had the necessary infrastructure Table 1 Distribution of Jordanian and Palestinian health professionals (n = 579) according to department/unit, 2017 Department/unit % Medical department 25.7 Allied health department 19.2 Surgical department 16.2 High dependency unit 12.8 Paediatrics department 4.5 Outpatient clinics 4.3 Administration 3.8 Emergency room 3.3 Maternity department 3.1 Operation room 2.1 Other 5.0 1326 EMHJ – Vol. 26 No. 11 – 2020Research article that enables clinicians while using HI and their institu- tions were committed to promoting and improving the implementation of HIT (68.3%). The needed health informatics skills The majority of health professionals reported that they were in need of HI skills (Table 7); 86.0% reported that they needed skills to monitor patients’ diagnosis and treatment, including access to clinical findings. Other skills needed included using shared hospital services (85.6%), using medical records (84.7%) and managing electronic patient data (84.5). Discussion The findings of this survey have several significant theoretical and practical implications for academia, the health care sector and government. Having experience in computer systems makes it easy to learn HI systems. In particular, our results show that the use of computer systems is not limited to a specific department. The ma- jority of the participants were using these systems in the medical laboratory and the pharmacy to run their daily transactions and operations. Electronic records and pre- scriptions were also used by almost two-thirds of the par- ticipants. Although one-third were using telemedicine, they demonstrated a strong willingness to expand the use to facilitate their daily tasks. In comparing the 2 countries, we found that nurses and allied health professionals in Palestine were more likely to report the use of the computer to accomplish their tasks. Additionally, nurses and allied health professionals in Palestine were more likely to report the use of electronic medical records than those in Jordan. Nevertheless, most Arab countries do not use HIT very often because of financial obstacles (7). In Jordan for example, the field of HIT started only recently, thus not many allied health professionals use it. However, it has been suggested that new health information technologies in the 21st century can transform the Arab world as societies need strong and efficient health policies to improve patient care (9). One possible reason for not widely adopting HIT in Jordan is the negative perception of its importance (16). Moreover, research in the region has shown that poor funding, lack of qualified personnel or their engagement in utilizing HIT are the main obstacle to implementing it (17). There are some limitations to this study. It was carried out as part of a large survey to determine the extent of usage of HIS, to assess the training needs, and to explore the attitudes of health professionals in Jordan and Palestine towards HI. So, we neither tested any hypothesis nor examined associations between variables. The Table 2 Distribution of Jordanian and Palestinian health professionals (n = 579) according to years of experience, 2017 Years of experience In health profession Using computer systems Using health information systems No. % No. % No. % 0 0 0 67 12 120 22 1–5 279 50 266 48 376 68 6–10 123 22 108 19 48 8 11–15 63 12 86 16 7 1 16–35 89 16 27 5 3 1 Total 554 100 554 100 554 100 Table 3 Use of computer systems according to department/ unit as reported by Jordanian and Palestinian health professionals (n = 579), 2017 Professional department/unit Computer use % Financial management 77 Medical laboratory 74 Pharmacy 71 Electronic records 67 Top management 66 Electronic prescription 59 Managing medical devices 50 Logistics 44 Telemedicine in education 32 Table 4 Computer use and training by Jordanian and Palestinian health professionals (n = 579), 2017 Profession Computer use Training in computer skills Jordan Palestine Jordan Palestine Physician 79.5 80.7 67.5 47.4 Nurse 56.6 84.7 50.0 63.7 Allied health professional 70.0 88.8 45.2 56.3 Overall 68.8 84.9 56.8 59.4 1327 EMHJ – Vol. 26 No. 11 – 2020Research article sample size included 20% of health employees working in the targeted 14 hospitals but these could not be described as representative of all health employees in Jordan and Palestine. The survey was carried out in a stressful period after the introduction of HIS in many of these hospitals, and thus participants might not have been typical of all those in the original populations. Comparisons and similarities with other studies noted must therefore be treated with caution. Many studies, especially in the Arab world, have reported that workloads were increased after the introduction of HIS, and clinicians mainly complained about the increased paperwork and IT workload (8,16,17). This raises the issue of the nonrespondents as they may perhaps have been the busiest professionals, those who could not find the time to fill out the questionnaire. In order to make it comprehensive, the questionnaire covered a wide range of topics concerning the training needs and impact of HI usage at individual, institutional and national levels: the cost of this broad scope was the inability to focus on specific clinical or professional areas such as electronic prescribing and electronic triaging. Furthermore, the questionnaire contained many questions about previous experience with using computers and HIS and training of health professionals and this could have produced some recall bias. Furthermore, we did not investigate the duration of using computers by health professionals nor the onset or frequency of using electronic medical records. This is of paramount importance because sustainability is a critical factor in achieving improvements in patient health outcomes (27). Another limitation of the analysis of this survey was the small numbers within the subgroups. About 74% of our respondents were working in 4 major medical, surgical, allied health and high dependency departments; only 21% were working in the remaining 6 departments, and 5% were working in various other departments. Nonetheless, the survey does throw up some potentially important issues concerning these subgroups. Our results showed that more than half the health care professionals had some training in computers skills, however, large proportions among different professions in both cohorts reported needing more specialized training in HI, suggesting a perceived lack of confidence in their current knowledge. Similarly, in an Australian study, almost half the nurses indicated a need for more computer training to better meet their job requirements (28). In the same study, nurses, as the largest users of computer technology, considered that employers often did not encourage information and computer technology training. Consistent with this, Trivedi and Joshi reported that more than three-quarters of health care professionals in a rural medical college in India had no formal computer training (29). Similar to the findings of a study in hospitals in Tehran (30), we found considerable variation in computer training experiences across the 3 groups of health professions in both settings. Physicians received more training than other health care professionals in the Jordan cohort, while nurses received more training in the Palestine cohort. Nurses in both countries perceived a higher need for more specialized training in HI and Table 5 Distribution of Jordanian and Palestinian health professionals (n = 579) perceptions of the health informatics situation in their institutions, 2017 Profession There is a need for specialized training Having a health information system is important Our health information system is adequate Jordan Palestine Jordan Palestine Jordan Palestine Physicians 49.4 50.9 81.7 75.4 59.8 52.6 Nurses 71.6 64.2 86.5 88.4 46.7 60.0 Allied health professionals 61.3 48.8 87.1 81.3 45.2 66.3 Overall 60.1 58.5 84.5 84.7 52.1 60.2 Table 6 Perception of institutional support for health informatics among Jordanian and Palestinian health professionals (n = 579), 2017 Institutional support for health informatics No. % The institution provides a supportive environment for health informatics programmes 425 73.6 The institution has the necessary infrastructure that enables clinicians while using health informatics 387 66.9 The institution is committed to promoting and improving the implementation of health information technology 390 68.3 My organization provides a supportive environment and/or provides the necessary training and resources for the proper use of electronic health systems 334 58.0 My organization has the necessary infrastructure, including staff and financial resources to support electronic health systems. 351 61.0 My organization promotes the use of health computing technology and/or provides sufficient encouragement for the use of electronic health systems 356 61.9 1328 EMHJ – Vol. 26 No. 11 – 2020Research article computer skills than other health care professionals. Other researchers have reported similar interest among nurses towards enhancing their careers (13). Finally, and although the majority of health care professionals in both setting believed that it was important to have HI system in their institutions, obstacles to implementing HIT still exist. To overcome the challenges related to the use of HIT by health professionals in Jordan and Palestine, medical and health curricula need to be revised to include and integrate HIT, and new programmes in HI need to be developed (16). Indeed, some universities in Jordan and Palestine have recently started to provide HI training and courses for undergraduate students in the health and IT fields. In conclusion, health professionals in Palestine and Jordan are in need for training in HI and therefore educational programmes in the area of HI are strongly recommended. Table 7 Key health informatics skills needed reported by health professionals in Jordan and Palestine (n = 579), 2017 Skill % Monitoring patient diagnosis and treatment 86.0 Using shared hospital services 85.6 Using patient medical records 84.7 Managing electronic patient data 84.5 Using patient medical records for clinical research 83.4 Telemedicine and tele-care services 75.9 Using coding standards 73.8 Acknowledgement This paper reflects the views only of the authors; the European Commission cannot be held responsible for any use which may be made of the information contained therein. The authors would like to thank the HiCure members who contribut- ed to this study. Funding: This study, part of the HiCure project, has been funded with support from the European Commission. Competing interests: None declared. Systèmes d'information sanitaire en Jordanie et en Palestine : nécessité d'une formation en informatique sanitaire Résumé Contexte : Certains pays arabes disposent de systèmes d’information sanitaire, mais ils manquent de personnel informatique bien formé. Une mauvaise gestion et un manque d'appréciation de l'importance des systèmes d'information sanitaire constituent des obstacles majeurs au développement et à l'adoption de ces systèmes dans les hôpitaux du monde arabe. Objectifs : La présente recherche fait partie d'une enquête menée pour déterminer l'utilisation de l'informatique sanitaire et pour évaluer les besoins de formation des professionnels de santé en Jordanie et en Palestine. Méthodes : En 2017, une enquête a été menée auprès des employés de toutes les professions de santé de 14 hôpitaux en Jordanie et en Palestine afin d’évaluer leur utilisation du système d’information sanitaire et de déterminer les compétences en informatique sanitaire nécessaires dans les deux pays. Résultats : La majorité des répondants ont indiqué que leurs services hospitaliers utilisaient des systèmes informatiques pour gérer leurs services. Plus de la moitié a reçu une formation en informatique, mais la moitié a également déclaré avoir besoin d'une formation spécialisée en informatique sanitaire. Entre 58,0 % et 73,6 % ont convenu que leurs hôpitaux fournissaient le soutien nécessaire au fonctionnement des systèmes d'information sanitaire. La grande majorité (86,0 %) des professionnels de santé ont déclaré avoir besoin de compétences pour surveiller le diagnostic et le traitement, y compris l'accès aux résultats cliniques. Les autres compétences nécessaires comprenaient le recours aux services hospitaliers partagés (85,6 %), l'utilisation des dossiers médicaux (84,7 %), la gestion électronique des données des patients (84,5 %), l'utilisation des dossiers médicaux des patients pour mener des recherches cliniques (83,4 %) et l'utilisation efficace des services et des technologies de télésoins (75,9 %). Conclusions : Les professionnels de santé en Palestine et en Jordanie ont besoin d'une formation en informatique sanitaire. Par conséquent, des programmes d’éducation dans ce domaine sont fortement recommandés. 1329 EMHJ – Vol. 26 No. 11 – 2020Research article References 1. Benbrahim H, Hachimi H, Amine A. Moroccan Electronic Health Record System. In: Proceedings of the International Conference on Industrial Engineering and Operations Management Paris, France, July 26–27, 2018. 2. Nadri H, Rahimi B, Timpka T, Sedghi S. The top 100 articles in the medical informatics: a bibliometric analysis. J Med Syst. 2017 Aug 19;41(10):150. doi:10.1007/s10916-017-0794-4 3. Almuayqil S, Atkins AS, Sharp B. Ranking of E-health barriers faced by Saudi Arabian citizens, healthcare professionals and IT specialists in Saudi Arabia. Health. 2016;08(10):1004–13. doi:10.4236/health.2016.810104 4. Cresswell K, Sheikh A.. Organizational issues in the implementation and adoption of health information technology innova- tions: an interpretative review. Int J Med Inform. 2013 May;82(5):e73–86. doi:10.1016/j.ijmedinf.2012.10.007 5. Badran MF. Electronic health records prospects in Egypt: a demand-side perspective. Paper presented at the 28th European Regional ITS Conference, Passau, Germany, 30 July–2 August 2017. 6. Bouhaddou O, Bennani Othmani M, Diouny S. Medical informatics in Morocco. Yearb Med Inform. 2013;8:190–6. PMID:2397457 7. Alsadan M, El Metwally A, Ali A, Jamal A, Khalifa M, Househ M. Health information technology (HIT) in Arab countries: a sys- tematic review study on HIT progress. J Health Inform Dev Countries. 2015;9(2):32–49. 8. Alrawabdeh W, Salloum A, Mingers J. Key factors influencing the diffusion of information and communication technology (ICT) in the Arab world. A comparative study. Brit J Econom, Finance Management Sci. 2012;5(2):45–59. 9. Almulhim DA, Househ MS. A perspective on the influence of health policy on health technology use within the Arab world. J Health Inform Dev Countries. 2012;6(1):375–84. 10. Berg M. Implementing information systems in health care organizations: myths and challenges. Int J Med Inform. 2001;64:143– 56. doi:10.1016/s1386-5056(01)00200-3 11. Waterson P, Hoonakker PL, Carayon P. Special issue on human factors and the implementation of health information technolo- gy (HIT): comparing approaches across nations. Int J Med Inform. 2013;82: 277–80. doi:10.1016/j.ijmedinf.2013.01.001 12. Gewald H, Núñez A, Gewald C, Vriesman LJ. An international comparison of factors inhibiting Physicians’ use of hospital infor- mation systems. In: Proceedings CONF-IRM. 2017 (http://aisel.aisnet.org/confirm2017/1, accessed 2 April 2010). 13. Eldin AS, Saad D, Samie GA. Evaluation of electronic health records adoption in Egypt. Int J Engineer Res Applications. 2013;3(1):1131–4. 14. Abd Ghani MK, Neamah AF. Electronic health records challenges and barriers in Iraq. Comp Engineer Intelligent Systems. 2017;7(6):1–7. ةيحصلا ةيتامولعلما لامج في بيردتلا لىإ ةجالحا :ينطسلفو ندرلأا في ةيحصلا تامولعلما مظُن ليوطلا لداع ،ضرخ فسوي ،نيرابج ينسح ةصلالخا .تامولعلما ايجولونكت لامج في ًاديج ًابيردت ينب َّردلما ينلماعلا لىإ رقتفت انهأ لاإ ،ةمئاق ةيحص تامولعم مُظُن ةيبرعلا نادلبلا ضعب في دجوي :ةيفللخا نادلبلاب تايفشتسلما في اهدماتعاو ةيحصلا تامولعلما مظُن ريوطت مامأ ةيسيئر قئاوع ةيحصلا تامولعلما مظُن ةيهمأ ريدقت مدعو ةرادلإا ءوس ل ِّكشيو .ةيبرعلا ندرلأا في ينيحصلا ينينهلما بيردت تاجايتحا مييقتلو ،ةيحصلا ةيتامولعلما مادختسا ديدحتل يرجُأ ٍحسم نم ًاءزج ثحبلا اذه لّثمي :فادهلأا .ينطسلفو تامولعلما مظنل مهمادختسا مييقتل ينطسلفو ندرلأا في ىفشتسم 14 في ةيحصلا نهلما عيجم في ينلماعلل 2017 ماع في حسم يرجُأ :ثحبلا قرط .نْيَدلبلا لاك في ةيحصلا ةيتامولعلما لامج في اهيلإ نوجاتيح يتلا تاراهلما مييقتو ةيحصلا ًابيردت مهفصن نم رثكأ ى َّقلت دقو .تامدلخا ليغشتل ةيبوساح ًماظُن مدختست اهيف نولمعي يتلا تايفشتسلما ماسقأ نأب ينبيجتسلما ةيبلاغ دافأ :جئاتنلا لىع %73.6و %58.0 ينب ام قفاوو .ةيحصلا ةيتامولعلما لامج في ٍصصختم ٍبيردت لىإ ةجاحب منهإ اولاق ًاضيأ مهفصن نكلو ،بوسالحا تاراهم لىع ينيحصلا ينينهلما نم )%86.0( ىمظعلا ةيبلاغلا تدافأو .ةيحصلا تامولعلما مظُن ليغشتل مزلالا معدلا تم َّدق اهيف نولمعي يتلا تايفشتسلما نأ تامدخ مادختسا ةمزلالا ىرخلأا تاراهلما تلمشو .ةيريسرلا جئاتنلا ةحاتإ كلذ في ماب ،جلاعلاو صيخشتلا دصرل تاراهم لىإ نوجاتيح منهأ تلاجسلا مادختساو ،)%84.5( ةينوتركللإا ضىرلما تانايب ةرادإو ،)%84.7( ةيبطلا تلاجسلا مادختساو ،)%85.6( ةكترشلما تايفشتسلما .)%75.9( ةيلاعفب دعُب نع ةياعرلا تامدخو تايجولونكت مادختساو ،)%83.4( ةيريسرلا ثوحبلا ءارجلإ ضىرملل ةيبطلا في ةيميلعت جمارب يرفوتب ةدشب َصوي كلذلو ،ةيحصلا ةيتامولعلما لامج في بيردتلا لىإ ندرلأاو ينطسلف في نويحصلا نوينهلما جاتيح :تاجاتنتسلاا .لاجلما اذه 1330 EMHJ – Vol. 26 No. 11 – 2020Research article 15. Moghaddasi H, Mohammadpour A, Bouraghi H,. Azizi A, Mazaherilaghab H. Hospital information systems: the status and ap- proaches in selected countries of the MiddleEast. Electron Physician. 2018;10(5):6829–35. doi:10.19082/6829 16. Hayajneh YA, Zaghloul AA. Barriers to the adoption of health information technology in arab countries’ hospitals: practitioners’ perspective. In: 24th International Conference of the European Federation for Medical Informatics: Quality of Life through Quality of Information, 2012. 28–29 August 2012. 17. Alkraiji AI, El-Hassan O, Amin FA. Health informatics opportunities and challenges: preliminary study in the Cooperation Coun- cil for the Arab States of the Gulf. J Health Inform Dev Countries. 2014;8(1):36–45. 18. Sabur DS, Neamah AF. Electronic health in Iraq. Int J Adv Res. 2016;4(8):295–305. 19. Shaban S, Eid HO, Barka E, Abu-Zidan FM. Towards a national trauma registry for the United Arab Emirates. BioMed Central Res Notes. 2010;3(1):187. doi:10.1186/1756-0500-3-187 20. Ajami S, Arab-Chadegani R. Barriers to implement electronic health records (EHRs). Mater Sociomed. 2013;25(3): 213–5. doi:10.5455/msm.2013.25.213-215 21. Wageih MA, Marcano-Cedeño A, Gómez EJ, Mantas J. Toward more successful biomedical informatics education programs and ecosystems in the Arab world. Studies in health technology and informatics, 2015;213:111–4. PMID:26152967 22. Palestinian health capacity project. Palestinian ministry of health saves money, improves quality of care through its health in- formation system. Washington DC: USAID and Intrahealth International; 2018 ( https://www.intrahealth.org/sites/ihweb/files/ attachment-files/phcphistubasimplementation.pdf, accessed 27 March 2020). 23. Khader Y, Jabareen H, Alzyoud S, Awad S, Rumeileh NA, Manasrah N, et al. Perception and acceptance of health informatics learning among health-related students in Jordan and Palestine. In: IEEE/ACS 15th International Conference on Computer Sys- tems and Applications (AICCSA), 1 October 2018:1 6. (http://sites.birzeit.edu/hicure/, accessed 20 August 2019). 24. Al-Shorbaji, N. Househ, M. Taweel, A. Alanizi, A. Mohammed, B.O. Abaza, H. et al. (2018). Middle East and North African Health Informatics Association (MENAHIA): Building Sustainable Collaboration. Yearbook Med Informatics. 27(1):286–91. 25. Sligo J, Gauld R, Roberts V, Villa L. A literature review for large-scale health information system project planning, implementa- tion and evaluation. Int J Med Inform. 2017;97:86–97. doi:10.1016/j.ijmedinf.2016.09.007 26. Aggelidis VP, Chatzoglou PD. Methods for evaluating hospital information systems: a literature review. EuroMed J Bus. 2008;3:99–118. 27. Luna D, Almerares A, Mayan JC, González Bernaldo de Quirós F, Otero C. Health informatics in developing countries: going be- yond pilot practices to sustainable implementations: a review of the current challenges. Healthc Inform Res. 2014 Jan;20(1):3–10. doi:10.4258/hir.2014.20.1.3 28. Eley R, Fallon T, Soar J, Buikstra E, Hegney D. The status of training and education in information and computer technology of Australia’s nurses : a national survey. J Clin Nurs. 2008 Oct;17(20):2758–67. doi:10.1111/j.1365-2702.2008.02285.x 29. Trivedi MJ, Joshi A. Computer and internet use by health care professionals in a rural medical college in India. Libr Philos Pract. 2008. 30. Sadoughi F, Hemma, M, Valinejadi A, Mohammadi A, Majdabadi HA. Assessment of health information technology knowledge, attitude, and practice among healthcare activists in Tehran hospitals. Int J Computer Sci Network Security. 2017;17(1):155–8. 1331 EMHJ – Vol. 26 No. 11 – 2020Research article Barriers to initiation of insulin therapy in poorly controlled type 2 diabetes based on self-determination theory Armin Rajab,1 Pegah Khaloo,1 Soghra Rabizadeh,1 Hamid Alemi,1 Salome Salehi,1 Reza Majdzadeh,2 Hossein Mirmiranpour,1 Assadollah Rajab, Alireza Esteghamati1 and Manouchehr Nakhjavani1 1Endocrinology and Metabolism Research Center, Vali-Asr Hospital, School of Medicine, Tehran University of Medical Sciences, Tehran, Islamic Re- public of Iran (Correspondence to: M. Nakhjavani: nakhjavanim@tums.ac.ir). 2Department of Epidemiology and Biostatistics, School of Public Health, Tehran University of Medical Sciences, Tehran, Islamic Republic of Iran. Abstract Background: Proper glycaemic control can slow progression of diabetes complications. One of the main causes of poor glycaemic control is delayed initiation of insulin therapy. Aims: To explain the reasons for delayed insulin initiation based on a behavioural model using patients’ innate psycho- logical needs. Methods: We enrolled 151 patients with type 2 diabetes who had indications for insulin therapy. Thirty general practi- tioners (GPs) were included as care providers. Patients were studied by questionnaires evaluating components of self de- termination theory, such as competency, relatedness and autonomy. We also evaluated patients’ attitudes towards insulin therapy using the Insulin Treatment Appraisal Scale questionnaire. GPs’ attitudes towards insulin therapy were assessed with a different questionnaire. Results: Competency of patients was scored as acceptable (14.44/20). Relatedness score was low at around 15.63/30. The findings suggested that the patients’ intrinsic motivation was less than their extrinsic motivation (8.41/15 vs 15.03/20). The main barrier to insulin therapy on the patients’ side was rejection of severity of illness (67.5%). According to GPs, low compliance (96.7%) was the main cause of delayed insulin prescription. Conclusions: We observed that patients do not have a proper understanding about their illness. Due to the low score of relatedness as a representative of patients and care providers’ relationship, we highlight the importance of educating both about insulin therapy and how they can have the most effective relationship in this process. Keywords: type 2 diabetes, insulin therapy, psychological aspects, self-determination theory, education Citation: Rajab A; Khaloo P; Rabizadeh S; Alemi H; Salehi S; Majdzadeh R; et al. Barriers to initiation of insulin therapy in poorly controlled type 2 diabetes based on self-determination theory. East Mediterr Health J. 2020;26(11):1331–1338. https://doi.org/10.26719/emhj.20.027 Received: 11/6/18; accepted: 25/03/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo) Introduction Diabetes has become one of the biggest global health issues, due to its rapidly growing prevalence, complica- tions and high burden of disease (1). Analyses from differ- ent countries including the Islamic Republic of Iran have demonstrated poor glycaemic control of diabetes in most patients (2). One of the main causes of poor glycaemic control is delayed initiation of insulin therapy (3–5). The negative attitudes expressed by patients are fear of needles, self-blame for needing insulin, cost of insulin and doubts about efficacy (6,7). In addition, doctors prefer to postpone prescription of insulin, which is a manifestation of clinical inertia (6,8,9). Clinical inertia is defined as the failure of providers to alter therapy in the face of clear indications (6), and it is suggested causes include limited experience and knowledge along with lack of standardized guidelines (10,11). Epidemiological shift from acute to chronic diseases has required a new vision in treatment (12). In the acute-care system, patients surrender a lot of control to healthcare providers. Diabetes care requires patients and healthcare providers to collaborate in development of self-management plans (13). Therefore, in the current study, we aimed to investigate Iranian patients’ and care providers’ attitudes that delay initiation of insulin therapy. There are many behavioural models that can be used to analyse patients’ compliance and self-management ability. Self-determination theory (SDT) is one the most appropriate models that can be used in patients with diabetes. SDT is an approach to human motivation and personality that uses innate psychological needs, including competence, relatedness and autonomy, which appear to be essential for optimal functioning and social well-being (14,15). Previous research based on SDT has revealed an association between medication adherence and autonomy and competence in chronic disease. In the present study, we used SDT to explain why patients refuse to initiate insulin therapy. Methods Participants This was a descriptive cross-sectional study in 5 general medical clinics in Tehran, Islamic Republic of Iran in the summer of 2016. We randomly recruited 151 patients with poorly controlled type 2 diabetes who were candidates for insulin therapy according to the 2016 American Dia- 1332 EMHJ – Vol. 26 No. 11 – 2020Research article betes Association/European Association for the Study of Diabetes (ADA/EASD) guidelines. There were 89 women (58.9%) and 62 men (41.1%), aged 29–88 (mean 56.7) years. Despite their need to start insulin therapy, none of the patients had begun. Thirty general practitioners (GPs) (18 female, 12 male; mean age 37.6 years, age range 29–54 years) were selected randomly from general medical clin- ics to estimate their attitude towards insulin therapy. We chose GPs rather than specialists because in the Islamic Republic of Iran, most people go to general medical clin- ics and are assessed by GPs. Also, national clinical guide- lines recommend that GPs are responsible for initiating insulin therapy or referring patients to specialists. There- fore, most patients with type 2 diabetes are under treat- ment by GPs in the Islamic Republic of Iran. Study design The study was divided into 2 parts: evaluation of patients’ compliance with insulin therapy, and evaluation of GPs attitudes to insulin therapy. All of the questionnaires were filled out by patients in the presence of a trained GP to help them better understand the questionnaires. Data for demographic characteristics, pertinent clinical information and exposure to insulin therapy were col- lected from the patients. GPs were also asked to fill in a questionnaire in the presence of another GP trained in the procedure of filling in the form. This study received ethical approval from the Medical Research and Ethics Committee, Tehran University of Medical Science. Patients’ questionnaires To measure the components of SDT, some well-known questionnaires were used. (1) Perceived Competence in Diabetes Scale (PCDS), which measures by 4 items the competency of patients in controlling their diabetes (16,17). (2) Health Care Climate Questionnaire (HCCQ), which measures by 15 items the support of the health system to improve patients’ autonomy. It also has a short form (containing 6 items) that is strongly correlated with the full version and highly reliable (16,17). (3) The Treat- ment Self-Regulation Questionnaire (TSRQ) contains 2 sections (19 items) to estimate the intrinsic and extrinsic motivators of controlling diabetes, displaying autono- mous versus controlled regulation of behaviour (17). (5) Insulin Treatment Appraisal Scale (ITAS) contains 20 items measuring the positive and negative attitude of pa- tients to insulin therapy (18). Revising the questionnaires All of the above questionnaires were in English and had not been translated into Persian before this study; there- fore, after translation, a qualitative study was done to consider the effects of social and cultural differences. Then, a group session with 15 type 2 diabetes patients was held to decide about the appropriate questions and items. After summarizing the discussions by the trained GP, participants were asked to confirm the items in or- der to increase the credibility and conformability of the items. To perform content validation for development of new questionnaires, we followed Abdollahpour et al. (19). Five experts including endocrinologists, diabetes educators and an epidemiologist gave their opinion re- garding comprehensiveness, relevance and clarity of the questionnaires. They were asked to evaluate the content validity ratio for the necessity of each item in the ques- tionnaires, and to investigate the specificity and clarity of each item using a 4-point Likert scale. After these steps, updated questionnaires were arranged in the presence of 5 patients attending the initial session and the question- naires were reviewed for revalidation. Changes in questionnaires and validation The PCDS questionnaire was unchanged. It was decided to use the short form of the HCCQ questionnaire. The TSRQ questionnaire was revised to “Estimation of In- ternal Motives and Control Motives”, including 3 items for determining internal motivations and 4 to estimate external (control) motivations. In these 3 questionnaires it was intended to use the 5-point instead of 7-point Lik- ert scale. Thus, the total score of the questionnaires was as follows: PCDS (4–20); short HCCQ (6–15); internal motivation (3–15); and external motivation (4–20). For the ITAS questionnaire, positive attitudes (4 items) were excluded, and instead, only 1 item was asked from the patients to answer in the 3-point Likert scale. Two more negative attitudes were added to other items: (1) perhaps in the future insulin will become rare, for example, as a consequence of economic sanctions, and (2) insulin ther- apy is more expensive than oral therapy), and patients were asked to agree or disagree. This questionnaire is also called R-ITAS (a modified version) and can illustrate the frequency of negative aspects of insulin therapy, which are the issues that patients feel or believe would act as barriers to accepting insulin therapy. Some additional questions were added based on the results of the above process and literature review (20), including whether patients were recommended to accept insulin therapy by their healthcare provider, and whether any of their acquaintances were receiving insulin therapy. The other question was based on the number of visits of patients to physicians in a year. Their attitude towards seriousness of diabetes and risk of developing its complications was measured on a 5-point Likert scale (1, very low and 5, very high). GPs’ questionnaire The same procedure was used for GPs to prepare a ques- tionnaire containing questions including: (1) trying to understand the reasons for not accepting insulin therapy from the patients’ perspective when GPs put themselves in the patients’ place; (2) reasons for delaying or not ad- ministering insulin therapy by physicians; and (3) GPs familiarizing themselves with clinical guidelines in this area and determining the therapeutic goals based on the guidelines. The validity and reliability of the GPs’ ques- tionnaire was assessed by the above-mentioned proce- dure. The internal consistency of the questionnaire for 10 GPs was measured α = 0.78 (P < 0.001). The trained GP who was involved in the patients’ questionnaires study was not included in the GPs’ questionnaire study. 1333 EMHJ – Vol. 26 No. 11 – 2020Research article Data analysis Data were collected and analysed by SPSS version 21 sta- tistical software. The relationships between psychomet- ric factors and demographic findings were evaluated by appropriate statistical tests, such as t test, Mann–Whit- ney test and Spearman correlation analysis for nonpara- metric variables, and Pearson correlation analysis. Tests of normality were performed by Kolmogorov–Smirnov test. Many of the distributions were not normal, includ- ing PCDS, HCCQ, and intrinsic motivation and extrinsic motivation (obtained from modified TSRQ), (all P < 0.001), so nonparametric signed rank tests were run. P < 0.05 was considered statistically significant. Results Baseline characteristics The average number of checkups per year for the 151 patients was 3.14, with significantly more in women (P = 0.002). The mean time that doctors spent on a patient in a usual visit was < 5 minutes. Only 73 patients (48.3%) were recommended to initiate insulin therapy. This number was significantly lower in patients who were un- der supervision of a GP [odds ratio (OR) = 5.56, P < 0.001, χ2 = 25.4]. Forty (26.5%) patients developed diabetes com- plications (Table 1). Baseline fasting blood sugar, haemoglobin A1c concentration, body mass index (BMI), systolic and diastolic blood pressure, and duration of diabetes are shown in Table 2. Mean BMI was significantly higher in female patients (P < 0.001, t = 10.99). We asked patients to score the seriousness of diabetes out of 5 (1, very low and 5, very high). The mean score was 3.44. Only 17 (11.3%) patients described diabetes as having low seriousness. We also asked patients to evaluate their risk of developing diabetes complications by giving a risk score of 1 (very low), 2 (low), 3 (moderate), 4 (high) or 5 (very high). The number of patients in each category was 73 (48.3%), 52 (34.4%), 12 (7.9%), 13 (8.6%) and 1 (0.7%), respectively. Mean score was 1.79. Psychometric findings The results for PCDS, HCCQ and modified TSRQ (intrin- sic and extrinsic motivation) are shown in Table 3. There was no significant difference in controlled motivation, self-motivation and perceived competence between pa- tients who were under supervision of a GP compared with an internist. HCCQ scores were significantly higher in patients working with an internist (P = 0.003, z = 2.94) and had a positive correlation with the time doctors spent in a usual visit (P < 0.001, z = 5.78). Self-motivation scores were also higher in this group. Age had an inverse cor- relation with PCDS (P < 0.001) and a positive correlation with controlled motivation (P < 0.001; r = 0.31). HCCQ, self-motivation and PCDS scores were lower in patients with diabetes complications (P < 0.001, z = 2.4; P < 0.001, z = 2.55; P = 0.011, z = 4.47, respectively). Patient attitudes toward insulin therapy Based on the findings of the R-ITAS questionnaire de- veloped from modification of ITAS, the following results were obtained. Only 39 (25.8%) patients believed in insu- lin efficacy; 82 (54.3%) did not have enough knowledge; and 30 (19.9%) thought that insulin was ineffective for treatment of diabetes. The most common reasons for patients refusing insulin therapy are shown in Figure 1. The main reason was that they did not believe in the se- verity of their disease. Only 26 (17.2%) patients were wor- ried about hypoglycaemia after initiating insulin therapy and it was not one of the common reasons of avoiding insulin therapy. Fear of needles (P = 0.02, t = 3.25) and difficulty with injecting the right amount of insulin (P = 0.035, t = 3.16) were more prevalent in older patients. The number of patients who agreed with insulin efficacy (n = 16; 27.1%) was significantly lower among those who knew another patient using insulin (P < 0.001). Patients with lower level of education, including illiteracy and be- low diploma level (n = 53; 55.8%), believed that they did not have enough knowledge about whether insulin was effective, compared with patients with higher level of education (n = 23; 41.0%) (P = 0.049, χ2 = 9.57). Patients with lower level of education agreed more than others that managing insulin injections takes a lot of time and energy (OR = 2.44, 95% confidence interval = 1.14–5.2, P = 0.012, χ2 = 6.30). Table 1 Demographic findings Characteristics No. % Sex Male 62 41.1 Female 89 58.9 Education Illiterate 10 6.6 Below diploma 85 56.3 Diploma 34 22.5 Higher education 22 14.6 Mean annual income (US$) < 2400 14 9.3 2400–4200 82 54.3 > 4200 55 36.4 Physician level of practice General practitioner 83 55 Internist (Endocrinologist) 68 45 Long-term complications Yes 40 26.5 No 111 73.5 Family history of type 2 diabetes Yes 99 65.5 No 52 34.4 Family history of insulin treatment Yes 59 39.1 No 92 60.9 1334 EMHJ – Vol. 26 No. 11 – 2020Research article GP attitudes From the GPs’ perspective, the most common reasons for patient refusal of insulin are shown in Figure 2. The most common reasons for GPs delaying insulin prescrip- tion were: expectation of low patient compliance (n = 29; 96.7%); fear of hypoglycaemia (n = 25; 83.3%); insulin is the last choice of therapy (n = 20; 66.7%); and lack of suit- able guidelines and training (n = 16; 53.3%). Twenty-four GPs believed they needed to refer patients to an internist or endocrinologist for insulin initiation, while the other 6 believed that they were able to initiate insulin therapy by themselves. Only 11 GPs managed diabetes based on standardized guidelines (ADA/EASD), and only 5 had read up-to-date guidelines. Discussion We aimed to determine the barriers to initiation of in- sulin therapy in patients with type 2 diabetes, who had clear indications for starting insulin therapy but were not using insulin. Our results showed that more than half of our patients did not have enough knowledge about in- sulin efficacy. This is one manifestation of inadequate knowledge of patients about diabetes (21). Only 17.2% of patients found themselves at risk of diabetes compli- cations. Denial of the severity of the disease and denial of the failure of oral agents were the main reasons for refusing insulin therapy among patients. It means that they cannot accept that their disease has progressed. This can be explained by the chronic nature of diabetes and the delayed appearance of complications. In other words patients do not worry about the future consequences of the disease because they feel healthy at the present time. Patients’ belief that injecting insulin is embarrassing was another noticeable reason for their delaying initiation of insulin therapy. This highlights the importance of injec- tion stigma and how it affects the proper treatment of di- abetes. Other reasons expressed by participants included difficulties in fulfilling daily responsibilities and difficul- ties with injecting the right amount of insulin. Doctors assumed that fear of needles and the pain caused by injection were the main reasons for rejecting insulin therapy among patients; however, as mentioned before, patients feel embarrassed about insulin injection and are not necessarily frightened of it. In our study, 51.7% of patients were not recommended to take insulin at all despite the medical indicators showing a significant need for it. Care providers should endeavour to explain more thoroughly the benefits of taking insulin and try to make patients comfortable with starting treatment early. This would break the cycle of clinical inertia. This is more common among patients who are under supervision of a GP (22). Our study showed similar results. Low compliance and high possibility of hypoglycaemia were the main reasons GPs expressed for avoiding insulin prescription. We observed that just 17.2% of patients were worried about hypoglycaemia as a result of insulin therapy compared to 83.3% of GPs. The reason could be lack of knowledge of people about insulin therapy and the risk of hypoglycaemia. It requires more complete future investigation in the future. Since most patients with diabetes attend GPs for treatment (23) we need to educate GPs and establish standardized guidelines. A significant majority (63.3%) of our participants complained about a lack of national guidelines for diabetes treatment. There are clinical guidelines discussing the principles needed to be considered by doctors in every visit, although many doctors do not have access to these guidelines; therefore, they may be neglected in usual visits (10,24). Table 2 Baseline characteristics SDMeanMaximumMinimumCharacteristics 74.96183.7359754FBS (mg/dl) 1.119.4214.607.50HbA1c (%) BMI (kg/m2) 4.3927.0854.0913.29Male 5.1729.3854.0915.06Female 20.64132.9920791SBP (mmHg) 11.7679.5313051DBP (mmHg) 5.618.46301Duration of diabetes (yr) BMI = body mass index; DBP = diastolic blood pressure; FBS = fasting blood sugar; HbA1c = haemoglobin A1c; SBP = systolic blood pressure; SD = standard deviation. Table 3 Psychometric findings SDMeanMaximumMinimum 3.1514.44194PCDS 4.6715.63286HCCQ 2.518.41144Intrinsic motivation 2.715.03195Extrinsic motivation HCCQ = Health Care Climate Questionnaire; PCDS = Perceived Competence in Diabetes Scale; SD = standard deviation. 1335 EMHJ – Vol. 26 No. 11 – 2020Research article Psychometric findings showed a reasonable score (14.4/20) for PCDS. This index indicates the competency of diabetic patients to control their disease (25). Our results demonstrated an inverse association between age and PCDS score. This indicates that ageing has a negative impact on how patients feel about their competence in self-management. These results confirm previous studies (26). Older patients need more help in order to enhance this index, especially for complicated procedures such as insulin therapy. Contrary to PCDS index, the mean HCCQ score was not positive enough. It was almost half of the maximum score. This index reflects the relationship between care providers and patients and how healthcare systems support patients for self-management (27). Numerous studies have suggested that low HCCQ score and poor communication with care providers could be one of the reasons for patients refusing insulin therapy (27). Moreover, low HCCQ score aggravates the negative attitudes toward insulin therapy among patients (28). A well-established relationship between doctors and patients improves the latter’s attitude (27). A low HCCQ score indicates poor insight of doctors about diabetes treatment and the fact that they should play the role of counsellor for patients (13). We found similar results in our study since the reasons patients expressed for refusing insulin therapy varied from those doctors assumed. Grant et al. described the negative impacts of these prejudgments on initiation of insulin therapy (29). We studied the components of TSRQ index including intrinsic motivation and extrinsic motivation separately. Extrinsic motivations were significantly higher compared to intrinsic motivations; however, it is reported that intrinsic motivation is more important for initiation of and adherence to insulin therapy (17). Extrinsic motivation on its own can be counterproductive. It is necessary for the internalization of the requested behaviour but if the internalization does not occur properly it can make the patients disappointed rather than motivated (14,15). In our study intrinsic motivations had the lowest scores among SDT parameters. That is because patients do not find themselves at risk of diabetes complications. Figure 1 Common reasons for patients refusing insulin 0 10 20 30 40 50 60 70 80 Denial of disease severity Denial of failure of treatment with oral agents Embarrassment about injection Interference with activities of daily living Difficulty of timely and correctly use of insulin Ongoing loss of personal health Data presented in bars as percentage (number of cases) % Figure 2 Common reasons for patients refusing insulin from general practitioners’ point of view 0 20 40 60 80 100 Fear of injection Difficulty of adjusting insulin dose Interference with responsibilities Pain of injection Denial of disease severity % Data presented in bars as percentage (number of cases) 1336 EMHJ – Vol. 26 No. 11 – 2020Research article It is reported that this issue can minimize patients’ motivation to initiate more-intensive therapies such as insulin therapy (30). One particularly interesting finding of our study was that patients who knew a person using insulin did not believe in insulin efficacy. It can be explained by the poor glycaemic control of patients when initiating insulin therapy (3,31). In other words, patients who accept insulin therapy are those who are at later stages of the disease but people suppose that insulin has caused complications. We showed that participants with lower compared with higher educational level had uncertainty about the efficacy of insulin therapy. However, agreement about efficacy of insulin therapy among participants with higher education level was not as high as we expected. Perhaps these findings are based on the fact that patients with low and high education level had little knowledge about diabetes and its treatment, although the latter group had more definite opinions. In addition, highly educated patients disagreed more with the idea that managing insulin injections takes a lot of time and energy. These results may explain the self-confidence of patients with higher level of education. There were some limitations to our study. First, benchmark scores for the questionnaires have not yet been determined in the Islamic Republic of Iran. Therefore, we tried just to show the components of SDT among Iranian patients themselves. Second, we were not able to evaluate the general knowledge of our participants about diabetes because of low cooperation and large number of questionnaires. Third, we only included GPs as care providers. In conclusion, initiation of insulin therapy is dependent on multiple factors. We observed that patients do not have a proper understanding of their illness and require improved intrinsic motivation. Care providers do not have an accurate understanding of what the patients’ psychological barriers to treatment are. Our study highlights the importance of educating both patients and care providers about insulin therapy and how they can have effective communication in this relationship. Acknowledgement The authors wish to thank all patients for their participation and kind cooperation. Funding: None. Competing interests: None declared. Obstacles face à la mise en place de l’insulinothérapie dans le diabète de type 2 mal contrôlé selon la théorie de l’autodétermination Résumé Contexte : Un contrôle adéquat de la glycémie permet de ralentir la progression des complications du diabète. Le retard dans la mise en place de l’insulinothérapie constitue l’une des principales causes d’un mauvais contrôle de la glycémie. Objectifs : Expliquer les raisons du retard de la mise en place du traitement par insuline sur la base d’un modèle comportemental fondé sur les besoins psychologiques innés des patients. Méthodes : Nous avons recruté 151 patients atteints de diabète de type 2 pour lesquels l’insulinothérapie était indiquée. Une trentaine de médecins généralistes ont été inclus en qualité de prestataires de soins. L ’ étude a été menée au moyen de questionnaires évaluant des composantes de la théorie de l’autodétermination, telles que la compétence, l’affiliation et l’autonomie. Nous avons également évalué les attitudes des patients vis-à-vis de l’insulinothérapie à l’aide du questionnaire de l’échelle d’évaluation de l’insulinothérapie (Insulin Treatment Appraisal Scale). Les attitudes des médecins généralistes vis- à-vis de l’insulinothérapie ont été évaluées au moyen d’un questionnaire différent. Résultats : La compétence des patients a été jugée acceptable (14,44/20). Le score se rapportant à l’affiliation était faible, à environ 15,63/30. Les résultats suggèrent que la motivation intrinsèque des patients était inférieure à leur motivation extrinsèque (8,41/15 contre 15,03/20). Du côté des patients, le rejet de la gravité de la maladie (67,5 %) constituait le principal obstacle à la mise en place de l’insulinothérapie. Selon les médecins généralistes, une faible observance (96,7 %) était la principale cause de retard dans la prescription d’insuline. Conclusions : Nous avons observé que les patients ont une mauvaise compréhension de leur maladie. En raison de la faiblesse du score concernant l’affiliation, en tant qu’élément représentant la relation entre les patients et les prestataires de soins, nous soulignons l’importance d’une meilleure sensibilisation des malades et soignants à l’insulinothérapie et de la manière de mettre en place une relation efficace. 1337 EMHJ – Vol. 26 No. 11 – 2020Research article References 1. IDF diabetes atlas. Brussels: International Diabetes Federation; 2015. 2. ADVANCE Collaborative Group, Patel A, MacMahon S, Chalmers J, Neal B, Billot L, et al. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med. 2008 Jun 12;358(24):2560–72. http://dx.doi.org/10.1056/NEJ- Moa0802987 PMID:18539916 3. Calvert MJ, McManus RJ, Freemantle N. Management of type 2 diabetes with multiple oral hypoglycaemic agents or insulin in primary care: retrospective cohort study. Br J Gen Pract. 2007 Jun;57(539):455–60. PMID:17550670 4. Russell-Jones D, Pouwer F, Khunti K. Identification of barriers to insulin therapy and approaches to overcoming them. Diabetes Obes Metab. 2018 Mar;20(3):488–496. http://dx.doi.org/10.1111/dom.13132. PMID:29053215 5. Ates E, Set T, Saglam Z, Tekin N, Karatas Eray I, Yavuz E, et al. Insulin initiation status of primary care physicians in Turkey, bar- riers to insulin initiation and knowledge levels about insulin therapy: a multicenter cross-sectional study. Prim Care Diabetes. 2017 Oct;11(5):430–6. http://dx.doi,org/10.1016/j.pcd.2017.05.003 PMID:28579057 6. Peyrot, M., Rubin RR, Lauritzen T, Skovlund SE, Snoek FJ, Matthews DR, et al. Resistance to insulin therapy among patients and providers: results of the cross-national Diabetes Attitudes, Wishes, and Needs (DAWN) study. Diabetes Care. 2005 Nov;28(11):2673–9. http://dx.doi.org/10.2337/diacare.28.11.2673 PMID:16249538 7. Krall, J., Gabbay R, Zickmund S, Hamm ME, Williams KR, Siminerio L. Current perspectives on psychological insulin resistance: primary care provider and patient views. Diabetes Technol Ther. 2015 Apr;17(4):268–74. http://dx.doi.org/10.1089/dia.2014.0268 PMID:25551737 8. American Diabetes Association. Standards of medical care in diabetes. Diabetes Care, 2015 Jan;38(Suppl 1). https://care.diabetes- journals.org/content/suppl/2014/12/23/38.Supplement_1.DC1/January_Supplement_Combined_Final.6-99.pdf 9. Brunton, SA. Implementing treatment guidelines for type 2 diabetes in primary care. Postgrad Med. 2009 Apr;121(2):125–38. http://dx.doi.org/10.3810/pgm.2009.03.1985 10. Bernard AM, Anderson L, Cook CB, Phillips LS. What do internal medicine residents need to enhance their diabetes care? Diabe- tes Care. 1999 May;22(5):661–6. http://dx.doi.org/10.2337/diacare.22.5.661 PMID:10332662 11. Drass J, Kell S, Osborn M, Bausell B, Corcoran J Jr, Moskowitz A, et al. Diabetes Care for Medicare Beneficiaries: Attitudes and behaviors of primary care physicians. Diabetes Care, 1998 Aug;21(8):1282–7. http://dx.doi.org/10.2337/diacare.21.8.1282 PMID:9702434 12. Anderson RM, Funnell MM. Compliance and Adherence are Dysfunctional Concepts in Diabetes Care. Diabetes Educ. 2000 Jul– Aug;26(4):597–604. http://dx.doi.org/10.1177/014572170002600405 PMID:11140071 لىإ ًادانتسا ًاديج هطبض نكمي لا يذلا نياثلا طمنلا نم ير َّك ُّسلا ءادل ينلوسنلأاب جلاعلا ءدب نود لو َُت يتلا قئاوعلا يرصلما ريرقت ةيرظن اضر ليع ،بجر للها دسأ ،روبنايرميرم ينسح ،هدازدمج اضر ،يلحاص يمولاس ،يميلعلا ديحم ،هداز بيار ىرغص ،ولاخ اجيب ،بجر ينمرأ نياوجخن رهچونم ،تياماغتسإ ةصلالخا بابسلأا دحأ ينلوسنلأاب جلاعلا ءدب رخأت دعُيو .ير َّك ُّسلا تافعاضم مقافت ءاطبإ لىإ حيحص ٍوحن لىع مدلا ركس طبض يدؤي نأ نكمي :ةيفللخا .مدلا ركسل ديلجا طبضلا نود لوتح يتلا ةيسيئرلا .ةيرطفلا ةيسفنلا ضىرلما تاجايتحا لىإ دنتسي يكولس جذومن لىع ًءاِنب ينلوسنلأاب جلاعلا ءدب ر ُّخأت بابسأ حشر لىإ ةساردلا هذه تفده :فادهلأا ًاماع ًاسرامم ينثلاث تلمشو .ينلوسنلأاب جلاعلا متهلااح يعدتست نياثلا طمنلا نم ير َّك ُّسلاب ًاضيرم 151 ةساردلا هذه تلمش :ثحبلا قرط طابترلااو ةءافكلا لثم ،يرصلما ريرقت ةيرظن تان ِّوكم مييقتل تانايبتسا قيرط نع ةساردلل ضىرلما عضخو .ةياعرلا تامدخ يم ِّدقُم رودب نوعلطضي ينسرمالما فقاوم تَمّيُقو .ينلوسنلأاب جلاعلا مييقت سايقم نايبتسا مادختساب ينلوسنلأاب جلاعلا هاتج ضىرلما فقاولم ًماييقت انيرجأ ماك .ةيللاقتسلااو .فلتمخ نايبتسا مادختساب ينلوسنلأاب جلاعلا هاتج ين ِّماعلا تراشأو .30/15.63 وحنب ر َّدقُت ةضفخنم ةجرد لىع طابترلاا لصح مايف ،)20/14.44( ةلوبقم ةجرد لىع ضىرلما ةءافك تلصح :جئاتنلا مادقإ نود لوُيح يذلا سييئرلا قئاعلا لَّثَتو .)20/15.03 لباقم 15/8.41( ةيجرالخا مهعفاود نم َّلقأ ناك ضىرملل تياذلا عفادلا نأ لىإ جئاتنلا ببسلا )%96.7( لاثتملاا ضافخنا ناك ،ين ِّماعلا ينسرماملل ًاقْفَوو .)%67.5( ضرلما ةدشب فاترعلاا مدع في ينلوسنلأاب جلاعلا لىع ضىرلما .ينلوسنلأا فصو ر ُّخأتل سييئرلا تامدخ يم ِّدقُمو ضىرلما ينب ةقلاعلل لثممك طابترلاا ةجرد ضافخنلا ًارظنو .مهضرلم ٍحيحص ٍمهفب نوعتمتي لا ضىرلما نأ انظحلا :تاجاتنتسلاا راطإ في ينفرطلا ينب ةلاّعف ةقلاع لضفأ لىإ لوصولا نكمي فيكو ،ينلوسنلأاب جلاعلا نأشب ماهنم ٍلك فيقثت ةيهمأ لىع َءوضلا طّلسُن اننإف ،ةياعرلا .ةيلمعلا هذه 1338 EMHJ – Vol. 26 No. 11 – 2020Research article 13. Anderson RM, Funnell MM. Patient empowerment: reflections on the challenge of fostering the adoption of a new paradigm. Patient Educ Couns. 2005 May;57(2):153–7. http://dx.doi.org/10.1016/j.pec.2004.05.008 PMID:15911187 14. Ryan RM, Deci EL. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. Am Psychol. 2000 Jan;55(1):68–78. http://dx.doi.org/10.1037//0003-066x.55.1.68 PMID:11392867 15. Vallerand RJ. Deci and Ryan’s self-determination theory: a view from the hierarchical model of intrinsic and extrinsic motiva- tion. Psychol Inq. 2000;11(4):312–8. https://www.jstor.org/stable/1449629?seq=1 16. Williams GC. Patrick H, Niemiec CP, Williams LK, Divine G, Lafata JE, et al. Reducing the health risks of diabetes: how self-de- termination theory may help improve medication adherence and quality of life. Diabetes Educ. 2009 May–Jun;35(3):484–92. http://dx.doi.org/10.1177/0145721709333856 PMID:19325022 17. Williams GC, McGregor HA, Zeldman A, Freedman ZR, Deci EL. Testing a self-determination theory process model for promot- ing glycemic control through diabetes self-management. Health Psychol, 2004 Jan;23(1):58–66. http://dx.doi.org/10.1037/0278- 6133.23.1.58 PMID:14756604 18. Snoek FJ, Skovlund SE, Pouwer F. Development and validation of the insulin treatment appraisal scale (ITAS) in patients with type 2 diabetes. Health Qual Life Outcomes. 2007 Dec 20;5:69. http://dx.doi.org/10.1186/1477-7525-5-69 PMID:18096074 19. Abdollahpour I, Nedjat S, Noroozian M, Majdzadeh R. Performing content validation process in development of questionnaires. Iran J Epidemiol. 2011;6(4):66–74. http://irje.tums.ac.ir/browse.php?a_id=70&sid=1&slc_lang=en 20. Kunt T, Snoek FJ. Barriers to insulin initiation and intensification and how to overcome them. Int J Clin Pract Suppl. 2009 Oct;164:6–10. http://dx.doi.org/10.1111/j.1742-1241.2009.02176.x PMID:19751453 21. Rezaaei J, Rezaaei M, Razlaansari HA, Khaaledi B, Taymoori B, Khaatooni AR, et al. Self-care in insulin therapy among type I dia- betics referring to the Diabetes Research Center (Kermanshah,1999). J Kermanshah Univ Med Sci. 2003;6(4):e81165 (in Persian). https://jkums.com/en/articles/81165.html 22. Ziemer DC, Miller CD, Rhee MK, Doyle JP, Watkins C Jr, Cook CB et al. Clinical inertia contributes to poor diabetes control in a primary care setting. Diabetes Educ. 2005 Jul–Aug;31(4):564–71. http://dx.doi.org/10.1177/0145721705279050 PMID:16100332 23. Cobble ME. Initiating and intensifying insulin therapy for type 2 diabetes: why, when, and how. Am J Ther. 2009 Jan– Feb;16(1):56–64. http://dx.doi.org/10.1097/MJT.0b013e3181966bf0 PMID:19142156 24. Giangola J, Olohan K, Longo J, Goldstein JM, Gross PA. Barriers to hyperglycemia control in hospitalized patients: a descriptive epidemiologic study. Endocr Pract. 2008 Oct;14(7):813–9. http://dx.doi.org/10.4158/EP.14.7.813 PMID:18996808 25. Anderson RM, Funnell MM, Fitzgerald JT, Marrero DG. The Diabetes Empowerment Scale: a measure of psychosocial self-effi- cacy. Diabetes Care. 2000 Jun;23(6):739–43. http://dx.doi.org/10.2337/diacare.23.6.739 PMID:10840988 26. Nam S, Chesla C, Stotts NA, Kroon L, Janson SL. Factors associated with psychological insulin resistance in individuals with type 2 diabetes. Diabetes Care. 2010 Aug;33(8):1747–9. http://dx.doi.org/10.2337/dc10-0099 PMID:20435797 27. Heisler M, Vijan S, Anderson RM, Ubel PA, Bernstein SJ, Hofer TP. When do patients and their physicians agree on diabetes treatment goals and strategies, and what difference does it make? J Gen Intern Med. 2003 Nov;18(11): 893–902. http://dx.doi. org/10.1046/j.1525-1497.2003.21132.x PMID:14687274 28. Korytkowski M. When oral agents fail: practical barriers to starting insulin. Int J Obes Relat Metab Disord. 2002 Sep;26 Suppl 3:S18–24. http://dx.doi.org/10.1038/sj.ijo.0802173 PMID:12174319 29. Grant R, Adams AS, Trinacty CM, Zhang F, Kleinman K, Soumerai SB. Relationship between patient medication adherence and subsequent clinical inertia in type 2 diabetes glycemic management. Diabetes Care. 2007 Apr;30(4):807–12. http://dx.doi. org/10.2337/dc06-2170 PMID:17259469 30. Anderson RJ, Freedland KE, Clouse RE, Lustman PJ. The Prevalence of Comorbid Depression in Adults With Diabetes. Diabetes Care. 2001 Jun;24(6):1069–78. http://dx.doi.org/10.2337/diacare.24.6.1069 PMID:11375373 31. Khunti K, Wolden ML, Thorsted BL, Andersen M, Davies MJ. Clinical inertia in people with type 2 diabetes. Diabetes Care. 2013 Nov;36(11):3411–7. https://doi.org/10.2337/dc13-0331 1339 EMHJ – Vol. 26 No. 11 – 2020Research article Prevalence of non-reporting of hospital medical errors in the Islamic Republic of Iran Mehrdad Askarian,1 Seyyed M. Sherafat,2 Maryam Ghodsi,3 Zahra Shayan,4 Charles Palenik,5 Nahid Hatam6 and Yavor Enchev7 1Department of Community Medicine, School of Medicine, Shiraz University of Medical Sciences, Shiraz, Islamic Republic of Iran; Health Behavior Science Research Center, Shiraz University of Medical Sciences, Shiraz, Islamic Republic of Iran. 4Trauma Research Center, Department of Community Medicine; 2Student Research Committee; 6Department of Health Service Administration, School of Management and Information Sciences, Shiraz University of Medical Sciences, Shiraz, Islamic Republic of Iran (Correspondence to: Mehrdad Askarian: askariam@sums.ac.ir). 3Shiraz University of Medical Sciences, Shiraz, Islamic Republic of Iran. 5School of Dentistry, Indiana University, Indianapolis, United States of America. Shiraz University of Medical Sciences, Shiraz, Islamic Republic of Iran. 7Department of Neurosurgery, University Hospital St. Marina, Medical University of Varna, Varna, Bulgaria.. Abstract Background: Medical errors frequently occur in health care facilities, jeopardizing patient safety and increasing associ- ated costs. Aims: This cross-sectional investigation examined the rates of and reasons for non-reporting of medical errors at Nema- zee Hospital, Shiraz, Islamic Republic of Iran. Methods: Self-administered questionnaires were completed by 283 staff members, including physicians, nurses and medical students. One-way analysis of variance, Fisher’s least significant difference post hoc, Spearman correlation coef- ficient and intraclass correlation tests were used for statistical analyses. Results: Almost all (95.8%) participants had observed at least 1 medical error during the previous year, with over half (50.5%) observing 3–10 errors. The preferred method for reporting medical errors among physicians and medical students was verbal and informal (40.3% and 41.8% respectively), while nurses preferred written forms (45.7%). The results indicat- ed significant differences between groups concerning individual and organizational barriers in general, and among all sub-categories (P < 0.001). Conclusion: Concerns of legal entanglements and confidentiality issues were recognized as the main barriers to report- ing medical errors. Keywords: medical errors, non-reporting, patient safety, hospitals, health care workers Citation: Askarian M; Sherafat SM; Ghodsi M; Shayan Z; Palenik C; Hatam N; et al. Prevalence of non-reporting of hospital medical errors in the Islam- ic Republic of Iran. East Mediterr Health J. 2020;26(11):1339–1346. https://doi.org/10.26719/emhj.19.050 Received: 20/06/17; accepted: 30/07/18 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO li- cense (https://creativecommons.org/licenses/by-nc-sa/3.0/igo) Introduction More than 16 years after the landmark report by the Insti- tute of Medicine, serious concerns about patient safety continue to exist (1–3). Medical errors are estimated to be the third leading cause of death in the United States of America (4). There have been 4 major studies conducted since the Institute of Medicine study concerning deaths associated with medical errors. Together they cover over 37 million patient admissions with a 3.1% rate of adverse events and a 0.7% mortality rate (5–7). Efficiency, security of care, care giver reactivity and patient contentment are safety parameters that are at the core of health care quality (3). The most crucial factor, however, is a comprehensive understanding of medical errors (8). Gathering meticulous evidence via transparent incident reporting, free sharing of data and the creation of a culture of learning from our mistakes are indispensable in the development of medical error reduction plans and improving patient safety (2,5,9). The backbone of the movement towards an enhanced culture of safety is a well-organized error reporting system (10). A survey in 6 South Australian hospitals reported that even though 98.3% of respondents were aware of their facility’s incident reporting systems, more than 40% had never filed a report (11). Another study reported 84.3% of 338 internal medicine physicians and residents believed that reporting medical errors improved quality. However, their rate for reporting minor errors was only 16.9% and for major errors 3.8% (12). Understanding the multiple factors that influence reporting errors among health care workers is crucial to supplying missing elements of an effective communication programme. Such barriers can damage the transparency of a safety climate and a culture of learning from errors (8,13,14). A perfectionist belief is that only poor physicians make mistakes (2,13). Several studies indicate that there is a lack of knowledge concerning reportable incidences. Many physicians and nurses often do not consider near misses and medication omissions as being reportable (2). In one study, 25% of participants did not know how to retrieve their facility’s incident reporting form (11). At the institutional level, the safety climate often determines front-line provider attitudes (13). Loss of 1340 EMHJ – Vol. 26 No. 11 – 2020Research article malpractice insurance coverage, fear of punitive actions, time constraints, poorly designed reporting systems, negative feedback, lack of confidentiality and a power hierarchy within professional groups are factors that negatively affect error reporting (2,8,13,14). Other factors can be even more specific, such as fear of disciplinary action and threats to positive evaluations and promotion, especially among nurses (2,13,15). In a study of 20 hospitals in the north of the Islamic Republic of Iran, 182 (0.06% of 317 966 admissions) medical errors were reported. The lack of a reliable reporting system, negative attitudes toward reporting among staff and managers and a punitive culture were mentioned as the potential causes (16). The rate of non-reporting was estimated at 78.9% within the nursing staff of a teaching hospital in Kermanshah (17). There were comparable results in 2 other independent surveys among nurses at Imam Khomeini Hospital in Tehran (18,19). Barriers to medical error reporting were investigated among health care providers in the Islamic Republic of Iran in 2012. The study reported a high incidence of mishaps with lower rates of reporting (< 50%). The absence of an effective medical error reporting system, an insufficient supporting atmosphere among peers, lack of sufficient knowledge regarding the importance of error reporting and fear of malpractice litigation were listed as the most common impediments (20). In the summer of 2014, a web-based error reporting system was launched at Nemazee Hospital in Shiraz. Reporting forms were revised and multiple training courses presented. Following these efforts, error reports increased by approximately 2000 cases per month. However, topical specialists felt that there was still room for improvement. Also, there was a strong need to compare the effectiveness of the Nemazee Hospital programme with others currently operating in the Islamic Republic of Iran. Therefore, the aims of this study were to determine the reasons behind non-reporting by hospital staff physicians, nurses and medical students and how well the facility error reporting is actually operating. Methods Study setting We conducted a cross-sectional descriptive study from September–December 2015 in Nemazee Hospital, Shiraz, Islamic Republic of Iran. The facility is a general, special- ty and subspecialty teaching hospital with 750 beds and is considered as the main referral centre in the south of the country. Questionnaires Data were collected from 283 participants (82% response rate), including physicians, nurses and medical students using a self-administered questionnaire. Samples were selected with a confidence level of 95%, standard devia- tion 2 (score of barriers against error reporting based on a pilot study), precision of 0.25, population of 2000 and considering withdrawal proportion of 25%. The study population was selected randomly from available personnel in all 32 wards in the hospital and during all shifts. Questionnaires were presented in written form and completed immediately during a break in the shift and took 10 minutes on average. The questionnaire used in this study was developed by a hospital study group. A group of topical experts then assessed the questionnaire for content validity. The questionnaire was piloted on 19 medical students and 17 nurses in the study group. Cronbach’s alpha calculation was used to assess reliability. The resulting value was 0.819. The first section of the questionnaire covered demographic characteristics, including sex, work experience and department. The second section contained questions about participant knowledge concerning medical errors, past in-service training, number of witnessed errors during the previous year, preferred method for reporting errors and response to medical errors committed, either by themselves or their peers. Barriers to medical error reporting were assessed through 13 statements involving personal and organizational opinions and behaviours. Participants reported their level of agreement using a scale of 1–10. Unfamiliarity with medical errors or the reporting process, fear of punishment, cultural issues within a group, lack of confidentiality, a time-consuming reporting system, lack of proper feedback and fear of malpractice lawsuits were considered as organizational factors. There were 4 additional questions that addressed respondent knowledge about medical errors. Statistical analyses Analysis included descriptive analysis, 1-way analysis of variance, Fisher’s least significant difference post hoc, in- traclass correlation (ICC) test. SPSS, version 23, was used. P-value < 0.05 was considered significant. Ethical considerations The protocol of this study was approved by the Shiraz University of Medical Sciences research ethics commit- tee (IR.SUMS.MED.REC.1394.S01). Participation was vol- untary. Results Participants included 151 nurses (53.3%), 77 physicians (27.2%) and 55 medical students (19.4%). The largest group of participants [53 nurses (35.1%), 41 physicians (53.2%) and 23 medical students (41.8%)] worked in internal med- icine units; 62 nurses (41.1%) and 31 physicians (40.2%) had less than 5 years of work experience. Witnessing at least 1 error during the previous year was reported by 271 (95.5%) respondents. More than half (143, 50.5%) had ob- served 3–10 medical errors in the previous year (Table 1). 1341 EMHJ – Vol. 26 No. 11 – 2020Research article Physicians (n = 31, 40.3%) and medical students (n = 23, 41.8%) preferred to report errors via a telephone call or in person to a supervisor. However, nurses favoured using written forms to report an error (n = 69, 45.7%). Among physicians, 32 (41.6%) stated that “they would report only themselves, believing reporting superiors would be considered offensive.” Among nurses, 54 (35.8%) believed that “It is mandatory to report all witnessed errors to a supervisor”. The largest group of medical students (n = 21, 38.2%) indicated that “It is mandatory to disclose if the committed error is significant (e.g. vitally important)”. Knowing the person making an error would not influence the decision to report for 38 physicians (49.4%), 120 nurses (79.5%) and 31 medical students (56.4%). To measure agreement of actual information level and the self-perception of participants, we calculated group ICC indices: most correlations were weak. The degree of consistency among physicians (ICC = 0.06) was less than for nurses (ICC = 0.39) and medical students (ICC = 0.38). The questionnaire also assessed the impact that hospital-provided medical error training courses had on study participants; 55 physicians (71.4%) and 24 medical students (43.6%) reported having no or inadequate training on errors(n = 59, 76.6%) and error reporting (n = 39, 70.9%). Conversely, most nurses reported that they had received adequate training on error identification (n = 89, 58.9%) and error reporting (n = 98, 64.9%). The second part of the questionnaire contained 13 statements regarding possible obstacles to error reporting. The analysis of variance test indicated significant differences between groups concerning individual barriers in general and among all sub- categories, with nurses having the highest mean score for all categories (Table 2). The least significant difference post-hoc analysis showed statistically significant differences between nurses and physicians (P < 0.001) and nurses and medical students (P < 0.001). Similarly, significant differences existed among the 3 study groups concerning organizational barriers in general and among all sub-categories (again, nurses had the highest mean score in each category) (Table 3). Barriers most often mentioned by physicians were “Worrying about the revelation of a colleague’s identity”, Table 1 Distribution of observed medical errors among physicians, nurses and medical students (n = 283) during 2015 in Nemazee Hospital, Shiraz , Islamic Republic of Iran Occupation No. of times seen Not seen (%) 1–2 (%) 3–10 (%) > 10 (%) Physician 1.3 23.4 54.5 20.8 Nurse 7.3 31.1 45.0 16.6 Medical student 0.0 20.0 60.0 20.0 Total 4.2 26.9 50.5 18.4 Table 2 Score distribution for individual barriers to error reporting among physicians, nurses and medical students (n = 283) in Nemazee Hospital, Shiraz, Islamic Republic of Iran, 2015 Barrier Mean (range 0–5) SD P-value Individual (overall) Physicians 1.8 0.4 < 0.001 Nurses 3.6 1.7 Medical students 1.9 0.3 Unfamiliarity with medical error/reporting process Physicians 1.7 0.5 < 0.001 Nurses 3.1 1.9 Medical students 1.8 0.5 Fear of punishment Physicians 2.1 0.7 < 0.001 Nurses 4.6 3.0 Medical students 2.3 0.6 Cultural issues Physicians 1.6 0.5 < 0.001 Nurses 3.2 1.9 Medical students 1.5 0.4 P-value is estimated based on 1-way analysis of variance test. SD = standard deviation. 1342 EMHJ – Vol. 26 No. 11 – 2020Research article “Fear of a malpractice lawsuit” and “An inefficient error reporting system”. The top issue for nurses and medical students was “Worry about being criticized by peers or supervisors” (Table 4). Discussion Medical errors remain a serious health concern world- wide and require special attention by health care admin- istrators and policy-makers (5,10). Health care providers are often reluctant to report patient safety problems, which could jeopardize proper medical error reporting (13). This study was designed to assess the main impedi- ments against error reporting among physicians, nurses and medical students in Nemazee Hospital. We found that almost 96% of respondents had observed an error at least once during the previous year. This agrees with an American study in which 94% of physicians and 88.7% of other health care workers witnessed one or more medical errors in their department (12). The majority (98%) of anaesthetists who participated in a similar study conducted in Switzerland disclosed being involved with a medical error (7). As in other studies, we found that nurses were more inclined to report medical errors than were physicians and students. Physicians perceived more barriers against proper reporting (21,22). One study indicated that only 42% of physicians routinely reported medical errors (23). Conversely, the findings of a Saudi Arabian study suggested physicians were more likely to report major medical errors than other health care workers (9). Our findings are comparable to surveys in that demographic characteristics, especially work experience and assigned wards among nurses, did not influence medical error reporting (13). The informal reporting style of physicians has been noted in other studies (9,14). Nurses favoured more formal reporting schemes for all types of errors. All 3 of our groups indicated their rate of reporting was the same if the error was theirs or that of a colleague. However, Alsafi et al. indicated that almost one-third of physicians would not report a peer to preserve their relationship (9). In our study, the ranking of barriers was different between the groups. Fear of legal complaints, reproach, punishment, lack of positive feedback, concern for endangering a colleague’s occupational status or revelation of his/her identity were often reported. In contrast, items such as “Errors are unavoidable in medical practice” and “There was nothing serious to report” received the lowest scores. However, a study from Australia reported conflicting results (24). Fear of legal consequences was among the top reasons for non-reporting by physicians, nurses and medical students. Legal repercussions have been recognized as one of the most pervasive obstacles to open communication (25,26). A study of 733 nurses in Urmia, Islamic Republic of Iran, indicated that reproach and penalties were the most cited obstacles to proper reporting (26). A survey among Table 3 Score distribution of organizational barriers against error reporting among physicians, nurses and medical students (n = 283) in Nemazee Hospital, Shiraz, Islamic Republic of Iran, 2015 Barrier Mean SD P- value Organizational (overall) Physicians 2.1 0.5 < 0.001 Nurses 4.4 2.2 Medical students 2.1 0.4 No confidentiality Physicians 2.0 0.7 < 0.001 Nurses 4.1 3.1 Medical students 2.0 0.7 Time consuming reporting system Physicians 2.0 0.8 < 0.001 Nurses 4.3 3.1 Medical students 1.9 0.8 Lack of proper feedback Physicians 2.2 0.8 < 0.001 Nurses 4.5 2.2 Medical students 2.2 0.7 Fear of malpractice lawsuit Physicians 2.2 0.8 < 0.001 Nurses 4.7 3.4 Medical students 2.2 0.7 P-value is estimated based on 1-way analysis of variance test. SD = standard deviation. 1343 EMHJ – Vol. 26 No. 11 – 2020Research article nurses from hospitals affiliated with Tehran and Shiraz universities of medical sciences listed fear of legal action, job threats and a negative culture of blame as major impediments (27). However, some studies, including one from the Netherlands, indicated that fear of legal complaints was not a significant concern for internists and residents reporting medical errors (8,28). Research carried out in the United States of America revealed that when anonymous reporting was established, nurses reported both minor and major medical errors at higher rates (29). In a Saudi Arabian study, 60% of physicians noted that reporting would be easier when there was a heightened level of confidentiality (9). Fear of being identified was not a major concern in our study, however, it was listed as one of the most discouraging factors concerning medical error reporting. Time consumption was not a top-ranked factor against proper medical error in our study nor in another from the Islamic Republic of Iran (26). More than 50% of Australian physicians and 40% of nurses felt their cumbersome reporting system was an important barrier (30). A Swiss cross-sectional survey involving anaesthetists reported not only sluggish reporting systems, but also inadequate feedback and lack of information negatively affected error reporting (8). Polisena et al. determined that more than half of the physicians and nurses surveyed believed their reporting system was incompetent and that they failed to receive plausible and timely feedback (30). Our study showed comparable results, which indicated that ineffectiveness of the error reporting system was among the top 3 causes of non-reporting by physicians, nurses and medical students. The World Health Organization published a guideline for medical students in 2009 which included 11 topics on patient safety (31). A number of studies indicate that most medical students receive little information about medical errors or reporting techniques (32,33), and staff physicians and nurses do not receive adequate information (8). Throckmorton et al. showed nurses could not identify errors in 40% of cases (29), although no practical scenario was provided for the accurate assessment of medical error recognition and understanding in this study, and there was evidence that 73.5% of nurses and almost half of physicians and medical students had an inadequate background on the topic. Significant limitations of this study included voluntary participation, self-reporting of information and possible recall bias. Some participants might not have recalled their errors very well. This could result in underestimation of the actual rate of reported medical errors. Also, correlation measurements between causes of non-reporting and job titles of nurses, including supervisory, was not possible due to the limited number of supervisors and head nurses among our participants. Possible future studies could involve greater numbers of participants and other types of health care workers. Applying our questionnaire in other hospitals could also be worthwhile. It would be valuable if hospitals of various sizes were involved. Conclusions Concerns about legal entanglements, reporting method- ology and confidentiality issues were recognized as the main barriers to reporting medical errors in Nemazee Hospital. Nurses did report errors at higher rates and they were found to prefer more formal, extensive, written reporting forms and the inclusion of all types of errors. Physicians and medical students preferred the opposite. Furthermore, the lack of knowledge on identifying and reporting medical errors among physicians and medi- cal students requires attention by hospital administers. It appears that a formal reintroduction of the reporting system with accompanying in-service training is needed. Table 4 Barriers against error reporting among physicians, nurses and medical students (n = 283) in Nemazee Hospital, Shiraz, Islamic Republic of Iran, 2015 Items Physicians Nurses Medical students P-value Mean (SD) Mean (SD) Mean (SD) Not familiar with reportable issues 1.77 (0.74) 2.99 (2.67) 1.85 (0.7) < 0.001 Not familiar with error reporting ways 1.96 (0.79) 2.69 (2.64) 2.15 (0.8) 0.023 Worry about my identity revelation 1.95 (0.84) 3.70 (3.15) 1.82 (0.7) < 0.001 Worry about revelation of colleague’s identity 2.50 (0.81) 4.47 (3.42) 2.18 (0.61) < 0.001 It is a time-consuming process 2.00 (0.78) 4.34 (3.14) 1.93 (0.77) < 0.001 Fear of malpractice lawsuit 2.18 (0.84) 4.69 (3.36) 2.16 (0.69) < 0.001 Fear of penalty or job loss 2.13 (0.80) 4.43 (3.16) 2.27 (0.71) < 0.001 Worry about being criticized by peers or supervisor 2.03 (0.78) 4.73 (3.32) 2.33 (0.67) < 0.001 There was nothing serious to report 1.42 (0.59) 3.58 (3.06) 1.44 (0.57) < 0.001 Error reporting is not efficient 2.16 (0.80) 4.52 (3.25) 2.22 (0.74) < 0.001 Come on! Who wants to do such things? 1.70 (0.73) 3.08 (2.74) 1.64 (0.7) < 0.001 Error reporting is not a usual concern 1.81 (0.74) 3.62 (2.77) 1.62 (0.68) < 0.001 Error is unavoidable in medical practice 1.40 (0.61) 2.86 (2.61) 1.22 (0.50) < 0.001 SD = standard deviation. 1344 EMHJ – Vol. 26 No. 11 – 2020Research article Acknowledgment This article is the result of a research project that was funded by the Vice-Chancellor for Research, Shiraz University of Medical Sciences, and carried out by Seyyed Mohammad Mahdi Sherafat in partial fulfilment of the requirements for certification as a general practitioner at Shiraz University of Medical Sciences. Funding: Shiraz University of Medical Sciences, Grant No. 7696. Competing interests: None declared. Prévalence de la non-notification des erreurs médicales en milieu hospitalier en République islamique d’Iran Résumé Contexte : Les erreurs médicales sont fréquentes dans les établissements de soins, ce qui a pour effet de mettre en péril la sécurité des patients et d’augmenter les coûts associés. Objectifs : La présente étude transversale avait pour objectif d’examiner les taux et les raisons de la non‐notification des erreurs médicales à l’hôpital de Nemazee, à Chiraz en République islamique d’Iran. Méthodes : Des questionnaires auto‐administrés ont été remplis par 283 membres du personnel, parmi lesquels des médecins, des membres du personnel infirmier et des étudiants en médecine. Une analyse de variance à sens unique, le test post hoc de la différence la moins significative de Fisher, le coefficient de corrélation de Spearman et les tests de corrélation intraclasse ont été utilisés pour les analyses statistiques. Résultats : La quasi-totalité des participants (95,8 %) avait observé au moins une erreur médicale au cours de l’année précédente, plus de la moitié (50,5 %) ayant relevé entre 3 et 10 erreurs. Pour signaler les erreurs médicales, les médecins et les étudiants en médecine privilégiaient la méthode verbale et informelle (40,3 % et 41,8 % respectivement), tandis que le personnel infirmier préférait les formulaires écrits (45,7 %). Les résultats indiquent des différences significatives entre les groupes concernant les obstacles individuels et organisationnels en général, et entre toutes les sous-catégories (p < 0,001). Conclusion : L’étude a montré que la complexité juridique et les problèmes de confidentialité constituaient les principaux obstacles à la notification des erreurs médicales. ةيملاسلإا ناريإ ةيروهجم في تايفشتسلما في ةيبطلا ءاطخلأا نع غلابلإا مدع راشتنا لدعم فيشنإ روفاي ،متاح ديهان ،كينيلاب زلراشت ،ناياش ةرهز ،سيدق ميرم ،تفاشر ديس ،نايركسأ دادرهم ةصلالخا .كلذب ةطبترلما فيلاكتلا نم ديزيو ،رطخلل ضىرلما ةملاس ض ِّرعُي امم ،ةيحصلا ةياعرلا قفارم في ةيبطلا ءاطخلأا عقت ام ًايرثك :ةيفللخا ناريإ ةيروهمجب زايرش ةنيدم في يزمان ىفشتسم في ةيبطلا ءاطخلأا نع غلابلإا مدع تلادعم ثحب لىإ يعطقلما ءاصقتسلاا اذه فده :فادهلأا .كلذ بابسأو ،ةيملاسلإا ليلحتلا مِدخُتساو .ةيبطلا تايلكلا في بلاطو نوضرمم وأ تاضرممو ءابطأ مهنيب نم ،ًايتاذ تانايبتسلاا ًافظوم 283 لمكتسا :ثحبلا قرط ضارغلأ ،ةئفلا لخاد طباترلا لماعم رابتخاو ،طباترلل ناميربس لماعم رابتخاو ،يونعم قرف لقلأ يدعَبلا شريف رابتخاو ،قورفلل هاتجلاا يداحأ .ةيئاصحلإا تلايلحتلا ثودح )%50.5( مهفصن نم رثكأ ظحلاو ،قباسلا ماعلا للاخ لقلأا لىع ًادحاو ًايبط ًأطخ )%95.8( ًابيرقت ينكراشلما عيجم ظحلا :جئاتنلا ةيمسرلا يرغو ةيوفشلا ةقيرطلا يه ةيبطلا ءاطخلأا نع غلابلإل ةيبطلا تايلكلا بلاطو ءابطلأا ىدل ةل َّضفُلما ةقيرطلا تناكو .ءاطخأ 10-3 ينب ةيربك تافلاتخا دوجو لىإ جئاتنلا تراشأو .)%45.7( ةبوتكلما َجذمانلا تاضرملماو نوضرملما ل َّضف ينح في ،)لياوتلا لىع %41.8و %40.3( .)P<0.001( ةيعرفلا تائفلا عيجم ينبو ،ةماع ٍةفصب ةيميظنتلاو ةيدرفلا زجاولحاب قلعتي مايف تاعومجلما .ةيبطلا ءاطخلأا نع غلابلإا ضترعت يتلا ةيسيئرلا تابقعلا يه ةيسرلا اياضقو ةينوناقلا تاكباشتلاب ةقلعتلما لغاوشلا نأ تُبث :تاجاتنتسلاا 1345 EMHJ – Vol. 26 No. 11 – 2020Research article References 1. Kohn LT, Corrigan JM, Donaldson MS. To err is human: building a safer health system. Washington, DC: National Academy Press; 1999. 2. Mahajan RP. Critical incident reporting and learning. Br J Anaesth. 2010 Jan;105(1):69–75. doi:10.1093/bja/aeq133. PMID:20551028 3. Garrouste-Orgeas M, Philippart F, Bruel C, Max A, Lau N, Misset B. Overview of medical errors and adverse events. Ann Inten- sive Care. 2012 Jan;2(1):2. doi:10.1186/2110-5820-2-2. PMID:22339769 4. Deaths: final data for 2013. National Vital Statistics Report. Atlanta: Centers for Disease Control and Prevention; (http:///www. cdc.gov/nchs/fastats/ leading-causes-of-death.htm, accessed 27 May 2018). 5. Makary MA, Daniel M. Medical error – the third leading cause of death in the US. BMJ. 2016 May;2139:1–5. doi:10.1136/bmj.i2139. PMID:27143499 6. James JT. A New, Evidence-based estimate of patient harms associated with hospital care. J Patient Safety. 2013 Sep;9(3):122–8. doi:10.1097/PTS.0b013e3182948a69. PMID:23860193 7. Wilson RM, Michel P, Olsen S, et al. Patient safety in developing countries: retrospective estimation of scale and nature of harm to patients in hospital. BMJ. 2012 Mar;344:e832. doi:10.1136/bmj.e832. PMID:22416061 8. McLennan SR, Engel-Glatter S, Meyer AH, Schwappach DLB, Scheidegger DH, Elger BS. Disclosing and reporting medical errors. Eur J Anaesthesiol. 2015 July;32(7):471–6. doi:10.1097/EJA.0000000000000236. PMID:26389547 9. Alsafi E, Bahroon SA, Tamim H, Al-Jahdali HH, Alzahrani S, Al Sayyari A. Physicians’ attitudes toward reporting medical errors-an observational study at a general hospital in Saudi Arabia. J Patient Safety. 2011 Mar;7(3):143–6. doi:10.1097/PTS.0b013e31822c5a82. PMID:21857239 10. Poorolajal J, Rezaie S, Aghighi N. Barriers to medical error reporting. Int J Prev Med. 2015 Oct 7;6:97. doi:10.4103/2008- 7802.166680. PMID:26605018 11. Evans SM, Berry JG, Smith BJ, Esterman A, Selim P, O’Shaughnessy J, DeWit M. Attitudes and barriers to incident reporting: a collaborative hospital study. Qual Saf Health Care. 2006 Feb;15(1):39–43. doi:10.1136/qshc.2004.012559. PMID:16456208 12. Smith KS, Harris KM, Potters L Sharma R, Mutic S, Gay HA, Wright J, Samuels M, Ye X, Ford E, Terezakis S. Physician at- titudes and practices related to voluntary error and near-miss reporting. J Oncol Pract. 2014 Sept;10(5):e350–7. doi:10.1200/ JOP.2013.001353. PMID:25095825 13. Castel ES, Ginsburg LR, Zaheer S, Tamim H. Understanding nurses’ and physicians’ fear of repercussions for reporting errors: clinician characteristics, organization demographics, or leadership factors? BMC Health Serv Res. 2015 Aug;15:326. doi:10.1186/ s12913-015-0987-9.PMID:26272228 14. Perez B, Knych S a, Weaver SJ, Liberman A, Abel EM, Oetjen D, Wan TT. Understanding the barriers to physician error reporting and disclosure: a systemic approach to a systemic problem. J Patient Saf. 2014 Mar;10(1):45–51. doi:10.1097/PTS. 0b013e31829e4b68. PMID:24553443 15. Kaldjian LC, Jones EW, Rosenthal GE. Facilitating and impeding factors for physicians’ error disclosure: A structured literature review. Jt Comm J Qual Patient Saf. 2006 Apr;32(4):188–98. PMID:16649649 16. Saravi B, Mardanshahi A, Ranjbar M, Siamian H, Azar MS, Asghari Z, Motamed N. Rate of Medical Errors in Affiliated Hos- pitals of Mazandaran University of Medical Sciences. Mater Socio Medica. 2015 Feb;27(1):31. doi:10.5455/msm.2014.27.31–4. PMID:25870528 17. Mirzaei M, Safari Faramani R, Sepahvand E. [Prevalence, types of medication errors and barriers to reporting errors by nursing in an educational hospital in Kermanshah]. Hayat. 2013 Mar;19(3):28–37 (in Farsi). 18. Cheraghi MA, Nikbakhat Nasabadi AR, Mohammad Nejad E, Salari A, Ehsani Kouhi Kheyli SR. Medication errors among nurses in intensive care units (ICU). J Mazand Univ Med Sci. 2012 Jan;22(Suppl. 1):115–9. 19. MohammadNejad E, Ehsani SR, Salari A. Sajjadi A, Hajiesmaeelpour A. Refusal in reporting medication errors from the perspec- tive of nurses in emergency ward. J Res Dev Nurs Midwifery. 2013 Jan;10(1):61–8. 20. Pfeiffer Y, Briner M, Wehner T, Manser T. Motivational antecedents of incident reporting: Evidence from a survey of nurses and physicians. Swiss Med Wkly. 2013 Nov;143:1–11. doi:10.4414/smw.2013.13881 PMID:24222585 21. Kingston MJ, Evans SM, Smith BJ, Berry JG. Attitudes of doctors and nurses towards incident reporting: a qualitative analysis. Med J Aust. 2004 Jul;181(1):36–9. PMID:15233611 22. Zaheer S, Ginsburg L, Chuang Y-T, Grace SL. Patient safety climate (PSC) perceptions of frontline staff in acute care hospitals: Examining the role of ease of reporting, unit norms of openness, and participative leadership. Health Care Manage Rev. 2013 Jan–Mar;40(1):13–23. doi:10.1097/HMR.0000000000000005. PMID:24378403 23. Waring JJ. Beyond blame: cultural barriers to medical incident reporting. Soc Sci Med. 2005 May;60(9):1927–35. PMID:15743644 24. Iedema R, Allen S, Sorensen R, Gallagher TH. What prevents incident disclosure, and what can be done to promote it? Jt Comm J Qual Patient Saf. 2011 Sep;37:409–17. PMID:21995257 25. Heard GC, Sanderson PM, Thomas RD. Barriers to adverse event and error reporting in anesthesia. Anesth Anal. 2012 Mar;114:604–14. doi:10.1213/ANE.0b013e31822649e8. PMID:21821515 1346 EMHJ – Vol. 26 No. 11 – 2020Research article 26. Bayazidi S, Zarezadeh Y, Zamanzadeh V, Parvan K. Medication error reporting rate and its barriers and facilitators among nurs- es. J Caring Sci. 2012 Nov 27;1(4):231–6. doi:10.5681/jcs.2012.032. PMID:25276700 27. Hashemi F, Nasrabadi AN, Asghari F. Factors associated with reporting nursing errors in Iran: a qualitative study. BMC Nurs. 2012 Oct;11(1):20. doi:10.1186/1472-6955-11-20. PMID:23078517 28. Linthorst GE, Kallimanis-King BL, Douwes Dekker I, Hoekstra JB, de Haes JC. What contributes to internists’ willingness to disclose medical errors? Neth J Med. 2012 Jun;70(5):242–8. PMID:22744929 29. Throckmorton T, Etchegaray J. Factors affecting incident reporting by registered nurses: the relationship of perceptions of the environment for reporting errors, knowledge of the nursing practice act, and demographics on intent to report errors. J Peri Anesth Nurs. 2007 Dec;22(6):400–12. PMID:18039512 30. Polisena J, Gagliardi A, Urbach D, Clifford T, Fiander M. Factors that influence the recognition, reporting, and resolution of incidents related to medical devices and other healthcare technologies: a systematic review. Syst Rev. 2015 Mar;4:37. doi:10.1186/ s13643-015-0028-0. PMID:25875375 31. WHO patient safety curriculum guide for medical schools. Geneva: World Health Organization; 2009] (http://www.who.int/ patientsafety/activities/technical/medical_curriculum/, accessed 18 September 2019) 32. Flin R, Patey R, Jackson J, Mearns K, Dissanayaka U. Year 1 medical undergraduates? Knowledge of and attitudes to medical error. Med Educ. 2009 Dec;43(12):1147–55. doi:10.1111/j.1365-2923.2009.03499.x 33. Li L, Duan Y, Chen P, Li J, Mao X, Barraclough BH, Zhang M. Knowledge, skills, and attitudes of medical students to patient safety: A cross-sectional pilot investigation in China. J Evid Based Med. 2012 Aug;5(3):124–33. doi:10.1111/j.1756-5391.2012.01187.x. PMID:23672219. 1347 EMHJ – Vol. 26 No. 11 – 2020Research article Assessment of nurses’ patient safety culture in 30 primary health-care centres in Tunisia Mohamed Ayoub Tlili,1,2 Wiem Aouicha,1 Mohamed Ben Dhiab3 and Manel Mallouli4 1Laboratoire de Recherche LR12ES03, Faculté de Médecine de Sousse, Université de Sousse, Sousse, Tunisia (Correspondence to: Mohamed Ayoub Tlili: medtlili@hotmail.fr). 2Ecole Supérieure des Sciences et Techniques de la Santé de Sousse, Université de Sousse, Sousse, Tunisia. 3Vice-Dean, Faculté de Médecine de Sousse, Université de Sousse, Sousse, Tunisia. 4Département de Médecine Familiale et Communautaire, Laboratoire de Recherche LR12ES03, Faculté de Médecine de Sousse, Université de Sousse, Sousse, Tunisia Abstract Background: Ensuring patient safety and health-care quality remain priorities and challenges worldwide and the role of nurses is essential to meet these challenges. Developing patient safety culture is a key component to improve patient safety and health-care quality. Aims: To assess nurses’ patient safety culture in primary health-care centres in Tunisia and to determine its associated factors. Methods: This was a multicentre, cross-sectional descriptive study conducted across 30 primary health-care centres in Tunisia, using the French validated version of the Hospital Survey on Patient Safety Culture questionnaire. All the nurses working in these centres were invited to participate in the study (n = 158). Results: The response rate for participation in the study was 87.3%. The dimension of “teamwork within units” had the highest score (70.6%). Three safety dimensions had low scores: “frequency of event reporting” (27.6%), “staffing” (34.76%) and “nonpunitive response to errors” (36.5%). Two factors were associated with patient safety culture: participation in risk management committees, and district of the primary care centre. Conclusions: The level of nurses’ patient safety culture needs to be improved in primary health-care centres in Tunisia. Strategies to nurture patient safety culture should focus upon building leadership capacity that supports open communi- cation, blame-free environment, teamwork and continuous organizational learning. Keywords: Patient safety culture, nurses, patient safety, public health, primary care Citation: Tlili MA; Aouicha W; Ben Dhiab M; Mallouli M. Assessment of nurses’ patient safety culture in 30 primary health-care centres in Tunisia. East Mediterr Health J. 2020;26(11):1347-1354. https://doi.org/10.26719/emhj.20.026 Received: 12/06/19; accepted: 19/11/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Introduction Adverse events (AEs) remain a global challenge, and millions of people are prone to death or injury due to preventable medical errors (1) and several studies have shown the severity of these AEs in terms of cost, frequen- cy and serious consequences (2). As a result, enhancing patient safety has become a priority for healthcare sys- tems and providers (2). The situation is more difficult and serious in developing countries with a higher risk of patient harm due to the limitation of resources and lack of adequate infrastructure. In Tunisia, a study in the town of Sousse showed that the rate of AEs varied between 10% and 11.3% (3). The rates of AEs vary according to the sectors, the nature of the services and levels of care, depending on the specific countries (4). Health care in Tunisia is divided into public and private structures. The public structures are organized on 3 levels of care. The front-line (first-line) structures include primary healthcare centres serving the community with primary and essential care. The second- level structures (regional hospitals) refer to a second tier of the healthcare system, in which patients from primary health care are referred to specialists in higher-level hospitals for treatment and specialized diagnosis. The third level of care provides specialized consultative care, usually on referral from primary and secondary care, along with teaching and research functions. Primary healthcare centres provide the first contact for patients and quality and patient safety in these facilities are vital (1,5). However, since severe and complicated cases requiring special treatment are handled in hospitals, both healthcare providers and local communities tend to underestimate the importance of primary healthcare services (1). In fact, it has been identified that a significant proportion of safety incidents arising in hospitals originate at the earlier levels of care (5), and some errors in primary care can result in severe consequences (1,6,7). A study in Spain deemed that 64.3% of AEs in primary care were preventable (8). To prevent such harm, there is a growing recognition of the importance of establishing a patient safety culture (PSC) (2,9–12). PSC is defined as the product of individual and group values, attitudes, perceptions, competencies and patterns of behaviour that determine commitment to the style and proficiency of an organization’s safety management (13). Furthermore, it should be noted that development of PSC starts with evaluation of its 1348 EMHJ – Vol. 26 No. 11 – 2020Research article existing level (1,14). In particular, nurses as healthcare providers believe that patient safety is primarily a nursing responsibility (14,15), and it has been estimated that > 90% of potential medication errors are discovered by nurses (14). Thus, nurses are considered key to safety improvement and play a vital role in enhancing quality of care (12,14,16). The nature of work carried out by nurses and the roles they perform provide them with opportunities to reduce AEs and intercept healthcare errors before they occur (12). Thus, assessment of nurses’ PSC is crucial to identify the strengths and weaknesses of their safety culture and to help units’ caregivers identify the patient safety problems that they have. It has been shown that the level of PSC in Tunisia is low (10,17,18). This can be explained by the lack of professional involvement in training sessions on patient safety and the late introduction of the concept of PSC in Tunisia. For example, a study in Tunisian operating rooms showed that all dimensions of PSC needed improvement (10). Another study showed that no dimension was considered as developed (17). To our knowledge, no studies have assessed nurses’ PSC in Tunisia, specifically in crucial frontline healthcare facilities that deliver essential care. Therefore, we assessed nurses’ PSC in Tunisian primary healthcare centres and determined its associated factors. Methods Study design, setting, duration and participants This was a cross-sectional multicentre study from Janu- ary to April 2016 in all 30 primary healthcare centres in Sousse, Kasserine and Kairouan, Tunisia. These centres are partners of the Faculty of Medicine of University of Sousse and carry out consultations 4 days a week. Each centre has 1 or 2 doctors and an average of 4 nurses. All 158 nurses involved in the selected primary healthcare centres were invited to participate in the study and 138 provided survey feedback. Nurses were divided into registered nurses and specialized nurses with specific additional training for particular specialties (emergency, paediatric and geriatric care). Nurses who were not involved in healthcare practices and those with < 1 month’s experience were excluded. This exclusion criterion was recommended by the questionnaire’s user guide provided by the Coordination Committee of the Clinical Evaluation and Quality in Aquitaine (CCECQA), which was responsible for validation of the French version of the questionnaire (19). Questionnaire The current study used the French version of the Hospi- tal Survey on Patients Safety Culture (HSOPSC) question- naire, which was translated and validated by the CCCEQA (20). It is the most broadly used instrument to evaluate PSC because of its favourable psychometric properties; it is a valid and reliable instrument that allows the study concept (PSC) to be measured appropriately (21). The Cronbach a was 0.88 for the whole questionnaire and varied between 0.46 and 0.84 for the individual dimen- sions (20). Ten PSC dimensions were explored by the French version through 45 items. The questionnaire user’s guide defined and described the 10 dimensions related to PSC as follows (19). (D1) Overall perceptions of patient safety: procedures and systems are good at preventing errors and there is a lack of patient safety problems. (D2) Frequency of events reported: mistakes of the following types are reported: (1) mistakes caught and corrected before affecting the patient; (2) mistakes with no potential to harm the patient; and (3) mistakes that could harm the patient but do not. (D3) Supervisor/ manager expectations: supervisors/managers consider staff suggestions for actions promoting and improving patient safety; praising staff for following patient safety procedures; and do not overlook patient safety problems. (D4) Organizational learning – continuous improvement: mistakes have led to positive changes and changes are evaluated for effectiveness. (D5) Teamwork within units: staff support each other, treat each other with respect, and work together as a team. (D6) Communication openness: staff freely speak up if they see something that may negatively affect a patient and feel free to question those with more authority. (D7) Nonpunitive response to errors: staff feel that their mistakes and event reports are not held against them and that mistakes are not kept in their personnel file. (D8) Staffing: there are enough staff to handle the workload, which is appropriate to provide the best care for patients. (D9) Management support for patient safety: hospital management provides a work climate that promotes patient safety and shows that patient safety is a top priority. (D10) Teamwork across units: hospital units cooperate and coordinate with one another to provide the best care for patients. The questionnaire assesses 10 dimensions deemed related to PSC in a way that if the professionals have a dysfunction in one or more dimensions (score < 50%) it reflects a failing PSC. For example, if professionals work in a punitive environment (D7) or have dysfunctional teamwork (D5), it means PSC is failing. To have a well- developed PSC, the 10 dimensions must be developed (score > 75%) (19). The survey also explored nurses’ perception of patient safety quality (1 item), and the number of AEs reported during the last 12 months (1 item), which referred to the number of events that the nurses responding to the survey reported. The questionnaire also included a section on general information on the participants, which was used to determine the factors associated with PSCeus, namely: professional title (specialty), sex, age, work experience, participation in risk management committees, and district of the primary healthcare centre. 1349 EMHJ – Vol. 26 No. 11 – 2020Research article A Likert scale of 5 points was used to explore participants’ PSC perception ranging from ‘strongly disagree’ to ‘strongly agree’ or from ‘never’ to ‘always’ depending on the nature of the item. Data collection and ethical considerations After obtaining institutional ethics committee approval and administrative authorization from different centres’ management, a self-reported paper-based questionnaire was distributed to the participants. The investigator went to the centre and distributed the questionnaire after ex- plaining the aims and outcomes of the study to all the nurses meeting the inclusion criteria and who agreed to respond. They could freely and anonymously fill in the questionnaire and return their responses directly to the investigator. The investigators did not work in the centres and only went to distribute the questionnaires. The data entry and analysis were confided to another re- searcher. Data analysis Data analysis was performed using SPSS version 20 and Epi info 6.04d for Windows. Descriptive statistical analysis such as frequencies and percentages of positive responses for each item and dimension were used to ex- amine professionals’ perceptions about PSC. Items were worded in both positive and negative terms. For items with a positive formulation, answers “strongly agree/ agree” or “most of the time/always” were considered pos- itive. For items with a negative formulation, the answers “strongly disagree/disagree” or “never/rarely” responses were considered positive for PSC. Items with negative formulation were identified according to the questionnaire’s user guide (19) and were coded conversely. According to the user guide, if none of the dimensions’ sections was entirely filled, the questionnaire would not be taken into account (19). Also, if fewer than half of the items in the questionnaire were completed, or the same answers were given to all the items, the questionnaire was considered ineligible and excluded (19). A bivariate analysis was carried out to highlight the associations between the sociodemographic and professional data and the different dimensions of PSC. Percentages were compared by Pearson’s χ2 test. The materiality threshold was set at 0.05. Results Participant characteristics A total of 138 participants provided survey feedback and the response rate was 87.3%; 92 (67.7%) were reg- istered nurses and 46 (33.3%) were specialized nurses (Table 1). Most respondents (n = 102; 73.9%) were female, and the male:female ratio was 0.35. One hundred and three (74.6%) nurses had work experience of > 10 years. Nurses’ perception of patient safety quality and frequency of reported AEs Nurses’ perception of patient safety quality in the prima- ry healthcare centres was ranked as acceptable in 57.2% of cases and poor in 17.4% (Table 2). Ninety-eight (71%) of the participants declared that they did not report any AE in the last 12 months. PSC dimensions Overall perception of patient safety had an average pos- itive score of 53.65% (Table 3). The percentage of positive Table 1 Characteristics of participants Characteristics n % Professional title/specialty Specialized nurses 46 33.3 Registered nurses 92 67.7 Total 138 100 Sex Female 102 73.9 Male 36 26.1 Total 138 100 Age > 40 years 90 65.2 ≤ 40 years 48 34.8 Total 138 100 Work experience < 10 years 35 25.4 ≥ 10 years 103 74.6 Total 138 100 Participation in risk management committees Yes 24 17.4 No 114 82.6 Total 138 100 Location of primary healthcare centre Urban 102 73.9 Rural 36 26.1 Table 2 Nurses perception of patient safety quality and number of reported adverse events during the last 12 months Nurses perception of patient safety quality n % Excellent 9 6.5 Very good 25 18.2 Acceptable 79 57.2 Poor 24 17.4 Failing 1 0.7 No. of events reported n % None 98 71 1 or 2 20 14.5 3–5 8 5.8 6–20 > 20 5 7 3.6 5.1 1350 EMHJ – Vol. 26 No. 11 – 2020Research article Table 3 Scores and items of the 10 dimensions of patient safety culture Items of patient safety culture dimensions Absolute frequency (n) Average positive response (%) D1: Overall perceptions of safety 53.65 Patient safety is never sacrificed to get more work done 90 65 Our procedures and systems are good at preventing errors from happening 83 59.8 It is just by chance that more serious mistakes do not happen around here 67 48.9 We have patient safety problems in this facility 56 40.9 D2: Frequency of events reported 27.7 When a mistake is made, but is caught and corrected before affecting the patient, it is reported 40 29.2 When a mistake is made, but has no potential to harm the patient, it is reported 33 24 When a mistake is made that could harm the patient, but does not, it is reported 41 29.9 D3: Supervisor/manager expectations and actions promoting patient safety 53.47 Manager says a good word when he/she sees a job done according to established patient safety procedures 85 61.3 Manager seriously considers staff suggestions for improving patient safety 74 53.3 Whenever pressure builds up, my manager wants us to work faster, even if it means taking shortcuts 71 51.1 My manager overlooks patient safety problems that happen over and over 67 48.2 D4: Organizational learning and continuous improvement 48.66 We are actively doing things to improve patient safety 94 67.9 Mistakes have led to positive changes here 77 55.5 After we make changes to improve patient safety, we evaluate their effectiveness 106 76.6 We are given feedback about changes put into place based on event reports 14 10.2 We are informed about errors that happen in the facility 47 34.3 In this facility, we discuss ways to prevent errors from happening again 66 47.5 D5: Teamwork within units 70.6 People support one another in this facility 93 67.1 When a lot of work needs to be done quickly, we work together as a team to get the work done 109 78.8 In facility, people treat each other with respect 95 68.6 When one area in this unit gets really busy, others help out 94 67.9 D6: Communication openness 42.13 Staff will freely speak up if they see something that may negatively affect patient care 70 50.4 Staff feel free to question the decisions or actions of those with more authority 39 28.5 Staff are afraid to ask questions when something does not seem right 66 47.5 D7: Nonpunitive response to error 36.5 Staff feel like their mistakes are held against them 49 35.8 When an event is reported, it feels like the person is being written up, not the problem 56 40.9 Staff worry that mistakes they make are kept in their personnel file 45 32.8 D8: Staffing 34.76 We have enough staff to handle the workload 67 48.9 Staff in this facility work longer hours than is best for patient care 22 16 We work in crisis mode trying to do too much, too quickly 54 39.4 D9: Management support for patient safety 51.07 Management provides a work climate that promotes patient safety 60 43.8 The actions of management show that patient safety is a top priority 79 56.9 Management seems interested in patient safety only after an adverse event happens 58 42.3 Units work well together to provide the best care for patients 85 61.3 1351 EMHJ – Vol. 26 No. 11 – 2020Research article responses was highest for teamwork within units (70.6%). The lowest scores were for frequency of event reporting (27.7%), staffing (34.76%) and nonpunitive response to er- ror (36.5%). Factors associated with PSC The dimensions of PSC were not significantly associat- ed with sex, professional title, or work experience. Two factors were associated with PSC dimensions: frequency of AEs reported was significantly higher among partic- ipants involved in risk management committees (P = 0.02); and overall perception of safety was significantly higher among nurses working in urban compared with rural districts (P = 0.03). Discussion To our knowledge, there have been no studies of PSC among nurses working in primary healthcare in Tuni- sia. Therefore, the present study was conducted to as- sess nurses’ PSC in Tunisian primary healthcare centres. The dimension of teamwork within units had the highest score (70.6%). Three dimensions had low scores, namely, frequency of event reporting (27.6%), staffing (34.76%) and nonpunitive response to errors (36.5%). Two factors were associated with PSC: participation in risk manage- ment committees, and district of the primary care centre. Recently, patient safety in primary care has been given increasing attention (22), and many studies have shown a high level of AEs with negative consequences (1,6,7,23). Given the importance of assessing PSC to enhance patient safety in primary care, several studies have sought to determine professionals’ PSC in this setting (1,5,9,24–26). Many studies have focused on nurses, in the belief that understanding nurses’ perceptions are crucial for policy- makers to address PSC in relation to nurses’ staffing policies (12,14,15). The dimension of overall perception of safety had a score of 53.65%. This reflects the lack of safety standards in the primary healthcare centres and the need to implement corrective measures to increase awareness of this issue among professionals. Indeed, 59.1% of nurses confirmed that they had problems with security in their workplace. We found that the dimension of teamwork within units had the highest score (70.6%) and this was similar to previous studies (1,25,27). However, when it came to critical care areas such as operating rooms (10) and intensive care units (28), this dimension had a low score (41.7% and 46.99%, respectively). This may be due to the fact that primary healthcare centres are small buildings with fewer staff compared to hospitals and critical care units and are unsophisticated environments that encourage teamwork (29). Staffing had a positive score of 34.76%, and most nurses reported that they did not have enough staff to handle the workload, and that they worked longer hours than are best for patient care. This situation may have severe negative consequences for patient safety and quality of care. O’Brien-Pallas et al. investigated the relationship between nurse staffing, workload and patient outcomes. They found that nurse staffing (fewer registered nurses), increased workload, and an unstable nursing environment was linked to negative patient outcomes, including falls and medication errors (30). They also reported that when nursing demand/supply levels exceeded 80%, negative outcomes increased for nurses themselves and hospitals, as well as patients. The dimension that had the lowest score was frequency of events reported (27.7%). This under-reporting can be explained by the fact that the commission of error is always considered to indicate lack of skill and rarely seen as a learning opportunity. Several barriers exist to reporting AEs, including insufficient time to report, lack of feedback, fear of blame and damage to reputation in a competitive environment, and loss of patient confidence (11,31). This dimension was similar to nonpunitive response to error, which also had a low score (36.5%). Nurses reported that they felt that their mistakes were held against them and their involvement in the AE was being highlighted rather than the AE itself. This problem of under-reporting AEs must be taken into consideration and treated with vigilance; nurses should be encouraged to report AEs and even rewarded for so doing. It is essential to establish a culture in which individuals are supported to identify and report AEs without threat of punitive action or blame. Reporting of AEs is an integral part of a continuous cycle of improving patient safety and quality of care that includes error identification, reporting, analysis and corrective actions (32). Items of patient safety culture dimensions Absolute frequency (n) Average positive response (%) D10: Teamwork across units 45.95 There is good cooperation among units that need to work together 75 54.7 Units do not coordinate well with each other 55 40.1 It is often unpleasant to work with staff from other units 55 40.1 Things “fall between the cracks” when transferring patients from one unit to another 53 38.7 Important patient care information is often lost during shift changes 83 59.8 Problems often occur in the exchange of information across units 58 42.3 Table 3 Scores and items of the 10 dimensions of patient safety culture (Concluded) 1352 EMHJ – Vol. 26 No. 11 – 2020Research article In our study, participants who were engaged in risk management committees had a significantly higher score for the dimension frequency of events reported (37.2% vs 17.7%; P = 0.02). This finding agrees with results from the PSC survey that was conducted in operating rooms in Tunisia (10). Risk management describes a dynamic process that includes all measures for systematic identification, analysis, assessment, surveillance and control of risks. An effective risk management should not start only after the evaluation of an incident but when failure can still be avoided and damage can be prevented. Overall perception of safety was significantly more developed among nurses working in urban than in rural areas (60.1% vs 40.2%; P = 0.03). This difference can be explained by the fact that, in Tunisia, urban healthcare institutions are better equipped with more sophisticated equipment and better human and material resources. We recommend systematic improvement of staff qualifications by providing training opportunities and educational interventions to promote better understanding of the principles of teamwork, help staff acknowledge each other’s roles and perspectives, and develop effective communication strategies. At the level of practice, policy-making, administration, research and curriculum, we recommend improved training of nurses regarding patient safety. To this end, the World Health Organization has published 2 guides: the first is for students, to be integrated into health universities (Patient Safety Curriculum Guide for Medical Schools) (33); and the second is designed for health professionals as part of continuing education (Patient Safety Curriculum Guide Multi-professional Edition) (34). Improvement of quality of care and patient safety by implementing a quality management system is essential, with managerial training concerning communication and nursing leadership. Nurses need to feel protected and encouraged to report errors and AEs, and introduction of an anonymous reporting system that protects the reporter is therefore recommended. Our study had some limitations. First, assessment of PSC using a self-administered questionnaire could have been associated with declaration bias. A self-administered questionnaire may influence the responses of those who, for fear of reprisal or prosecution, give desirable answers that do not reflect reality. Second, HSOPSC did not allow us to calculate an overall score for PSC for all the targeted centres, which would have allowed rapid comparison of quality of care and safety culture between healthcare organizations. By following the questionnaire guidelines, we were only able to calculate a score for each dimension without calculating a mean score for all the dimensions combined. Third, there was possible recall bias, specifically when remembering the number of AEs reported, resulting in possible under- or overestimation of reported results. Finally, even though we included all the training centres of the targeted region, the sampling technique did not allow us to assume that these included settings were representative of the entire primary healthcare system in Tunisia. Conclusions Our findings demonstrated that none of the PSC dimen- sions were developed in our primary healthcare centres. We highlighted different areas of concern, such as fre- quency of events reported, nonpunitive response to er- ror, and staffing. More attention should be paid to PSC in primary healthcare because changing values and atti- tudes needs time and motivation through training and improving risk management skills among nurses. Also, the results highlight the necessity of implementation of quality management systems in Tunisian primary healthcare centres. Strategies to nurture PSC should fo- cus upon building leadership capacity that supports open communication, blame-free environment, teamwork and continuous organizational learning. Acknowledgement We thank the professionals in all the centres where the study was conducted for their cooperation and the supportive working conditions that they offered. Funding: None. Competing interests: None declared. Évaluation de la culture de la sécurité des patients chez le personnel infirmier dans 30 centres de soins de santé primaires en Tunisie Résumé Contexte : Garantir la sécurité des patients et la qualité des soins de santé demeure une priorité et un défi dans le monde entier face auxquels le personnel infirmier joue un rôle essentiel. Le développement de la culture de la sécurité des patients constitue un élément clé pour améliorer la sécurité de ces derniers ainsi que la qualité des soins de santé. Objectifs : Évaluer la culture de la sécurité des patients chez le personnel infirmier dans les centres de soins de santé primaires tunisiens et déterminer les facteurs qui y sont associés. 1353 EMHJ – Vol. 26 No. 11 – 2020Research article References 1. Bodur S, Filiz E. A survey on patient safety culture in primary healthcare services in Turkey. Int J Qual Health Care. 2009 Oct;21(5):348–55. https://doi.org/10.1093/intqhc/mzp035. 2. Occelli P, Quenon J-L, Hubert B, Hoarau H, Pouchadon M-L, Amalbert R, et al. La culture de sécurité en santé: un concept en pleine émergence. Risques Qualite. 2007;5:207–12 (in French). 3. Bouafia N, Bougmiza I, Bahri F, Letaief M, Astagneau P, Njah M. Ampleur et impact des évènements indésirables graves liés aux soins: Étude d’incidence dans un hôpital du Centre-Est tunisien. Pan Afr Med J. 2014;16 (in French). https://doi.org/10.11604/ pamj.2013.16.68.1161. 4. Zegers M, De Bruijne MC, Spreeuwenberg P, Wagner C, Van Der Wal G, Groenewegen PP. Variation in the rates of adverse events between hospitals and hospital departments. Int J Qual Health Care. 2011 Apr;23(2):126–33. https://doi.org/10.1093/intqhc/ mzq086. 5. Webair HH, Al-assani SS, Al-haddad RH, Al-Shaeeb WH, Bin Selm MA, Alyamani AS. Assessment of patient safety culture in primary care setting, Al-Mukala, Yemen. BMC Fam Pract. 2015;16:Article number 136. https://doi.org/10.1186/s12875-015-0355-1. 6. Dovey SM, Meyers DS, Phillips RL, Green LA, Fryer GE, Galliher JM, et al. A preliminary taxonomy of medical errors in family practice. Qual Saf Health Care. 2002 Sep;11(3):233–8. http://dx.doi.org/10.1136/qhc.11.3.233 PMID:12486987 7. Rosser W, Dovey S, Bordman R, White D, Crighton E, Drummond N. Medical errors in primary care: results of an international study of family practice. Can Fam Physician. 2005 Mar;51:386–7. PMID:16926931 Méthodes : Il s’agissait d’une étude descriptive transversale multicentrique menée dans 30 centres de soins de santé primaire tunisiens à l’aide de la version française validée du questionnaire de l’enquête hospitalière sur la culture de la sécurité des patients. L’ ensemble du personnel infirmier travaillant dans ces centres a été invité à participer à l’étude (n = 158). Résultats : Le taux de réponse concernant la participation à l’étude a été de 87,3 %. La dimension « travail d’équipe au sein des unités » a obtenu le score le plus élevé (70,6 %). Trois dimensions de la sécurité présentaient des scores faibles : « fréquence de notification des événements » (27,6 %), « dotation en personnel » (34,76 %) et « réponse non punitive à l’erreur » (36,5 %). Deux facteurs étaient associés à la culture de la sécurité des patients : la participation aux comités de gestion des risques et le district du centre de soins primaires. Conclusions : Le niveau de culture de la sécurité des patients chez le personnel infirmier doit être amélioré dans les centres de soins de santé primaires en Tunisie. Les stratégies visant à développer la culture de la sécurité des patients doivent porter sur le renforcement des capacités d’encadrement qui favorisent une communication ouverte, un environnement professionnel bienveillant, le travail d’équipe et un apprentissage organisationnel continu. سنوت في ةيل َّولأا ةيحصلا ةياعرلا زكارم نم ًازكرم 30 في ضيرمتلا مقاوط ىدل ضىرلما ةملاس ةفاقث مييقت ليولم لانم ،بايذ نب دممح ،ةشيوع مائو ،لييلت بويأ دممح ةصلالخا نع ىنغ لاو ،لماعلا ءاحنأ عيجم في ةحصلا عاطق هجاوت يتلا تايدحتلاو تايولولأا نم ةيحصلا ةياعرلا ةدوجو ضىرلما ةملاس نماض لازي لا :ةيفللخا .ةيحصلا ةياعرلا ةدوجو ضىرلما ةملاس ينسحتل ًايساسأ ًاصرنع ضىرلما ةملاس ةفاقث شرن دَعُيو .تايدحتلا هذه لىع بلغتلل ضيرمتلا مقاوط رْوَد لماوعلا ديدتحو ،سنوت في ةيل َّولأا ةيحصلا ةياعرلا زكارم في ضيرمتلا مقاوط ىدل ضىرلما ةملاس ةفاقث مييقت لىإ ةساردلا هذه تفده :فادهلأا .ابه ةطبترلما مادختساب ،سنوت في ةيل َّولأا ةيحصلا ةياعرلا زكارم نم ًازكرم 30 لىع تيرجُأ زكارلما ةددعتم ةيعطقم ةيفصو ةسارد هذه تناك :ثحبلا قرط زكارلما هذه في ينلماعلا ضيرمتلا مقاط دارفأ عيجم يعُدو .ضىرلما ةملاس ةفاقث نايبتسا للاخ نم تايفشتسلما حْسَم نم ةزاجُلما ةيسنرفلا ةخسنلا .)158 = ددعلا( ةساردلا في ةكراشملل تلصحو .)%70.6( تاجردلا لىعأ »تادحولا لخاد يعمالجا لمعلا« دعُب ق َّقحو .%87.3 ةساردلا في ةكراشملل ةباجتسلاا لدعم غلب :جئاتنلا يرغ ةباجتسلاا»و ،)%34.76( »فيظوتلا»و ،)%27.6( »ثادحلأا نع غلابلإا رتاوت« :يهو ،ةضفخنم تاجرد لىع ةملاسلاب ةقلعتم داعبأ ةثلاث .ةيلولأا ةياعرلا زكرم اهيف عقي يتلا ةقطنلماو ،رطاخلما ةرادإ نالج في ةكراشلما :ضىرلما ةملاس ةفاقثب نلاماع طبتراو .)%36.5( »ءاطخلأل ةيباقعلا ز ِّكرت نأ يغبنيو .سنوت في ةيل َّولأا ةيحصلا ةياعرلا زكارم في ضيرمتلا مقاوط ىدل ضىرلما ةملاس ةفاقث ىوتسم ينستح لىإ ةجاح ة َّمَث :تاجاتنتسلاا ،يعمالجا لمعلاو ،موللا ءاقلإ نم ةيلاخ ةئيب ةئيتهو ،حتفنُلما لصاوتلا معدت يتلا ةيدايقلا تاردقلا ءانب لىع ضىرلما ةملاس ةفاقث زيزعت تايجيتاترسا .رمتسلما يميظنتلا م ُّلعتلاو 1354 EMHJ – Vol. 26 No. 11 – 2020Research article 8. Aranaz-Andrés JM, Aibar C, Limón R, Mira JJ, Vitaller J, Agra Y, et al. A study of the prevalence of adverse events in primary healthcare in Spain. Eur J Public Health. 2012 Dec;22(6):921–5. https://doi.org/10.1093/eurpub/ckr168 9. Chen I-C, Li H-H. Measuring patient safety culture in Taiwan using the Hospital Survey on Patient Safety Culture (HSOPSC). BMC Health Serv Res. 2010;10:Article number 152. https://doi.org/10.1186/1472-6963-10-152 10. Mallouli M, Tlili MA, Aouicha W, Ben Rejeb M, Zedini C, Salwa A, et al. Assessing patient safety culture in Tunisian operating rooms: a multicenter study. Int J Qual Health Care. 2017 Apr;29(2):176–82. https://doi.org/10.1093/intqhc/mzw157 11. El-Jardali F, Jaafar M, Dimassi H, Jamal D, Hamdan R. The current state of patient safety culture in Lebanese hospitals: a study at baseline. Int J Qual Health Care. 2010 Oct;22(5):386–95. https://doi.org/10.1093/intqhc/mzq047 12. Ammouri AA, Tailakh AK, Muliira JK, Geethakrishnan R, Al Kindi SN. Patient safety culture among nurses. Int Nurs Rev. 2015 Mar;62(1):102–10. http://dx.doi.org/10.1111/inr.12159 PMID: 25495946 13. Nieva VF, Sorra J. Safety culture assessment: a tool for improving patient safety in healthcare organizations. Qual Saf Health Care. 2003;12 Suppl 2:ii17–23. http://dx.doi.org/10.1136/qhc.12.suppl_2.ii17 14. Khater WA, Akhu-Zaheya LM, AL-Mahasneh SI, Khater R. Nurses’ perceptions of patient safety culture in Jordanian hospitals. Int Nurs Rev. 2015 Mar;62(1):82–91. http://dx.doi.org/10.1111/inr.12155 PMID:25439981 15. Abdou HA, Saber KM. A baseline assessment of patient safety culture among nurses at student university hospital. World J Med Sci. 2011 Jan;6(1):17–26. 16. Richardson A, Storr J. Patient safety: a literature [corrected] review on the impact of nursing empowerment, leadership and collaboration. Int Nurs Rev. 2010 Mar;57(1):12–21. https://doi.org/10.1111/j.1466-7657.2009.00757.x 17. Cheikh AB, Bouafia N, Mahjoub M, Ezzi O, Nouira A, Njah M. Patient’s safety culture among Tunisian healthcare workers: results of a cross sectional study in university hospital. Pan Afr Med J 2016 3 Aug;24:Article 299. https://doi.org/10.11604/ pamj.2016.24.299.8466 18. Ben Rejeb M, Chebil D, Merzougui L, Kacem B, Khefacha-Aissa S, Dhidah L, et al. Patient safety culture in a Tunisian teaching tertiary care hospital. Arch Nurs Pract Care. 2017;3(1):077–83. http://dx.doi.org/10.17352/2581-4265.000030 19. Comité de Coordination de l’évaluation Clinique et de la Qualité en Aquitaine. Mesure de la culture de sécurité des soins dans les établissements de santé: guide d’utilisation; 2015. 20. Occelli P, Quenon J-L, Kret M, Domecq S, Delaperche F, Claverie O, et al. Validation of the French version of the Hospital Survey on Patient Safety Culture questionnaire. Int J Qual Health Care 2013 Sep;25(4):459–68. https://doi.org/10.1093/intqhc/mzt047 21. Giai J, Boussat B, Occelli P, Gandon G, Seigneurin A, Michel P, et al. Hospital survey on patient safety culture (HSOPS): variabili- ty of scoring strategies. Int J Qual Health Care 2017 Oct;29(5):685–92. https://doi.org/10.1093/intqhc/mzx086 22. Verbakel NJ, Zwart DL, Langelaan M, Verheij TJ, Wagner C. Measuring safety culture in Dutch primary care: psychometric characteristics of the SCOPE-PC questionnaire. BMC Health Serv Res 2013;13:Article number 354. https://doi.org/10.1186/1472- 6963-13-354 23. Sandars J, Esmail A. The frequency and nature of medical error in primary care: understanding the diversity across studies. Fam Pract 2003 Jun;20(3):231–6. https://doi.org/10.1093/fampra/cmg301 24. Smits M, Christiaans-Dingelhoff I, Wagner C, Wal G van der, Groenewegen PP. The psychometric properties of the “Hos- pital Survey on Patient Safety Culture” in Dutch hospitals. BMC Health Serv Res 2008;8: Article number 230. https://doi. org/10.1186/1472-6963-8-230 25. Mohamed AM, Ali MS, Gewaifel GI. Assessment of Patient Safety Culture in Primary Healthcare Services in Alexandria, Egypt. Glob J Epidemiol Public Health 2015;2:5–14. http://dx.doi.org/10.12974/2313-0946.2015.02.01.1 26. Tabrizchi N, Sedaghat M. The first study of patient safety culture in Iranian primary health centers. Acta Med Iran 2012;50(7):505–10. PMID:22930384 27. Ghobashi MM, El-ragehy HAG, Ibrahim HM, Al-Doseri FA. Assessment of patient safety culture in primary health care settings in Kuwait. Epidemiol Biostat Public Health 2014;11(3). https://doi.org/10.2427/9101 28. Minuzzi AP, Salum NC, Locks MOH. Avaliação da cultura de segurança do paciente em terapia intensiva na perspectiva da equi- pe de saúde. Texto Contexto Enferm 2016;25(2) (in Portuguese). https://doi.org/10.1590/0104-07072016001610015 29. Wilson T, Pringle M, Sheikh A. Promoting patient safety in primary care. BMJ 2001 Sep 15;323(7313):583–4. http://dx.doi. org/10.1136/bmj.323.7313.583 PMID:11557689 30. Duffield C, Diers D, O’Brien-Pallas L, Aisbett C, Roche M, King M, et al. Nursing staffing, nursing workload, the work environ- ment and patient outcomes. Appl Nurs Res. 2011 Nov;24(4):244–55. https://doi.org/10.1016/j.apnr.2009.12.004 31. Rea D, Griffiths S. Patient safety in primary care: incident reporting and significant event reviews in British general practice. Health Soc Care Community 2016 Jul;24(4):411–9. https://doi.org/10.1111/hsc.12221 32. Cooke DL, Dunscombe PB, Lee RC. Using a survey of incident reporting and learning practices to improve organisational learn- ing at a cancer care centre. BMJ Qual Saf. 2007;16:342–8. https://doi.org/10.1136/qshc.2006.018754 33. Patient safety curriculum guide for medical schools 2009. Geneva: World Health Organization; 2009 (https://apps.who.int/iris/ bitstream/handle/10665/44091/9789241598316_eng.pdf?sequence=1&isAllowed=y, accessed 25 April 2020). 34. Patient safety curriculum guide: multi-professional edition. Geneva: World Health Organization; 2011 (https://apps.who.int/iris/ handle/10665/44641, accessed 25 April 2020). 1355 EMHJ – Vol. 26 No. 11 – 2020Research article Travel burden and geographic access to health care among children with cancer in Saudi Arabia Abdulrahman Alsultan,1,2 Abdullah Aljefri,3 Mouhab Ayas,3 Musa Alharbi,4 Nawaf Alkhayat,5 Faisal Al-Anzi,6 Fawwaz Yassin,7 Fawaz Alkasim,8 Qasim Alharbi,9 Shaker Abdullah,10 Mohammed Abrar10 and Wasil Jastaniah10,11 1Department of Pediatrics, College of Medicine, King Saud University, Riyadh, Saudi Arabia (Correspondence to: A. Alsultan: aalsultan1@ksu.edu.sa). 2Department of Pediatric Hematology/Oncology, King Abdullah Specialist Children’s Hospital, Riyadh, Saudi Arabia. 3Department of Pediatric Hematol- ogy/Oncology, King Faisal Specialist Hospital & Research Center, Riyadh, Saudi Arabia. 4Department of Pediatric Hematology/Oncology, Cancer Center, King Fahad Medical City, Riyadh, Saudi Arabia. 5Department of Pediatrics, Prince Sultan Military Medical City, Riyadh, Saudi Arabia. 6Prince Faisal Bin Bandar Cancer Center, Qassim, Saudi Arabia. 7Department of Pediatric Hematology/Oncology, King Faisal Specialist Hospital & Research Center, Jeddah, Saudi Arabia. 8Department of Pediatric Hematology and Oncology, King Saud Medical City, Riyadh, Saudi Arabia. 9Department of Pediatric Hematology/Oncology, King Fahad Specialist Hospital, Dammam, Saudi Arabia. 10Department of Oncology, Princess Noorah Oncology Center, King Saud Bin Abdulaziz University and King Abdulaziz Medical City, Jeddah, Saudi Arabia. 11Department of Pediatrics, Faculty of Medicine, Umm Al-Qura University, Makkah, Saudi Arabia. Abstract Background: Travel burden has a substantial psychosocial impact and financial strain on childhood cancer patients and their families. Aims: To study the geographic distribution of childhood cancer and assess the travel burden for care in Saudi Arabia. Methods: This was a cross-sectional multi-institutional study that enrolled 1657 children with cancer who were diag- nosed between 2011 and 2014. Cancer type/stage, city/region of residence, and city/region of treating centre were recorded. Travel burden was measured based on a 1-way distance in kilometres from the city centre to the treatment institution. This study was supported by Sanad Children’s Cancer Support Association. Results: Diagnosis was leukaemia (45.2%), non-CNS solid tumours (30.2%), lymphoma (12.3%), CNS tumours (11.8%) and histiocytosis (0.5%). Childhood cancer centres were in the same city as where the patients lived in 652 (39.3%) cases, same region but different city in 308 (18.6%), different regions in 613 (37%), and not known in 84 (5.1%). The mean 1-way travel dis- tance for patients who lived in different regions was 790 (range, 116–1542) km. A total of 536 (32%) patients lived ≥ 400 km and 216 (13%) > 1000 km from the treatment centre. Among 642 patients with acute lymphoblastic leukaemia who required 2–3 years of therapy, 197 (31%) lived ≥ 400 km and 94 (15%) >1000 km from the treatment centre. Conclusions: Nearly two thirds of patients with childhood cancer lived in different cities than the treatment centres, in- cluding one third of patients who lived ≥ 400 km away. There is a need to develop strategies to improve access to childhood cancer care. Keywords: cancer, geographic distribution, paediatrics, Saudi Arabia, travel burden Citation: Alsultan A; Aljefri A; Ayas M; Alharbi M; Alkhayat N; Al-Anzi F; et al. Travel burden and geographic access to health care among children with cancer in Saudi Arabia. East Mediterr Health J. 2020;26(11):1355-1362. https://doi.org/10.26719/emhj.20.020 Received: 10/04/19; accepted: 13/11/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo) Introduction The outcome of childhood cancer has improved signif- icantly in recent decades with current estimated 5-year survival rates of 80%. This progress has been due to suc- cessful clinical trials conducted by collaborative research groups, such as the Children’s Oncology Group, com- bined with advances in supportive care (1,2). However, during progression to cure, other concerns might arise in families of children with cancer, such as employment disruption, high out-of-pocket spending, travel burden, and psychosocial difficulties (3,4). These concerns are infrequently addressed in depth with families given that the focus of medical teams is primarily on delivering optimal treatment. Governmental financial support for patients and their families, active involvement of social service teams at cancer centres, and support from non- profit organizations are ways to address some of these concerns in Saudi Arabia. Childhood cancer in Saudi Arabia affects 1 in 10 000 children. The 2015 Saudi Cancer Registry Report showed that leukaemia was the most common childhood cancer (35%) followed by brain tumours (12.2%) and non-Hodgkin’s lymphoma (12.2%) (5). Centres that treat childhood cancer are mainly in 4 (Riyadh, Makkah, Eastern and Qassim) out of the 13 regions in Saudi Arabia. Patients and their families in the remaining 9 regions need to travel long distances by land or air for their initial diagnosis and treatment. Increased travel distance between the residence of patients and the treatment centre is associated with increased financial burden, work interruption, and residence relocation (6). In a large study in the United Kingdom of Great Britain and Northern Ireland, travel burden was associated with survival disadvantage among cancer patients (7). The travel burden and its impact on cancer outcome have not been studied in Saudi Arabia. In this study, 1356 EMHJ – Vol. 26 No. 11 – 2020Research article we examined the geographic distribution of childhood cancer in different regions in Saudi Arabia, assessed the burden of travel among patients and their families, and evaluated the influence of travel burden on the initial cancer staging in solid tumours. Methods Patient population We performed a cross-sectional multi-institutional study in 10 centres that treat most cases of childhood cancer in Saudi Arabia. Informed consent was obtained from par- ents of all participants and the study was approved by the institutional review boards in all participating insti- tutions. We enrolled 1657 patients: 917 (55%) male and 740 (45%) female. aged ≤ 14 years who were diagnosed with cancer between January 2011 and December 2014. We collected the following information: cancer type, cancer stage, city/region of residence, and city/region of treat- ment centre. Data were recorded remotely using RED- Cap (Research Electronic Data Capture) electronic data capture tools hosted and stored centrally in a secure Mi- crosoft SQL database (8). The study was organized by the Saudi Arabian Pediatric Hematology Oncology Society (SAPHOS) as part of a study to determine the prevalence of hereditary cancer syndromes, as described previously (9). It was supported by Sanad Children’s Cancer Support Association. Geographic distribution and travel burden of childhood cancer Number of patients, sex and characteristics of cancer were described for each region. The proportion of child- hood cancer in each region was compared to the pro- portion of normal children aged ≤ 14 years living in the same region, using data from the demographic survey performed in 2016 by the Saudi General Authority of Sta- tistics (10). Travel burden was assessed using Google map based on a 1-way distance in kilometres from the city cen- tre where the patients lived, to the treatment institution. Data analysis Descriptive analyses were presented as mean (standard deviation) values for continuous data and as frequencies for categorical data. A t test was used to compare 2 means and χ2 or Fisher’s exact test to compare proportions of 2 groups. Patients who lived in the same city as the treat- ment institution were used as a reference group. P < 0.05 was considered to be statistically significant. Stata Statis- tical Software Release 12 was used for all analyses (Stat- aCorp LP, College Station, TX, USA). Results Cancer epidemiology in different regions Cancer classification and geographic distribution of pa- tients who were enrolled in the study are summarized in Table 1. The total number of patients (n =1657) enrolled in our study represented 50% of all childhood cancer cases expected to be diagnosed during the study period, based on the Saudi Cancer Registry (5). A total of 1501 (91%) pa- tients were Saudi and the remaining 156 (9%) were from other nationalities. Leukaemia was the most common diagnosis (45.2%), followed by non-CNS solid tumours (30.2%), lymphoma (12.3%), CNS tumours (11.8%) and last- ly histiocytosis (0.5%). There was no marked difference in the pattern of cancer among regions. Figure 1 shows the proportion of children with cancer in each region as well as the proportion of normal children aged ≤ 14 years. The proportion of children with cancer who lived in Riyadh Region was 30.8% of all patients enrolled in our study, while the proportion of normal children who lived in Ri- yadh Region was 24.7% of all normal Saudi children (P < 0.0001). Travel burden and access to cancer care in different regions Treatment institutions were in the same city for only 652 (39.3%) patients (Table 2). The treatment centre was in the same region but different city for 308 (18.6%) pa- tients and the average 1-way travel distance among those patients was 159 (range, 19–737) km. The remaining 613 (37%) patients lived in different regions from the treat- ment centres with a mean 1-way travel distance of 790 (range, 116–1542) km. A total of 536 (32%) patients lived ≥ 400 km and > 3 hours travel time from the treatment centres. Among those, 216 (13%) patients lived > 1000 km from the treatment centre. Patients with acute lymphoblastic leukaemia (ALL) required prolonged therapy and frequent visits to cancer centres for several years. There were 642 patients with ALL in our study: 283 (44%) lived in the same city as the treatment centre; 103 (16%) lived in the same region but different city; 230 (36%) lived in a different region; and the address was unknown for 26 (4%). The average travel distance for ALL patients who lived in different regions was 792 (range, 280–1542) km. A total of 197 (31%) ALL patients lived ≥ 400 km from the treatment centre, and 94 (15%) of those lived > 1000 km distant. Regional referral pattern in childhood cancer Most patients living in Riyadh (99%), Makkah (90%) and Qassim (90%) Regions were treated in the same region. Nearly half of patients living in the Eastern Region (45%) had to be treated in Riyadh. Childhood cancer centres in Riyadh were the main referral centres for most regions, except Madinah and Albaha, and centres in Jeddah treat- ed most patients from these 2 regions (Table 3, Figure 2). There were 613 patients who lived in different regions than the cancer centres: 453 (74%) were treated in Riyadh, 139 (23%) in Jeddah and 21 (3%) in Qassim. Discussion In this study, we described the geographic distribution of childhood cancer and assessed the travel burden among our patients and their families. Nearly two thirds of 1357 EMHJ – Vol. 26 No. 11 – 2020Research article Ta bl e 1 G eo gr ap hi c d is tr ib ut io n an d ch ar ac te ri st ic s o f c hi ld ho od ca nc er in S au di A ra bi a D ia gn os is Re gi on s ( pa ti en t’s h om e) To ta l Ri ya dh M ak ka h Ea st er n M ad in ah As ir Ja za n Q as si m H ai l Al jo uf Ta bu k N aj ra n Al ba ha N or th er n bo rd er N D Le uk ae m ia 26 3 14 3 69 43 52 28 35 15 20 14 11 12 13 31 74 9 (4 5. 2% ) Ac ut e ly m ph ob la st ic le uk ae m ia 23 8 11 8 42 40 48 25 27 15 18 12 10 11 12 26 64 2 Ac ut e m ye lo id le uk ae m ia 11 20 2 1 4 2 5 0 2 1 1 1 1 2 53 Ac ut e le uk ae m ia , N O S 13 3 23 2 0 1 3 0 0 1 0 0 0 3 49 JM M L 1 0 1 0 0 0 0 0 0 0 0 0 0 0 2 Ch ro ni c m ye lo id le uk ae m ia 0 1 1 0 0 0 0 0 0 0 0 0 0 0 2 M ye lo dy sp la st ic sy nd ro m e 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 N on -C N S s ol id tu m ou rs 14 0 11 0 62 36 25 15 12 23 14 14 12 6 4 27 50 0 (3 0. 2% ) W ilm s’ tu m ou r 30 23 5 9 2 2 6 7 3 3 1 0 2 6 99 Re tin ob la st om a 22 7 13 3 2 4 0 3 2 3 2 0 1 7 69 Ew in g’ s s ar co m a/ PN ET 20 15 5 4 5 1 1 2 3 3 4 1 0 4 68 N eu ro bl as to m a 14 16 8 3 3 4 4 4 1 0 1 1 0 2 61 Rh ab do m yo sa rc om a 19 9 5 5 9 1 1 2 1 0 0 0 1 4 57 H ep at ob la st om a 10 8 2 3 0 2 0 2 0 0 1 0 0 0 28 O st eo sa rc om a 4 5 4 2 0 1 0 0 2 3 0 2 0 2 25 Re na l c el l c ar ci no m a 8 0 9 3 0 0 0 0 1 0 0 0 0 0 21 G er m ce ll tu m ou r 2 11 3 1 2 0 0 1 0 0 0 0 0 0 20 Th yr oi d ca rc in om a 0 1 0 1 0 0 0 0 0 1 0 2 0 0 5 O th er sa rc om a 4 10 3 1 1 0 0 2 0 0 2 0 0 2 25 O th er ca rc in om a 3 4 2 0 0 0 0 0 0 1 0 0 0 0 10 O th er so lid tu m ou r 4 1 3 1 1 0 0 0 1 0 1 0 0 0 12 Ly m ph om a 49 39 22 9 17 15 12 9 4 5 3 1 1 17 20 3 (12 .3 %) H od gk in ’s ly m ph om a 32 15 14 7 9 8 6 6 2 5 1 0 1 9 11 5 N on -H od gk in ’s ly m ph om a 17 24 8 2 8 7 6 3 2 0 2 1 0 8 88 CN S t um ou rs 56 43 22 18 7 8 3 10 6 6 2 4 2 9 19 6 (11 .8 %) M ed ul lo bl as to m a 23 25 10 9 2 3 3 5 3 2 0 2 0 4 90 G lio m a 22 13 6 7 4 2 0 5 3 2 1 2 2 2 71 Ep en dy m om a 6 2 4 2 0 0 0 0 0 1 1 0 0 0 16 AT RT 3 0 0 0 1 2 0 0 0 1 0 0 0 0 7 G er m ce ll tu m ou r 1 0 2 0 0 0 0 0 0 0 0 0 0 1 4 O th er C N S tu m ou r 1 3 0 0 0 1 0 0 0 0 0 0 0 2 8 H ist io cy to sis 3 1 0 2 1 0 1 0 0 0 0 0 0 1 9 (0 .5 %) To ta l 51 1 ( 30 .8 %) 33 6 (2 0. 2% ) 17 5 (10 .6 %) 10 8 (6 .5 %) 10 2 (6 .2 %) 67 (4 .0 %) 63 (3 .8 %) 57 (3 .4 %) 44 (2 .7 %) 39 (2 .4 %) 28 (1 .7 %) 23 (1 .4 %) 20 (1 .2 %) 84 (5 .1% ) 16 57 N OS = n ot ot he rw ise sp ec ifi ed ; N D = no t d et er m in ed ; J M M L = ju ve ni le m ye lo m on oc yt ic leu ka em ia ; A TR T = at yp ica l t er at oi d rh ab do id tu m ou r; PN ET = p er ip he ra l n eu ro ec to de rm al tu m ou r. 1358 EMHJ – Vol. 26 No. 11 – 2020Research article patients lived in cities other than the city of the treatment centre, including one third of patients who lived in differ- ent regions. The average travel burden was 1-way travel of 790 km for patients living in different regions than the treatment centres. There was a higher proportion of childhood cancer patients who lived in Riyadh Region (30.8%) compared to 24.7% of normal children living in the region. This was probably caused by residence relo- cation to Riyadh by some families to be closer to cancer treatment centres. The government covers the cost of airline tickets for patients and their parents; however, on many occasions families need to drive long distances be- cause of fully booked flights. Additionally, the travel bur- den is exacerbated by the limited accessibility to assigned local primary care physicians that is a common practice in Saudi Arabia. Thus, most of our patients’ health care is provided at cancer centres. Travel burden has multiple negative effects on cancer patients and their families. One study showed that childhood cancer patients living in rural areas were at higher risk of missing more school days, and their caregivers missed more work days and spent more out- of-pocket travel expenses compared to urban residents (6). Travel burden was highest for patients living in rural areas in Australia and was associated with significant financial strains (11). For colon cancer patients in the United States of America (USA), increased travel distance to cancer centres was associated with advanced stage at diagnosis and lower possibility of receiving adjuvant chemotherapy within 90 days of colectomy (12,13). Treatment outcome and survival were not assessed in our study. Thus, it is possible that patients living in remote areas might have worse outcome due to delay in managing cancer or treatment-related complications such as febrile neutropenia. Nevertheless, there is probably a survival advantage for patients traveling to more experienced childhood cancer centres in Saudi Arabia. This is supported by the survival benefit that was observed in the USA among cancer patients receiving treatment at National Cancer Institute (NCI)-designated cancer centres (14). Establishing satellite facilities of the main NCI cancer centres has improved geographic access to high-quality cancer care, with nearly 85% of the American population living within 3 hours of either a parent or satellite facility (15). There are currently limited numbers of satellite facilities that are administered by large childhood cancer centres in Saudi Arabia. Therefore, establishing satellite facilities or affiliated medical centres should be a priority to improve geographic access to cancer care among Saudi patients. In addition, incorporating survival data in the current Saudi Cancer Registry is essential. There is a need to develop strategies to improve access to cancer care in Saudi Arabia. Formation of a national referral system to coordinate between different healthcare sectors will facilitate timely access to childhood cancer centres. The integration of local primary care physicians (PCPs) in the care of children with cancer is essential (16). Paediatric oncologists should encourage parents to have local PCPs for their children. Additionally, there is a need to conduct regular workshops to train local PCPs and other local healthcare providers on various topics in childhood cancer, to enable them to recognize cancer at an early stage, refer patients promptly to cancer centres, and provide appropriate management of potential complications (17). It is also Table 2 Travel burden among children with cancer and their families Locations of treatment centre No. of patients (%) Same city 652 (39.3%) Same region but different city 308 (18.6%) Different region 613 (37.0%) < 200 km 3 (0.1%) 200 – < 400 km 74 (4.5%) 400 – < 600 km 122 (7.4%) 600 – < 800 km 104 (6.3%) 800 – < 1000 km 94 (5.7%) 1000 – < 1200 km 124 (7.5%) 1200 – < 1400 km 89 (5.4%) 1400 – < 1600 km 3 (0.1%) Unknown 84 (5.1%) Distance is based on 1-way travel. Figure 1 Regional distribution of childhood cancer in Saudi Arabia. Proportion of childhood cancer (C) in each region in relation to the total number of childhood cancer cases (n = 1657) is shown. Proportion of normal children aged ≤ 14 years (N) in each region in relation to the total number of normal children aged ≤ 14 years in Saudi Arabia (n = 7 864 928) is also shown. 1359 EMHJ – Vol. 26 No. 11 – 2020Research article important to involve PCPs in the long-term care of cancer survivors (18). Availability of 24-hour helpline at childhood cancer centres is necessary to support PCPs and give caregivers direct communication with oncologists at any time. Travel and accommodation support should be integrated into the cancer care of children in Saudi Arabia. Our study was limited by the lack of data on the date of first appearance of symptoms and signs of cancer compared to dates of diagnosis and starting treatment in order to measure accurately the impact of travel burden on the time to initiate cancer treatment. There are inconsistent reports on the association between time to diagnosis or treatment and poor survival in childhood cancer (19,20). Another study limitation was enrolling only patients who were treated at cancer centres. Thus, we could not assess potential early mortality among children with cancer living in rural areas prior to their acceptance in cancer centres. Early death within the first month of diagnosis in childhood cancer was associated with age < 1 year, low socioeconomic status, and certain cancers such as acute myeloid leukaemia (21). Conclusion The travel burden on children with cancer and their fam- ilies in Saudi Arabia is substantial. Approximately two thirds of patients live in cities different from where the cancer centres are located. One third of patients are > 3 hours away (≥ 400 km) from cancer centres. Our findings might guide policy-makers to develop national strategies to improve access to childhood cancer care in Saudi Ara- bia. Future studies are needed to assess the impact of liv- ing remotely from cancer centres on different outcomes such as event-free and overall survival. Funding: The study was funded by Sanad Children’s Cancer Support Association Research Grant Programme. Competing interests: None declared. Ta bl e 3 Re gi on al re fe rr al p at te rn in ch ild ho od ca nc er in S au di A ra bi a Fe at ur es Re gi on s ( pa ti en t’s h om e) Ri ya dh M ak ka h Ea st er n M ad in ah As ir Ja za n Q as si m H ai l Al jo uf Ta bu k N aj ra n Al ba ha N or th er n bo rd er N D N o. of p at ie nt s 51 1 33 6 17 5 10 8 10 2 67 63 57 44 39 28 23 20 84 Tr ea tm en t c en tr e – sa m e c ity 43 3 15 5 23 0 0 0 41 0 0 0 0 0 0 na Tr ea tm en t c en tr e – sa m e r eg io n bu t d iff er en t c ity 73 14 9 70 0 0 0 16 0 0 0 0 0 0 na Tr ea tm en t c en tr e – d iff er en t r eg io n 5 32 82 10 8 10 2 67 6 57 44 39 28 23 20 na Ri ya dh na 31 78 36 89 49 6 42 40 31 25 6 20 66 M ak ka h 3 na 1 71 13 18 0 2 3 8 3 17 0 8 Ea st er n 0 0 na 0 0 0 0 0 0 0 0 0 0 0 Q as si m 2 1 3 1 0 0 na 13 1 0 0 0 0 10 Figure 2 Regional referral pattern of childhood cancer in Saudi Arabia. Childhood cancer centres are present in 4 circled cities. Arrows represent common referral patterns from each region. The weight of the arrow corresponds to the proportion of patients referred to a specific region. 1360 EMHJ – Vol. 26 No. 11 – 2020Research article Poids des déplacements et accès géographique aux soins de santé pour les enfants atteints de cancer en Arabie saoudite Résumé Contexte : Le poids des déplacements a un impact psychosocial et financier considérable sur les enfants atteints de cancer et sur leurs familles. Objectifs : La présente étude avait pour objectif d’étudier la répartition géographique du cancer chez l’enfant et d’évaluer le poids des déplacements pour les soins en Arabie saoudite. Méthodes : Il s’agissait d’une étude transversale multi-institutionnelle portant sur 1657 enfants atteints de cancer, diagnostiqués entre 2011 et 2014. Le type/stade du cancer, la ville/région de résidence et la ville/région du centre de traitement ont été enregistrés. Le poids des déplacements a été mesuré sur la base de la distance en kilomètres, dans un sens, entre le centre de la localité et l’établissement de soins. Cette étude a reçu le soutien de Sanad Children’s Cancer Support Association. Résultats : Les diagnostics concernaient la leucémie (45,2 %), les tumeurs solides hors système nerveux central (30,2 %), le lymphome (12,3 %), les tumeurs du système nerveux central (11,8 %) et l’histiocytose (0,5 %). Les centres de lutte contre le cancer de l’enfant se trouvaient dans la même ville que celle où les patients vivaient dans 652 cas (39,3 %), dans la même région, mais dans des villes différentes dans 308 cas (18,6 %) et dans des régions différentes dans 613 cas (37 %). Ce lieu n’était pas connu dans 84 cas (5,1 %). La distance moyenne parcourue par trajet pour les patients qui vivaient dans des régions différentes était de 790 km (distance comprise entre 116 et 1542 km). Au total, 536 patients (32 %) vivaient à 400 km du centre de traitement et 216 (13 %) à plus de 1000 km. Parmi les 642 patients atteints de leucémie lymphoblastique aiguë ayant nécessité deux à trois ans de traitement, 197 (31 %) vivaient à une distance supérieure ou égale à 400 km du centre de traitement et 94 (15 %) à plus de 1000 km. Conclusions : Près des deux tiers des patients atteints d’un cancer de l’enfant vivaient dans des villes différentes des centres de traitement, dont un tiers des patients à une distance supérieure ou égale à 400 km. Il est nécessaire d’élaborer des stratégies visant à améliorer l’accès aux soins des enfants atteints de cancer. ةيبرعلا ةكلملما في ناطسرلاب نوباصلما لافطلأا اهيف شيعي يتلا ةيفارغلجا قطانلماب ةيحصلا ةياعرلا ةحاتإو رفسلا ءبع ةيدوعسلا ركاش ،بيرلحا مساق ،مساقلا زاوف ،ينساي زاوف ،يزنعلا لصيف ،طايلخا فاون ،بيرلحا ىسوم ،سايإ باهم ،يرفلجا للها دبع ،ناطلسلا نحمرلا دبع هينتسج لصاو ،راربأ دممح ،للها دبع ةصلالخا .مهسرُأو لافطلأا ناطسر ضىرم لىع ةيلام ًاطوغض ضرفيو ،ًايربك ًايعماتجاو ًايسفن ًايرثأت رفسلا ءبع رثؤي :ةيفللخا .ةيدوعسلا ةيبرعلا ةكلملما في ةياعرلا همزلتست يذلا رفسلا ءبع مييقتو لافطلأا ناطسرل فيارغلجا عيزوتلا ةسارد لىإ ةساردلا هذه تفده :فادهلأا ْيَماع ينب ضرلماب متهاباصإ تص ِّخُش ،ناطسرلاب ًاباصم ًلافط 1657 ت َّمض ماك ةددعتم تاسسؤم في ةيعطقلما ةساردلا هذه تيرجأ :ثحبلا قرط لىع رفسلا ءبع سايق متو .جلاعلا زكرم اهيف عقي يتلا ةقطنلما /ةنيدلماو ةماقلإا ةقطنم /ةنيدمو ناطسرلا ةلحرم /عون ل ِّجُسو .2014و 2011 معدل ةييرلخا دنس ةيعجم نم َمعدلا ةساردلا ت َّقلتو .جلاعلا ي ِّقلت ةسسؤم لىإ ةنيدلما طسو نم تاترموليكلاب ةر َّدقم هاتجلاا ةيداحأ ةفاسم ساسأ .ناطسرلاب ضىرلما لافطلأا اموفمللاو ،)٪30.2( يزكرلما يبصعلا زاهلجا يرغ ىرخأ ًةزهجأ بيصت يتلا ةبلصلا مارولأاو ،)٪45.2( مدلا ناطسر صيخشتلا لمش :جئاتنلا يتلا اهسفن ةنيدلما في عقت لافطلأا ناطسر زكارم تناكو .)٪0.5( ةجسنلما ايلالخا ةرثكو )٪11.8( يزكرلما يبصعلا زاهلجا ماروأو ،)٪12.3( في ةفلتمخ قطانم في عقتو ،)٪18.6( تلااح 308 في ةفلتمخ ةنيدم في نكلو اهسفن ةقطنلما في عقتو ،)٪39.3( ةلاح 652 في ضىرلما اهيف شيعي نيذلا ضىرملل هاتجلاا ةيداحأ رفسلا ةفاسم طسوتم ناكو .)٪5.1( ةلاح 84 في ًافورعم لافطلأا ناطسر زكرم ناكم نكي لمو ،)٪37( ةلاح 613 ،جلاعلا زكرم نم مك 400 ≥ دعُب لىع )٪32( ًاضيرم 536 هعوممج ام شيعيو .ًاترموليك )116-1542 ،ىدلما( 790 ةفلتمخ قطانم في نوشيعي دالحا يوافميللا مدلا ناطسرب ينباصم ًاضيرم 642 ينب نمو .جلاعلا زكرم نم مك 1000 نع ديزت ةفاسم دعُب لىع )٪13( ًاضيرم 216 شيعيو شيعيو ،جلاعلا زكرم نم مك 400 ≥ دعُب لىع نوشيعي مهنم )٪31( 197 ناك ،ماوعأ ةثلاث لىإ ْينَماع ينب حواترت جلاع ةدم لىإ نوجاتيح نيذلاو .جلاعلا زكرم نم مك 1000 نع ديزت ةفاسم دعُب لىع مهنم )٪15( 94 ثلث شيعي ءلاؤه ينب نمو ،جلاعلا زكرم اهيف عقي يتلا ةنيدلما يرغ في لافطلأا ناطسرب ينباصلما ضىرلما يثلث نم برقي ام شيعي :تاجاتنتسلاا .ناطسرلاب ينباصلما لافطلأل ةياعرلا ةحاتإ ينسحتل تايجيتاترسا عضو لىإ ةجاح ة َّمَث اذل .جلاعلا زكرم نم ترموليك 400 ≥ دعُب لىع ضىرلما 1361 EMHJ – Vol. 26 No. 11 – 2020Research article References 1. Siegel RL, Miller KD, Jemal A. Cancer Statistics, 2017. CA Cancer J Clin. 2017 Jan;67(1):7–30. http://dx.doi.org/10.3322/caac.21387 PMID:28055103 2. Adamson PC. Improving the outcome for children with cancer: development of targeted new agents. CA Cancer J Clin. 2015 May–Jun;65(3):212–20. http://dx.doi.org/10.3322/caac.21273 PMID:25754421 3. Steele AC, Mullins LL, Mullins AJ, Muriel AC. Psychosocial interventions and therapeutic support as a standard of care in pediat- ric oncology. Pediatr Blood Cancer. 2015 Dec;62 Suppl 5:S585–618. http://dx.doi.org/10.1002/pbc.25701 PMID:26700919 4. Nipp RD, Kirchhoff AC, Fair D, Rabin J, Hyland KA, Kuhlthau K, et al. Financial burden in survivors of childhood cancer: a report from the Childhood Cancer Survivor Study. J Clin Oncol. 2017 Oct 20;35(30):3474–81. http://dx.doi.org/10.1200/JCO.2016.71.7066 PMID:28817372 5. Saudi Health Council, National Health Information Center, Saudi Cancer Registry. Cancer incidence report Saudi Arabia 2015 (https://nhic.gov.sa/eServices/Documents/E%20SCR%20final%206%20NOV.pdf, accessed 17 April 2020). 6. Fluchel MN, Kirchhoff AC, Bodson J, Sweeney C, Edwards SL, Ding Q, et al. Geography and the burden of care in pediatric can- cers. Pediatr Blood Cancer. 2014 Nov;61(11):1918–24. http://dx.doi.org/10.1002/pbc.25170 PMID:25131518 7. Turner M, Fielding S, Ong Y, Dibben C, Feng Z, Brewster DH, et al. A cancer geography paradox? Poorer cancer outcomes with longer travelling times to healthcare facilities despite prompter diagnosis and treatment: a data-linkage study. Br J Cancer. 2017;117(3):439–49. https://doi.org/10.1038/bjc.2017.180 8. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap) – a metadata-driv- en methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009 Apr;42(2):377–81. http://dx.doi.org/10.1016/j.jbi.2008.08.010 PMID:18929686 9. Jastaniah W, Aljefri A, Ayas M, Alharbi M, Alkhayat N, Al-Anzi F, et al. Prevalence of hereditary cancer susceptibility syndromes in children with cancer in a highly consanguineous population. Cancer Epidemiol. 2018 Aug;55:88–95. http://dx.doi.org/10.1016/j. canep.2018.05.006 PMID:29859499 10. Demographic Survey 2016. General Authority for Statistics. https://www.stats.gov.sa/sites/default/files/en-demographic-re.- search-2016_4.pdf, accessed 17 April 2020). 11. Zucca A, Boyes A, Newling G, Hall A, Girgis A. Travelling all over the countryside: travel-related burden and financial difficul- ties reported by cancer patients in New South Wales and Victoria. Aust J Rural Health. 2011 Dec;19(6):298–305. http://dx.doi. org/10.1111/j.1440-1584.2011.01232.x PMID:22098213 12. Lin CC, Bruinooge SS, Kirkwood MK, Olsen C, Jemal A, Bajorin D, et al. Association between geographic access to cancer care, insurance, and receipt of chemotherapy: geographic distribution of oncologists and travel distance. J Clin Oncol. 2015 Oct 1;33(28):3177–85. http://dx.doi.org/10.1200/JCO.2015.61.1558 PMID:26304878 13. Massarweh NN, Chiang YJ, Xing Y, Chang GJ, Haynes AB, You YN, et al. Association between travel distance and metastat- ic disease at diagnosis among patients with colon cancer. J Clin Oncol. 2014 Mar 20;32(9):942–8. http://dx.doi.org/10.1200/ JCO.2013.52.3845 PMID:24516014 14. Onega T, Duell EJ, Shi X, Demidenko E, Gottlieb D, Goodman DC. Influence of NCI cancer center attendance on mor- tality in lung, breast, colorectal, and prostate cancer patients. Med Care Res Rev. 2009 Oct;66(5):542–60. http://dx.doi. org/10.1177/1077558709335536 PMID:19454624 15. Onega T, Alford-Teaster J, Wang F. Population-based geographic access to parent and satellite National Cancer Institute Cancer Center facilities. Cancer. 2017 Sep 1;123(17):3305–11. http://dx.doi.org/10.1002/cncr.30727 PMID:28464212 16. Rubin G, Berendsen A, Crawford SM, Dommett R, Earle C, Emery J, et al. The expanding role of primary care in cancer control. Lancet Oncol. 2015 Sep;16(12):1231–72. http://dx.doi.org/10.1016/S1470-2045(15)00205-3 PMID:26431866 17. Radhi M, Fulbright JM, Ginn KF, Guest EM. Childhood cancer for the primary care physician. Prim Care. 2015 Mar;42(1):43–55. http://dx.doi.org/10.1016/j.pop.2014.09.006 PMID:25634704 18. Signorelli C, Wakefield CE, Fardell JE, Wallace WHB, Robertson EG, McLoone JK, et al. The impact of long-term follow-up care for childhood cancer survivors: a systematic review. Crit Rev Oncol Hematol. 2017 Jun;114:131–8. http://dx.doi.org/10.1016/j. critrevonc.2017.04.007 PMID:28477741 19. Lethaby CD, Picton S, Kinsey SE, Phillips R, van Laar M, Feltbower RG. A systematic review of time to diagnosis in children and young adults with cancer. Arch Dis Child. 2013 May;98(5):349–55. http://dx.doi.org/10.1136/archdischild-2012-303034 PMID:23476000 20. Neal RD, Tharmanathan P, France B, Din NU, Cotton S, Fallon-Ferguson J, et al. Is increased time to diagnosis and treatment in symptomatic cancer associated with poorer outcomes? Systematic review. Br J Cancer. 2015 Mar 31;112 Suppl 1:S92–107. http:// dx.doi.org/10.1038/bjc.2015.48 PMID:25734382 21. Furutani E, Rodriguez-Galindo C, Green AL. Early death in pediatric cancer: remaining questions and next steps. Oncotarget. 2017 Nov 1;8(57):96478–9. http://dx.doi.org/10.18632/oncotarget.22257 PMID:29228545 1362 EMHJ – Vol. 26 No. 11 – 2020Research article Prevalence of refractive error and visual impairment among school-age children of Hargesia, Somaliland, Somalia Zahra Abdi Ahmed,1 Saif Hassan Alrasheed2,3 and Waleed Alghamdi3 1Department of Primary Eye Care, Faculty of Optometry and Visual Sciences, Al-Neelain University, Khartoum, Sudan. 2Department of Binocular Vision, Faculty of Optometry and Visual Sciences, Al-Neelain University, Khartoum, Sudan (Correspondence to: Saif Alrasheed: s.rasheed@qu.edu.sa). 3Department of Optometry, College of Applied Medical Sciences, Qassim University, Qassim, Saudi Arabia. Abstract Background: Childhood visual impairment is a global public health problem, especially in low and middle-income coun- tries. Its most common causes are avoidable by early diagnosis and treatment. Aims: To assess prevalence of refractive error and visual impairment among school-aged children in Hargeisa, Somali- land, Somalia. Methods: This was a cross-sectional study of 1204 students (aged 6–15 years) in 8 randomly selected primary schools in Hargeisa from November 2017 to January 2018. We used the modified Refractive Error Study in Children to determine prevalence of refractive error and visual impairment, including the following investigations: distance visual acuity, as- sessed by Snellen Tumbling E-chart; refraction, assessed by retinoscope binocular vision assessment; and examination of anterior and posterior segments. Results: Prevalence of uncorrected, presenting and best-corrected visual impairment of 6/12 or worse was 13.6%, 7.6% and 0.75%, respectively. Only 16 of 91 (17.6%) children were using spectacles and the rest were unaware of the problem. Refrac- tive error was the cause of visual impairment in 76.8% of participants, amblyopia in 22.0%, trachoma in 2.4%, and corneal opacity and cataract in 0.6%. Anterior segment abnormalities were found in 8.3%, mainly vernal keratoconjunctivitis, while posterior abnormalities were observed in 0.7%. Prevalence of myopia was 9.1%, hypermetropia 2.7% and astigmatism 3.9%. Prevalence of visual impairment because of Refractive Error was associated with increasing age, but there was no significant association with school grade or sex. Conclusion: Prevalence of visual impairment among school-aged children in Hargeisa was high, and the leading cause was uncorrected Refractive Error. There are barriers to care and it is critical that they are overcome. Keywords: refractive error, childhood visual impairment, myopia, hypermetropia, vernal keratoconjunctivitis Citation: Abdi Ahmed Z; Alrasheed SH; Alghamdi W. Prevalence of refractive error and visual impairment among school-age children of Hargesia, Somaliland, Somalia. East Mediterr Health J. 2020;26(11):1362-1370. https://doi.org/10.26719/emhj.20.077 Received: 18/03/19; accepted: 19/11/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Introduction Globally, it is estimated that there are 36 million people who are blind, 216.6 million have moderate to severe visual impairment (VI) and 188.5 million have mild VI. The leading cause of VI is uncorrected refractive error (RE) (1,2). Furthermore, 90% of people with VI live in de- veloping countries. Almost 19 million children aged < 15 years have VI globally. In developing countries, 7–31% of childhood blindness is avoidable, 10–58% is treatable, and 3–28% is preventable (3). RE is an eye condition in which light from a distant object is not focused on the retina; it might be focused in front of or behind the retina. There are 3 types of RE: myopia, hypermetropia and astigma- tism. The exact cause of ametropia remains unknown with common risk factors being hereditary, nutritional and environmental (4). Population-based studies on VI and RE in children have been conducted on populations with different racial backgrounds and environments in Africa. These studies have shown that the prevalence of VI among children was 2.15% in South Africa (5), 5.5% in Khartoum, Sudan (6), 4.4% in South Darfur, Sudan (7), and 9.5% in Ethiopia (8). World Health Organization (WHO) defines VI in children as presenting with visual acuity (VA) less than 6/12 in the better eye. However, uncor- rected VA (UVA) is defined as VA less than 6/12 in one or both eyes (7). VI among children in developing countries is a priority of eye health programmes, including Vision 2020: the Right to Sight Initiative (9). Globally, the prin- cipal cause of VI is uncorrected REs (43%) and cataracts (33%) (1). Special attention should be given to children, because VI restricts their education and general perfor- mance, personality development, future quality of life and career opportunities (10). The Refractive Error Study in Children (RESC) protocol was developed by WHO in collaboration with, and under financial support from the National Eye Institute, National Institutes of Health and the United States of America to assess the prevalence of VI and RE worldwide, as well as to assess the effect of childhood VI due to uncorrected RE (11,12). Somaliland, Somalia has a population of 4.5 million, with estimated urban poverty of 29%, which is similar to 26% in Ethiopia. Only about half of children aged 6–13 years go to primary school in Somaliland, in stark contrast to 87% in neighbouring Ethiopia (13,14). No 1363 EMHJ – Vol. 26 No. 11 – 2020Research article studies have assessed VI and RE among school-aged children in Hargeisa, Somaliland. The aims of this study were to assess the common causes of VI, types of RE, and differences in prevalence according to sex, age and school grade. Methods Study design This was a cross-sectional, school-based study of VI and RE among children from Hargeisa, Somaliland. Accord- ing to the Ministry of Education, the overall number of students enrolled in public and private primary schools in Hargeisa during 2017–2018 was 243 485, comprising 127 829 boys and 115 656 girls. The modified RESC pro- tocol was used to assess the prevalence of VI and RE in these children. Noncycloplegic refraction was used to as- sess the prevalence of RE, which is defined as follows: (1) myopia ≥ −0.5 D in one or both eyes; (2) hypermetropia ≥ 2.0 D ; and (3) astigmatism ≥ 0.75 D cylindrical refraction (11). Inclusion and exclusion criteria Children aged 6–15 years who attended school on the days of examination and their parents agreed to partic- ipate in the study. Children unable to provide parental consent were excluded. Study sample The study sample was selected through stratified multi- stage sampling. We assumed a prevalence of RE of 5% according to the estimated prevalence of childhood RE in Africa (5%), Sudan (6.8%) (5) and Kenya (5.1%) (15, 16). Considering a prevalance of RE of 5%, 95% confidence in- terval and maximum acceptable random sampling error of 1.5%, a sample size of 811, based on the formula below, was estimated. Considering the design effect = 1.5, a final sample of 1216 was estimated. n = (z2 pq)/d2 = (1.962 × 0.05 × 0.95)/0.0152 = 811 => 811 × 1.5 = 1216 Considering a nonresponse rate = 10%, the final sample size was 1351 schoolchildren. The study sample com- prised 8 schools (4 for boys and 4 for girls) that were ran- domly selected from 22 districts of Hargesia. One class from each grade (1–8) with a minimum of 21 children was randomly chosen. Ethical considerations Ethical permission for the study was obtained from Al- Neelain University, Khartoum, Sudan because of un- availability of an ethics committee in Somaliland. The study was conducted according to the Declaration of Hel- sinki guidelines. Informed consent was obtained from all participants. All forms and data sheets were shredded as soon as the details were entered into the database system for analysis. Clinical investigation The clinical examinations were performed using the modified RESC protocol. Demographic information was collected from participants, and VA at distance was measured using the Snellen Tumbling E-chart with E’s of standard size at a 6-m distance. Participants with VA ≤ 6/12 were assessed by pinhole test, and if their vision im- proved, they underwent retinoscopy without cycloplegia and subjective refraction. All children were examined by a penlight and low-power hand magnifier to assess any anterior-segment abnormalities in the eyelids, conjunc- tiva, cornea, pupils and pupillary reflex reaction. A cover test was conducted for heterophoria or heterotropia and the angle of deviation was measured using the corneal light reflex (Hirschberg test) and the Prism Cover Test at distance and near fixation, respectively. The ocular mo- tility test was performed to assess eye muscle function. Subjective refraction was determined using a standard refraction trial set to achieve best correct vision for chil- dren whose vision improved with the pinhole test. Chil- dren with VA ≤ 6/12 whose vision did not improve by pinhole test had outer eye and fundus examination by direct ophthalmoscopy, and any abnormal findings were recorded as causes of VI. Data analysis Data for each participant were analysed descriptively using standard deviations and percentages with SPSS version 22. The relationship between measures was de- termined using correlation, cross-tabulations and χ2 anal- ysis. For all statistical determinations, significance levels were established at P = 0.05. Results Study population A total of 1351 children were selected to participate in the study and 1204 (89%) were actually entered into the study. Demographic characteristics of participants The 1204 participants were aged 6–15 years, with a mean of 11.18 [standard deviation (SD); 2.45] years (Table 1). There were 658 (54.7%) boys and 546 girls (45.3%). The mean (SD) age of the boys and girls was 11.15 (2.47) and 11.21 (2.44) years, respectively. Most participants were aged 11 (14.2%) and 12 (13.7%) years, respectively. The ages with the fewest participants were 6 (3.2%) and 7 (5.1%) years, respectively. There was no significant difference in mean age between the boys and girls (ANOVA: F = 0.167, P = 0.683), although there was a significant differ- ence in mean ages of the children according to school grades (ANOVA: F = 341.733, P = 0.01). Distribution of ocular signs and symptoms A total of 943 (78.8%) participants did not complain of any ocular symptoms; 153 (12.7%) complained of blurred vision; 87 (7.2%) had itching and redness; and 15 (1.2%) had pain and photophobia. 1364 EMHJ – Vol. 26 No. 11 – 2020Research article VA A total of 1044 children presented with normal vision (6/6) in the right eye; 1034 had normal vision in the left eye; and 1071 had normal vision in the better eye (Table 2). Thirty-six, 38 and 42 children had uncorrected vision (6/9) in the right, left and better eye, respective- ly. An uncorrected VI was found in 164 children (13.6%, 95% CI, 11.7–15.5%), while 91 (7.6%, 95% CI, 6.1–9.1%) chil- dren had VI. With best-corrected VA, this decreased to 9 (0.75%, 95% CI, 0.3–1.2%) children. Prevalence of VI The prevalence of presenting VI was 91 (7.6%, 95% CI, 6.1–9.1%) and only 16 (17.6%) of these children were wear- ing spectacles. There were no significant association be- tween prevalence of VI and age (P = 0.209), sex (P = 0.060) and school grade (P = 0.393). Girls had a higher prevalence of VI (4.6%, 95% CI, 3.4–5.8) than boys had (2.6%, 95% CI, 1.7–3.5%). Younger children age 6–7 years had lower prev- alence of VI (2%, 95% CI, 1.2–2.8%) than those aged 10–11 years (3.4%, 95% CI, 2.4–4.4%), and the highest prevalence was in children aged 14–15 years (4.4%, 95% CI, 3.2–5.7%). Binocular anomalies Tropia was found in 9 (0.7%) children: 4 with esotropia and 5 with exotropia. Anterior-segment examination A total of 1104 children (91.7%, 95% CI, 90.1–93.3%) had no abnormalities detected in the right eye and 1103 (91.6%, 95% CI, 90.0–93.2) had no abnormalities in the left eye. Ninety-seven children (8.1%, 95% CI, 6.6–9.6%) had vernal keratoconjunctivitis in both eyes. Three children (0.25%, 95% CI, 0.03–0.5%) had trachoma in the left eye and 2 (0.2%, 95% CI, 0.1–0.5%) had trachoma in the right eye. One child had cataract in the right eye (0.1%, 95% CI, −0.1 to 0.3%) and one (0.1%, 95% CI, 0–0.3%) had corneal opaci- ty in the left eye. Table 1 Demographic characteristic of participants Age (years) Sex Total Male Female n % n % n % 6 20 3.0 18 3.3 38 3.2 7 35 5.3 27 4.9 62 5.1 8 58 8.8 40 7.3 98 8.1 9 58 8.8 57 10.4 115 9.6 10 95 14.4 62 11.4 157 13.0 11 92 14.0 79 14.5 171 14.2 12 80 12.2 85 15.6 165 13.7 13 83 12.6 66 12.1 149 12.4 14 71 10.8 58 10.6 129 10.7 15 66 10.1 54 9.9 120 10.0 Total 658 546 1204 Table 2 Distribution of uncorrected visual acuity for right, left and better eye by percentage and confidence interval UVA Right eye Left eye Better eye Best-corrected VA n % (95% CI) n % (95% CI) n % (95% CI) n % (95% CI) 6/6 1044 86.7 (84.8–88.6) 1034 85.9 (83.9–87.9) 1071 89.0 (87.2–90.8) 1182 98.1 (97.5–99.0) 6/9 36 3.0 (2.0–4.0) 38 3.2 (2.2–4.2) 42 3.5 (2.5–4.5) 13 1.08 (0.5–1.7) 6/12 21 1.7 (1.0–2.4) 28 2.3 (1.5–3.2%) 24 2.0 (1.2–2.8) 3 0.25 (0.03–0.53) 6/18 32 2.7 (1.8–3.6) 32 2.7 (1.8–3.6) 31 2.6 (1.7–3.5) 2 0.17 (0.0–0.4) 6/24 29 2.4 (1.5–3.3) 32 2.7 (1.8–3.6%) 19 1.6 (0.9–2.3) 2 0.17 (0.0–0.4) 6/36 23 1.9 (1.13–2.67) 19 1.6 (0.9–2.3) 9 0.7 (0.2–1.2) 2 0.17(0.0–0.4) 6/60 12 1.0 (0.4–1.6) 13 1.1 (0.5–1.7) 4 0.3 (0.01–0.61) — — CF 6 0.5 (0.1–0.9) 7 0.6 (0.16–1.04) 4 0.3 (0.01–0.61) — — HM 1 0.1 (0.0–0.3) 1 0.1 (0.0–0.3) — — — — Total 1204 100.0% 1204 100% 1204 100.0% 1204 %100.0 VA ≥ 6/12 91 7.6 (6.1–9.1) 9 0.75(0.3–1.2) CF = count fingers; CI = confidence interval; HM = hand movement; UVA = uncorrected visual acuity; VA = visual acuity. 1365 EMHJ – Vol. 26 No. 11 – 2020Research article Prevalence of RE A total of 189 children (15.7%, 95% CI, 13.7–17.8%) had REs, and 1015 (84.3%, 95% CI, 82.3–86.4%) were emme- tropic (Table 3). Myopia had the highest prevalence (n = 110, 9.1%), followed by astigmatism (n = 47, 3.9%) and hypermetropia (n = 32, 2.7%). The prevalence of RE was significantly associated with age (P = 0.011) but not sex (P = 0.073) or school grade (P = 0.168). Prevalence of REs was higher among girls (n = 100, 18.3%) than boys (n = 89, 13.5). Prevalence of REs significantly increased with age. Chil- dren aged 15 years had the highest prevalence (n = 21; 17.5%), compared to those aged 8 years (n = 14, 14.3%), 7 years (n = 10, 16.1%) and 6 years (n = 5, 13.2%). Children in school grade 4 had the highest prevalence of REs (n = 33, 21.9%), and those in grade 2 had the lowest prevalence (n = 16, 10.7%). The prevalence of myopia was increase with age; it was more common in children aged 15 years (10.8%) than in those aged 6 (5.3%), 7 (8.1%) and 8 (7.1%) years. In contrast, prevalence of hypermetropia was highest in children aged 6 years (5.3%), and lowest in those aged 15 years (0.8%). According to the gender the prevalence of myopia, hypermetropia and astigmatism was higher in girls at 10.3, 2.7 and 5.3%, respectively, than in boys at 8.2, 2.6 and 2.7%, respectively. Posterior-segment examination Posterior-segment examination revealed that 1196 chil- dren (99.3%, 95% CI, 98.8–99.8%) had no abnormalities. Ocular media and fundus abnormalities were seen in 8 (0.7%) children. Retinal disorders were found in 6 (0.5%) children and media opacity in 2 (0.2%). Principal causes of VI The causes of UVA of 6/12 or worse at least in 1 eye are presented in Table 4. RE was the main cause of VI in 126 (76.8%) affected children, followed by amblyopia (n = 36, 22.0%) and corneal opacity and cataract (n = 1, 0.6%). Schoolchildren who received eye drops or were referred One hundred and forty-two children (11.8%, 95% CI, 10.0– 13.6%) had uncorrected REs and were referred to Manhal Specialist Hospital, Hargeisa. Two children were referred for further examination and treatment of media opacity. Eighty-seven children (7.2%, 95% CI, 5.7–8.7%) were pre- scribed eye drops, and 15 (1.2%, 95% CI, 0.6–1.8%) received only advice for their complaints. Discussion Childhood blindness and VI are priority conditions tar- geted in Vision 2020: the Right to Sight Initiative of WHO Table 3 Prevalence of refractive error in one or both eyes by age, sex and school grade Category Emmetropia (95% CI, 82.3–86.4) Myopia (95% CI, 7.5–10.7) Hypermetropia (95% CI, 1.8–3.6) Astigmatism (95% CI, 2.8–5.0) no funds reflex (95% CI, 1.4–1.6) Total n % n % n % n % n % n % Age ( P = 0.011) 6 33 86.8 2 5.3 2 5.3 1 2.6 0 0 38 3.2 7 52 83.9 5 8.1 3 4.8 2 3.2 0 0 62 5.1 8 84 85.7 7 7.1 5 5.1 2 2.0 0 0 98 8.1 9 100 87.0 9 7.8 5 4.3 1 0.9 0 0 115 9.6 10 132 84.1 14 8.9 4 2.5 7 4.5 0 0 157 13.0 11 145 84.8 17 9.9 2 1.2 7 4.1 0 0 171 14.2 12 136 82.4 17 10.3 6 3.6 6 3,6 1 0.6 165 13.7 13 129 86.6 13 8.7 2 1.3 5 3.4 0 0 149 12.4 14 105 81.4 13 10.1 2 1.6 9 7.0 0 0 129 10.7 15 99 82.5 13 10.8 1 0.8 7 5.8 0 0 120 10.0 Sex ( P = 0.073) M 569 90.6 54 8.2 17 2.6 18 2.7 0 0 658 54.7 F 445 81.5 56 10.3 15 2.7 29 5.3 1 0.2 546 45.3 Class level ( P = 0.168) 1 129 87.2 6 4.1 7 4.7 6 4.1 0 0 148 12.3 2 134 89.3 10 6.7 3 2.0 3 2.0 0 0 150 12.5 3 122 81.3 17 11.3 7 4.7 4 2.7 0 0 150 12.5 4 118 78.1 22 14.6 3 2.0 7 4.6 1 0.7 151 12.5 5 130 86.1 12 8.0 4 2.6 5 3.3 0 0 151 12.5 6 130 85.5 11 7.2 3 2.0 8 5.3 0 0 152 12.6 7 122 80.8 18 11.9 3 2.0 8 5.3 0 0 151 12.5 8 129 85.4 14 9.3 2 1.3 6 4.0 0 0 151 12.5 Total 1015 84.3 110 9.1 32 2.7 47 3.9 1 0.1 1204 100% CI = confidence interval. 1366 EMHJ – Vol. 26 No. 11 – 2020Research article (17). Knowledge of the prevalence of RE and VI among school age children can help the relevant authorities to plan and provide eye care services in the particular geo- graphical area. The present study attempted to provide this information, as well as being the first study in So- maliland to assess the prevalence of the VI and RE among school-aged children. Noncycloplegic refraction was used to assess REs in this study, similar to studies of school-aged children in Nigeria (18) and South Africa (19). Noncycloplegic refraction was chosen so as not to interfere with the academic activity of the children. The prevalence of VI in the present study was 7.6%, which is lower than 10.1% in Malaysia (20) and 10.3% in China (21), but higher compared with 1.2% in South Africa (5) 1.2%, 2.67% in South America (22) and 3.5% in the Islamic Republic of Iran (23). These results indicate that VI among school-aged children requires urgent intervention by the community and nongovernmental organizations. The results also reflect lack of childhood eye care services in this region as well as lack of community awareness about the consequences of childhood VI. In the present study, the prevalence of VI was higher among girls than boys (4.6% vs 2.6%), which agrees with a study in Ethiopia (3.2% for girls and 2.6% for boys) (24). This might have been due to socioeconomic factors that contributed to better access to health services for boys. However, the difference was not significant. The prevalence of RE in either eye was 15.7%, which is lower than that in Ghana (25.6%) (25), India (25.1%) (26), Egypt (22.1%) (27) and Qatar (19.7%) (28), but higher than in Uganda (11.6%) (29), Ghana (13.3%) (30) and Saudi Arabia (13.7%) (31). The prevalence of RE in our study was similar to that in Viet Nam (16.3%) (32) and Saudi Arabia (16.3%) (33). This variation may be related to the type of sampling method used, size of population screened, and variation in geographic location. We found no significant association between prevalence of RE and school grade or sex. However, we did show that the prevalence of VI caused by uncorrected RE increased significantly with age. Nevertheless, we found that prevalence of RE was higher among girls than boys (81.3% vs 13.5%), which, as mentioned above, might have been due to better access to health care for boys in this culture. This is consistent with a similar study in Saudi Arabia (34). The prevalence of myopia was 9.1%, which is higher than 6.0% in Ethiopia (24) but lower than 14.1% in Ghana (30). In our study, older school children had a higher prevalence of myopia, which was similar to a study in Viet Nam (32). Alrasheed et al. (7) attributed this age- associated increase in myopia with decreased outdoor activity of many children and this has been reported as an issue in other studies (5,25,32). The prevalence of hypermetropia in this study was 2.7%, which is significantly lower than that reported in studies in Ethiopia 26.4% (24) and Saudi Arabia 6.9% (35). However, it is higher than in South Africa (1.8%) (10) and China (1.6%) (36) but similar to Tunisia (2.61%) (37). The lower prevalence of hypermetropia in our study might have been due to use of noncycloplegic refraction, which could have missed a significant number of cases of hypermetropia. The prevalence of hypermetropia decreased with age and was higher in children aged 6 and 7 years compared with 14 and 15 years. This result agreed with Chebil et al. (37), who reported that this variation could be related to a decrease in the dioptric power of the lens (it goes form 23 D at age 3 years to 20 D at 14 years), or with an increase in the optical density of the crystalline cortex. The prevalence of astigmatism in the current study was 3.9%. This is lower than that found in the Islamic Republic of Iran (6.6%) (38) and South Africa (14.6%) (5) but similar to that in Poland (4%) (39). The prevalence of manifest strabismus was 0.7%, which is similar to that among children in the United Republic of Tanzania (0.5%) (40) but lower than in Iranian school children (1.2%) (41). In this study, uncorrected RE was the most common cause of VI among children, and was responsible for 76.8% of cases. This is similar to other studies that used RESC protocol, such as in Ethiopia (77.3%) (24) and India (77%) (26) but lower than in Malaysia (87.0%) (20) and in the Islamic Republic of Iran (87.3%) (23). Alrasheed et al. (7) suggested that this could have been because of genetic differences as well as different lifestyles in terms of outdoor activities. The second most frequent cause of VI among children was amblyopia at 22.0%, which is higher than in Sudan (5.6%) (6) and South Africa (9.6%) (4). This may be due to the high rate of poverty and illiteracy in Somaliland and the poor health system in the country. Table 4 Causes of uncorrected visual acuity 6/12 or worse Causes Children with VA 6/12 or worse in one or both eyes Prevalence in the population in one or both eyes, % (95% CI) n % Refractive error 126 76.8 10.5 (8.8–12.2) Amblyopia 36 22.0 3.0 (2.04–4.0) Corneal opacity 1 0.6 0.08 (−0.08 to 0.24) Cataract 1 0.6 0.08 (0.08–0.24) Any cause 164 100.0 13.6 (11.7–15.5) CI = confidence interval; VA = visual acuity. 1367 EMHJ – Vol. 26 No. 11 – 2020Research article In the present study, out of 91 children with VI, only 16 (17.6%) were already using spectacles, while the rest were not aware of the problem. This may have been due to lack of child and parental awareness of the vision problem, attitudes regarding the need for spectacles, cost of spectacles, cosmetic appearance, peer pressure and concerns that wearing glasses may cause progression of RE (42,43). This study had several limitations. First, a large number of schools were not registered with the Ministry of Education in Hargeisa, so the study sample did not include all schools. Second, almost half of school-age children were not attending school due to poverty, thus the study only included children who attended school. Third, places of study and examination differed among schools in terms of lighting, ventilation and comfort. Fourth, distribution of children’s ages at school levels was not uniform, so older children were not only in the eighth and seventh grades, and children aged 6 and 7 years were less prevalent in this study, because many children in Somaliland start school later than the recommended 6 years. Fifth, Log Mar charts were not available, so we used Snellen Tumbling E-charts, and slit lamps and fundus biomicroscopy were not easy to transport between locations, so they were replaced by torch and magnifier, and ophthalmoscopy. Finally, RE was assessed by noncycloplegic refraction, which could have missed a significant number of cases of hypermetropia. Conclusions The prevalence of VI among school-aged children in Har- geisa, Somaliland was high and the commonest causes were uncorrected REs. There are barriers to care and it is critical that they are overcome. Acknowledgement We are grateful to all schools included in this study and their staff for help with data collection. We also thank all the students who participated in this study and their parents, as well as the students of the Faculty of Optometry and Visual Science, University Of Hargeisa. We are grateful to Manhal Specialist Hospital ,which provided us with the required in- strumentation, and free treatment for any student who was referred to the hospital. Funding: None. Competing interests: None declared. Prévalence du vice de réfraction et de la déficience visuelle chez les enfants d’âge sco- laire de Hargesia, au Somaliland (Somalie) Résumé Contexte : Les troubles visuels chez l’enfant constituent un problème de santé publique mondial, en particulier dans les pays en développement. Ses causes les plus courantes sont évitables par un diagnostic précoce et la mise en place rapide d’un traitement. Objectifs : La présente étude avait pour objectif d’évaluer la prévalence du vice de réfraction et de la déficience visuelle chez les enfants d’âge scolaire à Hargeisa, au Somaliland (Somalie). Méthodes : Il s’agissait d’une étude transversale menée auprès de 1204 élèves (âgés de 6 à 15 ans) de huit écoles primaires sélectionnées de manière aléatoire à Hargeisa, entre novembre 2017 et janvier 2018. Nous avons utilisé l’étude modifiée sur le vice de réfraction chez les enfants afin de déterminer la prévalence de cette affection et de la déficience visuelle, comprenant les examens suivants : acuité visuelle à distance, évaluée par l’échelle de Snellen avec des E ou C directionnels ; la réfraction, évaluée par examen de la vision binoculaire par rétinoscopie ; et examen des segments antérieurs et postérieurs. Résultats : La prévalence de la déficience visuelle non corrigée, détectée et la mieux corrigée de 6/12 ou un score inférieur était respectivement de 13,6 %, de 7,6 % et de 0,75 %. Seuls 16 enfants sur 91 (17,6 %) portaient des lunettes tandis que les autres n’avaient pas conscience du problème. Le vice de réfraction était la cause de la déficience visuelle chez 76,8 % des participants, l’amblyopie dans 22,0 % des cas, le trachome chez 2,4 % des enfants, et l’opacité cornéenne et la cataracte chez 0,6 % d’entre eux. Des anomalies du segment antérieur ont été observées chez 8,3 % des participants à l’étude, principalement des kératoconjonctivites vernales, tandis que des anomalies du segment postérieur ont été observées dans 0,7 % des cas. La prévalence de la myopie était de 9,1 %, celle de l’hypermétropie de 2,7 % et celle de l’astigmatisme de 3,9 %. La prévalence de la déficience visuelle due à un vice de réfraction était associée à un âge supérieur, sans toutefois que l’on puisse noter de corrélation significative avec le niveau scolaire ou le sexe. Conclusion : La prévalence de la déficience visuelle chez les enfants d’âge scolaire de Hargesia était élevée, principalement en raison d’un vice de réfraction non corrigé. Il existe des obstacles aux soins et il est essentiel de les surmonter. 1368 EMHJ – Vol. 26 No. 11 – 2020Research article References 1. Bourne RRA, Flaxman SR, Braithwaite T, Cicinelli MV, Das A, Jonas JB et al. Magnitude, temporal trends, and projections of the global prevalence of blindness and distance and near vision impairment: a systematic review and meta-analysis. Lancet Glob Health. 2017;5(9):e888–97. http://dx.doi.org/10.1016/S2214-109X(17)30293-0 PMID:28779882 2. Flaxman SR, Bourne RRA, Resnikoff S, Ackland P, Braithwaite T, Cicinelli MV et al. Global causes of blindness and distance vision impairment 1990–2020: a systematic review and meta-analysis. Lancet Glob Health. 2017;5(12):e1221–34. http://dx.doi. org/10.1016/S2214-109X(17)30393-5 PMID:29032195 3. Kong L, Fry M, Al-Samarraie M, Gilbert C, Steinkuller PG. An update on progress and the changing epidemiology of causes of childhood blindness worldwide. J AAPOS. 2012 Dec;16(6):501–7. http://dx.doi.org/10.1016/j.jaapos.2012.09.004 PMID:23237744 4. Benjamin WJ. Borish’s clinical refraction E-book. Elsevier Health Sciences; 2006 5. Naidoo KS, Raghunandan A, Mashige KP, Govender P, Holden BA, Pokharel GP, Ellwein LB. Refractive error and visual impair- ment in African children in South Africa. Invest Ophthalmol Vis Sci. 2003 Sep;44(9):3764–70. http://dx.doi.org/10.1167/iovs.03- 0283. PMID:12939289 6. Zeidan Z, Hashim K, Muhit MA, Gilbert C. Prevalence and causes of childhood blindness in camps for displaced persons in Khar- toum: results of a household survey. East Mediterr Health J. May–Jun 2007;13(3):580–5. PMID:17687831 7. Alrasheed SH, Naidoo KS, Clarke-Farr PC. Prevalence of visual impairment and refractive error in school-aged children in South Darfur State of Sudan. Afr Vis Eye Health. 2016 Oct 27;75(1):1–9. https://doi.org/10.4102/aveh.v75i1.355 8. Mehari ZA, Yimer AW. Prevalence of refractive errors among schoolchildren in rural central Ethiopia. Clin Exp Optom. 2013 Jan;96(1):65–9. http://dx.doi.org/10.1111/j.1444-0938.2012.00762.x PMID:22784031 9. Courtright P, Hutchinson AK, Lewallen S. Visual impairment in children in middle-and lower-income countries. Arch Dis Child. 2011 Dec 1;96(12):1129–34. http://dx.doi.org/10.1136/archdischild-2011-300093 PMID:21868404 10. Holden BA. The role of optometry in Vision 2020. Community Eye Health. 2002;15(43):33–6. PMID:17491876 11. Logan N. The development of refractive error. In: Optometry: science, techniques and clinical management. 2nd ed. Oxford: Elsevier Health Sciences; 2009 Jun 22:159–71. 12. Assessment of the prevalence of visual impairment attributable to refractive error or other causes in school children. Protocol and manual of procedures. Geneva: World Health Organization; 2007 (https://www.who.int/blindness/causes/RESCProtocol. pdf?ua=1, accessed 20 May 2020). )لاموصلا( دنلايلاموص ،اسيجره في ةساردلا نس في لافطلأا ينب صربلا فعضو يراسكنلاا أطلخا راشتنا لدعم يدماغلا ديلو ،ديشرلا فيس ،دحمأ يدبع ةرهز ةصلالخا .لخدلا ةضفخنلما نادلبلا في ماَّيس لاو ،يلماعلا ديعصلا لىع ةماعلا ةحصلا تلاكشم نم ًةلكشم ةلوفطلا ةلحرم في صربلا فعض ُّدَعُي :ةيفللخا .نْيَركبُلما جلاعلاو صيخشتلا قيرط نع ًاعويش رثكلأا هبابسأ بنتج نكميو دنلايلاموص ،اسيجره في ةساردلا نس في لافطلأا ينب صربلا فعضو يراسكنلاا أطلخا راشتنا لدعم مييقت لىإ ةساردلا هذه تفده :فادهلأا )لاموصلا( في ًايئاوشع ةراتمخ ةيئادتبا سرادم 8 في )ةنس 15و تاونس 6 ينب مهرماعأ حواترت( بلاط 1204 ةيعطقلما ةساردلا هذه تلمش :ثحبلا قرط لافطلأا ىدل ةل َّدعلما يراسكنلاا أطلخا ةسارد انمدختساو .2018 نياثلا نوناك/رياني لىإ 2017 نياثلا نيشرت /برمفون نم ةترفلا للاخ اسيجره ططمخ ةطساوب اهمييقت مت يتلا ،دعُب نع صربلا ةدح :ةيلاتلا تاءاصقتسلاا كلذ في ماب ،صربلا فعضو يراسكنلاا أطلخا راشتنا لدعم ديدحتل .ةيفللخاو ةيماملأا ءازجلأا صحفو ،ةيكبشلا راظنمب يننيعلا ةيؤر مييقت للاخ نم همييقت مت يذلا ،راسكنلااو ،Snellen Tumbling E-chart لىع %0.75و %7.6و %13.6 هجو لضفأ لىع ح َّحصُلماو نِلعتسُلماو ح َّحصُلما يرغ لقأ وأ 12/6 ةميقب صربلا فعض راشتنا لدعم غلب :جئاتنلا وه صربلا فعض ببس ناكو .ةلكشملل ينكردم يرغ نوقابلا ناكو تاراظنلا نومدختسي )%17.6( ًلافط 91 ينب نم طقف 16 ناكو .لياوتلا .ينكراشلما نم %0.6 في ّداسلاو ةَِّينْرَقلا ةماَتَعو ،مهنم %2.4 في اموكاترلاو ،مهنم %22.0 في شَمَغلاو ،ينكراشلما نم %76.8 في يراسكنلاا أطلخا تاهوشت تظحول ينح في ،يعيبرلا ةمحتللماو ةَِّينْرَقلا باهتلا صوصلخا هجو لىعو ،ينكراشلما نم %8.3 في يماملأا ءزلجا في تاهوشت تدجُوو مجانلا صربلا فعض راشتنا لدعم طبتراو .%3.9 ةيرؤبلالاو ،%2.7 صربلا دمو ،%9.1 َسَلحا راشتنا لدعم غلبو .مهنم %0.7 في يفللخا ءزلجا في .سنلجا عون وأ سيردلما فصلاب مهم طابترا كانه نكي لم نكلو ،رمعلا ةدايزب يراسكنلاا أطلخا نع يرغ يراسكنلاا أطلخا هءارو سييئرلا ببسلا ناكو ،ًاعفترم اسيجره في ةساردلا نس في لافطلأا ينب صربلا فعض راشتنا لدعم ناك :تاجاتنتسلاا .اهيلع بلغتلا متي نأ ناكمب ةيهملأا نمو ،ةياعرلا ميدقت نود لوتح قئاوع ة َّمَثو .ح َّحصُلما 1369 EMHJ – Vol. 26 No. 11 – 2020Research article 13. Somalia: Somaliland, including government structure, security, and presence of internally displaced persons (IDPs) from Somalia (2016–March 2018) [website]. Geneva: United Nations High Commissioner for Refugees (https://www.refworld.org/ docid/5ad453514.html, accessed 20 May 2020). 14. New world Bank GDP and poverty estimates for Somaliland [website]. Washington DC: World bank; 2014 (https://www.worldP- bank.org/en/news/press-release/2014/01/29/new-world-bank-gdp-and-poverty-estimates-for-somaliland, accessed 20 May 2020). 15. Muma MK, Kimani K, Kariuk, Wanyike MM. Prevalence and significant refractive errors in primary school children of a rural district of Kenya. East Afr J Ophthalmol. 2007;13(2):48–51. 16. Minassian D. Sample size calculation for eye surveys: a simple method. Community Eye Health. 1997;10(23):42–44. https://www. cehjournal.org/article/sample-size-calculation-for-eye-surveys-a-simple-method/ 17. The World Health Report 1998: life in the 21st century a vision for all. Geneva: World Health Organization; 1998 (https://www. who.int/whr/1998/en/, accessed 20 May 2020). 18. Adegbehingbe BO, Oladehinde MK, Majemgbasan TO, Onakpoya HO, Osagiede EO. Screening of adolescents for eye diseases in Nigerian high schools. Ghana Med J. 2005 Dec;39(4):138–42. PMCID:PMC1790830 19. Mabaso RG, Oduntan AO, Mpolokeng MB. Refractive status of primary school children in Mopani district, Limpopo Province, South Africa. Afr Vis Eye Health. 2006 Dec 19;65(4):125–33. https://doi.org/10.4102/aveh.v65i4.267 20. Goh PP, Abqariyah Y, Pokharel GP, Ellwein LB. Refractive error and visual impairment in school-age children in Gombak Dis- trict, Malaysia. Ophthalmology. 2005 Apr 1;112(4):678–85. http://dx.doi.org/10.1016/j.ophtha.2004.10.048 PMID:15808262 21. He M, Huang W, Zheng Y, Huang L, Ellwein LB. Refractive error and visual impairment in school children in rural southern China. Ophthalmology. 2007 Feb ;114(2):374–82. http://dx.doi.org/10.1016/j.ophtha.2006.08.020 PMID:17123622 22. Furtado JM, Lansingh VC, Carter MJ, Milanese MF, Peña BN, Ghersi HA, et al. Causes of blindness and visual impairment in Latin America. Surv Ophthalmol. 2012 Mar–Apr;57(2):149–77. http://dx.doi.org/10.1016/j.survophthal.2011.07.002 PMID:22137039 23. Fotouhi A, Hashemi H, Khabazkhoob M, Mohammad K. The prevalence of refractive errors among schoolchildren in Dezful, Iran. Br J Ophthalmol. 2007 Mar;91(3):287–92. http://dx.doi.org/10.1136/bjo.2006.099937 PMID:17035280 24. Yared AW, Belaynew WT, Destaye S, Ayanaw T, Zelalem E. Prevalence of refractive errors among school children in Gondar Town, Northwest Ethiopia. Middle East Afr J Ophthalmol. 2012 Oct;19(4):372–6. http://dx.doi.org/10.4103/0974-9233.102742 PMID:23248538 25. Ovenseri-Ogbomo GO, Omuemu VO. Prevalence of refractive error among school children in the Cape Coast Municipality, Ghana. Clin Optom. 2010 Jul 3;2:59–66. 26. Padhye AS, Khandekar R, Dharmadhikari S, Dole K, Gogate P, Deshpande M. Prevalence of uncorrected refractive error and other eye problems among urban and rural school children. Middle East African journal of ophthalmology. 2009 Apr;16(2):69–74. http://dx.doi.org/10.4103/0974-9233.53864 PMID:20142964 27. El-Bayoumy BM, Saad A, Choudhury AH. Prevalence of refractive error and low vision among schoolchildren in Cairo. East Mediterr Health J. 2007 May–Jun;13(3):575–80. PMID:17687830 28. Al-Nuaimi AA, Salama RE, Eljack IE. Study of refractive errors among school children Doha. World Fam Med J. 2010;8(7):41–8. 29. Kawuma M, Mayeku R. A survey of the prevalence of refractive errors among children in lower primary schools in Kampala district. Afr Health Sci. 2002 Aug;2(2):69–72. PMID:12789105 30. Ovenseri-Ogbomo GO, Assien R. Refractive error in school children in Agona Swedru, Ghana. Afr Vis Eye Health. 2010 Dec 11;69(2):86–92. https://doi.org/10.4102/aveh.v69i2.129 31. Al Wadaani FA, Amin TT, Ali A, Khan AR. Prevalence and pattern of refractive errors among primary school children in Al Hassa, Saudi Arabia. Glob J Health Sci. 2012 Nov 11;5(1):125–34. http://dx.doi.org/10.5539/gjhs.v5n1p125 PMID:23283044 32. Paudel P, Ramson P, Naduvilath T, Wilson D, Phuong HT, Ho SM, et al. Prevalence of vision impairment and refractive error in school children in Ba Ria –Vung T au province, Vietnam. Clin Exp Ophthalmol. 2014 Apr;42(3):217–26. http://dx.doi.org/10.1111/ ceo.12273 PMID:24299145 33. Aldebasi YH. Prevalence of correctable visual impairment in primary school children in Qassim Province, Saudi Arabia. J Optom. 2014 Jul–Sep;7(3):168–76. http://dx.doi.org/10.1016/j.optom.2014.02.001 PMID:25000873 34. Dandona R, Dandona L, Srinivas M, Sahare P, Narsaiah S, Munoz SR, et al. Refractive error in children in a rural population in India. Invest Ophthalmol Vis Sci. 2002 Mar;43(3):615–22. PMID:11867575 35. Al-Rowaily MA. Prevalence of refractive errors among pre-school children at King Abdulaziz Medical City, Riyadh, Saudi Arabia. Saudi Journal of Ophthalmology. 2010 Apr;24(2):45–8. http://dx.doi.org/10.1016/j.sjopt.2010.01.001 PMID:23960874 36. Li Z, Xu K, Wu S, Lv J, Jin D, Song Z, Wang Z, Liu P. Population‐based survey of refractive error among school‐aged children in rural northern China: the Heilongjiang Eye Study. Clin Exp Ophthalmol. 2014 May–Jun;42(4):379–84. http://dx.doi.org/10.1111/ ceo.12198 PMID:23952961 37. Chebil A, Jedidi L, Chaker N, Kort F, Limaiem R, Mghaieth F, et al. Characteristics of astigmatism in a population of Tuni- sian school-children. Middle East Afr J Ophthalmol. 2015 Jul–Sep;22(3):331–4. http://dx.doi.org/10.4103/0974-9233.150635 PMID:26180472 1370 EMHJ – Vol. 26 No. 11 – 2020Research article 38. Khalaj M, Gasemi M, Zeidi IM. Prevalence of refractive errors in primary school children [7–15 years] of Qazvin City. Eur J Sci Res. 2009;28(2):174–85. 39. Czepita D, Mojsa A, Ustianowska M, Czepita M, Lachowicz E. Prevalence of refractive errors in schoolchildren ranging from 6 to 18 years of age. Ann Acad Med Stetin. 2007;53(1):53–6. PMID:18561610 40. Wedner SH, Ross DA, Balira R, Kaji L, Foster A. Prevalence of eye diseases in primary school children in a rural area of Tanzania. Br J Ophthalmol. 2000 Nov;84(11):1291–7. http://dx.doi.org/10.1136/bjo.84.11.1291 PMID:11049957 41. Jamali P, Fotouhi A, Hashemi H, Younesian M, Jafari A. Refractive errors and amblyopia in children entering school: Shahrood, Iran. Optom Vis Sci. 2009 Apr;86(4):364–9. http://dx.doi.org/10.1097/OPX.0b013e3181993f42 PMID:19289975 42. Alrasheed SH, Naidoo KS, Clarke-Farr PC. Attitudes and perceptions of Sudanese high-school students and their parents towards spectacle wear. Afr Vis Eye Health. 2017 Apr 11;77(1):1–7. https://doi.org/10.4102/aveh.v77i1.392 43. Alrasheed SH, Naidoo KS, Clarke-Farr PC, Binnawi KH. Building consensus for the development of child eye care services in South Darfur State in Sudan using the Delphi technique. Afr J Prim Health Care Fam Med. 2018 Oct 24;10(1):1–9. http://dx.doi. org/10.4102/phcfm.v10i1.1767 PMID:30456975 1371 EMHJ – Vol. 26 No. 11 – 2020Research article Risk factors associated with worse outcomes in COVID-19: a retrospective study in Saudi Arabia Anas Khan,1,2 Saqer Althunayyan,3 Yousef Alsofayan,2 Raied Alotaibi,4 Abdullah Mubarak,4 Mohammed Arafat,1 Abdullah Assiri5 and Hani Jokhdar5 1Department of Emergency Medicine, College of Medicine, King Saud University, Riyadh, Saudi Arabia. 2Global Center for Mass Gatherings Medicine, Ministry of Health, Riyadh, Saudi Arabia. 3Department of Accident and Trauma, Prince Sultan Bin Abdulaziz College for Emergency Medical Services, King Saud University, Riyadh, Saudi Arabia. 4Department of Basic Science, Prince Sultan Bin Abdulaziz College for Emergency Medical Services, King Saud University, Riyadh, Saudi Arabia. 5Ministry of Health, Riyadh, Saudi Arabia. (Correspondence to: Yousef Alsofayan: y-m-alsofayan@hotmail.com; yalsofayan@moh.gov.sa). Abstract Background: The rapid emergence of the novel coronavirus disease 2019 (COVID-19) has resulted in millions of infected patients and hundreds of thousands of deaths worldwide. Health care services delivery is being compromised due to the surge in the number of infected patients during this pandemic. Aims: This study aimed to assess the risk factors associated with poor prognosis among COVID-19 patients in Saudi Arabia. Methods: This was a multi-centre retrospective cohort study that included all laboratory-confirmed COVID-19 cases with definitive outcomes in Saudi Arabia during March 2020. Demographic, clinical history, comorbidity and outcomes data were retrieved from the National Health Electronic Surveillance Network (HESN) database. We used logistic regression models to calculate crude and adjusted odds ratios (OR) to explore risk factors for critical outcomes (intensive care unit admission or death) among COVID-19 cases. Results: We included 648 COVID-19-positive patients with a median age of 34 years. Of these, 11.9% were in the critical group. Risk factors associated with worse outcomes included males (OR=1.92), age >60 years (OR=3.65), cardiac diseases (OR=3.05), chronic respiratory diseases (OR=2.29), and cases with two or more comorbidities (OR=2.57) after adjusting for age and sex; all had significant P-values <0.05. Conclusions: Independent risk factors for critical outcomes among COVID-19 cases include old age, males, cardiac patients, chronic respiratory diseases, and the presence of two or more comorbidities. We recommend designing a unique multi-item scale system to prognosticate COVID-19 patients. Keywords: COVID-19, risk factors, ICU admission, mortality, Saudi Arabia Citation: Khan A; Althunayyan S; Alsofayan Y; Alotaibi R; Mubarak A; Arafat M; et al. Risk factors associated with worse outcomes in COVID-19: a retro- spective study in Saudi Arabia. East Mediterr Health J. 2020;26(11):1371–1380. https://doi.org/10.26719/emhj.20.130 Received: 24/06/20; accepted: 14/09/20 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Introduction The rapid emergence of the novel Coronavirus Disease 2019 (COVID-19) has resulted in millions of infected pa- tients and hundreds of thousands of deaths worldwide. In Saudi Arabia, the total number of confirmed COVID-19 cases as of 13 June 2020 reached 123 308 confirmed cas- es and 932 deaths and is increasing daily (1). Its spread is becoming difficult to control and efforts should focus on effective mitigation measures to minimize the disease impact on those prone to developing adverse outcomes. Internationally, health-care services delivery is being compromised due to the surge in the number of infected patients during this COVID-19 pandemic. Overwhelming the health-care system will lead to an unexpected rise in morbidity and mortality of various treatable conditions. Therefore, it is critical to risk stratify COVID-19 patients based on their predicted outcomes and guide appropriate management and disposition accordingly. The clinical manifestation of COVID-19 is broad and ranges from asymptomatic and mild upper respiratory tract symptoms to severe illnesses with multiorgan failure and death (2–4). Furthermore, it is challenging to predict the clinical course or determine patients at risk of deterioration. Previous reports showed that old age and male gender are risk factors for disease severity and mortality (5,6). Other medical comorbidities are associated with poor prognoses such as cardiovascular disease, diabetes mellitus, chronic respiratory disease, and hypertension (7,8). Moreover, distinct signs and symptoms or even laboratory findings are correlated with worse outcomes (5). Nevertheless, these studies’ results are difficult to generalize in the Eastern Mediterranean Region because the reported clinical predictors of mortality were studied in different population groups, and clinical characteristics can be different (9). In this cohort study, we assessed the risk factors of ICU admission or death among COVID-19 patients 1372 EMHJ – Vol. 26 No. 11 – 2020Research article in Saudi Arabia including clinical features, common comorbidities, and a number of laboratory findings. Methods This retrospective cohort study collected data from all laboratory-confirmed COVID-19 cases located in health care facilities across all regions of Saudi Arabia in March 2020. Health-care facilities are mandated to enter the demographic, clinical, laboratory and outcomes data of COVID-19-positive patients in the National Health Electronic Surveillance Network (HESN) database by the health-care providers under the supervision of the Ministry of Health (MoH). All confirmed cases of COV- ID-19 were screened and only patients with definitive outcomes were included; those in the active phase were excluded. Demographic, clinical, laboratory, comorbidity, and outcomes data of COVID-19 positive patients were retrieved from the HESN database and extracted into electronic sheets by two data collectors. Any discrepancies were solved by a 3rd independent reviewer based on the medical reports. In parallel with the World Health Organization (WHO) protocols, COVID-19 was diagnosed based on the results of quantitative RT-PCR testing from nasopharyngeal samples (10). We analyzed age in three different forms: continuous, binary and 20- year intervals based on the risk stratification for severe disease in patients with COVID-19 (11). Fever was defined as a temperature of 38 ֯C or higher, high respiratory rate was defined as more than 24 breaths per minute and low oxygen saturation was defined as less than 94% (5,9,12). Comorbidities were classified based on the International Classification of Diseases, Revision 10 (ICD-10) diagnostic codes, then we reported any comorbidity, one or more comorbidity, and two or more comorbidities (8,13). Lymphocytopenia was defined as a lymphocyte count of less than 1500 per cubic millimetre (14). The primary endpoint of our study was ICU admission, death or recovery. The secondary endpoint was the in- hospital length of stay (LOS) in days. Patients included in our study were classified into critical and non-critical groups. Critical cases were defined as patients with ICU admission or death. Non-critical cases were defined as recovered patients with hospital discharge without ICU admission in accordance with MoH coronavirus disease guidelines (15). This study was approved by the MoH Institutional Review Board Central Committee (Approval number 20-75 M). Data privacy and confidentiality were maintained throughout the study as subjects were labeled with unique identification numbers; data concealment was maintained throughout the study by generating strong passwords for the electronic system and limiting access to designated investigators after signing non- disclosure agreement forms. Descriptive statistics were used to describe categorical variables that were presented by counts and percentages. In contrast, continuous variables were based on the median and interquartile range (IQR) since the normality test was significant using the Kolmogorov-Smirnov test and the Shapiro-Wilk test. The non-parametric Mann Whitney U-test was used to compare two numerical groups. Categorical variables underwent a test of association using the Chi-square test or Fisher exact test when the number of cases was small. Univariable and multivariable logistic regression models were used to obtain the crude and the adjusted odds ratio (OR) and their associated 95% confidence interval (CI). All percentages were rounded to one decimal place. The statistical significance was set to a P-value of < 0.05. The analysis was done using Statistical Package for the Social Sciences 24 (IBM-SPSS-24). Results Between 1–31 March 2020 a total of 1519 COVID-19-posf- itive cases were screened, 648 patients with definitive outcomes were included in the analysis; 11.9% (n=77) of them were critical while 88.1% (n=571) were non-criti- cal (Figure 1). Out of the 77 critical patients, 15.6% (n=12) patients have died and 84.4% (n=65) have recovered (Figure 1). The distribution of gender varied between critical and non-critical groups where males constituted 67.5% (n=52) of the critical group versus 50.8% (n=290) of the non-critical group, with a statistically signifi- cant association P = 0.006. The median and IQR of age differed significantly P = 0.001 across critical cases vera- sus non-critical cases with 37 years (27) and 33 years (18), respectively. Smoking was not associated with the worse outcomes, with a P-value of 0.943 (Table 1). Additionally, the extracted outcomes were presented by age and gen- der (Figure 2). Comorbidities were seen in a high percentage of cases where 29% (n=188 of 648) had one or more. Comorbidities were higher in the critical group with 42.9% (n=33) having one or more comorbidity, and 27.3% (n=21) having two or more comorbidities. On the other side, 27.1% (n=155) of the non-critical patients had one or more comorbidity, and 10% (n=57) had two or more comorbidities. The association between the presence of one or more comorbidity was significantly associated with outcomes with P=0.004. The comorbidities included diabetes mellitus (DM), hypertension (HTN), cardiac diseases, chronic respiratory diseases (CRD), cancer, immunodeficiency, and chronic kidney diseases (CKD). Diabetes was seen in 11.3% (n=73 of 648) of the cases with 20.8% (n=16) in the critical arm versus 10% (n=57) in the non-critical arm (P=0.005). Cardiac and cancer/ immunodeficient patients were 3.5% (n=23 of 648) and 2.8% (n=18 of 648) of the cases, respectively. Cardiac patients were present in 10.4% (n=8) of the critical group versus 2.6% (n=15) in the non-critical group P=0.001. Similarly, cancer and immunodeficiency were seen in 6.5% (n=5) of the critical arm versus 2.3% (n=13) in the non-critical arm. Among cases, CRD was reported in 12.5% (n=81 of 648) of the population with 22.1% (n=17) in the critical patients arm versus 11.2% (n=64) in the non- 1373 EMHJ – Vol. 26 No. 11 – 2020Research article critical arm with statistically significant differences P=0.007 (Table 1). Fever was seen in 85.3% (n=163 of 191), with a slightly higher percentage 94.7% in the critical group (n=36) versus 83% (n=127) in the non-critical group. However, there was no statistical difference across the two groups. Cough was reported in 89% (n=203 of 228) of the total group with a significant difference across the two groups outcomes P=0.049. Sore throat, runny nose, and headache were reported in 79.3% (n=115 of 145), 74.5% (n=73 of 98) and 27.8% (n=140 of 503) respectively. None of those symptoms showed significant differences across critical and non-critical arms. Gastrointestinal (GI) symptoms and myalgia were reported by 14.1% (n=71 of 503) and 28.8% (n=145 of 503), respectively. In addition, there were no statistical differences observed across the two arms outcomes (Table 1). Vital signs were evaluated for the population based on their continuous scale or predefined categories. However, for vital signs and laboratory results, there were missing data for a high number of cases. There was no statistical difference in both continuous and categorical presentation where heart rate ≥ 100 beats per minute occurred in 36.8% (n=7) of the critical group versus 20.8% (n=20) of the non-critical group. Similarly, the respiratory rate had a median (IQR) of 20 (2) breaths per minute. The cut-off for the categories was taken as 24 breaths per minute. No significant difference occurred in the groups’ outcomes. Oxygen saturation had a median (IQR) of 98 (3) % with a cut-off value of 94%. The comparison across the two groups outcomes yielded no significant differences either. Finally, systolic and diastolic blood pressures had a median (IQR) of 125 (22) mmHg and 74 (12) mmHg, respectively, with no significant differences in the critical versus non-critical groups (Table 1). Neutrophils as well as WBC total count and lymphocytes percentages had no statistically significant differences across the groups of these variables (Table 1). The overall in-hospital length of stay (LOS) in days had a median (IQR) of 5 (14) days. There was a significant difference in the LOS P=0.001 as critical patients had longer LOS with a median (IQR) of 11.5 (11) days versus 4 (12) days for non-critical cases (Table 1). The LOS was illustrated for different risk groups (Figure 3). Older patients (age ≥65), diabetic, hypertensive, CKD, and cancer patients/immunodeficient had a longer median LOS of 10 days. Patients without comorbidities had the lowest median LOS with only two days (Figure 3). Several risk factors were explored using logistic regression with outcomes being binary as critical or non- critical. Crude and adjusted ORs were calculated using age as a continuous variable and gender (Table 2). Male gender was found to be a statistically significant risk factor P=0.012 with a crude OR and 95% CI of 2.01 (1.22– 3.34). The OR of male gender became 1.92 (1.15–3.20) after adjusting for age. Older age was found to be significant on the continuous and categorical scales. Those with an age of ≥ 65 years had 3.15 (1.40–7.09) higher odds ratios of experiencing ICU admission or death with significant P=0.007. Moreover, age categories showed an increasing trend of being in the critical group where the age group 41–60 years had an OR=1.90 (0.69–5.25) and those >60 years had OR=4.04 (1.32–12.36) against the 1–20 years reference group. Being a smoker had an OR=0.98 (0.51–1.88) with a non- significant P-value that remained non-significant even after adjusting for age and sex. Comorbidities showed significantly increased odds of being in the critical group with one or more comorbidity having OR=2.01 (1.24– 3.28), two or more comorbidities had OR=3.38 (1.91–5.99), DM 2.37 (1.28–4.37), and HTN 2.28 (1.23–4.20). These comorbidities had their adjusted OR decreased indicating that the age and gender are potential confounders. Patients with CKD had an OR=1.50 (0.32–6.96) that remained non-significant when adjusted for age and sex. On the other hand, CRD was considered to be a significant risk factor with an adjusted OR=2.29 (1.24– Figure 1 Study flowchart for COVID-19 laboratory-confirmed cases with definitive outcomes in Saudi Arabia, 1–31 March, 2020 COVID-19 cases with definitive outcomes N=648 Excluded Active COVID-19 cases N=871 Non Critical N=571 Critical N=77 ICU admission and Recovery N=65 ICU admission and Death N=12Confirmed COVID-19 cases N=1519 1374 EMHJ – Vol. 26 No. 11 – 2020Research article Table 1 Baseline characteristics of the COVID-19 cases based on their reported outcomes Characteristic Total Patients (N=648) Critical (N=77) Non-critical (N=571) P-value* Sex* 0.006 Male 342 (52.8%) 52 (67.5%) 290 (50.8%) Female 306 (47.2%) 25 (32.5%) 281 (49.2%) Age* (MD, IQR) -years 34 (19) 37 (27) 33 (18) 0.001 Age* Categories -years 0.001 1-20 69 (10.6%) 5 (6.5%) 64 (11.2%) 21-40 359 (55.4%) 38 (49.4%) 321 (56.2%) 41-60 170 (26.2%) 22 (28.6%) 148 (25.9%) >60 50 (7.7%) 12 (15.6%) 38 (6.7%) Age* (binary) -years 0.001 <65 616 (95.1%) 68 (88.3%) 548 (96.0%) ≥65 32 (4.9%) 9 (11.7%) 23 (4.0%) Smoking Status (N=647) 0.943 Yes 104 (16.1%) 12 (15.8%) 92 (16.1%) No 543 (83.9%) 64 (84.2%) 479 (83.9%) Occupation 0.443 Working in health care facilities × 101 (15.6%) 14 (18.2%) 87 (15.2%) Military 21 (3.2%) 4 (5.2%) 17 (3.0%) Others 526 (81.2%) 59 (76.6%) 467 (81.8%) Comorbidities Diabetes Mellitus* 73 (11.3%) 16 (20.8%) 57 (10.0%) 0.005 Hypertension* 75 (11.6%) 16 (20.8%) 59 (10.3%) 0.007 Chronic kidney disease 12 (1.9%) 2 (2.6%) 10 (1.8%) 0.605 Chronic respiratory diseases*‡ 81 (12.5%) 17 (22.1%) 64 (11.2%) 0.007 Cancer/Immunodeficiency* 18 (2.8%) 5 (6.5%) 13 (2.3%) 0.035 Cardiac diseases* † 23 (3.5%) 8 (10.4%) 15 (2.6%) 0.001 No comorbidity 382 (59.0%) 23 (29.8%) 359 (62.9%) 0.001 1 or more Comorbidity* 188 (29.0%) 33 (42.9%) 155 (27.1%) 0.004 2 or more Comorbidity* 78 (12.0%) 21 (27.3%) 57 (10.0%) 0.001 Length of stay* (MD, IQR) -days 5 (14) 11.5 (11) 4 (12) 0.001 Symptoms Fever (N=191) 163 (85.3%) 36 (94.7%) 127 (83.0%) 0.067 Cough* (N=228) 203 (89.0%) 41 (97.6%) 162 (87.1%) 0.049 Sore Throat (N=145) 115 (79.3%) 18 (85.7%) 97 (78.2%) 0.433 Runny Nose (N=98) 73 (74.5%) 7 (63.6%) 66 (75.9%) 0.381 Headache (N=504) 140 (27.8%) 10 (18.2%) 130 (29.0%) 0.092 GI Symptoms ǂ (N=504) 71 (14.1%) 11 (20.0%) 60 (13.4%) 0.182 Myalgia (N=504) 145 (28.8%) 17 (30.9%) 128 (28.5%) 0.710 Vital Signs Temperature (N=268) -֯C 0.132 <38 219 (81.7%) 34 (73.9%) 185 (83.3%) ≥38 49 (18.3%) 12 (26.1%) 37 (16.7%) Heart rate (N=115) -beats/min 0.133 <100 88 (76.5%) 12 (63.2%) 76 (79.2%) ≥100 27 (23.5%) 7 (36.8%) 20 (20.8%) Respiratory rate (MD, IQR) - breaths/min 20 (2.0) 20 (4.0) 20 (2.0) 0.260 Respiratory rate (N=106) 0.187 ≤24 101 (95.3%) 17 (89.5%) 84 (96.6%) > 24 5 (4.7%) 2 (10.5%) 3 (3.4%) 1375 EMHJ – Vol. 26 No. 11 – 2020Research article 4.25). Cancer/immunodeficient patients had a significant crude OR=2.98 (1.03–8.61) that became non-significant when adjusted with OR=2.24 (0.73–6.87). Finally, cardiac patients showed the highest OR=4.30 (1.76–10.50) that remained significant even after adjusting for age and sex with an OR=3.05 (1.16–8.02). None of the symptoms showed a significant association with the outcomes even after adjusting for age and sex (Table 2). Discussion To the best of our knowledge, this is one of the first studies in the Eastern Mediterranean Region to assess the association between common comorbidities, clinical manifestations and laboratory results for critical COV- ID-19 patients. We found an association between gender, age, diabetes, HTN, chronic respiratory diseases, cardiac diseases, cancer patients/immunodeficiency with the outcomes of clinical interest. There was no significant as- sociation among tobacco smokers or patients presenting with specific signs and symptoms. The gender distribution of our patients was 53% males and 47% females. In addition, males were seen more in the critical group 67.5% versus 32.5% in females. These findings are consistent with previous evidence that suggests male patients have a higher severity and mortality (6,16). The median age of our sample was 34 years; this could be attributed to the young population of Saudi Arabia (17). When age was analyzed as a continuous variable, it showed a significant association with increased risk by almost 3% each year. Consistent with the literature, cases aged 65 years or older had an increased risk of being admitted to ICU or dying from COVID-19 (18,19). The results were also significant in patients older than 60 years of age with an increase in risk by 3.65 times (95% CI: 1.18–11.27) in relation to those 1–20 years-old cases. Remarkably, 15.6% of the critical cases were over the age of 60. These findings confirmed the previous evidence reporting age as a risk factor for poor outcomes (5,20). However, with a lower cutoff age of 60 years when compared with recent studies (2,18). This might be related to comorbidities appearing at an early age in our population (21). Age-related responses with weak immune systems are probable contributing factors for adverse outcomes of the disease (18). Characteristic Total Patients (N=648) Critical (N=77) Non-critical (N=571) P-value* SBP (MD, IQR) -mmHg 125 (22.0) 121 (20.0) 125 (21.0) 0.336 DBP (MD, IQR) -mmHg 74 (12.0) 74 (10.0) 74.5 (14.0) 0.450 Oxygen saturation (MD, IQR) -% 98 (3.0) 98 (3.3) 98 (3.0) 0.659 Oxygen saturation (N=259) 0.610 < 94 28 (10.8%) 4 (8.7%) 24 (11.3%) ≥94 231 (89.2%) 42 (91.3%) 189 (88.7%) Blood Laboratory testing WBC means (MD, IQR) -103 /µL 5.5 (3.3) 6.5 (7.8) 5.5 (3.0) 0.460 WBC (N=34) 0.906 <4 5 (14.7%) 1 (12.5%) 4 (15.4%) 4-12 26 (76.5%) 6 (75.0%) 20 (76.9%) >12 3 (8.8%) 1 (12.5%) 2 (7.7%) Neutrophils (MD, IQR) -% 62.1 (67.9) 68 (47.3) 60.2 (67.3) 0.405 Neutrophils (N=28) 0.777 <55 10 (35.7%) 2 (28.6%) 8 (38.1%) 55-70 9 (32.1%) 2 (28.6%) 7 (33.3%) >70 9 (32.1%) 3 (42.9%) 6 (28.6%) Lymphocytes (MD, IQR) -% 21.8 (13.7) 22.4 (31.2) 21.8 (13.0) 0.919 Lymphocytes (N=23) 0.923 <20 10 (43.5%) 3 (50.0%) 7 (41.2%) 20-40 10 (43.5%) 2 (33.3%) 8 (47.1%) >40 3 (13.0%) 1 (16.7%) 2 (11.8%) MD=median; IQR=interquartile range; GI=gastrointestinal; SBP=systolic blood pressure; DBP=diastolic blood pressure; WBC=white blood cells * Significant result at ɑ=0.05 × Involves physicians, nurses, pharmacists, lab technicians, cleaners, and other workers in health care facilities ‡ Includes asthma, COPD, interstitial lung disease, bronchiectasis, lung cancer, and others † Includes Ischemic heart diseases and heart failure ǂ Involves abdominal pain, vomiting, or diarrhoea Table 1 Baseline characteristics of the COVID-19 cases based on their reported outcomes (concluded) 1376 EMHJ – Vol. 26 No. 11 – 2020Research article Both DM and HTN were associated with worse outcomes with a crude OR around 2.37 and 2.28, respectively. This is supported by previous cohort studies confirming similar findings (5,6). However, the adjusted OR became non-significant (P = 0.304, P = 0.443 for DM and HTN, respectively), which may indicate that the crude OR was biased. On the other hand, patients with a history of cardiac diseases had significantly higher adjusted OR, strongly suggesting that cardiac diseases were an independent risk factor for ICU admission and mortality. This concurs with findings reported by previous studies that cardiac diseases were associated with worse outcomes among COVID-19 patients (5,6). The impact of cardiovascular diseases in COVID-19 could be related to the impaired cardiovascular compensatory mechanism or the direct cardiac injury reported to be associated with a higher incidence of worse outcomes (22). Furthermore, a study conducted in Wuhan, China, found cardiac injuries in almost 20% of their COVID-19 patients and more frequent in cardiac patients with higher mortality (23). In our study, CRD was a significant factor for ICU admission and mortality as adjusted with an OR of 2.29 (CI: 1.24–4.25) and P = 0.008. This was supported by similar findings from a recent observational Figure 2 Age and gender distribution per outcomes in 648 laboratory-confirmed COVID-19 cases with definitive outcomes in Saudi Arabia, 1–31 March, 2020 >60 41-60 21-40 1-20 200 150 100 50 0 50 100 150 200 Number of Patients Male Ag e in y ea rs Female Non-critical Critical Figure 3 Median length of in-hospital stay (LOS) for a total of 648 COVID-19 patients according to different risk factors and outcomes 10 10 10 10 10 9 9 9 7 5 4 2 11.5 4 0 2 4 6 8 10 12 CKD HTN DM Cancer/Immodeficiency Age≥65 Patients ≥1 comorbidity Patients ≥2 comorbidity Cardiac diseases CRD Smoking Age<65 Patients without comorbidity Critical Non-critical Median LOS(Days) According to Outcomes According to Risk Factors CKD=chronic kidney diseases; HTN=hypertension; DM=diabetes mellitus; CRD=chronic respiratory diseases (d 1377 EMHJ – Vol. 26 No. 11 – 2020Research article study carried out in the UK indicating that CRD was a risk factor for in-hospital mortality in COVID-19 patients (6). Having two or more comorbidities are independent risk factors for ICU admission and mortality with an adjusted OR=2.57. Comorbidities should be considered when risk stratifying patients with COVID-19 as supported by the nationwide analysis from China (8). Although initial symptoms and vital signs were not associated with worse outcomes, it is crucial to not fully rely on the initial clinical manifestations in predicting outcomes for COVID-19 patients because the disease might have an unpredictable course. The missing data variables in our study contributed broadly to these differences in that we retrieved only 5.2% of WBCs, 4.3% of neutrophils, and 3.5% of lymphocytes results; these failed to generate a robust analysis to assess the association. Tobacco smoking is a controversial factor in COVID-19. There has been a strong relation between angiotensin-converting enzyme 2 (ACE2) expression in the lung tissue and the spread of COVID-19 disease. Table 2 Risk factors associated with intensive care unit (ICU) admission or death among COVID-19 patients Crude OR (95% CI) P-value Adjusted OR* (95%CI) P-value Sex (Male) 2.01 (1.22-3.34) 0.006 1.92 (1.15-3.20) 0.012 Age (continuous) -years 1.03 (1.01-1.05) 0.001 1.03 (1.01-1.04) 0.001 Age (≥65) - years 3.15 (1.40-7.09) 0.005 3.07 (1.35-6.96) 0.007 Age Categories - years 1-20 Reference - Reference 21-40 1.52 (0.57-3.99) 0.401 1.33 (0.50-3.55) 0.564 41-60 1.90 (0.69-5.25) 0.214 1.63 (0.59-4.54) 0.35 >60 4.04 (1.32-12.36) 0.014 3.65 (1.18-11.27) 0.024 Smoker 0.98 (0.51-1.88) 0.943 0.79 (0.40-1.58) 0.517 Comorbidities One or more comorbidity 2.01 (1.24-3.28) 0.005 1.51 (0.87-2.62) 0.141 Two or more Comorbidity 3.38 (1.91-5.99) 0.001 2.57 (1.33-4.97) 0.005 Diabetes Mellitus 2.37 (1.28-4.37) 0.006 1.45 (0.72-2.93) 0.304 Hypertension 2.28 (1.23-4.20) 0.009 1.37 (0.64-2.80) 0.443 CRD 2.25 (1.23-4.08) 0.008 2.29 (1.24-4.25) 0.008 Chronic Kidney diseases 1.50 (0.32-6.96) 0.608 0.98 (0.21-4.68) 0.981 Cardiac Diseases 4.30 (1.76-10.50) 0.001 3.05 (1.16-8.02) 0.024 Cancer/Immunodeficiency 2.98 (1.03-8.61) 0.043 2.24 (0.73-6.87) 0.158 Symptoms Fever 3.69 (0.84-16.27) 0.085 2.78 (0.61-12.59) 0.185 Cough 6.07 (0.79-46.23) 0.081 5.05 (0.65-39.02) 0.121 Sore Throat 1.67 (0.46-6.09) 0.437 1.28 (0.34-4.88) 0.715 Runny Nose 0.56 (0.15-2.09) 0.386 0.50 (0.13-1.94) 0.317 Headache 0.55 (0.27-1.12) 0.096 0.62 (0.30-1.28) 0.199 GI Symptoms 1.62 (0.79-3.31) 0.185 1.80 (0.87-3.74) 0.113 Myalgia 1.12 (0.61-2.06) 0.711 1.18 (0.63-2.18) 0.610 Vital Signs Temperature (≥38) -֯C 1.77 (0.84-3.72) 0.136 1.64 (0.77-3.51) 0.200 Heart Rate ≥100 -beats/min 2.22 (0.77-6.36) 0.139 1.99 (0.67-5.88) 0.216 Respiratory Rate (Continuous) -breaths/min 1.13 (0.93-1.36) 0.212 1.10 (0.91-1.33) 0.307 Respiratory rate (>24) 3.29 (0.51-21.24) 0.210 2.34 (0.35-15.84) 0.384 SBP (continuous) -mmHg 0.99 (0.96-1.02) 0.595 0.99 (0.96-1.02) 0.365 DBP (continuous) -mmHg 0.97 (0.93-1.03) 0.319 0.96 (0.90-1.02) 0.143 Oxygen saturation % 1.02 (0.93-1.13) 0.652 1.04 (0.93-1.15) 0.513 Oxygen saturation (< 94) 0.75 (0.25-2.28) 0.611 0.75 (0.24-2.33) 0.623 OR=odds ratio; CI=confidence interval CRD=chronic respiratory diseases; GI=gastrointestinal; SBP=systolic blood pressure; DBP=diastolic blood pressure * Adjusted for age and gender 1378 EMHJ – Vol. 26 No. 11 – 2020Research article Smoking is remarkably associated with dose-dependent upregulation of ACE2 expression causing more harm in developing critical outcomes among the patients (24,25). On the contrary, some reports have not found smoking to be associated with COVID-19 severity (26). Surprisingly, one paper proposed smoking to be a protective factor against COVID-19 in developing less serious infections and hypothesized the pathophysiological explanation could be related to the nicotine effect (27). Nevertheless, our result confirmed the poor association between smoking and worse outcomes encouraging more structured studies in that subject to provide conclusive evidence. Remarkably, the median LOS was 11.5 days for critical COVID-19 patients with only four days for the non-critical arm. This was expected considering the severity of the disease and the associated comorbidity; patients without comorbidities had a span of two days while those with comorbidities had a span of nine days. A recent study in China reported a median LOS of 14.5 days for critical patients, higher than the LOS seen in our results. This could be explained by their higher median age of critical cases compared to our cases, 63 years versus 37 years respectively (14). A better understanding of this disease is critical to reducing the impact of the pandemic. Many of our findings are concordant with reported risk factors for COVID-19 disease. In many ways, this study will help risk-stratify, prioritize the detection, and guide clinical management and disposition effectively based on their demographic data, clinical symptoms, and associated comorbidities. Furthermore, it will support decision- makers to unify clinical guidelines and describe predictors for mortality of COVID-19 in the Eastern Mediterranean Region. Based on the current evidence of risk factors and predictors, we suggest designing a special multi-item scale system to prognosticate COVID-19 patients. We encourage designing a comprehensive assessment tool that contains clinical symptoms, risk factors, radiological features, and laboratory findings of COVID-19 patients to anticipate the clinical courses and guide future management. Although some of the essential risk factors associated with poor outcomes have been described in this study, further prospective studies in the region are recommended to investigate various radiological features and expected laboratory values as predictors of COVID-19 prognosis. This study identified many risk factors associated with adverse outcomes in COVID-19 patients, but several limitations were present. First, due to the retrospective nature of this study, we could not eliminate missing variables especially, in the vital signs and laboratory results. Therefore, no data analysis or interpretation could be withdrawn from these variables. Second, radiological studies were not available in the HESN database. This would have an additive value in risk-stratifying patients based on their radiological abnormalities. Third, many laboratory results were not retrieved including renal function tests, liver function tests, D-dimer, coagulation profile, and troponin levels. These could guide clinical practice if correlated with COVID-19 disease severity and related outcomes. Conclusions In subjects with COVID-19, age, male gender, cardiac dis- eases, CRD, and having two or more comorbidities were independent risk factors for ICU admission and mortali- ty. Although the young Saudi population has limited risk, we found that age more than 60 years was associated with worse outcomes. Finally, neither the initial sign and symptoms nor tobacco smoking were linked to adverse outcomes. Based on the current evidence of risk factors and predictors, we suggest designing a unique mul- ti-item scale system to prognosticate COVID-19 patients. Acknowledgment We would like to acknowledge the contribution of all MoH staff, mainly the General Directorate of Statistics and Infor- mation Management, and Directorate General of Infectious Diseases. Funding: None. Competing interests: None declared. Facteurs de risque associés à une détérioration des résultats pour la COVID-19 : étude rétrospective en Arabie saoudite Résumé Contexte : L’ émergence rapide de la maladie à nouveau coronavirus 2019 (COVID-19) a entraîné l’infection de millions de patients et des centaines de milliers de décès dans le monde entier. La prestation des services de soins de santé est mise à mal en raison de l’augmentation massive du nombre de patients infectés pendant cette pandémie. Objectifs : La présente étude visait à évaluer les facteurs de risque associés à un mauvais pronostic chez les patients atteints de COVID-19 en Arabie saoudite. 1379 EMHJ – Vol. 26 No. 11 – 2020Research article References 1. COVID-19 Daily Updates. Saudi Arabia. Riyadh: Ministry of Health; 2020 (https://covid19.moh.gov.sa/, accessed 14 June 2020). 2. Huang C, Wang Y, Li X, Ren L, Zhao J, Hu Y, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. The Lancet. 2020;395(10223):497-506. https://doi.org/10.1016/S0140-6736 (20)30183-5. 3. Chen N, Zhou M, Dong X, Qu J, Gong F, Han Y, et al. Epidemiological and clinical characteristics of 99 cases of 2019 novel coro- navirus pneumonia in Wuhan, China: a descriptive study. The Lancet. 2020;395(10223):507-513. https://doi.org/10.1016/S0140-6736 (20)30211-7. 4. Wang D, Hu B, Hu C, Zhu F, Liu X, Zhang J, et al. Clinical characteristics of 138 hospitalized patients with 2019 novel coronavi- rus–infected pneumonia in Wuhan, China. JAMA. 2020;323(11):1061-1069. https://doi:10.1001/jama.2020.1585. 5. Zhou F, Yu T, Du R, Fan G, Liu Y, Liu Z, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. The Lancet. 2020;395 (10229):1054-1062. https://doi.org/10.1016/S0140-6736 (20)30566-3. 6. Docherty AB, Harrison EM, Green CA, Hardwick HE, Pius R, Norman L, et al. Features of 20 133 UK patients in hospital with COVID-19 using the ISARIC WHO clinical characterization protocol: prospective observational cohort study. BMJ. 2020 May 22;369:m1985. https://doi.org/10.1136/bmj.m1985. ةيدوعسلا ةيبرعلا ةكلملما في يعجر رثأب ةسارد :19-ديفوك ضرلم ةئيسلا جئاتنلاب ةطبترلما رطلخا لماوع رادخوج نياه ،ييرسع للها دبع ،تافرع دممح ،كرابم للها دبع ،يبيتعلا دئار ،نايفصلا فسوي ،ناينثلا رقص ،ناخ سنأ ةصلالخا في تايفولا نم فلالآا تائم عوقوو ضىرلما نم ينيلالما ةباصإ لىإ )19-ديفوك( 2019 دجتسلما انوروك سويرفل عيسرلا راشتنلاا ىدأ :ةيفللخا .ةحئالجا للاخ ينباصلما ضىرلما دادعأ في ئجافلما دايدزلال ًارظن ًابلس ةمدقلما ةيحصلا ةياعرلا تامدخ ترثأت دقف .لماعلا ءاحنأ عيجم .ةيدوعسلا ةيبرعلا ةكلملما في 19-ديفوك ضىرلم ةئيسلا جئاتنلاب ةطبترلما رطلخا لماوع مييقت لىإ ةساردلا هذه تفده :فادهلأا ةيئانه جئاتن ملهو ةيبرخلما ليلاحتلاب مهجئاتن ةدكؤلما 19-ديفوك تلااح عيجم تلمش زكارلما ةددعتم ةيعجر ةيدشح ةسارد انيرجأ :ثحبلا قرط ةدعاق نم جئاتنلاو ةبحاصلما ضارملأاو يكينيلكلإا خيراتلاو ةيناكسلا تانايبلا عجم متو .ةيدوعسلا ةيبرعلا ةكلملما في راذآ /سرام رهش للاخ فاشكتسلا ةلّدعلماو مالخا ةيحجرلأا بسن باسلح يتسيجوللا رادحنلاا جذمان مادختسا متو .)نصح( يئابولا دصترلل نيوتركللإا ماظنلا تانايب .)ةافولا وأ ةزكرلما ةيانعلا ةدحو في ميونتلا( 19-ديفوك تلاالح ةجرلحا جئاتنلل ةبحاصلما رطلخا لماوع رطلخا لماوع تلمشو .ةجرلحا ةعومجلما في %11.9 مهنم .اًماع 34 رماعأ طسوتمب 19-ديفوكب اًيبايجإ اًضيرم 648 ةساردلا تنمضت :جئاتنلا ،)3.05=ةيحجرلأا ةبسن( بلقلا ضارمأ ،)3.65=ةيحجرلأا ةبسن( اًماع 60 >رمعلا ،)1.92=ةيحجرلأا ةبسن( روكذلا ةئيسلا جئاتنلاب ةطبترلما ليدعت دعب )2.57=ةيحجرلأا ةبسن( رثكأ وأ ينبحاصم ينضرمب ةطبترلما ةباصلما تلاالحاو ،)2.29=ةيحجرلأا ةبسن( ةنمزلما ةيسفنتلا ضارملأا .0.05 < ةيئاصحإ ةيهمأ تاذ مهنم لكل ةيلماتحلاا ةميقلا تناكو ؛سنلجاو رمعلا ضارملأاو بلقلا ضىرمو روكذلاو نسلا في مدقتلا :19-ديفوك ضىرلم ةجرلحا جئاتنلاب ةطبترلما ةلقتسلما رطلخا لماوع لمشت :تاجاتنتسلاا ةيحصلا ةلالحاب ؤبنتلل صرانعلا ددعتم ديرف سايقم ماظن ميمصتب صيونو .رثكأ وأ ينبحاصم ينضرمب ةطبترلما ةباصلما تلاالحاو ةنمزلما ةيسفنتلا .19-ديفوك ضىرلم Méthodes : Il s’agissait d’une étude de cohorte rétrospective multicentrique qui incluait tous les cas de COVID-19 confirmés en laboratoire, avec des résultats définitifs en Arabie saoudite en mars 2020. Les données démographiques, les antécédents cliniques, la comorbidité et les résultats ont été extraits de la base de données du National Health Electronic Surveillance Network. Nous avons utilisé des modèles de régression logistique pour calculer les odds ratios bruts et ajustés (OR) afin d’étudier les facteurs de risque pour les résultats critiques (admission en unité de soins intensifs ou décès) parmi les cas de COVID-19. Résultats : Nous avons inclus 648 patients positifs pour la COVID-19, dont l’âge médian était de 34 ans. Parmi eux, 11,9 % appartenaient au groupe critique. Les facteurs de risque associés aux pires résultats comprenaient l’appartenance au sexe masculin (OR = 1,92), un âge supérieur à 60 ans (OR = 3,65), les maladies cardiaques (OR = 3,05), les maladies respiratoires chroniques (OR = 2,29) et les cas présentant deux comorbidités ou plus (OR = 2,57) après ajustement en fonction de l’âge et du sexe ; tous avaient des valeurs p inférieures à 0,05 significatives. Conclusions : Les facteurs de risque indépendants pour les résultats critiques parmi les cas de COVID-19 comprennent la vieillesse, l’appartenance au sexe masculin, les maladies cardiaques, les maladies respiratoires chroniques et la présence de deux comorbidités ou plus. Nous recommandons de mettre au point un système d’échelle unique multi-items pour pronostiquer les patients atteints de COVID-19. 1380 EMHJ – Vol. 26 No. 11 – 2020Research article 7. Wu Z, McGoogan J. characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) Outbreak in China. JAMA. 2020;323 (13):1239-1242. https://doi:10.1001/jama.2020.2648. 8. Guan W, Liang W, Zhao Y, Liang H, Chen Z, Li Y, et al. Comorbidity and its impact on 1590 patients with Covid-19 in China: A nationwide analysis. Eur Respir J. 2020;55 (6):2000547. https://doi.org/10.1183/13993003.00547-2020. 9. Alsofayan Y, Althunayyan S, Khan A, Hakawi A, Assiri A. Clinical characteristic of COVID-19 in Saudi Arabia: A national retrospec- tive study. J Infect Public Health. 2020;13(7):920-925.https://doi.org/10.1016/j.jiph.2020.05.026. 10. National laboratories. Geneva: World Health Organization; 2020 (https://www.who.int/emergencies/diseases/novel-coronavin- rus-2019/technical-guidance/laboratory-guidance/, accessed 8 June 2020). 11. Risk stratification for severe disease in patients with coronavirus disease 2019 (COVID-19). UpToDate; 2020 https://www.uptoo- date.com/contents/image/print?imageKey=PWYS%2F127891&topicKey=PC%2F127759&search=covid&source=outline_link&se- lectedTitle=5~150, accessed 1 June 2020). 12. Singer M, Deutschman C, Seymour C, Shankar-Hari M, Annane D, Bauer M, et al. The Third International Consensus Defini- tions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315 (8):801-810. https://doi:10.1001/jama.2016.0287. 13. ICD-10 for Mortality and Morbidity Statistics. Geneva: World Health Organization; 2019 (https://icd.who.int/browse10/2019/en, accessed 24 May 2020). 14. Guan W, Ni Z, Hu Y, Liang W, Ou C, He J, et al. Clinical characteristics of coronavirus disease 2019 in China. New Engl J Med. 2020;382 (18):1708-1720. https://doi:10.1056/NEJMoa2002032. 15. COVID-19 coronavirus disease guidelines. Riyadh: Ministry of Health; 2020 (https://www.moh.gov.sa/Ministry/MediaCenter/ Publications/Documents/Coronavirus-Disease-2019-Guidelines-v1.2.pdf, accessed 28 May 2020). 16. Chen T, Wu D, Chen H, Yan W, Yang D, Chen G, et al. Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study. BMJ. 2020;368:m1091. https://doi.org/10.1136/bmj.m1091. 17. Population by gender, age groups and nationality (Saudi/Non-Saudi). Riyadh: General Authority for Statistics; 2018 (https:// www.stats.gov.sa/en/5680/, accessed 27 May 2020). 18. Wu C, Chen X, Cai Y, Xia J, Zhou X, Xu S, et al. Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China. JAMA Intern Med. 2020;180(7):934-943. https://doi. org/10.1001/jamainternmed.2020.0994. 19. Yang X, Yu Y, Xu J, Shu H, Xia J, Liu H, et al. Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumo- nia in Wuhan, China: a single-centered, retrospective, observational study. Lancet Respir Med. 2020 May;8(5):475-481. https:// doi:10.1016/S2213-2600 (20)30079-5. 20. Grasselli G, Zangrillo A, Zanella A, Antonelli M, Cabrini L, Castelli A, et al. Baseline characteristics and outcomes of 1591 patients infected with SARS-CoV-2 admitted to ICUs of the Lombardy region, Italy. JAMA. 2020;323 (16):1574–1581. https://doi:10.1001/ jama.2020.5394. 21. Khoja AT, Aljawadi MH, Al-Shammari SA, Mohamed AG, Al-Manaa HA, Morlock L, et al. The health of Saudi older adults; results from the Saudi National Survey for elderly health (SNSEH) 2006–2015. Saudi Pharm J. 2018 Feb 1;26(2):292-300. https://doi. org/10.1016/j.jsps.2017.11.008. 22. Li B, Yang J, Zhao F, Zhi L, Wang X, Liu L, et al. Prevalence and impact of cardiovascular metabolic diseases on COVID-19 in China. Clin Res Cardiol. 2020;531–538. https://doi.org/10.1007/s00392-020-01626-9. 23. Shi S, Qin M, Shen B, Cai Y, Liu T, Yang F, et al. Association of cardiac injury with mortality in hospitalized patients with COV- ID-19 in Wuhan, China. JAMA Cardiol. 2020;5(7):802-810. https://doi:10.1001/jamacardio.2020.0950. 24. Smith JC, Sausville EL, Girish V, Yuan ML, Vasudevan A, John KM, et al. Cigarette smoke exposure and inflammatory signaling increase the expression of the SARS-CoV-2 receptor ACE2 in the respiratory tract. Dev Cell. 2020 May 16;53(5):514-529.e3. https:// doi.org/10.1016/j.devcel.2020.05.012. 25. Zheng Z, Peng F, Xu B, Zhao J, Liu H, Peng J, et al. Risk factors of critical & mortal COVID-19 cases: A systematic literature review and meta-analysis. J Infect. 2020 Aug;81(2):e16-e25 Epub 2020 https://doi.org/10.1016/j.jinf.2020.04.021. 26. Lippi G, Henry BM. Active smoking is not associated with severity of coronavirus disease 2019 (COVID-19). Eur J Intern Med. 2020 May;75:107-108. https://doi:10.1016/j.ejim.2020.03.014. 27. Miyara M, Tubach F, Pourcher V, Morelot-Panzini C, Pernet J, Haroche J, et al. Low rate of daily active tobacco smoking in pa- tients with symptomatic COVID-19. Qeios. 2020. https://doi:10.32388/WPP19W.4. 1381 EMHJ – Vol. 26 No. 11 – 2020Research article Quality utilization of antenatal care and low birth weight: evidence from 18 demographic health surveys Saverio Bellizzi1 and Susanna Padrini2 1Partnership for Maternal, Newborn and Child Health, Geneva, Switzerland (Correspondence to: S. Bellizzi: bellizzis@who.int). 2Associazione Italiana per la Solidarietà tra i Popoli (AISPO), Milano, Italy. Abstract Background: Low birthweight is a crucial factor in child mortality and morbidity and affects almost 20% of infants world- wide, mostly in low- and middle-income countries. Aims: To assess the relationship between access to and quality of antenatal care and occurrence of low birth weight. Methods: We analysed data from 18 demographic and health surveys, from 2005 to 2013, including 69 446 children. The main study outcome was birthweight < 2.5 kg, and access to and number of antematal care visits were exposure variables. Moreover, antenatal care attendants and time of visit (trimester) were considered. Multiple logistic regression adjusted for sampling at primary and country level was utilized. Results: At least 1 and ≥ 4 antenatal care consultations were both associated with decreased odds of low birth weight when compared to none and < 4 antenatal care consultations, respectively. Additional benefit stemmed from having skilled antenatal care attendants and the first antenatal care consultation during the first trimester. Conclusions: Proper antenatal care coverage during pregnancy is beneficial for preventing low birth weight in low- and middle-income countries. Keywords: antenatal care, demographic health survey, low birth weight, low-income country, middle-income country Citation: Bellizzi S; Padrini S. Quality utilization of antenatal care and low birth weight: evidence from 18 demographic health surveys. East Mediterr Health J. 2020;26(11):1381-1387. https://doi.org/10.26719/emhj.20.055 Received: 14/11/18; accepted: 24/11/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo) Introduction In 2012, the World Health Assembly endorsed a compre- hensive plan under Resolution 65.6 with specific global nutrition targets for 2025 (1). This policy included a 30% reduction in low birthweight (LBW) (2), corresponding to a reduction from 20 million to ~14 million neonates with birthweight < 2.5 kg (3) between 2012 and 2025. LBW affects almost one sixth of infants worldwide with > 95% of cases located in developing countries (3), and is recognized as 1 of the most influential factors on child mortality and morbidity. LBW increases mortality risk by 20–30 times (4), and contributes to 60–80% of all ne- onatal deaths worldwide (5,6). Surviving infants are at higher risk of pathological conditions such as infection immediately after birth and throughout the first year of life (7). LBW is also associated with morbidity later in life, such as psychosocial disorders (8), impaired cognitive function (9), coronary heart disease (10) and noninsulin dependent diabetes (11). Several risk factors are claimed to be associated with LBW, including maternal factors, pregnancy, multiple gestation, socioeconomic character- istics, drug treatment and body mass index (12–15). At least 4 antenatal care (ANC) consultations, with the first preferably in the first trimester (16), has been a worldwide recommended policy for the last 2 decades. However, there is still inconclusive evidence on its impact on maternal and neonatal outcomes in developing countries. Some studies have shown that ANC improves birthweight (17,18), while others have shown a lack of evidence for the effectiveness of content, frequency and timing of visits in standard ANC programmes on maternal and child health (19). Our research used data from demographic and health surveys (DHSs) in 18 countries and examined the association between adequate utilization of ANC and occurrence of LBW. Methods Study design This was a population-based study of data from 18 DHSs between 2005 and 2013, which reported birthweight for at least 80% of births over the 5 years preceding the survey: Albania 2008/2009, Armenia 2010, Congo (Brazzaville) 2011/2012, Dominican Republic 2013, Gabon 2012, Guyana 2009, Honduras 2011/2012, Indonesia 2012, Jordan 2012, Kyrgyzstan 2012, Maldives 2009, Republic of Moldova 2005, Peru 2012, Philippines 2013, Sao Tome and Princi- pe 2008/2009, Swaziland 2006/07, Tajikistan 2012, and Ukraine 2007. Detailed information on procedures and sampling techniques for all DHSs have been published elsewhere (20). Face-to-face interviews were carried out for a total of 213 752 women. 1382 EMHJ – Vol. 26 No. 11 – 2020Research article Study population The study population consisted of all the latest singleton live births (n = 77 809) during the 5 years preceding the DHS in each country. After excluding 8363 (10.7%) indi- viduals for whom we had missing data on BW, the final sample included 69 446 babies. Information on BW was obtained through birth certificates and maternal recall for 21 334 (30.7%) and 48 112 (69.3%) infants, respectively. Outcome, exposure and control variables The main outcome was LBW, which was defined as < 2.5 kg. Characteristics of ANC were the exposure variables, which were defined as follows: “ANC”, if any ANC con- sultation was performed; “provider”, classified as skilled (doctors, nurses or other trained attendants) or unskilled attendant (traditional attendants or others); “number of ANC consultations”, subdivided into < 4 or ≥ 4 ANC vis- its; “ANC timing”, dichotomized into ANC first consulta- tion in the first trimester or after the first trimester; and “quality of ANC”, with ≥ 4 ANC consultations started in the first trimester with a skilled attendant on 1 side, and all the others on the other side. A series of socioeconom- ic, pregnancy and maternal characteristics were evalu- ated as possible confounders, including age, education, wealth, place of residence, birth interval, birth order, wanted pregnancy and child sex. Statistical analysis Statistical analysis was performed using STATA 13.1 SE (StataCorp, College Station, TX, USA). The “svy” com- mand was used to adjust for clustering by primary sam- pling unit. Number of total livebirths and LBW by coun- try were tabulated with relative percentages. All the study categorical confounding variables were tested against LBW using the χ2 test. Furthermore, we used the χ2 test to examine the association between the quality of ANC and the following socioeconomic variables: wealth sta- tus, maternal age and education, and place of residence. The Metaprop syntax (21) was used in the pooled meta-analysis of all country datasets, which generated weighted subgroup and overall pooled estimates with inverse-variance weights obtained from a random-effects model. In this model, no residual heterogeneity was assumed. The final model included wealth, age, birth order, birth spacing, education, wanted pregnancy, child sex, and rural/urban residence; the factors primary sampling unit and country were added with random effect. Stepwise logistic regression analysis of LBW on the 5 ANC exposure variables was conducted adjusting for socioeconomic, maternal and pregnancy characteristics. P < 0.05 was considered statistically significant. Ethical approval This study used existing data obtained from ORC Mac- ro (Calverton, MD, USA) through formal request mech- anisms (https://dhsprogram.com). No additional ethical review for the secondary analysis was required since each country and the Institutional Review Board of ORC Macro approved the DHS data collection procedures. Results Overall, 6238 (9.0%) newborns with LBW were record- ed, ranging from a minimum of 36 (2.8%) of 1281 in Al- bania to a maximum of 883 (20.8%) of 4238 newborns in the Philippines (Table 1). Data on ANC were missing for 1404 individuals, corresponding to 2% of the total study population. Most mothers had ANC (n = 66 513; 97.7%) and half of them (n = 33 038) had the first consultation during the first trimester (Table 2). Only 6517 (10%) wom- en had < 4 consultations. Almost all pregnant women (97.2%) consulted a trained operator. Less than half of them (n = 31 372) had a good quality of ANC according to World Health Organization (WHO) criteria. Table 3 shows a clear trend toward increasing prevalence of LBW with decreasing wealth, poorer education and shorter birth intervals, in addition to higher risk in unwanted pregnancies and female sex. There were associations between wealth status and education and the quality of ANC. The richest and most educated women, in addition to those living in urban areas, were more likely to have ≥ 4 ANC consultations performed by skilled attendants, with the first consultation during the first trimester. The adjusted logistic regression showed a significant benefit of having any ANC consultation when compared Table 1 Numbers of live births and low birth weight infants in 18 low- and middle-income countries Country, year Live births Low birth weight (%) Albania 2008/2009 1281 36 (2.8) Armenia 2010 1139 1,438 (6.0) Congo (Brazzaville 2011/2012) 5355 467 (8.7) Dominican Republic 2013 2847 378 (13.3) Gabon 2012 3485 445 (12.8) Guyana 2009 1294 167 (12.9) Honduras 2011/2012 7062 654 (9.3) Indonesia 2012 13 045 840 (6.4) Jordan 2012 6612 817 (12.4) Kyrgyzstan 2012 3089 147 (4.8) Maldives 2009 3206 328 (10.2) Republic of Moldova 2005 1350 63 (4.7) Peru 2012 7385 479 (6.5) Philippines 2013 4238 883 (20.8) Sao Tome and Principe 2008/2009 1159 79 (6.8) Swaziland 2006/2007 1788 116 (6.5) Tajikistan 2012 2955 197 (6.7) Ukraine 2007 2156 74 (3.4) Total 69 446 6238 (9.0) Results are total number of newborns and number of low birthweight infants among the last births for each woman in the preceding 5 years. Results from 18 demographic health surveys. 1383 EMHJ – Vol. 26 No. 11 – 2020Research article to no ANC (OR 1.2; 95% CI 1.0–1.4) (Table 4). Among infants of women who underwent ANC, having < 4 consultations, first consultation after the first trimester, being attended by an unskilled operator and not meeting WHO quality criteria were associated with 1.5 (95% CI 1.4–1.7), 1.1 (95% CI 1.0–1.2), 1.2 (95% CI 1.1–1.4) and 1.1 (95% CI 1.0–1.2) increased ORs of LBW, respectively. Discussion This secondary analysis of DHS data from 18 countries showed that the absence of ANC consultation increased the risk of LBW. All WHO criteria, separately and com- bined, for adequate antenatal consultations resulted in significant protection against LBW. We compared our re- sults on the country incidence of LBW with other sourc- es and found no substantial differences. Estimates from the United Nations Children’s Fund and WHO global and country reports on LBW confirm the smallest percentage (3%) for Albania up to the highest (20%) in the Philippines (22). Our findings on the influence of maternal education on LBW are not surprising. A study in the Islamic Republic of Iran showed that the prevalence of LBW in infants born to women with no education was 16.9%, which decreased to 5.4% in women educated to a higher level (23). The explanation may lie in greater access to ANC and better nutritional behaviour. Similarly, parity and birth spacing have been detected as important determinants for LBW. One study showed that mothers with very short interpregnancy intervals (IPIs; < 3 months) and high parity had a higher risk of having LBW infants when compared to those with very short IPI but low parity (24). The explanation for these differences may be depleted nutritional reserves in women with high parity and short IPI. Other DHSs from single countries have reported the benefit of an early start to ANC and the importance of a sufficient number of consultations. A study from Nepal showed how women with no ANC were twice as likely to have LBW infants when compared to mothers with ≥ 4 ANC consultations (25). A study from Colombia reported that having the first ANC after the first trimester was associated with an increased OR for LBW when compared with first visits at the first trimester (26). Similar findings were reported in a study in Kenya (27), indicating a positive effect of ANC, which influences dietary behaviour and treatment from any illness that may have negative effects on the health of the fetus. Although our secondary analysis had advantages, such as large sample size and use of standardized questionnaires that limited the risk of intercountry variation, it had some limitations. First, we considered only the 18 DHSs with at least 80% of data on BW, but we cannot exclude bias for all remaining women not able to report information, which may have led to underestimation of LBW. Second, two thirds of the information on BW relied on maternal recall, therefore presenting a particular type of misreporting called heaping. Heaping consists of rounding and reporting weights as multiple of 500 g, which makes interpretation difficult when infants are reported as weighing 2.5 kg, and thus likely to be misclassified as having normal weight (28). Third, several possible confounding variables such as genetics and maternal history of diseases were not available. Finally, we had no information on nutritional status of women to exclude maternal factors that would increase risk of LBW. Table 2 Distribution of ANC variables among 69 446 low birthweight and normal weight livebirths in 18 low- and middle-income countries in 2005–2013 ANC variables Low birth weight Normal weight P n (%) n (%) χ² ANC No 177 (11.6) 1,352 (88.4) < 0.001 Yes 5948 (8.9) 60 565 (91.1) Time of first ANC consultation First trimester 2795 (8.5) 30 243 (91.5) < 0.001 After first trimester 3248 (9.4) 31 406 (90.6) No. of ANC consultations ≥ 4 5108 (8.5) 54 888 (91.5) < 0.001 < 4 840 (12.9) 5677 (87.1) ANC attendant Skilled 5908 (8.9) 60 291 (91.1) < 0.001 Unskilled 217 (11.8) 1626 (88.2) ANC highest quality Yes 2628 (8.4) 28 744 (91.6) < 0.001 No 3610 (9.5) 34 464 (90.5) ANC = antenatal care. 1384 EMHJ – Vol. 26 No. 11 – 2020Research article In conclusion, our study reinforces the need to encourage pregnant women to attend ANC to reduce LBW, with its short- and long-term consequences. Policies should in particular address access to and quality of ANC among disadvantaged socioeconomic groups, which are at higher risk of LBW. Uneducated mothers are less likely to understand health messages and to be concerned about their health and nutritional status. Poorer women are less likely to afford the cost of ANC and transportation in areas where health infrastructure is distant. Funding: None. Competing interests: None declared. Table 3 Characteristics of mothers of 69 446 low birthweight and normal weight infants in 18 low- and middle-income countries in 2005–2013 Maternal characteristics Low birth weight Normal weight P n (%) n (%) χ² Maternal age (years) 15–19 568 (13.2) 3745 (86.8) < 0.001 20–24 1541 (9.7) 14 305 (90.3) 25–29 1538 (8.1) 17 383 (91.9) 30–34 1155 (7.7) 13 756 (92.2) 35–39 838 (8.5) 8996 (91.5) 40–44 479 (10.4) 4127 (89.6) 45–49 119 (11.7) 896 (88.3) Birth order 1 2,227 (10.6) 19 920 (89.9) < 0.001 > 1 4,011 (8.5) 43 288 (91.5) Preceding birth interval (months) < 18 397 (12.1) 2871 (87.8) < 0.001 18–23 438 (9.2) 4337 (90.8) 24–35 863 (8.4) 9347 (91.5) > 35 2313 (8.0) 26 733 (92.0) Place of residence Urban 2936 (8.7) 30 829 (91.3) 0.01 Rural 3302 (9.2) 32 379 (90.7) Education No education 248 (11.7) 1869 (88.3) < 0.001 Primary 1901 (10.1) 16 897 (89.9) Secondary 3094 (8.7) 32 338 (91.3) Higher 992 (7.6) 12 063 (92.4) Wealth index Poorest 1882 (11.3) 14 780 (88.7) < 0.001 Poorer 1540 (9.4) 14 748 (90.5) Middle 1202 (8.5) 13 009 (91.5) Richer 960 (7.8) 11 378 (92.2) Richest 654 (6.6) 9293 (93.4) Wanted pregnancy Wanted 4077 (8.5) 44 109 (91.5) < 0.001 Not wanted 2156 (10.2) 19 057 (89.8) Child sex Male 2964 (8.2) 33 031 (91.8) < 0.001 Female 3274 (9.8) 30 177 (90.2) 1385 EMHJ – Vol. 26 No. 11 – 2020Research article Table 4 Odds ratios for low birthweight in 69 446 singleton births ANC OR (95% CI) unadjusted OR (95% CI) adjusteda No ANC visit 1.3 (1.1–1.6) 1.2 (1.0–1.4) < 4 ANC visits 1.6 (1.5–1.7) 1.5 (1.4–1.7) ANC visit after first trimester 1.1 (1.0–1.2) 1.1 (1.0–1.2) No Skilled ANC 1.4 (1.2–1.6) 1.2 (1.1–1.4) No Quality ANC 1.1 (1.0–1.2) 1.1 (1.0–1.2) aAdjusted for wealth, age, birth order, birth spacing, education, wanted pregnancy, child sex, and rural/urban residence. ANC = antenatal care; CI = confidence interval; OR = odds ratio. Utilisation qualitative des soins prénatals et faible poids de naissance : données issues de 18 enquêtes démographiques sur la santé Résumé Contexte : Le faible poids à la naissance est un facteur crucial de la mortalité et de la morbidité infantiles et touche près de 20 % des nourrissons dans le monde, principalement dans les pays à revenu faible et intermédiaire. Objectifs : La présente étude avait pour objet d’évaluer le lien entre l’accès aux soins prénatals et leur qualité d’une part, et le faible poids de naissance d’autre part. Méthodes : Nous avons analysé les données de 18 enquêtes démographiques et sanitaires, de 2005 à 2013, portant sur 69 446 enfants. Le principal résultat de l’étude concernait un poids de naissance inférieur à 2,5 kg. L ’ accès aux visites prénatales et le nombre de consultations étaient des variables d’exposition. En outre, le personnel de consultation prénatale et le calendrier des visites (trimestrielles) ont été pris en compte. La régression logistique multiple ajustée pour l ’ échantillonnage aux niveaux primaire et national a été utilisée. Résultats : Deux facteurs, en l’occurrence le fait d’avoir au moins une consultation prénatale et un nombre de visites supérieur ou égal à quatre, ont été associés à une diminution de la probabilité de faible poids de naissance par rapport à l’absence de visite et à un nombre de consultations prénatales inférieur à quatre, respectivement. La mise à disposition de personnel de consultation prénatale qualifié et la première consultation prénatale prévue au cours du premier trimestre de la grossesse constituaient également des avantages. Conclusions : Une couverture adéquate des soins prénatals pendant la grossesse est bénéfique pour prévenir le faible poids de naissance dans les pays à revenu faible et intermédiaire. ًاحسم 18 نم ةدَمتسم لئلاد :ةدلاولا دنع نزولا ضافخناب اتهدوجو ةدلاولل ةقباسلا ةياعرلا لىع لوصلحا ةقلاع ًايناكس ًايحص ينيرداب انازوس ،يزيليب وييرفاس ةصلالخا نم ٪20 نم برقي ام لىع رّثؤيو ،ملهلاتعاو لافطلأا تاَيَفَو لىإ يدؤت دق يتلا ةمسالحا لماوعلا دحأ ةدلاولا دنع نزولا ضافخنا دعُي :ةيفللخا .لخدلا ةطسوتلماو ةضفخنلما نادلبلا في مهمظعمو ،لماعلا ءاحنأ عيجم في ع َّضُرلا .ةدلاولا دنع نزولا ضافخناو اتهدوجو ةدلاولل ةقباسلا ةياعرلا لىع لوصلحا ينب ةقلاعلا مييقت لىإ ةساردلا هذه تفده :فادهلأا ةجيتن تناكو .ًلافط 446 69 تلمش ،2013 لىإ 2005 نم ةترفلا في ًايناكسو ًايحص ًاحسم 18 نم ةدَمتسم تانايبل ًلايلتح انيرجأ :ثحبلا قرط تايرغتم ةباثمب ابه ةقلعتلما تارايزلا ددعو ةدلاولل ةقباسلا ةياعرلا لىع لوصلحا ناكو ،مجك 2.5 نم لقأ ةدلاولا دنع نزولا نأ ةيسيئرلا ةساردلا رادحنا مدخُتساو ،)رهشأ ةثلاث لك لملحا ةلحرم( ةرايزلا تقوو ،ةدلاولل ةقباسلا ةياعرلا ييئاصخأ رابتعلاا ينعب ةساردلا تذخأ ماك .ض ُّرعتلل .يرْطُقلاو ليولأا ينيوتسلما لىع تانيعلا ذخأ ةاعارلم هحيحصت مت ددعتم يتسجول نزولا ضافخنا تلاماتحا ةلقب ةدلاولل ةقباسلا ةياعرلاب قلعتي مايف رثكأ وأ ةيبط تارايز 4و ةدحاو ةيبط ةرايز نع لقي لا ام ءارجإ طبترا :جئاتنلا ميدقت في نيرهام ينيئاصخأ ر ُّفوت نم ةيفاضإ ةدئاف تأشنو .لياوتلا لىع ،ةيبط تارايز 4 نم لقأ ءارجإو ةيبط تارايز ءارجإ مدعب ًةنراقم ةدلاولا دنع .لملحا نم لولأا ثلثلا ءانثأ ةدلاولل ةقباسلا ةياعرلاب قلعتي مايف ةيبط ةرايز لوأ ءارجإو ،ةدلاولل ةقباسلا ةياعرلا ةطسوتلماو ةضفخنلما نادلبلا في ةدلاولا دنع نزولا ضافخنا نم ةياقولل ةديفم لملحا ءانثأ ةدلاولل ةقباسلا ةياعرلاب ةبسانلما ةيطغتلا :تاجاتنتسلاا .لخدلا 1386 EMHJ – Vol. 26 No. 11 – 2020Research article References 1. Comprehensive implementation plan on maternal, infant and young child nutrition. WHA 65/6. Geneva: World Health Organi- zation; 2012 (https://www.who.int/nutrition/publications/CIP_document/en/, accessed 6 May 2020). 2. Global targets 2025. To improve maternal, infant and young child nutrition. Geneva: World Health Organization. (https://www. who.int/nutrition/global-target-2025/en/, accessed 6 May 2020). 3. Guidelines on Optimal feeding of low birthweight infants in low- and middle-income countries. Geneva: World Health Organi- zation; 2011 (https://www.who.int/maternal_child_adolescent/documents/infant_feeding_low_bw/en/, accessed 6 May 2020). 4. Romero C, Duke J, Dabelea D, Romero T, Ogden LG. Does the epidemiologic paradox hold in the presence of risk factors for low birth weight infants among Mexican-born women in Colorado? J Health Care Poor Underserved. 2012 May;23(2):604–14. http:// dx.doi.org/10.1353/hpu.2012.0065 PMID:22643610 5. Care of the preterm and/or low-birth-weight newborn. World Prematurity Day – 17 November 2018 [website]. World Health Organization (http://www.who.int/maternal_child_adolescent/newborns/prematurity/en/, accessed 6 May 2020). 6. Lawn J, Blencowe H, Oza S, You D, Lee AC, Waiswa P, et al. Every newborn: progress, priorities, and potential beyond survival. Lancet. 2014 Jul 12;384(9938):189–205. http://dx.doi.org/10.1016/S0140-6736(14)60496-7 PMID:24853593 7. Varga P, Berecz B, Gasparics A, Dombi Z, Varga Z, Jeager J, et al. Morbidity and mortality trends in very-very low birth weight premature infants in light of recent changes in obstetric care. Eur J Gynecol Reprod Biol. 2017 Apr;211:134–9. http://dx.doi. org/10.1016/j.ejogrb.2017.01.051 PMID:28258032 8. Nomura Y, Wickramaatne PJ, Pilowsky DJ, Newcorn JH, Bruder-Costello B, Davey C, et al. Low birthweight and risk of affective disorders & selected medical illness in offspring at high and low risk for depression. Compr Psychiatry. 2007 Sep–Oct;48(5):470– 8. http://dx.doi.org/10.1016/j.comppsych.2007.04.005 PMID:17707257 9. Farajdokht F, Sadigh-Eteghad S, Dehghani R, Mohaddes G, Abedi L, Bughchechi R, et al. Very low birth weight is associated with brain structure abnormalities and cognitive function impairments: a systematic review. Brain Cogn. 2017 Nov;118:80–9. http:// dx.doi.org/10.1016/j.bandc.2017.07.006 PMID:28802183 10. Wang SF, Shu L, Sheng J, Mu M, Wang S, Tao XY, et al. Birth weight and risk of coronary heart disease in adults: a meta-anal- ysis of prospective cohort studies. J Dev Orig Health Dis. 2014 Dec;5(6):408–19. http://dx.doi.org/10.1017/S2040174414000440 PMID:25263759 11. Ova J, Nakagami T, Kurita M, Yamamoto Y, Hasegawa Y, Tanaka Y, et al. Association of birthweight with diabetes and insulin sensitivity or secretion in the Japanese general population. J Diabetes Investig. 2015 Jul;6(4):430–5. http://dx.doi.org/10.1111/ jdi.12325 PMID:26221521 12. Gao W, Paterson J, Carter S, Percival T. Risk factors for preterm and small-for-gestational-age babies: a cohort from the Pacif- ic Islands Families Study. J Paediatr Child Health. 2006 Dec;42(12):785–92. http://dx.doi.org/10.1111/j.1440-1754.2006.00978.x PMID:17096714 13. Nguyen N, Savitz DA, Thorp JM. Risk factors for preterm birth in Vietnam. Int J Gynaecol Obstet. 2004 Jul;86(1):70–8. http:// dx.doi.org/10.1016/j.ijgo.2004.04.003 PMID:15207686 14. Sebayang SK, Dibley MJ, Kelly PJ, Shankar AV, Shankar AH; SUMMIT Study Group. Determinants of low birthweight, small-for- gestational-age and preterm birth in Lombok, Indonesia: analysis of the birthweight cohort of the SUMMIT trial. Trop Med Int Health. 2012 Aug;17(8):938–50. http://dx.doi.org/10.1111/j.1365-3156.2012.03039.x PMID:22943372 15. Mpharm SA, Hababeh M, Kitamura A, Hajat S. Factors associated with low birthweight in Palestinian refugees at Al-Wehdat camp, Jordan: a cross-sectional study using routine data. Lancet. 2019;393(S8). https://doi.org/10.1016/S0140-6736(19)30594-X 16. Provision of effective antenatal care: integrated management of pregnancy and child birth (IMPAC). Standards for maternal and neonatal care (1.6), Department of Making Pregnancy Safer, World Health Organization; 2006 (https://www.who.int/reproduc- tivehealth/publications/maternal_perinatal_health/effective_antenatal_care.pdf, accessed 6 May 2020). 17. Nazim NH, Fan L. Does prenatal healthcare improve child birth weight outcomes in Azerbaijan? Results of the national demo- graphic and health survey. Econ Hum Biol. 2011 Jan;9(1):56–65. http://dx.doi.org/10.1016/j.ehb.2010.08.003 PMID:20851064 18. Jewell RT, Triunfo P. The impact of prenatal care on birth weight: the case of Uruguay. Health Econ. 2006 Nov;15(11):1245–50. http://dx.doi.org/10.1002/hec.1121 PMID:16786548 19. Reichman NE, Corman H, Noonan K, Dave D. Infant health production functions: what a difference the data make. Health Econ. 2009 Jul;18(7):761–82. http://dx.doi.org/10.1002/hec.1402 PMID:18792077 20. The DHS Program. Demographic and health surveys [website] (Dhsprogram.com/publications/index.cfm, accessed 6 May 2020). 21. Nyaga VN, Arbyn M, Aerts M. Metaprop: a Stata command to perform meta-analysis of binomial data. Arch Public Health. 2014 Nov 10;72(1):39. http://dx.doi.org/10.1186/2049-3258-72-39 PMID:25810908 22. Low birthweight: country, regional and global estimates. New York: United Nations Children’s Fund and World Health Organi- zation; 2004 (http://www.who.int/iris/bitstream/10665/43184/1/9280638327.pdf, accessed 6 May 2020). 23. F. Jafari, H. Eftekhar, A. Pourreza, J. Mousavi. Socio-economic and medical determinants of low birth weight in Iran: 20 years after establishment of a primary healthcare network. Public Health. 2010 Mar;124(3):153–8. http://dx.doi.org/10.1016/j. puhe.2010.02.003 PMID:20226486 1387 EMHJ – Vol. 26 No. 11 – 2020Research article 24. Merklinger-Gruchala A, Jasienska G, Kapiszewska M. Short interpregnancy interval and low birth weight: A role of parity. Am J Hum Biol. 2015 Sep–Oct;27(5):660–6. http://dx.doi.org/10.1002/ajhb.22708 PMID:25754897 25. Khanal V, Zhao Y, Sauer K. Role of antenatal care and iron supplementation during pregnancy in preventing low birth weight in Nepal: comparison of national surveys 2006 and 2011. Arch Public Health 2014 Feb;72(1):4. http://dx.doi.org/10.1186/2049-3258-72- 4 PMID:24499636 26. Pinzon-Rondon AM, Gutierrez-Pinzon V, Madrinan-Navia H, Amin J, Aguilera-Otalvaro P, Hoyos-Martínez A. Low birth weight and prenatal care in Colombia: a cross-sectional study. BMC Pregnancy Childbirth. 2015 May 20;15:118. http://dx.doi.org/10.1186/ s12884-015-0541-0 PMID:25989797 27. Awiti JO. A multilevel analysis of prenatal care and birth weight in Kenya. Health Econ Rev 2014 Dec;4(1):33. http://dx.doi. org/10.1186/s13561-014-0033-3 PMID:26208933 28. Blanc AK, Wardlaw T. Monitoring low birth weight: an evaluation of international estimates and an updated estimation proce- dure. Bull World Health Organ. 2005 Mar;83(3):178–85. http://dx.doi.org/S0042-96862005000300010 PMID:15798841 1388 EMHJ – Vol. 26 No. 11 – 2020Research article Macromineral enrichment of white bread reduces postprandial glycaemia without altering sensory properties: a crossover study Rania El Khoury,1 Noor El Solh,1 Ammar Olabi,1 Imad Toufeili,1 Sani Hlais1 and Omar Obeid1 1Department of Nutrition and Food Science, Faculty of Agricultural and Food Sciences, American University of Beirut, Beirut, Lebanon (Correspondence to: O. Obeid: omar.obeid@aub.edu.lb). Abstract Background: Metabolism of refined carbohydrates, which are associated with detrimental health effects, is known to be affected by macrominerals including P, Mg and K. Aims: To assess the impact of their addition to flour on the sensory properties of white pita bread and postprandial gly- caemia of healthy individuals. Methods: The study was conducted at the American University of Beirut (between February and October 2014). Plain, restored and fortified wheat flour, with macrominerals were used to prepare 3 types of bread: white pita bread (WP), restored white pita bread (WP-R) (premilling levels) and fortified white pita bread (WP-F) (double the premilling levels). Sensory characteristics of bread were assessed and postprandial glycaemia was determined using a single-blinded cross- over design whereby participants consumed 1 of the 3 different types of pita bread in random order. Results: No significant difference (P > 0.05) between the different types of bread was detected using the triangle and ac- ceptability tests, except for texture (P < 0.05). Macromineral enrichment of bread (WP-R and WP-F) significantly reduced postprandial glucose (P = 0.013) and triglyceride (P = 0.001) levels. Conclusions: Macromineral enrichment of refined carbohydrates may have a promising role in lowering postprandial glucose and triglycerides, and thus decrease their negative health consequences.. Keywords: sensory properties, glucose, triglyceride, white bread, macromineral enrichment Citation: El Khoury R; El Solh N; Olabi A; Toufeili I; Hlais S; Obeid O. Macromineral enrichment of white bread reduces postprandial glycaemia without altering sensory properties: a crossover study. East Mediterr Health J. 2020;16(11):1388–1395. https://doi.org/10.26719/2020.26.11.1388 Received: 16/05/18; accepted: 26/02/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Introduction Over the past few decades, there have gradual but signif- icant changes in eating behaviour worldwide in light of nutritional transition from traditional diets rich in com- plex carbohydrates to diets high in simple carbohydrates (1). These changes were associated with an increase in the prevalence of chronic diseases and accordingly, the recent dietary guidelines in the United States of America (2) strongly emphasize the importance of reducing sim- ple carbohydrates. Fibre, vitamin and mineral content of flour is drastically reduced by conventional milling and grain refinement processes. Macrominerals including P, Mg and K, are reduced by about 69%, 74% and 84%, re- spectively (3), and are known to improve postprandial glucose and insulin metabolism (4,5). P plays an essential role in carbohydrate metabolism via phosphorylation of glucose to glucose-6-phosphate; an essential step for glu- cose clearance and trapping into cells (6). The need for P is highest during the postprandial period, as indicated by its reduced level after glucose ingestion and by the im- provement in insulin sensitivity following its addition (7). Mg mediates glucose transport mechanisms into the cell membranes through its effect on insulin signalling via tyrosine kinase activity, phosphorylase B kinase ac- tivity and glucose transporter protein activity (8,9). K in its turn acts as a cofactor for several enzymes involved in carbohydrate phosphorylation and oxidation, such as protein kinases and phosphatases. K is also known to affect glucose tolerance (5). Thus, low availability of the above-mentioned macrominerals would be expected to delay postprandial cellular uptake of glucose, impair phosphorylation, and eventually hinder carbohydrate metabolism and energy production (10). These conditions would ultimately favour the onset and development of the different components of metabolic syndrome, espe- cially impaired glucose tolerance and diabetes (11). In this same context, diminished insulin sensitivity is known to promote hypertriglyceridaemia (12); therefore, serum tri- glyceride (TG) levels are expected to increase in a setting of low mineral availability. Refined white flour has received extensive worldwide acceptance, since it is used to produce baked goods that are more palatable, softer in texture and have extended freshness. White pita bread is heavily consumed in the Middle East and increasingly in Europe and North America. Therefore, the objective of this work was to assess the impact of P, Mg and K enrichment on sensory properties and postprandial glycaemia of white pita bread. 1389 EMHJ – Vol. 26 No. 11 – 2020Research article Methods Study design This study was conducted between February and October 2014, according to the Declaration of Helsinki and all pro- cedures involving human subjects were approved by the Institutional Review Board at the American University of Beirut (approval no. NUT0019). Written informed con- sent was obtained from all participants. The clinical trial was registered with Clinical Trial.gov, NCT02598986. Wheat flour (80% extraction; Bakalian Flour Mills, Beirut, Lebanon) was used and 2 levels of mineral supplementation were made. Restoration: minerals were added to white flour so that each kilogram contained 3.6 g MgCO3 (G&G Vitamins, East Grinstead, UK) and 12.5 g KH2PO4 (Dyets, Bethlehem, PA, United State of America). Fortification: minerals were added to white flour to almost double the original levels, so that each kilogram of white flour contained 7.2 g MgCO3 and 25 g KH2PO4. The amounts of added P and Mg were considered safe since both were lower than the tolerable upper limits set at 4 g/ day and 350 mg/day for P and Mg, respectively (13). After supplementation, different types of white pita bread were made and used for the different tests. White pita bread making Bread samples were prepared as previously described (14). Upon termination of the bread making process, 3 samples from each type of bread [white pita bread (WP), white pita bread-restored (WP-R) and white pita bread-fortified (WP-F)] were analysed for their mineral content by inductively coupled plasma mass spectrom- etry (ICP-MS) using the standard method EPA 200 – 7/8 (15). P, Mg and K contents of WP-R were 84%, 200% and 60% higher than those of WP, respectively (Table 1). P, Mg and K contents of WP-F were 260%, 410% and 230% higher than those of WP, respectively. P, Mg and K contents of WP-F were almost double those of the WP-R. Experiment 1: difference and acceptability sensory tests Twenty-four healthy untrained male volunteers partici- pated in a difference/discrimination test. Two triangular tests were conducted to compare WP versus WP-R or WP-F. Panellists were asked to indicate the odd sample in each set and to rinse their mouths before each sam- ple. A consumer acceptability test was conducted with 60 healthy randomly recruited panellists (29 women and 31 men, mean age 22 years, range 19–29 years) from the American University of Beirut as described previously (16). The 3 samples used in different tests were assessed. Ten grams of each type of white pita bread were prepared 2 hours prior to serving them and were stored in the re- frigerator (4°C). Panellists rated overall acceptability, ap- pearance, colour, odour, flavour and texture on a 9-point hedonic scale (17). Panellists were instructed to rinse their mouths before each sample. The order of the samples within each set was randomized among the panellists in both tests. Experiment 2: determination of postprandial glucose and triglyceride Independently from the first experiment, 11 healthy male volunteers were recruited and asked to maintain their regular dietary habits and physical activity during the entire study course, and to avoid alcohol consumption and unusual strenuous exercise 24 hours prior to each experimental session. Volunteers were aged 18–30 years (mean 24.5 years) with body mass index between 18.5 and 29.9 kg/m2, without significant medical or chronic diseas- es, with no regular use of medication that affected body weight, and without weight loss of ≥ 3% in the preceding 3 months. A single-blinded, randomized crossover study was conducted. Each participant consumed 1 of the 3 different types of pita bread on each of 3 visits. The order of meals was assigned randomly and the visits were separated by a minimum washout period of 10 days. In each session, overnight fasted participants were asked to ingest 90 g (containing 50 g carbohydrate) white pita bread within 10– 15 minutes and subsequently drink 200 ml water. Blood samples were collected at baseline (before ingestion) and at 15, 30, 45, 60, 90 and 120 minutes after ingestion. Blood samples were centrifuged for 15 minutes at 4°C at 2500 g and serum was stored in aliquots at −80°C until analysis. Serum glucose, TG, and total P, Mg and K were measured using the Vitros 350 Chemistry System (Ortho-Clinical Diagnostics, Johnson & Johnson, New York, United State of America). Fasting serum insulin was determined using an ELISA kit (Diametra Millipore Corporation, Billerica, MA, United State of America). Statistical analysis Experiment 1: data related to triangular tests were ana- lysed by checking the minimum number of correct re- sponses using a binomial table with P = 0.05 (17). As for the acceptability test, 2-way analysis of variance using the GLM procedure of SAS (version 9.02) was performed as described previously (16). In the statistical model for Table 1 Phosphorus, potassium and magnesium content of the different pita bread types Mineral Treatment WP (n=3) WP-R (n=3) WP-F (n=3) Phosphorus (g/kg) 3.20±0.01 5.90±0.00 11.60±0.00 Potassium (g/kg) 3.70±0.01 5.90±0.23 12.20 ±0.01 Magnesium (g/kg) 0.53±0.01 1.60±0.04 2.70±0.26 WP=white pita bread; WP-R=restored white pita bread; WP-F=fortified white pita bread. Results are expressed as the mean ± standard deviation (SD). 1390 EMHJ – Vol. 26 No. 11 – 2020Research article acceptability, the response variable was the specific ac- ceptability variable. Factors in the model were the pan- ellist and treatment (WP, WP-R and WP-F). The panel- list was included as a random effect and treatment as a fixed effect. Means were separated by Tukey’s honestly significant difference test. For all data, significance was established at P < 0.05. Experiment 2: The difference (Δ) in serum total P, Mg, K, TG and glucose was calculated. This represents the value at each time point minus the value at time 0. Repeated-measures analysis of variance was used to determine statistical significance with effects of bread type, time, and bread type × time interaction. Results Experiment 1 Difference test and hedonic acceptability In the triangular difference test, 13 correct answers out of the 24 responses were needed to show a significant difference. However, only 8 and 10 panellists responded correctly for the WP versus WP-R and WP versus WP-F tests, respectively (both P > 0.05). Therefore, the trian- gular tests did not detect any significant differences be- tween the different types of bread. The consumer acceptability test (Table 2) found no significant differences for most acceptability attributes (overall acceptability, appearance, colour, odour and flavour; P > 0.05). Texture, however, was significantly more liked than that of the WP-F bread (P < 0.05), although no significant difference was detected between WP and WP-R or WP-R and WP-F bread. Experiment 2 Participants’ characteristics Baseline fasting serum levels of glucose, insulin, homeo- stasis model assessment of insulin resistance (HOMA-IR) (18), TG and total P, K and Mg were within the normal ranges (Table 3), and these were found to be similar be- tween the different experimental sessions for each type of bread. Postprandial mineral responses Results were expressed as changes from baseline, which were the difference between the macromineral levels at each time point minus their corresponding values at baseline. Postprandial serum levels of the measured mac- rominerals were altered by food ingestion. Serum total P decreased following ingestion of all bread types, al- though this failed to reach statistical significance. How- ever, the changes in serum total P were significant be- tween bread types (P = 0.015), and serum P in WP-F bread returned to baseline by the end of the session (Figure 1A). Postprandial serum Mg levels experienced a gradual and significant increase with time (P = 0.027), although no significant difference was detected among the dif- ferent bread types (Figure 1B), despite their varied con- tent of Mg (Table 1). In contrast to Mg, postprandial K levels decreased with time, although not significantly (Figure 1C), and were significantly different among bread types (P = 0.001). Postprandial TG and glucose responses Results of TG and glucose were also expressed as differ- ences from baseline. The changes in postprandial serum TG (Figure 2A) differed significantly among bread types (P = 0.001), and WP-R and WP-F maintained lower lev- els at all time points. Similarly, changes in postprandial serum glucose (Figure 2B) differed significantly among bread types (P < 0.013) and over time (P < 0.001). Serum glucose levels peaked at 30–45 minutes after ingestion and the peaks were sooner with the enriched pita breads. Thereafter, the enriched breads exhibited a faster de- crease in serum glucose as compared to the WP bread, starting from 45 minutes until the end of the experiment. The magnitude of the decrease seemed to be synergisti- cally related to the mineral content of the bread. Discussion This study was designed to investigate the glycaemic response of macronutrient-enriched pita bread, as well as its sensory properties. Our results showed that the palatability of white pita bread was not affected by the addition of macrominerals, as indicated by the lack of dif- ferences in the triangular and acceptability tests. How- ever, a small difference in texture was detected between WP and WP-F but not the triangular test, which is known to be more attentive to differences. Hence, no major dif- ferences were observed when the bread was assessed in its entirety. Our findings are in line with other studies, in which addition of K, Ca and Mg salts as replacements for NaCl did not yield any differences in appearance, tex- ture and taste of brown bread (19). Therefore, it can be concluded that the addition of macrominerals to white wheat flour in an amount comparable to that found in Table 2 Hedonic acceptability variables for the different pita bread types Acceptability variables Overall acceptability Appearance Colour Odour Flavour Texture WP 6.27±1.33 6.22±1.17 6.32±1.08 6.08±1.34 6.38±1.54 6.35±1.72a WP-R 6.25±1.49 6.32±1.56 6.52±1.19 6.23±1.28 5.87±1.78 5.95±1.84ab WP-F 6.07±1.33 6.28±1.21 6.40±1.39 6.12±1.53 5.85±1.62 5.42±1.71b P value 0.601 0.861 0.542 0.795 0.066 0.004 Results are expressed as mean ± standard deviation (SD). a,bMeans with different superscripts are statistically significant (P < 0.05) as analysed by paired t-test. WP = white pita bread; WP-R = restored white pita bread; WP-F = fortified white pita bread. 1391 EMHJ – Vol. 26 No. 11 – 2020Research article whole wheat flour, and even in double quantities, does not significantly affect acceptability of white pita bread. The reduction in serum P following ingestion of the different types of bread was in line with other studies (7,20), and this is mediated by insulin, which is known to stimulate peripheral uptake of both glucose and P. Thus, insulin favours glucose phosphorylation (21) in a manner that mimics the action of glucokinase activators (22). The inability of WP and WP-R to normalize serum P (return to baseline at 120 minutes) unlike that of WP-F (7) implies that their P content was not sufficient to meet the needs of intracellular phosphorylation. Moreover, the observed nonsynergistic relation between Mg content of the bread and changes in postprandial serum Mg is likely to result from the ability of P to potentiate insulin sensitivity (7,23,24), which is known to stimulate Mg clearance (20). Furthermore, improvement in insulin sensitivity may have also been attributed to the nonsynergistic relation between K content of bread and changes in postprandial serum K levels. Likewise, Mg (25,26) and K (27) are reported to improve glucose clearance and insulin sensitivity. The ability of macrominerals (P, Mg and K) to enhance their own intracellular uptake may help to explain the reported inverse association between P intake and blood pressure (28). This further implies that postprandial levels of these macrominerals depend on a balance between their availability in the circulation and their capacity for intracellular uptake and storage, and this implies that their circulating level is not a good indicator of their bodily status. At the glycaemic level, the inverse association between the macromineral content of bread and postprandial glucose level, especially from time 60 minutes (7), may have been the outcome of an improvement in glucose clearance due to the capacity of the added minerals to improve glucose phosphorylation and insulin sensitivity. Furthermore, this capacity may have also contributed to the observed reduction in serum TG after ingestion of enriched bread (WP-R and WP-F). Our findings suggest that postprandial glucose and TG levels, especially from 60 minutes, are dependent on exogenous factors including P, Mg and K. In support, P status was reported to correlate with a favourable lipid profile, including increased high-density lipoprotein and decreased serum TG levels (10,12). In agreement, we have recently found that the addition of P to a high-fat meal was able to alter postprandial lipidaemia by increasing apolipoprotein B48 and decreasing apolipoprotein B100 (29). Besides, Mg supplementation also improves postprandial lipidaemic response in healthy individuals (30). Worldwide, daily consumption of wheat and wheat products, mainly in the form of bread and pasta, is about 180 g per capita and this contributes to about 20% of total energy intake (31). Even with a consumption of 500 g/ day of WP-R, which is considered an excessive amount compared to the reported daily consumption of about 150 g per day (32) and is equivalent to around 4604.6 kJ (1100 kcal), the upper limit for both P and Mg would not be reached. The high palatability of white pita bread makes it popular and a major contributor to overall glycaemic load, which increases the risk of development of diabetes, abnormal lipid profile and obesity (33,34). Nonetheless, the health benefits of whole wheat cereal products are reported not to be related to their fibre content (35); therefore, our findings may partially explain the benefits of whole wheat products that are known to have high content of macrominerals, specifically P, Mg and K. Even though the beneficial effects of whole grains have been widely publicized, the adoption of diets rich in whole grains is still facing resistance, probably due to their low palatability. The major limitation of this study was that postprandial insulin and other appetite hormone levels were not measured. In addition, the contribution of each mineral to the observed changes was not clear. Further studies are required to determine the postprandial response of prediabetic and diabetic patients, as well as the long-term impact of macronutrient enrichment on diabetes and different components of the metabolic syndrome. Conclusion White wheat flour enrichment with macrominerals (P, Mg and K) did not affect the palatability of white pita bread, while postprandial glucose and TG levels were re- duced. Furthermore, this supports the benefit of increas- ing the consumption of whole grain wheat, since it re- tains most of its mineral content. This study successfully identified the beneficial role of minerals in improving the glycaemic response of a simple carbohydrate product, white pita bread. The data may prove useful to ameliorate the detrimental potential effect of simple carbohydrates. Funding: The work was supported by the Farouk Jaber Innovative Biomedical Research Award from the Faculty of Medicine, American University of Beirut. Award num- ber 100410. The funder had no role in the design, analysis or writing of this article. Competing interests: None declared. Table 3 Baseline characteristics of the 11 participants Mean SD Age (yr) 23.7 2.4 Weight (kg) 85.6 10.6 Height (m) 1.80 0.04 BMI (kg/m2) 26.2 2.4 Fasting serum glucose (mg/dl) 95.0 7.8 Fasting serum insulin (μU/ml) 5.23 2.18 HOMA-IR 1.23 0.54 Fasting serum triglycerides (mg/dl) 92.5 29.4 Fasting serum phosphate (mg/dl) 3.76 0.53 Fasting serum potassium (mg/dl) 4.74 0.38 Fasting serum magnesium (mg/dl) 1.95 0.21 Results are expressed as mean and standard deviation (SD). BMI = body mass index; HOMA-IR = homeostatic model assessment of insulin resistance. 1392 EMHJ – Vol. 26 No. 11 – 2020Research article Figure 1 Postprandial changes in phosphorus (A), magnesium (B) and potassium (C) following the ingestion of different pita breads. All values are presented as mean and standard error of the mean. The difference reflects changes between the variable at each time point and the same variable at baseline (t = 0) 0.2 0.1 0 -0.1 -02 -0.3 -04 -0.5 -0.6 -0.7 0 15 30 45 60 75 90 105 120 Time (min) ∆ T ot al p ho sp ho ru s ( m g/ dl ) Two–way ANOVA Bread: P = 0.015 Time: P = 0.121 Group–Time: P = 0.896 A 0.25 0.2 0.15 0.1 0.5 0 -0.05 -0.1 -0.15 Time (min) ∆ T ot al m ag ne si um (m g/ dl ) Two–way ANOVA Bread: P = 0.001 Time: P = 0.083 Group–Time: P = 0.589 B 0 15 30 45 60 75 90 105 120 0.6 0.4 0.2 0 -02 -0.4 -0.6 -0.8 -1 Time (min) ∆ T ot al p ot as si um (m g/ dl ) Two–way ANOVA Bread: P = 0.001 Time: P = 0.257 Group–Time: P = 0.726 C 0 15 30 45 60 75 90 105 120 white pita bread white pita bread-fortifiedwhite pita bread-restored 1393 EMHJ – Vol. 26 No. 11 – 2020Research article Figure 2 Postprandial changes in triglycerides (A) and glucose (B) following the ingestion of different pita breads. All values are presented as mean and standard error of the mean. The difference reflects changes between the variable at each time point and the same variable at baseline (t = 0) 10 5 0 -5 -10 -15 -20 -25 0 15 30 45 60 75 90 105 120 Time (min) ∆ T ot al tr ig ly ce ri de s ( m g/ dl ) Two–way ANOVA Bread: P = 0.001 Time: P = 0.083 Group–Time: P = 0.589 A 35 30 25 20 15 10 5 0 -5 -10 -15 0 15 30 45 60 75 90 105 120 Time (min) ∆ T ot al g lu co se (m g/ dl ) Two–way ANOVA Bread: P = 0.013 Time: P = 0.001 Group–Time: P = 0.388 B white pita bread white pita bread-fortifiedwhite pita bread-restored Enrichissement en macro-minéraux du pain blanc et réduction de la glycémie postprandiale sans altération des propriétés sensorielles : étude croisée Résumé Contexte : On sait que le métabolisme des glucides raffinés, qui sont associés à des effets nocifs sur la santé, est affecté par les macro-minéraux, notamment le phosphore, le magnésium et le potassium. Objectifs : Évaluer l’impact de l’ajout de ces macro-minéraux à la farine sur les propriétés sensorielles du pain pita blanc et sur la glycémie postprandiale d’individus en bonne santé. Méthodes : La présente étude a été menée à l’Université américaine de Beyrouth (entre février et octobre 2014). De la farine nature, de blé germé et enrichie en macro-minéraux a été utilisée pour préparer 3 types de pain : du pain pita blanc, du pain pita blanc à la farine de blé germé (degrés de prémouture) et du pain pita blanc enrichi (degrés de prémouture multipliés par deux). Les caractéristiques sensorielles du pain ont été évaluées et la glycémie postprandiale a été déterminée à l’aide d’un modèle croisé en simple aveugle, dans lequel les participants ont consommé un des trois différents types de pain pita dans un ordre aléatoire. 1394 EMHJ – Vol. 26 No. 11 – 2020Research article References 1. Popkin BM, Gordon-Larsen P. The nutrition transition: worldwide obesity dynamics and their determinants. Int J Obes. 2004 Nov;28(Suppl 3):S2–S9. http://dx.doi.org/10.1038/sj.ijo.0802804 PMID:15543214 2. 2015–2020 Dietary guidelines for Americans, 8th edition. Washington DC: U.S. Department of Health and Human Services and U.S. Department of Agriculture; 2015. (http://health.gov/dietaryguidelines/2015/guidelines/, accessed 20 December 2019). 3. FoodData Central [website]. Washington DC: U.S. Department of Agriculture, Agricultural Research Service; 2019 (https://fdc.nal. usda.gov, accessed 20 December 2019). 4. Larsson SC, Wolk A. Magnesium intake and risk of type 2 diabetes: a meta‐analysis. J Intern Med. 2007 Aug;262(2):208–14. http:// dx.doi.org/10.1111/j.1365-2796.2007.01840.x PMID:17645588 5. He FJ, MacGregor GA. Beneficial effects of potassium on human health. Physiol Plant. 2008 Aug;133(4):725–35. http://dx.doi. org/10.1111/j.1399-3054.2007.01033.x PMID:18724413 6. Bouché C, Serdy S, Kahn CR, Goldfine AB. The cellular fate of glucose and its relevance in type 2 diabetes. Endocr Rev. 2004 Oct;25(5):807–30. http://dx.doi.org/10.1210/er.2003-0026 PMID:15466941 7. Khattab M, Abi-Rashed C, Ghattas H, Hlais S, Obeid O. Phosphorus ingestion improves oral glucose tolerance in healthy male subjects: a crossover experiment. Nutr J. 2015 Oct 29;14:1. http://dx.doi.org/10.1186/s12937-015-0101-5 PMID:26514124 8. Suarez A, Pulido N, Casla A, Casanova B, Arrieta FJ, Rovira A. Impaired tyrosine-kinase activity of muscle insulin receptors from hypomagnesaemic rats. Diabetologia. 1995 Nov;38(11):1262–70. http://dx.doi.org/10.1007/bf00401757 PMID:8582534 9. Barbagallo M, Dominguez LJ, Galioto A, Ferlisi A, Cani C, Malfa L, et al. Role of magnesium in insulin action, diabetes and cardio-metabolic syndrome X. Mol Aspects Med. 2003 Feb–Jun;24(1–3):39–52. http://dx.doi.org/10.1016/s0098-2997(02)00090-0 PMID:12537988 Résultats : Aucune différence significative (p > 0,05) entre les différents types de pain n’a été détectée à l’aide de la méthode triangulaire et des tests d’acceptabilité, à l’exception de la texture (p < 0,05). L’enrichissement en macro-minéraux du pain (à la farine de blé germé et à la farine enrichie) a permis de réduire significativement les taux de glucose (p = 0,013) et de triglycérides (p = 0,001) postprandiaux. Conclusions : L’enrichissement en macro-minéraux des glucides raffinés peut jouer un rôle prometteur dans la diminution du glucose et des triglycérides postprandiaux, et ainsi diminuer leurs conséquences néfastes sur la santé. ةسارد :ةيسلحا صئاصلخا يريغت نود لكلأا دعب مدلا في ركسلا ةبسن ض ِّفُي ةيربكلا نداعلماب ضيبلأا زبلخا ءانغإ ةبوانلماب ديبع رمع ،سيلح نياس ،لييفطلا دماع ،يبلُعلا رماع ،حل ُّصلا رون ،يرولخا اينار ةصلالخا اهنع مجني يتلا ةاّقنُلما تارديهوبركلل يئاذغلا ليثمتلا لىع رّثؤت مويساتوبلاو مويسينجلماو روفسوفلا اهيف ماب ةيربكلا نداعلما نأ فورعلما نم :ةيفللخا .ةراض ةيحص راثآ لكلأا دعب مدلا ركسو ضيبلأا بيرعلا زبخلل ةيسلحا صئاصلخا لىع قيقدلا لىإ ةيربكلا نداعلما ةفاضإ يرثأت مييقت لىإ ةساردلا هذه تفده :فادهلأا .ءاّحصلأا دارفلأا ىدل حمقلا قيقد مَدختسُيو .)2014 /لولأا نيشرت/ربوتكأو طابش/ريابرف ينب ةترفلا في( تويرب في ةيكيرملأا ةعمالجا في ةساردلا تيرجُأ :ثحبلا قرط تايوتسم( داعتسُلما ضيبلأا بيرعلا زبلخاو ،ضيبلأا بيرعلا زبلخا :زبلخا نم عاونأ ةثلاث دادعلإ ةيربكلا نداعلماب د َّوزلما ،م َّعدُلماو داعتسُلماو يداعلا مدلا في ركسلا ةبسن تد ِّدُحو زبخلل ةيسلحا صئاصلخا تَمّيُقو .)قَبسُلما نحطلا تايوتسم ةفعاضم( م َّعدُلما ضيبلأا بيرعلا زبلخاو ،)قَبسُلما نحطلا .يئاوشع ٍبيتترب ةفلتخلما ةثلاثلا بيرعلا زبلخا عاونأ ينب نم ًادحاو ًاعون نوكراشلما لوانت ثيح ةيمعتلا درفُم بيوانت ميمصت مادختساب لكلأا دعب ى َّدأو .)p<0.05( ماوقلا ءانثتساب ،ةيلوبقلماو ثلثلما تارابتخا مادختساب زبلخا عاونأ فلتمخ ينب )p>0.05( يربك نيابت ُّيأ فَشتكُي لم :جئاتنلا )p=0.013( لكلأا دعب زوكوللجا تايوتسم ضافخنا لىإ )م َّعدُلما ضيبلأا بيرعلا زبلخاو داعتسُلما ضيبلأا بيرعلا زبلخا( ةيربكلا نداعلماب زبلخا ءانغإ .ةظوحلم ٍةروصب )p=0.001( مدلا دييرسيلج يثلاثو دلحا لياتلابو ،لكلأا دعب مدلا دييرسيلغ يثلاثو زوكولغلا ضفخ في ٌدعاو ٌرود ةيربكلا نداعلماب ةاّقنُلما تارديهوبركلا ءانغلإ نوكي دق :تاجاتنتسلاا .ةحصلا لىع ةيبلسلا اهراثآ نم 1395 EMHJ – Vol. 26 No. 11 – 2020Research article 10. Kalaitzidis R, Tsimihodimos V, Bairaktari E, Siamopoulos KC, Elisaf M. Disturbances of phosphate metabolism: another feature of metabolic syndrome. Am J Kidney Dis. 2005 May;45(5):851–8. http://dx.doi.org/10.1053/j.ajkd.2005.01.005 PMID:15861350 11. Obeid O, Hachem D, Ayoub J. Refeeding and metabolic syndromes: two sides of the same coin. Nutr Diabetes. 2014 Jun 30;4:e120. http://dx.doi.org/10.1038/nutd.2014.21 PMID:24979149 12. Grundy SM. Hypertriglyceridemia, insulin resistance, and the metabolic syndrome. Am J Cardiol. 1999 May 13;83(9B):25F–9F. http://dx.doi.org/10.1016/s0002-9149(99)00211-8 PMID:10357572 13. Institute of Medicine. Dietary reference intakes for calcium, phosphorus, magnesium, vitamin d, and fluoride. Washington, DC: National Academies Press; 1997. 14. Toufeili I, Ismail B, Shadarevian S, Baalbaki R, Khatkar BS, Bell AE, et al. The role of gluten proteins in the baking of Arabic bread. J Cereal Sci. 1999 Nov;30(3):255–65. https://doi.org/10.1006/jcrs.1999.0286 15. Methods for the determination of metals in environmental samples. United States Environmental Protection Agency; 1991. 16. Lteif L, Olabi A, Baghdadi OK, Toufeili I. The characterization of the physicochemical and sensory properties of full-fat, re- duced-fat, and low-fat ovine and bovine Halloumi. J Dairy Sci. 2009 Sep;92(9):4135–45. http://dx.doi.org/10.3168/jds.2009-2070 PMID:19700674 17. Lawless HT, Heymann H. Acceptance testing. In: Sensory evaluation of food: principles and practices, 2nd edition. New York: Springer Science & Business Media; 2010:325–36. 18. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985 Jul;28(7):412–9. http:// dx.doi.org/10.1007/bf00280883 PMID:3899825 19. Charlton KE, Macgregor E, Vorster NH, Levitt NS, Steyn K. Partial replacement of NaCl can be achieved with potassium, magne- sium and calcium salts in brown bread. Int J Food Sci Nutr. 2007 Nov;58(7):508–21. http://dx.doi.org/10.1080/09637480701331148 PMID:17852502 20. Rosenbloom AL. Serum calcium and magnesium decline during oral glucose tolerance testing in children and adolescents with preclinical diabetes mellitus less than in normals. Metabolism. 1977 Sep;26(9):1033–9. http://dx.doi.org/10.1016/0026- 0495(77)90021-x PMID:895534 21. Amanzadeh J, Reilly RF. Hypophosphatemia: an evidence-based approach to its clinical consequences and management. Nat Clin Pract Nephrol. 2006 Mar;2(3):136–48. http://dx.doi.org/10.1038/ncpneph0124 PMID:16932412 22. Meininger GE, Scott R, Alba M, Shentu Y, Luo E, Amin H, et al. Effects of MK-0941, a novel glucokinase activator, on glycemic control in insulin-treated patients with type 2 diabetes. Diabetes Care. 2011 Dec;34(12):2560–6. http://dx.doi.org/10.2337/dc11-1200 PMID:21994424 23. Nowicki MI, Fliser DA, Fode PE, Ritz EB. Changes in plasma phosphate levels influence insulin sensitivity under euglycemic conditions. J Clin Endocrinol Metab. 1996 Jan;81(1):156–9. http://dx.doi.org/10.1210/jcem.81.1.8550745 PMID:8550745 24. Paolisso G, Scheen A, d’Onofrio F, Lefebvre P. Magnesium and glucose homeostasis. Diabetologia. 1990 Sep;33(9):511–4. http:// dx.doi.org/10.1007/bf00404136 PMID:2253826 25. Swaminathan R. Magnesium metabolism and its disorders. Clin Biochem Rev. 2003 May;24(2):47. PMID:18568054 26. Yajnik CS, Smith RF, Hockaday TD, Ward NI. Fasting plasma magnesium concentrations and glucose disposal in diabetes. Br Med J (Clin Res Ed). 1984 Apr 7;288(6243):1032–4. http://dx.doi.org/10.1136/bmj.288.6423.1032 PMID:6423182 27. Zillich AJ, Garg J, Basu S, Bakris GL, Carter BL. Thiazide diuretics, potassium, and the development of diabetes a quantitative review. Hypertension. 2006 Aug;48(2):219–24. http://dx.doi.org/10.1161/01.HYP.0000231552.10054.aa PMID:16801488 28. Nguyen TQ, Maalouf NM, Sakhaee K, Moe OW. Comparison of insulin action on glucose versus potassium uptake in humans. Clin J Am Soc Nephrol. 2011 Jul;6(7):1533–9. http://dx.doi.org/10.2215/CJN.00750111 PMID:21734082 29. Hazim J, Hlais S, Ghattas H, Shatila D, Bassil M, Obeid O. Phosphorus supplement alters postprandial lipemia in healthy male subjects: a pilot cross-over trial. Lipids Health Dis. 2014 Jul 8;13:1. http://dx.doi.org/10.1186/1476-511X-13-109 PMID:25002136 30. Kishimoto Y, Tani M, Uto-Kondo H, Saita E, Iizuka M, Sone H, Yokota K, et al. Effects of magnesium on postprandial serum lipid responses in healthy human subjects. Br J Nutr. 2010 Feb;103(4):469–72. http://dx.doi.org/10.1017/S0007114509992716 PMID:19941679 31. FAOSTAT. Food and agriculture data [website]. Food and Agriculture Organization of the United Nations (http://faostat3.fao.org/ download/FB/FBS/E, accessed 20 December 2019). 32. Nasreddine L, Hwalla N, Sibai A, Hamzé M, Parent-Massin D. Food consumption patterns in an adult urban population in Bei- rut, Lebanon. Public Health Nutr. 2006 Apr;9(2):194–203. http://dx.doi.org/10.1079/phn2005855 PMID:16571173 33. Chiu CJ, Liu S, Willett WC, Wolever TM, Brand-Miller JC, Barclay AW, et al. Informing food choices and health outcomes by use of the dietary glycemic index. Nutr Rev. 2011 Apr;69(4):231–42. http://dx.doi.org/10.1111/j.1753-4887.2011.00382.x PMID:21457267 34. Liu S, Manson JE, Stampfer MJ, Hu FB, Giovannucci E, Colditz GA, et al. A prospective study of whole-grain intake and risk of type 2 diabetes mellitus in US women. Am J Public Health. 2000 Sep;90(9):1409. http://dx.doi.org/10.2105/ajph.90.9.1409 PMID:10983198 35. Pereira MA, Jacobs DR, Pins JJ, Raatz SK, Gross MD, Slavin JL, et al. Effect of whole grains on insulin sensitivity in overweight hyperinsulinemic adults. Am J Clin Nutr. 2002 May;75(5):848–55. http://dx.doi.org/10.1093/ajcn/75.5.848 PMID:11976158 1396 Review EMHJ – Vol. 26 No. 11 – 2020 Rationing access to total hip and total knee replacement in the Islamic Republic of Iran to reduce unnecessary costs: policy brief Mohammad Soleimani,1 Shoresh Barkhordari,1 Farhad Mardani,2 Nasrin Shaarbafchizadeh3 and Fatemeh Naghavi-Al-Hosseini4 1Faculty of Medical Science, Isfahan University of Medical Sciences, Isfahan, Islamic Republic of Iran. 2Faculty of Dental Science, Isfahan University of Medical Sciences, Isfahan, Islamic Republic of Iran. 3Health Management and Economics Research Center, Faculty of Management and Medical Infor- mation, Isfahan University of Medical Sciences, Isfahan, Islamic Republic of Iran. 4Faculty of Pharmaceutical Science, Isfahan University of Medical Sciences, Isfahan, Islamic Republic of Iran. (Correspondence to: Fatemeh Naghavi-Al-Hosseini: f.naghavi.94@gmail.com). Abstract Rationing health services is an inseparable part of the health system of any country in order to achieve universal health coverage. Elective surgery for total hip and total knee replacement places a high financial burden on health systems. Such surgery should be done in a way to ensure that the people who most need it receive the service. Models for rationing total hip and knee replacement surgery were reviewed to suggest the best policy for rationing such surgery in the Islamic Re- public of Iran. We propose a system with three main tools: clinical guidelines, gate keepers and waiting lists, with shared decision-making as an auxiliary tool. Patients should be scored at the primary health care level based on clinical and radiographic examination, alternative treatments (conservative treatments) and risk factors, with a set threshold for re- ferral. Patients whose scores are above the threshold should be referred to secondary health care. These patients should be assessed again by specialists based on age, bone condition, surgery risk and other alternative treatments. Patients whose scores are above the threshold should be put on the waiting list for surgery. Keywords: arthroplasty, replacement, knee, hip, elective surgery, health services, policy, Iran Citation: Soleimani M; Barkhordari S; Mardani F; Shaarbafchizadeh N; Naghavi-Al-Hosseini F. Rationing access to total hip and total knee replace- ment in the Islamic Republic of Iran to reduce unnecessary costs: policy brief. East Mediterr Health J. 2020;26(11):1396–1402. https://doi.org/10.26719/ emhj.20.109 Received: 04/04/19; accepted: 17/10/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo) Introduction Health costs have increased faster than global economic growth over the past 15 years (1), which is an important issue for health systems. Financial resources allocated to health services are generally inadequate, especially in developing countries . As a result these services have only a small effect on public health and tend to benefit rich people more (2). One of the solutions to this problem that the World Health Organization has promoted as a prerequisite for achieving universal health coverage is rationing (3). Rationing has been referred to as not pro- viding services, which are considered to have benefits, to some people (4). Surgical operations for the total hip replacement and total knee replacement impose a large financial burden on the health system. Although these surgeries are mostly considered as the last-resort solution for treatment, studies show that nonsurgical treatments, such as physical therapy, can be more effective in hip and knee osteoarthritis compared with having no treatment (5). The increased average age of the world population and the higher prevalence of obesity and osteoarthritis together with increased health costs have led to concerns about the health system’s capacity to provide these procedures and consequently the need for rationing to ensure that people in most need have access to them (1,6–8). Main problem The fast-growing increase in osteoarthritis in low- and middle-income countries is similar to the increase in high-income countries. Some studies have reported a faster increase in some low- and middle-income coun- tries, which could be associated with low levels of educa- tion in these countries (9–11). According to a study in the Islamic Republic of Iran, the prevalence of osteoarthritis was 16.6% in urban areas and 20.5% in rural areas (12). Arthritis is the second leading disease causing long- term disability in individuals with the disease globally. The years spent with disability from arthritis increased by about 75% between 1990 and 2013 (13). In 2015, among 34 European and some Asian countries, on average 282 total hip replacements and total knee replacements were done per 100 000 people (14). According to reports of one of the main social insurance organizations in the Islamic Republic of Iran, primary orthopaedic surgery has been an obligation of insurance organizations since 2000. There are fewer than 10 000 knee surgeries a year in the country, but this figure is likely to reach to 30 000 in the next 5 years. The cost of this operation was high before the 2014 health system reform plan in the Islamic Republic of Iran, but the operation is now covered by insurance at a percentage rate of charge or free of charge (15). However, because of weaknesses in the plan, especially an inadequate referral system, cost issues and 1397 Review EMHJ – Vol. 26 No. 11 – 2020 ineffective negotiation with insurance companies, more reforms in health system policies are needed (16). Aim of the policy brief We aimed to develop a policy brief that helps the health system in the Islamic Republic of Iran to ration elective hip and knee joint replacement surgery in an equitable and clinically beneficial way. Methods We developed this policy brief through a literature review and group discussions among ourselves. We searched PubMed, Scopus, Web of Science, Embase, and Google Scholar up to 2019 using the keywords in English: “rationing”, “hip replacement”, “knee replacement” and “elective surgery”. In addition, the Guideline of American College of Rheumatology on hip and knee osteoarthritis (17) and the clinical guidelines of the Iranian Orthopaedic Society were used to assess the use of guidelines. Policy options Different tools have been used in rationing. These tools include waiting lists, clinical guidelines and gate-keeper systems. Shared decision-making is an auxiliary tool that has been effective in cost reduction but it has not been used as a tool on its own. In most countries, these tools have been used together, but to facilitate our analysis, we evaluated the tools separately. Clinical guidelines Clinical guidelines are used in New Zealand and the Unit- ed Kingdom of Great Britain and Northern Ireland at the micro-level where rationing is based on the views of phy- sicians of indications and contraindications for a medi- cal service (18). However, in developing countries, this method is used at a higher (meso-) level where insurance providers and hospitals determine the clinical guidelines. The main feature of the use of clinical guidelines is the use of evidenced-based medicine. However, there are many disagreements on the indications and contraindi- cations for total knee replacement surgery (19). In the studies we reviewed, rationing using clinical guidelines is not only considered an independent method of rationing but also an integral part of implementation of other rationing methods. For example, the American College of Rheumatology proposes conservative treatments such as water therapy and aerobic exercise for patients with osteoarthritis rather than surgery (17). The clinical criteria of age, bone status, surgical risk, preoperative procedures and motor limitations have been used to determine whether surgery is appropriate for patients with osteoarthritis or not (20). In 2006, the United Kingdom established a threshold of body mass index less than 30 kg/m2 for knee and hip surgeries. As a result, 8452 pelvic surgeries and 12 929 knee surgeries were eliminated, with a significant cost reduction (21). Obese people were 1.3 times more likely to have postoperative complications from shoulder, hip and knee surgery than people with normal weight (22). Obesity has also been associated with an increase in admission time in the hospital for people undergoing joint surgery (23). However, such a threshold would seem to deprive many people who are highly in need of this surgery of receiving it (24). Usually, total knee replacement is not done in people younger than 50 years or older than 80 years (25). In younger people, this is because of the potential complications of the surgery, and in older people this is because of their muscular condition and lack of movement and exercise, which can reduce the effectiveness of these operations (26). Gate-keeper system In a gate-keeper system, people cannot access second-lev- el services such as the hospital and specialist physician without referral by a general practitioner. This system has two main benefits: (i) cost control by reducing unnec- essary interventions, and (ii) use of effective secondary services because physicians are better informed than pa- tients about the quality of services provided by secondary providers (27). A study showed that 97% of people with severe knee problems, who were receiving secondary services, had initially seen a general practitioner (28). Another study showed that only 67% of orthopaedic referrals by general practitioners were appropriate (29). For appropriate referral, we need a referral threshold based on a clinical guideline that is available to general practitioners, such systems have been used in different countries. A review study proposed that general practitioners should consider four factors in referral for joint surgery: (i) Do clinical and radiographic characteristics of the patient justify the referral?; (ii) Has the patient had appropriate conservative treatments?; (iii) Does the patient have risk factors that might adversely affect the outcome of surgery?; and (iv) Can these risk factors be modified? (20) In a study in Switzerland in 2000, 20% cost reduction was observed as a result of the gate-keeper system (30). Waiting lists Two methods have been used to include the people on a waiting list. In the queue-based model, people are includ- ed on the waiting list based on the time of their referral, regardless of disease severity. This method is a chance- based prioritization. The other model is a scoring model where an individual’s position on the waiting list is based on specific scores for severity and need (6). Different scoring systems are used in different countries to accommodate individuals on the waiting list for joint replacement surgery. The most commonly used systems are the Oxford hip and knee score, reduced Western Ontario McMaster osteoarthritis index (WOMAC) score, New Zealand Orthopaedic Association score, clinical priority assessment criteria, and the score of the multi-attribute arthritis prioritization tool (6,31–33). According to the New Zealand Orthopaedic Association system, patients are scored from 0 to 100. After referral, each patient is scored by a consultant and 1398 Review EMHJ – Vol. 26 No. 11 – 2020 a nurse at the first visit with the specialist. Based on a defined threshold limit appropriate for conditions in the country, patients below the threshold are referred back to the general practitioner. Patients above the threshold limit are referred to the orthopaedic department and evaluated by a surgeon, who manages the waiting list. A study in New Zealand indicated that of 608 patients examined, 32% were referred back to the general practitioner based on this threshold, thus reducing the number of patients on the waiting list (33). In England, the use of the Oxford hip and knee score for knee surgery resulted in a cost reduction of £11.8 million a year (£ 1 = US$ 1.6041 in 2011, the date of the cited study) (31). The Oxford hip score questionnaire was translated into Farsi for use in the Islamic Republic of Iran for pre-operative total hip replacement patients (34). The adapted and validated Iranian version of the Oxford hip score questionnaire was found to be reliable and practicable for use with Iranian patients (34). Shared decision-making Clinical shared decision-making is not discussed as means of rationing, but can be considered an auxiliary tool for rationing. Shared decision-making can contribute to fair rationing along with other tools. Research has shown that patients are willing to share in the decision-making for their health care (6). In the United Kingdom, it was shown that individuals consider pain severity, inability to walk, costs and postoperative care some of the clinical criteria for joint replacement surgery (20) . Another study showed that 44–55% of people who required total knee replacement and total hip replacement were certainly or probably unwilling to have surgery (35). In another study, if individuals were consulted about their willingness to have surgery after the complications and conditions of the surgery were explained, a 36% cost reduction in joint replacement surgery was seen (36). Thus, prioritization of patients on the waiting list can be based on clinical cri- teria and the views of the patients themselves about the need for surgery (37). Policy recommendations Based on our evaluation of the various methods for ra- tioning surgical care (clinical guidelines, waiting lists, gate-keeper systems, and shared decision-making), Table 1 lists the disadvantages, benefits, and policy op- tions for each method. Clinical guidelines Clinical guidelines are the basis of the rationing method in many cases. Most clinical guidelines are based on ev- idence; however, clinical guidelines can also be based on consensus (38). The policy implemented in the United Kingdom to establish a threshold linked to body mass index is a special type of rationing based on clinical guidelines. Although this policy had many critics, it did reduce costs and time to admission to hospital (16) . Furthermore, the age of people can determine candidates for surgery. Clinical guidelines are necessary for rationing, and countries should develop guidelines relevant to their context. The Iranian Orthopaedic Association published a clinical guideline on joint replacement surgery in 2016. However, this guideline has not yet been implemented, so no cost–effectiveness assessment could be done (39). Waiting list Although a scoring-based waiting list is preferred to a list based on the time entered on the list, both methods have reduced costs. It should be noted that more developed countries have moved from a queuing model to a scoring model. Gate-keeper system As shown in Figure 1, gate keepers are the first line of ra- tioning. Since most people with serious joint problems go to general practitioners first and the referrals of general practitioners have been effective for accessing treatment (40), this rationing tool is recommended. In addition, an appropriate referral threshold can make the refer- rals more effective. Given the unsuccessful experience with an urban referral system in the Islamic Republic of Iran in 2005, a system should be designed with a refer- ral threshold based on clinical guidelines. This system should be piloted and the results evaluated. Proposed rationing system We propose a system summarized in Figure 1 with three main tools: clinical guidelines, gate keepers and waiting lists, with shared decision-making as an auxiliary tool. For patients to enter the rationing system for treatment through the gate keeper, they must be scored first by the primary health care units based on clinical and radio- graphic examination, alternative treatments (conserv- ative treatments) and risk factors. Patient whose scores are below the threshold, should be referred by general practitioners to physiotherapists for conservative treat- ment such as hydrotherapy and exercise. Patients whose scores are above the threshold, should be referred to secondary health care units. At this stage, patients are assessed by specialists and are scored on age, bone con- dition, risk of surgery and other alternative treatments. Patients whose scores are below this threshold will be re- ferred to physiotherapists again. Patients whose scores are above the threshold should be placed on a waiting list and prioritized according to age, sex, body mass index, occupation and history of total knee or hip replacement. As illustrated in Figure 1, decision-making should be ac- tively shared with patients; they should be encouraged to share their ideas about treatment and other options that may be available. Implementation of recommendations In order to implement the recommendations in our pol- icy brief at the mid-level in the Islamic Republic of Iran, the following process should be followed. First, establish a policy-makers’ group consisting of heads of universities of medical sciences, heads of hospitals, representative of the orthopaedic association, insurance providers, trusted 1399 Review EMHJ – Vol. 26 No. 11 – 2020 Ta bl e 1 A dv an ta ge s a nd d is ad va nt ag es o f m et ho ds fo r r at io ni ng To ol Ad va nt ag es D is ad va nt ag es Po lic y op ti on Im pl em en ta ti on le ve l Cl in ic al g ui de lin e • Si gn ifi ca nt co st re du ct io n in th e sh or t an d lo ng te rm • Lo ca l c lin ic al g ui de lin e • D is ag re em en ts o n ex is tin g in di ca tio ns • La ck o f a tt en tio n to sp ec ifi c p re fe re nt ia l tr ea tm en ts fo r p at ie nt s • Fa ilu re to re m ov e so m e in di ca tio ns d ue to la ck of p ro of o f i ne ff ec tiv en es s • Fa ilu re to u se a nd a ss es s t he n at io na l c lin ic al gu id el in es • Su rg ic al th re sh ol d co ns id er in g th e fo llo w in g fa ct or s, am on g ot he rs : Ag e Bo ne st at e • Re vi ew a nd m od ify th e na tio na l l oc al cl in ic al gu id el in es • Ir an ia n O rt ho pa ed ic A ss oc ia tio n • In su ra nc e pr ov id er s • U ni ve rs iti es o f m ed ic al sc ie nc es • H os pi ta ls W ai tin g lis t • Si gn ifi ca nt co st re du ct io n • D ec re as ed w ai tin g tim es in th e sh or t te rm • N eg at iv e ef fe ct o n Fa ir ne ss • La ck o f a v al id in de x to m ak e pr io ri tiz at io n on th e lis t m od el • Pr ol on ge d w ai tin g tim e • N eg at iv e ef fe ct s o f p ro lo ng ed w ai tin g tim e • Pr io ri tiz e ba se d on th e cl in ic al g ui de lin e • Cl ar ify p ri or iti za tio n pr oc es se s u si ng te le vi si on a dv er tis em en ts a nd p os te rs in ho sp ita ls • D em an d re du ct io n • U se th e pr iv at e se ct or fo r t o ge t o pe ra tio ns so on er (f or p eo pl e w ho d on ’t w an t t o be o n a w ai tin g lis t) • Ir an ia n O rt ho pa ed ic A ss oc ia tio n • In su ra nc e pr ov id er s • H os pi ta ls • U ni ve rs iti es o f m ed ic al sc ie nc es G at e- ke ep er sy st em • Be tt er q ua lit y of re fe rr al s • El im in at io n of u nn ec es sa ry se rv ic es a nd un ne ce ss ar y vi si ts to sp ec ia lis ts • In cr ea se d us e of co ns er va tiv e m ea su re s fo r t ho se n ot in n ee d of su rg er y • N ee d fo r l eg is la tiv e an d m ot iv at io na l in fr as tr uc tu re fo r i m pl em en ta tio n • D es pi te in cr ea se d qu al ity o f r ef er ra ls w ith cl in ic al g ui de lin e, n um be r o f r ef er ra ls d oe s no t c ha ng e • U ns uc ce ss fu l e xp er ie nc e of im pl em en tin g ur ba n re fe rr al sy st em in th e Is la m ic R ep ub lic of Ir an • N ee d to fo rm ul at e a re fe rr al th re sh ol d ba se d on th e cl in ic al g ui de lin es • Fo rm ul at e re fe rr al th re sh ol ds b as ed o n cl in ic al gu id el in es co ns id er in g th e fo llo w in g ite m s: Cl in ic al fe at ur es a nd ra di og ra ph y Co ns er va tiv e m ea su re s ( w at er th er ap y an d ae ro bi c e xe rc is e) Ri sk fa ct or s • Id en tif y fa ilu re fa ct or s o f t he Ir an ia n ur ba n re fe rr al sy st em p ro je ct • Ir an ia n O rt ho pa ed ic A ss oc ia tio n • In su ra nc e pr ov id er s • U ni ve rs iti es o f m ed ic al sc ie nc es Sh ar ed d ec is io n- m ak in g • Ef fe ct iv e co st re du ct io n • N o ne ed fo r n ew in fr as tr uc tu re • G re at er tr us t i n pr io ri tiz at io n • In ab ili ty o f p at ie nt to u nd er st an d cl in ic al de ci si on s • N ee d fo r a se pa ra te a pp oi nt m en t w ith th e ph ys ic ia n • Cl ar ify p ri or iti za tio n pr oc es se s a nd d efi ne th re sh ol ds in th e pr es en ce o f t he p at ie nt o r hi s/ he r r ep re se nt at iv e • Cl ar ify su rg ic al p ro ce ss es a nd p os to pe ra tiv e co m pl ic at io ns w ith th e pa tie nt • Ir an ia n O rt ho pa ed ic A ss oc ia tio n • In su ra nc e pr ov id er s 1400 Review EMHJ – Vol. 26 No. 11 – 2020 orthopaedic surgeons, representative of general practi- tioners and physiotherapists. Establish a research group, consisting of for example general practitioners, special- ists and statisticians, to evaluate and validate the clinical guidelines developed in 2016 by the Iranian Orthopaedic Association. Second, modify the clinical guidelines based on the results. Third, establish thresholds for referral and surgery based on the clinical guidelines. Fourth, reach an agreement with insurance providers and patient repre- sentatives on the cost of treatment that insurance covers. Funding: National Agency for Strategic Research in Medical Education, Tehran (grant no. 971935). Competing interests: None declared. Figure 1 Optimized policy recommendations for rationing total hip replacement (THR) and total knee replacement (TKR) surgery Put on a waiting list based on: • Age • Gender • BMI • job status • History of THR and TKR Patient assessment for referring to the specialist Primary health care unit (general practitioners) Secondary health care (Specialist) Patient assessment for THR or TKR Pathway for patient requiring THR and TKR Shared decision making Referral threshold lower threshold Lower treshold Surgery threshold Upper thresholdUpp er th resh old Conservative treatment REFER Limitation des interventions de prothèse totale de la hanche et du genou en République islamique d’Iran en vue de réduire les coûts inutiles : note d’orientation Résumé La limitation des services de santé est une composante indissociable du système de santé de tout pays dans l’objectif de parvenir à la couverture sanitaire universelle. La chirurgie programmée pour la prothèse totale de la hanche et du genou constitue une lourde charge financière pour les systèmes de santé. Cette chirurgie doit être effectuée de manière à ce que les personnes qui en ont le plus besoin en bénéficient. Les modèles de limitation de la chirurgie totale de la hanche et du genou ont été examinés afin de proposer la meilleure politique pour limiter ce type d’intervention en République islamique d’Iran. Nous proposons un système comportant trois outils principaux : les lignes directrices cliniques, les filtres et les listes d’attente, avec la prise de décision partagée comme outil auxiliaire. Les patients doivent être évalués au niveau des soins de santé primaires sur la base de l’examen clinique et radiographique, des traitements alternatifs (traitements conservateurs) et des facteurs de risque, avec un seuil défini pour l’orientation-recours. Les patients dont les scores sont supérieurs au seuil doivent être orientés vers les soins de santé secondaires. Ces patients doivent être réévalués par des spécialistes en fonction de leur âge, de leur condition osseuse, du risque chirurgical et d’autres traitements alternatifs. Les patients dont les scores sont supérieurs au seuil doivent être ensuite placés sur la liste d’attente des interventions chirurgicales. 1401 Review EMHJ – Vol. 26 No. 11 – 2020 References 1. Global spending on health: a world in transition. Geneva: World Health Organization; 2019 (https://apps.who.int/iris/bitstream/ handle/10665/330357/WHO-HIS-HGF-HF-WorkingPaper-19.4-eng.pdf?ua=1, accessed 14 June 2020). 2. James C, Carrin G, Savedoff W, Hanvoravongchai P. Clarifying efficiency-equity trade-offs through explicit criteria, with a focus on developing countries. Health Care Anal. 2005;13(1):33–51. https://doi.org/10.1007/s10728-005-2568-2 3. The world health report: health systems financing: the path to universal coverage. Geneva: World Health Organization; 2010 (https://www.who.int/whr/2010/en/, accessed 14 June 2020). 4. Brock DW. Health care resource prioritization and rationing: why is it so difficult? Soc Res. 2007;74(1):125–48. 5. Smidt N, de Vet HC, Bouter LM, Dekker J. Effectiveness of exercise therapy: a best-evidence summary of systematic reviews. Aust J Physiother. 2005;51(2):71–85. https://doi.org/10.1016/s0004-9514(05)70036-2 6. Cipriano LE, Chesworth BM, Anderson CK, Zaric G. An evaluation of strategies to reduce waiting times for total joint replace- ment in Ontario. Med Care. 2008;46(11):1177–83. https://doi.org/10.1097/MLR.0b013e31817925e8 7. He W, Goodkind D, Kowal P. An aging world: 2015. Washington, DC: US Government Publishing Office; 2016 (US Census Bu- reau, International Population Reports, P95/16-1). 8. The world health report: obesity and overweight: World Health Organization. (2018). 2019. 9. Cross M, Smith E, Hoy D, Carmona L, Wolfe F, Vos T, et al. The global burden of rheumatoid arthritis: estimates from the global burden of disease 2010 study. Ann Rheum Dis. 2014;73(7):1316–22. https://doi.org/10.1136/annrheumdis-2013-204627 10. Zhang Y, Jordan JM. Epidemiology of osteoarthritis. Clin Geriatr Med. 2010;26(3):355–69. https://doi.org/10.1016/j.cger.2010.03.001 11. Vina ER, Kwoh CK. Epidemiology of osteoarthritis: literature update. Curr Opin Rheumatol. 2018;30(2):160. https://doi. org/10.1097/BOR.0000000000000479 12. Davatchi F, Sandoughi M, Moghimi N, Jamshidi AR, Tehrani Banihashemi A, Zakeri Z, et al. Epidemiology of rheumatic diseases in Iran from analysis of four COPCORD studies. Int J Rheum Dis. 2016;19(11):1056–62. https://doi.org/10.1111/1756-185X.12809 13. Newton JN, Briggs AD, Murray CJ, Dicker D, Foreman KJ, Wang H, et al. Changes in health in England, with analysis by English regions and areas of deprivation, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2015;386(10010):2257–74. https://doi.org/10.1016/S0140-6736(15)00195-6 14. Health at a glance 2015. OECD indicators, Paris: OECD Publishing; 2015. https://dx.doi.org/10.1787/health_glance-2015-en 15. Esfandiaribayat M, Ghavam A, Yazdanpanah A. Comparing the cost of joint replacement surgery in patients receiving social security insurance in Shahid Beheshti and Shahid Chamran hospitals in Shiraz. Biosci Biotech Res Comm. 2017;(Suppl 1):139–45. 16. Olyaeemanesh A, Behzadifar M, Mousavinejhad N, Behzadifar M, Heydarvand S, Azari S, et al. Iran’s health system transforma- tion plan: a SWOT analysis. Med J Islam Repub Iran. 2018;32:39. https://doi.org/10.14196/mjiri.32.39 17. Hochberg MC, Altman RD, April KT, Benkhalti M, Guyatt G, McGowan J, et al. American College of Rheumatology 2012 recom- mendations for the use of non-pharmacologic and pharmacologic therapies in osteoarthritis of the hand, hip, and knee. Arthritis Care Res. 2012;64(4):465–74. https://doi.org/10.1002/acr.21596 18. Ham C, Coulter A. Explicit and implicit rationing: taking responsibility and avoiding blame for health care choices. J Health Serv Res Policy. 2001;6(3):163–9. https://doi.org/10.1258/1355819011927422 يرغ فيلاكتلا ليلقتل ةيملاسلإا ناريإ ةيروهجم في ةبكرلل ِّلُكلا لادبتسلااو كْرَولل ِّلُكلا لادبتسلاا تايلمع ديشرت تاسايسلا زجوم :ةيروضرلا ینیسلحا یوقن همطاف ،هداز یچفابرعش نیسرن ،نيادرم داهرف ،یرادروخرب شروش ،نيمایلس دممح ةصلالخا ةیرایتخلاا ةحارلجا لِّثُتو .ةلماشلا ةیحصلا ةیطغتلا قیقتح لجأ نم دلب يأ في يحصلا ماظنلا نم أزجتی لا ًاءزج ةیحصلا تامدلخا دیشرت دعُی لوصح نمضی ٍوحن لىع ةحارلجا هذه ءارجإ يغبنیو .ةیحصلا مظُّنلا لىع ًايربك ًایلام ًائبع ةبكرلل ِّلُكلا لادبتسلااو كْرَولل ِّلُكلا لادبتسلال ةبكرلل ِّلُكلا لادبتسلااو كرولل ِّلُكلا لادبتسلاا تاحارج دیشرت جذمان تضِرعُتساو .ةمدلخا لىع اهیلإ ةجالحا ِّسَمأ في مه نیذلا صاخشلأا ةیهیجوتلا ئدابلما :ةیسیئر تاودأ ثلاث لىع لمتشی ًاماظن حترقنو .ةیملاسلإا ناریإ ةیروهجم في تاحارلجا هذه لثم دیشترل ةسایس لضفأ حاترقلا ةیاعرلا ىوتسم لىع ضىرملل تاجرد ءاطعإ يغبنیو .ةدعاسم ةادأك ةكترشم رارق ذاتخا ةیلمع عم ،راظتنلاا مئاوقو ،تاباوبلا وبقارمو ،ةیریسرلا .ةلاحلإل ىندأ ٍّدح دیدتح عم ،رطلخا لماوعو ،)ةیظفحتلا تاجلاعلا( ةلیدبلا تاجلاعلاو ،يریوصتلاو يریسرلا صحفلا لىع ًءانب ةیل َّولأا ةیحصلا ينصصختلما لبِق نم ىرخأ ًةرم ضىرلما ءلاؤه مییقت بيجو .ةیوناثلا ةیحصلا ةیاعرلا لىإ ىندلأا َّدلحا متهاجرد زواجتت نیذلا ضىرلما ةلاحإ يغبنیو لىع ىندلأا دلحا متهاجرد زواجتت نیذلا ضىرلما عضو يغبنیو .ىرخلأا ةلیدبلا تاجلاعلاو ةیحارلجا رطاخلماو ماظعلا ةلاحو رمعلا ساسأ لىع .ةحارجلل عوضخلل راظتنلاا ةمئاق 1402 Review EMHJ – Vol. 26 No. 11 – 2020 19. Bryant JH. Health priority dilemmas in developing countries. In: Coulter A, Ham C, editors. The global challenge of health care rationing. Philadelphia: Open University Press; 2000. 20. Dowsey MM, Gunn J, Choong P. Selecting those to refer for joint replacement: who will likely benefit and who will not? Best Pract Res Clin Rheumatol. 2014;28(1):157–71. https://doi.org/10.1016/j.berh.2014.01.005 21. Coombes R. Rationing of joint replacements raises fears of further cuts. BMJ. 2005;331(7528):1290. https://doi.org/10.1136/ bmj.331.7528.1290 22. Jain NB, Guller U, Pietrobon R, Bond TK, Higgins LD. Comorbidities increase complication rates in patients having arthroplasty. Clin Orthop Relat Res. 2005;435:232–8. https://doi.org/10.1097/01.blo.0000156479.97488.a2 23. Jiganti JJ, Goldstein WM, Williams CS. A comparison of the perioperative morbidity in total joint arthroplasty in the obese and nonobese patient. Clin Orthop Relat Res. 1993;(289):175–9. 24. Davis W, Porteous M. Joint replacement in the overweight patient: a logical approach or new form of rationing? Ann R Coll Surg Engl. 2007;89(3):203–6. https://doi.org/10.1308/003588407X183247 25. Lonner JH, Hershman S, Mont M, Lotke PA. Total knee arthroplasty in patients 40 years of age and younger with osteoarthritis. Clin Orthop Relat Res. 2000;380:85–90. https://doi.org/10.1097/00003086-200011000-00012 26. Fortin PR, Clarke AE, Joseph L, Liang MH, Tanzer M, Ferland D, et al. Outcomes of total hip and knee replacement: preoperative functional status predicts outcomes at six months after surgery. Arthritis Rheum. 1999;42(8):1722–8. https://doi.org/10.1002/1529- 0131(199908)42:8<1722::AID-ANR22>3.0.CO;2-R 27. Brekke KR, Nuscheler R, Straume OR. Gatekeeping in health care. J Health Econ. 2007;26(1):149–70. https://doi.org/10.1016/j. jhealeco.2006.04.004 28. Tennant A, Fear J, Pickering A, Hillman M, Cutts A, Chamberlain M. Prevalence of knee problems in the population aged 55 years and over: identifying the need for knee arthroplasty. BMJ 1995;310(6990):1291–3. https://doi.org/10.1136/bmj.310.6990.1291 29. Fertig A, Roland M, King H, Moore T. Understanding variation in rates of referral among general practitioners: are inappro- priate referrals important and would guidelines help to reduce rates?. BMJ. 1993;307(6917):1467–70. https://doi.org/10.1136/ bmj.307.6917.1467 30. Feld LP, Reulier E. Strategic tax competition in Switzerland: evidence from a panel of the Swiss cantons. Ger Econ Rev. 2009;10(1):91–114. https://doi.org/10.1111/j.1468-0475.2008.00443.x 31. Dakin H, Gray A, Fitzpatrick R, MacLennan G, Murray D. Rationing of total knee replacement: a cost-effectiveness analysis on a large trial data set. BMJ Open. 2012;2(1):e000332. https://doi.org/10.1136/bmjopen-2011-000332 32. Gwynne-Jones DP, Iosua EE, Stout KM. Rationing for total hip and knee arthroplasty using the New Zealand Orthopaedic Association score: effectiveness and comparison with patient-reported scores. J Arthroplasty. 2016;31(5):957–62. https://doi. org/10.1016/j.arth.2015.11.022 33. Hudak PL, Clark JP, Hawker GA, Coyte PC, Mahomed NN, Kreder HJ, et al. “You’re perfect for the procedure! Why don’t you want it?” Elderly arthritis patients’ unwillingness to consider total joint arthroplasty surgery: a qualitative study. Med Decis Making. 2002;22(3):272–8. https://doi.org/10.1177/0272989X0202200315 34. Nourbakhsh M, Zarezadeh A, Shemshaki H, Etemadifar MR, Moezi M, Mazoochian F. Translation and cultural adaptation of the oxford hip score for Iranian population. Int J Prev Med. 2013;4(2):141–5. 35. Hawker GA, Wright JG, Coyte PC, Williams JI, Harvey B, Glazier R, et al. Determining the need for hip and knee arthroplasty: the role of clinical severity and patients’ preferences. Med Care. 2001;39(3):206–16. https://doi.org/10.1097/00005650-200103000- 00002 36. Jüni P, Dieppe P, Donovan J, Peters T, Eachus J, Pearson N, et al. Population requirement for primary knee replacement surgery: a cross‐sectional study. Rheumatology (Oxford). 2003;42(4):516–21. https://doi.org/10.1093/rheumatology/keg196 37. Kapiriri L, Martin DK. Priority setting in developing countries health care institutions: the case of a Ugandan hospital. BMC Health Serv Res. 2006;6(1):127. https://doi.org/10.1186/1472-6963-6-127 38. Woolhead G, Donovan J, Chard J, Dieppe P. Who should have priority for a knee joint replacement? Rheumatology. 2002;41(4):390–4. https://doi.org/10.1093/rheumatology/41.4.390 39. Aslani H, Nourbakhsh ST, Lahiji FA, Heydarian K, Jabalameli M, Ghazavi MT, et al. Iranian Joint Registry (Iranian National Hip and Knee Arthroplasty Registry). Arch Bone Jt Surg. 2016;4(2):192–6. 40. Winpenny EM, Miani C, Pitchforth E, King S, Roland M. Improving the effectiveness and efficiency of outpatient services: a scoping review of interventions at the primary–secondary care interface. J Health Serv Res Policy. 2017;22(1):53–64. https://doi. org/10.1177/1355819616648982 1403 Review EMHJ – Vol. 26 No. 11 – 2020 Application of geographic information systems in maternal health: a scoping review Leila Ahmadian,1 Fatemeh Salehi2 and Kambiz Bahaadinbeigy3 1Medical Informatics Research Centre, Institute for Futures Studies in Health, Kerman University of Medical Sciences, Kerman, Islamic Republic of Iran. 2Health Human Resources Research Centre, School of Management and Information Sciences, Shiraz University of Medical Sciences, Shiraz, Islamic Republic of Iran. 3Gastroenterology and Hepatology Research Centre, Institute of Basic and Clinical Physiology Sciences, Kerman University of Medical Sciences, Kerman, Islamic Republic of Iran. (Correspondence to: Fatemeh Salehi: fsalehi891@gmail.com). Abstract Background: Improving maternal health is a global health priority and requires accurate evaluation of factors affecting maternal health. Geographic information systems have been used to explore maternal health problems. Aims: The aim of this study was to identify studies that used geographic information systems in the field of maternal health care and to determine maternal health and mortality variables visualized on these systems. Methods: This was a scoping review in which we systematically searched PubMed and Science Direct for studies that used geographic information systems to evaluate maternal health care. We included all relevant cross-sectional studies published in English between December 1995 and December 2017. We extracted the following information from each study included: study year, region, objectives, type of geographic information system used, variables visualized by the geographic information system, and all other variables examined that related to maternal health. Results: Of 5240 articles initially retrieved, 40 were included for detailed review. Most of the studies (n = 32) were done in developing countries in Africa, Asia, and Latin America and the Caribbean. Most of the studies (n = 33) visualized mothers’ distance to health facilities and travel time to health care centres on geographic information systems. Other factors exam- ined included antenatal care capacity (n = 4) and capacity of maternal health services (n = 3). Conclusions: Comprehensive research on the application of geographic information systems in maternal care is lacking. Most studies applied simple descriptive mapping of spatial distribution patterns with a few relevant variables. Keywords: geographic information system; maternal health; maternal health services; health services research Citation: Ahmadian L; Salehi F; Bahaadinbeigy K. Application of geographic information systems in maternal health: a scoping review. East Mediterr Health J. 2020;26(11):1403–1414. https://doi.org/10.26719/emhj.20.095 Received: 28/06/18; accepted: 26/09/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Introduction Maternal health is a globally important health challenge. Maternal health refers to the health of women during pregnancy, childbirth and the postpartum period (1). Ac- cording to the World Health Organization (WHO) the global maternal mortality rate is unacceptably high and about 810 maternal deaths occur daily around the world with 94% of all maternal deaths occurring in developing countries (2). In 2015, 295 000 women died following pregnancy related complications – most of these deaths were in low-resource settings and most could have been prevented (2).The Sustainable Development Goals (SDGs) now call for an acceleration in progress in order to achieve a global maternal mortality rate of 70 mater- nal deaths per 100 000 live births, or less, by 2030 (3). Achieving this global goal will require countries to reduce their maternal mortality rate by at least 7.5% each year between 2016 and 2030 (3). Reducing maternal mortality was introduced globally as an important health priority in the United Nations Millennium Development Goals (4). Precise evaluation of the maternal mortality rate is the first step to reducing this rate. Extensive implementation of various software and information technology services in recent years has helped health experts to improve health care. These improvements have helped health care professionals work more efficiently and effectively (5,6). Recently, researchers have started to apply geographic information systems (GIS) to explore maternal and newborn health problems (7–9). GIS help to show regional variations and abnormal patterns of health characteristics. Through the use of GIS, researchers can gain insight into the use of health services and expose health problems and environmental risks. This technology can be easily used by non-geographers with basic computer literacy and map-reading skills (10). The use of GIS is a means to effectively link and analyse the range of data necessary to address complex questions in health promotion, public health, community medicine, epidemiology, and other fields (11). Despite a number of reviews on the use of GIS in maternal health (12–16), to the best of our knowledge, no review has collectively analysed the results of studies to determine maternal health and mortality factors visualized through GIS. Some studies have shown that the application of GIS in maternal health can help decision-making on policies to reduce maternal problems and improve maternal care outcomes (17,18). Therefore, 1404 Review EMHJ – Vol. 26 No. 11 – 2020 we did a scoping review to identify GIS studies related to maternal health care to determine maternal health and mortality factors (variables) visualized through GIS. We also summarized other variables investigated, but not visualized through GIS, to get a general overview of all evaluated variables in the included studies. The results of our review can be used to help in planning to reduce preventable causes of maternal health problems. Our results can also help determine gaps in the use of these types of data and provide a road map to guide more precise studies on various aspects of maternal health. Methods Search terms We searched for relevant articles in English from 1995 to 2017 using PubMed and Science Direct databases. We used two groups of key search terms. Group A in- cluded GIS-related terms: GIS; geographic information system; risk mapping; spatial analysis; spatial data; GPS; and health mapping. Group B included terms related to maternal health: maternal mortality; maternal health; maternal care; pregnancy; delivery; and maternal death. Our search strategy was as follows: first we used “OR” to combine the terms within each group A and B separately. Then, we combined keywords from the two groups using the “AND” operator to find all the studies that used GIS in the field of maternal care. We then limited the search in both databases to humans and studies published in Eng- lish. Inclusion and exclusion criteria The inclusion criteria were: cross-sectional study relat- ed to the use of GIS in maternal health care; published between December 1995 and December 2017; in English; and on a human population. The exclusion criteria were: review studies, editorials, commentaries and letters; studies that had not visualized data on GIS; and articles not in English. Review strategy Two reviewers independently screened and assessed the titles and abstracts of the retrieved articles. The review- ers met and reached consensus at the end of the screen- ing process. In cases of disagreement, a third independ- ent reviewer was consulted to resolve the disagreement. We compared the results of the independent screenings using the kappa statistic (kappa = 8.6). The reviewers were blinded to each other’s decisions to control for as- sessment bias. We also checked the reference lists of the articles retrieved for other relevant studies. We devel- oped a data collection form and extracted the following information from each paper: study year, region, objec- tives, GIS application type, variables visualized by GIS, and all other variables examined that related to maternal health. We examined the extraction forms for coverage, clarity and content validity in several meetings. We di- vided the extracted variables from the studies into two groups: subvariables and variables directly visualized on GIS. Based on expert consensuses and in order to better report on the variables, we classified the subvariables into six categories: (1) maternal factors, (2) socioeconomic factors, (3) health care service factors, (4) ecological deter- minant factors, (5) environmental factors, and (6) health related factors. Results Study selection In our initial search of the online databases we found 5240 articles (Figure 1). After our first screening of titles and abstracts based on out inclusion and exclusion crite- Figure 1 Flow diagram of the selection of papers 5240 studies retrieved: 4775 (91%) PubMed 465 (9%) Science Direct 1737 (33.1%) excluded because not on humans and not in English 1737 (33.1%) excluded because not on humans and not in English 3469 (66.2%) articles retained for title and abstract review 3268 (62.4%) excluded because did not meet the inclusion criteria 201 (3.8%) articles selected for full text review 161 (3.1%) excluded because did not meet inclusion criteria 40 (0.07%) articles included for final review 1405 Review EMHJ – Vol. 26 No. 11 – 2020 ria, we retained 201 eligible articles for further full-text review. Based on this review of full texts, we excluded 161 articles as they did not meet the inclusion criteria, and retained 40 articles for detailed analysis. Included studies Of the 40 articles we retained, all were journal papers. Table 1 gives a description of the 40 studies – publica- tion year, objectives, region and data sources. The oldest articles were published in 2004 (57,58). The number of studies investigating the application of GIS for mater- nal care has increased since 2010. Most of the studies (n = 29) were conducted in developing countries in Afri- ca, Asia, and Latin America and the Caribbean (19–23,26– 31,33,34,36,37,39–43,45–50,53,55,58) and 11 were conducted in developed countries in Europe, North America, Aus- tralia, New Zealand, and Japan (24,25,32,35,38,44,51,52,5 4,56,57). Nearly half of the studies (n = 16) were done in Africa because of its high maternal and infant mortality rates (21,23,26,28,31,36,39–42,46–50,58), nine were done in Asia (19,20,22,27,34,37,43,53,55), 11 in the Americas and the Caribbean (24,25,29,30,32,33,44,45,51,52,54), three in Europe (35,57,56) and one in Oceania (38). Each study presented data on one country/region, except one study which provided data on four countries (27). Multiple data resources were used to conduct the studies. The source of data in 15 of the studies was national statistics and censuses (Table 1). Other sources of data included, among others, results of other previous surveys (n = 11), interviews with women and health staff (n = 5) health registries (n = 6). Geographic access to health services was the most common factor examined (n = 22) in the included studies (20,22,25,26,28,30,35,37,39,40,41,43,44–47,49,51,52,56–58). Other common factors examined included antenatal care capacity (19,25,42,55) and capacity of maternal health services (21,48,53). Most studies (n = 28) used the spatial analysis of the GIS software such as network analysis, buffer, hot spots and Moran techniques (19,22,23,26–29,31–35,37–45,48– 50,52,53,55,56). Other studies (n = 10) used spatial mapping (20,21,24,25,30,36,51,54,57,58) and two used spatial modelling techniques (46,47). Many of the studies that used GIS in maternal health focused on potential geographic access to care on the basis of the spatial distribution of health facilities (27,35–37,44,46,47,51,52,56). Some investigated the effect of geographic access on mortality and care utilization (24,33,49,55). Other studies modelled the availability of and access to emergency obstetric care (22,28,43,50). The geospatial unit of analysis in 16 studies was the national level (27,28,29,31,35,36,39,40,44,46,48– 50,52,56,58). Most of the studies (n = 24) used ArcGIS (Esri, Redlands, California, United States of America) (19–21,24,25,27,31,32,34,36–39,42–47,50,51,53–55) to analyse the data, 10 studies used other types of GIS software – ArcView, QGIS, ArcInfo – (22,26,28–30,35,41,49,56,58), and six did not state which software was used to analyse the data (27,33,40,47,52,57). Variables extracted from the studies are shown in Table 2. Many studies visualized the variables: distance to health facilities (n = 16) and travel time to health care centres (n = 17). Other frequently mapped variables were spatial distribution of health services and emergency obstetric care (n = 11). Socioeconomic and sociocultural variables, such as women’s educational level (n = 10) and household wealth (n = 9), were often investigated. Discussion The findings of our study show that special attention has been paid to geographic access and travel time to health services in published literature on maternal health. Oth- er important reported variables included: maternal age, maternal educational level, household wealth, residential area, distribution of health services, and availability of emergency obstetric care facilities per population. Determining the distribution of human resources (obstetricians/gynaecologists, maternity nurses and midwives) can show imbalances in the distribution of health personnel. The findings of some research has shown that the educational level of women was strongly correlated with the maternal mortality rate (59–63). The results of our study also showed that mothers’ educational level was frequently reported in the included studies. Educational level has a positive influence on autonomy, awareness of health services, the health-seeking behaviour, responsibility and knowledge of self-care and healthy lifestyles (64) and can directly and indirectly contribute to a reduction in maternal mortality. Some believe that education may have a more important role compared to economic indicators, such as income, clean water supply and sanitary sewer access (65). Maternal age was examined as the underlying variable in many of the studies we reviewed, and is a key variable because older women are at a higher risk of death and complications during pregnancy or delivery (66,67). The risk of pregnancy-related death for mothers over 40 years between 1998 and 2005 has been reported to be six times higher compared with teenagers (68,69). Others research has found that young adolescents (< 15 years old) face a higher risk of complications and death as a result of pregnancy (70,71). The residential area of pregnant woman is also an important factor. Higher maternal mortality rates have been reported in women living in rural areas and poorer communities (2). The fact that most of the studies included in our review were conducted in developing countries, especially in Africa, is understandable as, according to WHO, 94% of all maternal deaths occur in developing countries (2) and factors that threaten the health of mothers are more common in these countries. Some research has provided evidence that comorbid conditions such as high blood pressure, diabetes and cardiac diseases contribute to maternal deaths (72). Others showed that direct pregnancy complications are 1406 Review EMHJ – Vol. 26 No. 11 – 2020 Table 1 Description of the included studies Author Publication year Objectives of the study Countrya Data sources Ansariade and Manderson, (19) 2015 Determine the influence of urban and rural settings on antenatal care and birthing decisions and investigate if women’s decision on antenatal care and birth assistance are geographically clustered Indonesia, Sulawesi Structured interview with women who delivered Jain, et al. (20) 2015 Assess the influence of economic and geo- graphic access to health facilities on institutional deliveries Pakistan Health facilities; and household surveys Tabatabai, et al. (21) 2014 Map and analyse the capacities of public and private hospitals to provide maternal health care United Republic of Tanzania, Ruvuma Hospital questionnaire; population census dataset Sabde, et al. (22) 2014 Identify potential areas for further interventions to increase the effectiveness of the emergency obstetric transport system India, Madhya Pradesh Interviews with parturient women Mwaliko, et al. (23) 2014 Determine the association between the place of delivery and the distance of a household from the nearest health facility, and assess the demographic characteristics of households with a delivery within a demographic surveillance system Kenya, western Database of Webuye health and demographic surveillance system; structured interviews with trained field assistants Detres, et al. (24) 2014 Examine how GIS maps can be used by local organizations to engage the community in the discussion of maternal and child health data to modify service delivery USA, Florida Florida vital statistics birth and infant death records Brown, et al. (25) 2014 Determine maternal ground transport times from community hospitals to the nearest hospital offering comprehensive (level III) neonatal care USA US census tract data; American hospital association annual survey Nesbitt, et al. (26) 2014 Compare methods to measure potential spatial access to delivery care in low- and middle-income countries Ghana Kintampo Health Research Centre surveillance data; health facility assessment; Ghana registered midwives association Tatem, et al. (27) 2014 Present methods to estimate women of childbearing age, and pregnancies and live births in relation to current health infrastructure Afghanistan Bangladesh, Ethiopia, United Republic of Tanzania Household survey data; United Nations statistics; new estimates of stillbirths, miscarriages and abortions from the Guttmacher Institute McKinnon, et al. (28) 2014 Assess the effect of distance to emergency obstetric and newborn care services on early neonatal mortality and examine whether proximity to services contributes to socioeconomic inequalities in early neonatal mortality Ethiopia Ethiopian Demographic and Health Survey; Ethiopian national emergency obstetric and newborn care needs assessment (Ethiopian Ministry of Health); questionnaire filled by women Wang, et al. (29) 2014 Estimate the influence of service readiness at health facilities on women’s use of facility delivery care for delivery Haiti Haiti Demographic and Health Survey; Haiti service provision assessment survey Gaspar, et al. (30) 2014 Evaluate the spatial distribution of public sector obstetric care Brazil, Belo Horizonte A system of obstetric information Municipal (SISMater®), Department of Health of Belo Horizonte (SMSA-BH);, cohort of 2956 newborns Benedict, et al. (31) 2014 Explore geographical patterns in the risk of not utilizing using a skilled birth attendant during childbirth in women of different socioeconomic backgrounds Ghana Ghana Demographic and Health Survey; interviews with women Blake, et al. (32) 2014 Explore the geographic relationships among between dairy farms, nitrate levels in drinking water, low birth weight and socioeconomic data at the Zip code level USA, California Zip codes, US census 1407 Review EMHJ – Vol. 26 No. 11 – 2020 Author Publication year Objectives of the study Countrya Data sources Almeida, et al. (33) 2014 Identify spatial patterns of in distribution of overall, early, and late neonatal mortality rates Brazil, São Paulo Department of information systems and information technology of the Brazilian national healthcare system Arslan, et al. (34) 2013 Determine the spatial patterns of perinatal mortality, examine whether regional differences exist and whether these differences are linked to regional risk factors Turkey, Kocaeli Registry of births and deaths Engjom, et al. (35) 2013 Assess the availability of obstetric institutions, the risk of unplanned delivery outside an institution and maternal morbidity in a national setting in which the number of institutions declined from 95 to 51 during over 30 years Norway Census data; Statistics Norway; medical birth registry Sudhof, et al. (36) 2013 Identify potential gaps in access to emergency obstetric care Rwanda, Kayonza Birth registries at in eight health centres and the district hospital Song, et al. (37) 2013 Assess spatial accessibility to maternity units China, Shenzhen Website of Shenzhen Health and Population and Family Planning Commission Chong, et al. (38) 2013 Assess the usefulness of geospatial methods in identifying communities at high risk of smoking during pregnancy and timing of the first antenatal visit Australia, New South Wales New South Wales health ministry Bowie C, et al. (39) 2013 Evaluate geographical access to health care facilities Malawi Malawi census; Ministry of Health facility surveys Masters, et al. (40) 2013 Estimate travel times between populations and health facilities using geospatial techniques Ghana Ghanaian Ministry of Health Yao, et al. (41) 2013 Present a geographical perspective on access to sexual and reproductive health care for rural women Mozambique, Gaza province Population survey data O’Meara, et al. (42) 2013 Assess spatial autocorrelation in uptake of antenatal care and relationship to individual, household and village-level factors Kenya, western Survey data Monyet al. (43) 2013 Investigate the availability and distribution of emergency obstetric care services in eight northern districts of Karnataka State in south India India, Karnataka Combination of self-reporting, record review and direct observation Brown, et al.(44) 2012 Determine the percentage of women of reproductive age living within a 30- and 60- minute drive time of the nearest tertiary care perinatal centre USA US census tract data Friedman, et al. (45) 2012 Evaluate the effect of an inverse relationship between health care use and distance to care related to emergency and essential surgical care Haiti, central district Retrospective review of operative logbooks; Haiti earthquake data portal Gething, et al. (46) 2012 Develop a uniquely detailed set of spatially-linked data and a calibrated geospatial model to undertake a national audit of geographical access to maternity care at birth Ghana Ghana Ministry of Health; University of Ghana; project by the Ghana Ministry of Health and Ghana Health Service Core; Welfare Indicator Questionnaire survey Blanford, et al. (47) 2012 Analyse the physical access of populations to health facilities with an emphasis on the effect of seasonal conditions and the implications of these conditions for availability of adequate health services, and provision of drugs and vaccinations Niger FAO Geo Network Portal; Niger Ministry of Health Table 1 Description of the included studies (Continued) 1408 Review EMHJ – Vol. 26 No. 11 – 2020 the leading causes of maternal deaths (73). Little attention was paid to these health factors in the studies in our review, which may be because the data sources used in these studies did not include these kind of health-related factors. The articles included in our study used and combined various data sources, which provides better results and allows greater understanding. Our review showed that the effects variables such as environmental factors, political policies, exposure to infectious diseases during pregnancy and nutritional status on maternal mortality are largely ignored. Evidence indicates important linkages between the water and sanitation environment and maternal and perinatal mortality (34,74). Although we believe there is a relation between the above-mentioned groups of variables and maternal mortality, it is difficult to know which variable is the strongest determinant. In addition, the strength of the association of these variables with maternal mortality may differ by region. However, proposing a dataset for research in this field would direct researchers to a unique guideline and standard data set (75). Despite the rapid growth of technologies and health information systems, most of health information systems do not merge patients’ records with external datasets. This fact can explain why isolated data systems cannot be used to recognize how the physical and environmental context of each patient influences his/her health choices Author Publication year Objectives of the study Countrya Data sources Massey, et al. (48) 2011 Identify priority regions for the expansion of human resources for health Senegal National agency for demography and statistics; WHO Gabrysch, et al. (49) 2011 Quantify the effects of distance to care and level of care on women’s use of health facilities for delivery Zambia National household data from the Zambian Demographic and Health Survey; national facility data from the Zambian health facility census Bailey, et al. (50) 2011 Provide a set of multicriteria decision analyses to help health planners make informed decisions about interventions to increase access to emergency services Ethiopia Ethiopian national survey on baseline assessment of emergency obstetric and newborn care; spatial population data from Land Scan™ population data Gjesfjeld & Jung. (51) 2011 Examine maternity care access for expectant mothers USA, Dakota North Dakota department of vital records Grzybowski, et al. (52) 2011 Systematically document newborn and maternal outcomes in terms of travel distance to access the nearest maternity services with caesarean section capability Canada, British Columbia British Columbia Perinatal Health Program Fisher and Myers (53) 2011 Test the appropriateness of new, inexpensive and simple GIS tools in poorly resourced areas of a developing country Indonesia, Nusa Tenggara Timur Cybertracker; health data collected by district and subdistrict health officer departments and clinics Bloch, et al. (54) 2011 Examine spatial patterns of neighbourhood contextual factors of stress with preterm birth and country of birth (USA or elsewhere) USA, Philadelphia Census data; de-identified geocoded Philadelphia birth records; publicly available Philadelphia police department crime statistics. Målqvist, et al. (55) 2010 Examine the association between distance from the mother’s home to the closest health facility and neonatal mortality, and investigating investigate the influence of distance on patterns of perinatal health care use Viet Nam, Quang Ninh Interviews with mothers and staff; medical records; VidaGIS database Pilkington, et al. (56) 2008 Describe the effect of maternity unit closures on distance and mean travel time between pregnant women’s homes and maternity units France French national perinatal surveys; vital statistics registries Dummer, et al. (57) 2004 Investigate whether geographical accessibility to hospitals affected the risk of infant mortality England, Cumbria Cumbrian births database Heard, et al. (58) 2004 Identify whether access to reproductive health services partly explains the use of modern contraception Malawi Malawi health facilities inventory; Malawi demographic and health survey WHO=World Health Organization; GIS=geographical information system; USA=United States of America; FAO=Food and Agriculture Organization. aWhere the region is not specified, the study done at the national level. Table 1 Description of the included studies (Concluded) 1409 Review EMHJ – Vol. 26 No. 11 – 2020 Table 2 Frequency of variables examined in the studies Background variables No. (%) (n = 40) Maternal factors Maternal age 11 (27.5) Antepartum haemorrhage 1 (25.0) Antenatal care visits 5 (12.5) Use of contraception (family planning) 2 (5.0) Number of children 1 (25.0) Parity 5 (12.5) Newborn outcomes 1 (25.0) Complications during last pregnancy 1 (25.0) Gestational age 2 (5.0) Birth weight 4 (10.0) Multiple births 2 (5.0) Type of delivery (normal or caesarean section, emergency or elective) 3 (7.5) Birth order and interval 3 (7.5) Socioeconomic factors Educational level of pregnant woman 10 (25.0) Parents’ educational level 3 (7.5) Household wealth 9 (22.5) Mother’s ethnicity 3 (7.5) Mother’s occupation 2 (5.0) Women’s autonomy within society 3 (7.5) Marital status 4 (10.0) Employment of head of household (employed/unemployed) 1 (25.0) Sex of head of household 1 (25.0) Sex of the newborn 2 (5.0) Sex of infants who have died 1 (25.0) Religion 1 (25.0) Exposure to media 2 (5.0) Exposure to family planning messages 1 (25.0) Residential area(urban, rural) 9 (22.5) Year of birth of mother 1 (25.0) Health care service factors Human resources 3 (7.5) Maternity and delivery beds 2 (5.0) Type of facility 4 (10.0) Level of delivery care (basic or comprehensive) 3 (7.5) Readiness of facilities to provide good delivery care 1 (25.0) Type of birth attendant (skilled or traditional) 2 (5.0) Place of delivery 3 (7.5) Ecological determinant factors Level of social vulnerability in catchment area 1 (25.0) Proportion of indigenous people in catchment area 1 (25.0) Environmental factors Drinking-water quality 1 (25.0) Health-related factors Chronic diseases (high blood pressure, diabetes, heart disease) 2 (5.0) Smoking and tobacco use during pregnancy 2 (5.0) Alcohol use during pregnancy 1 (25.0) HIV 1 (25.0) 1410 Review EMHJ – Vol. 26 No. 11 – 2020 and health outcomes. Therefore, the use of tools such as GIS is needed to evaluate these associations. Pregnant women’s access to health care centres and improvement in their health status are basic rights of women and can be thought of as an index of development in any country. Our study had some limitations. First, the variables examined in some of the studies were not clearly reported and may have been missing. Second, we classified the extracted variables based on expert consensus for a better reporting. As such, we may have misclassified some variables. Third, although we reported the effect of these variables on maternal care, we could not undertake a precise analysis because of the large number of descriptive studies and the many different objectives of the studies. We only included articles in English and searched only two databases which is another limitation as there might have been some relevant articles published in other languages and included in other databases. Conclusion Our review highlights the various applications of GIS in examining important variables in maternal care, and the need for programmes to improve the accessibility, use and quality of care for pregnancy and childbirth. Health care planners can use GIS to determine the best location and capacity of new health care facilities, and assess the costs. Furthermore, electronic health technologies, such as telemedicine, may be a way to overcome barriers of geographic access. Background variables No. (%) (n = 40) Geographic factors visualized on GIS Season of birth 3 (7.5) Distance to facility 16 (40.0) Travel time to facility and emergency obstetric care 17 (42.5) Type of transport taken to facility (on foot, vehicle, ambulance) 4 (10.0) Distribution health services and emergency obstetric care facilities per population 11 (27.5) Distribution of human resources 1 (25.0) Childbirths per region at health facilities, at home, or outside home or health facility (e.g. in car/ambulance) 5 (12.5) Distribution of childbirths occurring unassisted by health professionals 1 (25.0) Distribution of early neonatal, early fetal and late fetal deaths 4 (10.0) Distribution of woman receiving antenatal care 2 (5.0) Distribution of women who had caesarean sections 2 (5.0) Distribution of women of reproductive age 2 (5.0) Distribution of private and public maternity units 1 (25.0) Distribution of maternity beds 1 (25.0) Distribution of women with high-risk pregnancies 1 (25.0) Distribution of births with poor neonatal outcomes (national) 1 (25.0) Distribution of facility-based peripartum fetal care 1 (25.0) Table 2 Frequency of variables examined in the studies (Concluded) Acknowledgement We thank Dr Reza Khajouei and Dr Khodadad Sheikhzadeh for their comments. Funding: None. Competing interests: None declared. 1411 Review EMHJ – Vol. 26 No. 11 – 2020 Application des systèmes d’information géographique à la santé maternelle : étude exploratoire Résumé Contexte : L ’ amélioration de la santé maternelle est une priorité sanitaire à l’échelle mondiale et nécessite une évaluation précise des facteurs qui influent sur la santé des mères. Des systèmes d’information géographique ont été utilisés pour étudier les problèmes de santé maternelle. Objectifs : La présente étude avait pour objectif d’identifier les travaux de recherche faisant appel aux systèmes d’information géographique dans le domaine des soins de santé maternelle et de déterminer les variables en matière de santé et de mortalité maternelles que ces systèmes permettent de faire apparaître. Méthodes : Il s’agissait d’une étude exploratoire dans laquelle nous avons systématiquement recherché dans PubMed et Science Direct des études qui utilisaient des systèmes d’information géographique pour évaluer les soins de santé maternelle. Nous avons inclus toutes les études transversales pertinentes publiées en anglais entre décembre 1995 et décembre 2017. Nous avons extrait de chaque étude les informations suivantes : année d’étude, région, objectifs, type de système d’information géographique utilisé, variables mises en évidence par le système d’information géographique, et toutes les autres variables examinées concernant la santé maternelle. Résultats : Sur 5240 articles initialement récupérés, 40 ont fait l’objet d’une analyse détaillée. La plupart des études (n = 32) ont été réalisées dans des pays en développement en Afrique, en Asie, en Amérique latine et dans les Caraïbes. Grâce aux systèmes d’information géographique, la plupart des études (n = 33) ont permis de faire ressortir la distance entre le lieu d’habitation des mères et les établissements de santé ainsi que le temps de trajet pour se rendre dans les centres de soins. Les autres facteurs examinés portaient sur les capacités en matière de soins prénatals (n = 4) et de santé maternelle (n = 3). Conclusions : Il n’existe pas de recherche exhaustive sur l’application des systèmes d’information géographique aux soins maternels. La plupart des études ont appliqué une cartographie descriptive simple des schémas de répartition spatiale, complétée par l’ajout de quelques variables pertinentes. فياشكتسا ضارعتسا :ملأا ةحص لامج في ةيفارغلجا تامولعلما مُظُن قيبطت يجيب نيدلا ءابه زيبمق ،يلحاص ةمطاف ،نايدحمأ لىيل ةصلالخا مظُن تمِدخُتسا دقو .ملأا ةحص لىع رثؤت يتلا لماوعلل ًاقيقد ًماييقت بلطتيو ،يلماعلا ديعصلا لىع ةيحص ةيولوأ ملأا ةحص ينستح دعُي :ةيفللخا .ملأا ةحصب ةقلعتلما لكاشلما فاشكتسلا ةيفارغلجا تامولعلما تايرغتم ديدتحو ،تاهملأل ةيحصلا ةياعرلا لامج في ةيفارغلجا تامولعلما مظُن مدختست يتلا تاساردلا ديدتح لىإ ةساردلا هذه تفده :فادهلأا .مظُنلا هذه اهرهظُت يتلا تاهملأا تاَيَفَوو ملأا ةحص يتلا تاساردلا نع Science Directو PubMed تانايب دعاوق في يجهنم ٍبولسأب هيف انثحب ًايفاشكتسا ًاضارعتسا اذه ناك :ثحبلا قرط في ةيزيلجنلإا ةغللاب ةروشنلما ةلصلا ةقيثو ةيعطقلما تاساردلا عيجم انجردأو .تاهملأل ةيحصلا ةياعرلا مييقتل ةيفارغلجا تامولعلما مظُن مدختست ،ةساردلا ةنس :ةجَردُلما تاساردلا عيجم نم ةيلاتلا تامولعلما انصلختساو .2017 لولأا نوناك/برمسيدو 1995 لولأا نوناك/برمسيد ينب ةترفلا يتلا ىرخلأا تايرغتلما عيجمو ،ةيفارغلجا تامولعلما ماظن اهرهظُي يتلا تايرغتلماو ،مَدختسلما ةيفارغلجا تامولعلما ماظن عونو ،فادهلأاو ،ميلقلإاو .ملأا ةحصب ةقلعتلماو اهانسرد نادلب في )32 = ددعلا( تاساردلا مظعم تيرجُأو .ةيادبلا في تعجُترسا ةلاقم 5240 لصأ نم لييصفتلا ضارعتسلال ةلاقم 40 تجردُأ :جئاتنلا تقوو ةيحصلا قفارلما نع تاهملأا دعُب )33 = ددعلا( تاساردلا مظعم ترهظأو .يبيراكلا رحبلا ةقطنمو ةينيتلالا اكيرمأو ايسآو ايقيرفأ في ةيمان ةقباسلا ةياعرلا ميدقت لىع ةردقلا ةساردلل تعضخ يتلا ىرخلأا لماوعلا تلمشو .ةيفارغلجا تامولعلما مُظُن في ةيحصلا ةياعرلا زكارم لىإ رفسلا .)3 = ددعلا( تاهملأل ةيحصلا تامدلخا ميدقت لىع ةردقلاو ،)4 = ددعلا( ةدلاولل ةيفصو طئارخ مسر تاساردلا مظعم تقَّبطو .تاهملأا ةياعر لامج في ةيفارغلجا تامولعلما مظُن قيبطت نأشب ةلماش ثوحب دجوت لا :تاجاتنتسلاا .تايرغتلما نم ٍليلق ٍددع عم نياكلما عيزوتلا طمانلأ ةطيسب 1412 Review EMHJ – Vol. 26 No. 11 – 2020 References 1. Maternal health [website]. Geneva: World Health Organization; 2020 (https://www.who.int/maternal-health/en/, accessed 8 April 2020). 2. Maternal mortality: key facts. Geneva: World Health Organization; 2019 (https://www.who.int/en/news-room/fact-sheets/detail/ maternal-mortality, accessed 8 April 2020). 3. Trends in maternal mortality: 1990 to 2013. Estimates by WHO, UNICEF, UNFPA, The World Bank and the Unit- ed Nations Population Division. Geneva: World Health Organization; 2014. (https://apps.who.int/iris/bitstream/han - dle/10665/112682/9789241507226_eng.pdf?sequence=2, accessed 8 April 2020). 4. The Millennium Development Goals Report. New York: United Nations: 2015 (https://www.un.org/millenniumgoals/2015_ MDG_Report/pdf/MDG%202015%20rev%20(July%201).pdf, accessed 8 April 2020). 5. Ahmadian L, Salehi Nejad S, Khajouei R. Evaluation methods used on health information systems (HISs) in Iran and the effects of HISs on Iranian healthcare: a systematic review. Int J Med Inform. 2015;84(6):444–53. http://doi.org/10.1016/j.ijmedl- inf.2015.02.002 6. Ahmadian L, Khajouei R, Salehi Nejad S, Ebrahimzadeh M, Nikkar SE. Prioritizing barriers to successful implementation of hospital information systems. J Med Syst. 2014;38(12):151. http://doi.org/10.1007/s10916-014-0151-9 7. Gabrysch S, Cousens S, Cox J, Campbell OM. The influence of distance and level of care on delivery place in rural Zambia: a study of linked national data in a geographic information system. PLoS Med. 2011;8(1):e1000394. https://doi.org/10.1371/journal 8. Chen SC, Wang JD, Yu JK, Chiang TY, Chan CC, Wang HH, et al. Applying the global positioning system and google earth to evaluate the accessibility of birth services for pregnant women in northern Malawi. J Midwifery Womens Health. 2011;56(1):68– 74. https://doi.org/10.1111/j.1542-2011.2010.00005.x 9. Simões PP, Almeida RMVR. Geographic accessibility to obstetric care and maternal mortality in a large metropolitan area of Brazil. Int J Gynecol Obstet. 2011;112(1):25–9. https://doi.org/10.1016/j.ijgo.2010.07.031 10. Nykiforuk C, Flaman L. Exploring the utilization of geographic information systems in health promotion and public health (Technical report #08-001). Edmonton: University of Alberta; 2008 (http://bceohrngis.pbworks.com/f/BP+Exploring+GIS+in+A- Public+Health+2008.pdf, accessed 8 April 2020). 11. Nykiforuk CIJ, Flaman LM. Geographic information systems (GIS) for health promotion and public health: a review. Health Promot Pract. 2011:12(1):63–73. https://doi.org/10.1177/1524839909334624 12. Wangia V, Shireman TI. Review of Geographic Variation and Geographic Information Systems (GIS) Applications in Prescrip- tion Drug Use Research. Res Social Adm Pharm. 2013;9(6):666–87. https://doi.org/10.1016/j.sapharm.2012.11.006 13. Aimone AM, Perumal N, Cole DC. A systematic review of the application and utility of geographic information systems for ex- ploring disease relationships in paediatric global health research: the case of anaemia and malaria. Int J Health Geogr. 2013;12:1. https://doi.org/10.1186/1476-072X-12-1 14. Makanga PT, Schuurman N, von Dadelszen P, Firoz T. A scoping review of geographic information systems in maternal health. Int J Gynaecol Obstet. 2016:134(1):13–7. https://doi.org/10.1016/j.ijgo.2015.11.022 15. Ebener S, Guerra-Arias M, Campbell J, Tatem AJ, Moran AC, Johnson F, et al. The geography of maternal and newborn health: the state of the art. Int J Health Geogr. 2015;14(1):19. https://doi.org/10.1186/s12942-015-0012-x 16. Palaniyandi M. The role of remote sensing and GIS for spatial prediction of vector-borne diseases transmission: a systematic review. J Vector Borne Dis. 2012;49(4):197–204. 17. Ahmed N. Study reveals inadequate access to EmONC facilities in 24 districts of Bangladesh. Dhaka: icddr,b; 2014 (https://www. icddrb.org/news-and-events/news?id=634&task=view, accessed 8 April 2020). 18. Amoako-Johnson F, Frempong-Ainguah F, Matthews Z, Harfoot AJP, Nyarko P, Baschieri A, et al. Evaluating the impact of the community-based health planning and services initiative on uptake of skilled birth care in Ghana. PLoS One. 2015;10(3):e0120556. https://doi.org/10.1371/journal.pone.0120556 19. Ansariadi A, Manderson L. Antenatal care and women’s birthing decisions in an Indonesian setting: does location matter? Rural Remote Health. 2015;15:2959. 20. Jain AK, Sathar ZA, Haque M. The constraints of distance and poverty on institutional deliveries in Pakistan: evidence from georeference-linked data. Stud Fam Plann. 2015;46(1):21–39. https://doi.org/10.1111/j.1728-4465.2015.00013.x 21. Tabatabai P, Henke S, Susˇac K, Kisanga OME, Baumgarten I, Kynast-Wolf G, et al. Public and private maternal health service ca- pacity and patient flows in southern Tanzania: using a geographic information system to link hospital and national census data. Glob Health Action. 2014;7:22883. https://doi.org/10.3402/gha.v7.22883 22. Sabde Y, De Costa A, Diwan V. A spatial analysis to study access to emergency obstetric transport services under the public private “Janani Express Yojana” program in two districts of Madhya Pradesh, India. Reprod Health. 2014;11:57. https://doi. org/10.1186/1742-4755-11-57 23. Mwaliko E, Downing R, O’Meara W, Chelagat D, Obala A, Downing T, et al. Not too far to walk: the influence of distance on place of delivery in a western Kenya health demographic surveillance system. BMC Health Serv Res. 2014;14:212. https://doi. org/10.1186/1472-6963-14-212 24. Detres M, Lucio R, Vtucci J. GIS as a Community Engagement Tool: Developing a Plan to Reduce Infant Mortality Risk Factors. Matern Child Health J. 2014; 18:1049–1055. https://doi.org/10.1007/s10995-013-1337-3 1413 Review EMHJ – Vol. 26 No. 11 – 2020 25. Brown SA, Richards ME, Elwell EC, Rayburn WF. Geographical information systems for mapping maternal ground transport to level iii care neonatal centers. Am J Perinatol. 2014;31(4):287–92. https://doi.org/10.1055/s-0033-1348029 26. Nesbitt RC, Gabrysch S, Laub A, Soremekun S, Manu A, Kirkwood BR, et al. Methods to measure potential spatial access to delivery care in low- and middle-income countries: a case study in rural Ghana. Int J Health Geogr. 2014;13:25. https://doi. org/10.1186/1476-072X-13-25 27. Tatem AJ, Campbell J, Guerra Arias M, de Bernis L, Moran A, Matthews Z. Mapping for maternal and newborn health: the distributions of women of childbearing age, pregnancies and births. Int J Health Geogr. 2014;13(1):2. https://doi.org/10.1186/1476- 072X-13-2 28. McKinnon B, Harper S, Kaufman JS, Abdullah M. Distance to emergency obstetric services and early neonatal mortality in Ethio- pia. Trop Med Int Health. 2014;19(7):780–90. https://doi.org/10.1111/tmi.12323 29. Wang W, Winner M, Burgert C, Colston J. Influence of service readiness on use of facility delivery care: a study linking health facility data and population data in Haiti. (DHS working paper, No. 114). Rockville (MD): ICF International; 2014 (https://www. researchgate.net/publication/291343442_Influence_of_Service_Readiness_on_Use_of_Facility_Delivery_Care_A_Study_Link- ing_Health_Facility_Data_and_Population_Data_in_Haiti, accessed 8 April 20-20). 30. Gaspar JS, Sá T, Reis ZSN, Júnior RFN, Júnior MS, Gusmão RR. Use of geographic information system tools in research on neo- natal outcomes in a maternity school in Belo Horizonte – Brazil. Int J Adv Comput Sci Appl. 2014;5(12):91–6. 31. Asamoah BO, Agard A, Cromley EK. Spatial analysis of skilled birth attendant utilization in Ghana. Glob J Health Sci. 2014;6(4). https://doi.org/10.5539/gjhs.v6n4p117 32. Blake SB. Spatial relationships among dairy farms, drinking water quality, and maternal-child health outcomes in the San Joaquin Valley. Public Health Nurs 2014;31(6):492–9. https://doi.org/10.1111/phn.12166 33. Pilkington H, Blondel B, Carayol M, Breart G, Zeitlin J. Impact of maternity unit closures on access to obstetrical care: The French experience between 1998 and 2003. Soc Sci Med. 2008;67:1521–9. https://doi.org/10.1016/j.socscimed.2008.07.021 34. Arslan O, Cepni MS, Etiler N. Spatial analysis of perinatal mortality rates with geographic information systems in Kocaeli, Tur- key. Public health. 2013;127:369–79. https://doi.org/10.1016/j.puhe.2012.12.009 35. Engjom HM, Morken NH, Norheim OF, Klungsøyr K. Availability and access in modern obstetric care: a retrospective popula- tion-based study. BJOG. 2014;121:290–9. https://doi.org/10.1111/1471-0528.12510 36. Sudhof L, Amoroso C, Barebwanuwe P, Munyaneza F, Karamaga A, Zambotti G, et al. Local use of GIS mapping to improve rural C-section access. Trop Med Int Health. 2013;8:8–26. https://doi.org/10.1111/tmi.12016 37. Song P, Zhu Y, Mao X, Li Q, An L. Assessing spatial accessibility to maternity units in Shenzhen, China. PLoS ONE. 2013;8(7):e70227. https://doi.org/10.1371/journal.pone.0070227 38. Chong SH, Nelson M, Byun R, Harris L, Eastwood J, Jalaludin B. Geospatial analyses to identify clusters of adverse antenatal factors for targeted interventions. Int J Health Geog. 2013;12:46. https://doi.org/10.1186/1476-072X-12-46 39. Bowie C. Recent trends in geographical access to health care facilities in Malawi. Blantyre: Division of Community Health, Col- lege of Medicine, University of Malawi; 2013. 40. Masters SH, Burstein R, Amofah G, Abaogye P, Kumar S, Hanlon M. Travel time to maternity care and its effect on utilization in rural Ghana: a multilevel analysis. Soc Sci Med. 2013;93:147–54. https://doi.org/10.1016/j.socscimed.2013.06.012 41. Yao J, Murray AT, Agadjanian V. A geographical perspective on access to sexual and reproductive health care for women in rural Africa. Soc Sci Med. 2013;96:60–8. https://doi.org/10.1016/j.socscimed.2013.07.025 42. Mony PK, Krishnamurthy J, Thomas A, Sankar K, Ramesh BM, Moses S, et al. Availability and distribution of emergency ob- stetric care services in Karnataka State, south India: access and equity considerations. PLoS ONE. 2013;8(5):e64126. https://doi. org/10.1371/journal.pone.0064126 43. Almeida MCS, Gomes CMS, Nascimento LFC. Spatial analysis of neonatal mortality in the state of São Paulo, 2006–2010. Rev Paul Pediatr. 2014;32(4):374–80. https://doi.org/10.1016/j.rpped.2014.01.001 44. Brown S, Richards M, Rayburn W. Geographical information systems (GIS) mapping indicates access to tertiary care perinatal centers in the United States is regionally maldistributed. Am J Obstet Gynecol. 2012;206(1 Suppl):S304. https://doi.org/10.1016/j. ajog.2011.10.700 45. Friedman JM, Hagander L, Hughes CD, Nash KA, Linden AF, Blossom J, et al. Distance to hospital and utilization of surgical ser- vices in Haiti: do children, delivering mothers, and patients with emergent surgical conditions experience greater geographical barriers to surgical care. Int J Health Plann Mgmt. 2013;28(3):248–56. https://doi.org/10.1002/hpm.2134 46. Gething PW, Johnson FA, Frempong-Ainguah F, Nyarko P, Baschieri A, Aboagye P, et al. Geographical access to care at birth in Ghana: a barrier to safe motherhood. BMC Public Health. 2012;12:991. https://doi.org/10.1186/1471-2458-12-991 47. Blanford JI, Kumar S, Luo W, MacEachren AM. It’s a long, long walk: accessibility to hospitals, maternity and integrated health centers in Niger. Int J Health Geogr. 2012;11:24. https://doi.org/10.1186/1476-072X-11-24 48. Massey P. Reducing maternal mortality in Senegal: using GIS to identify priority regions for the expansion of human resources for health. World Health Popul. 2011;13(2):13–22. https://doi.org/10.12927/whp.2011.22633 49. Gabrysch S, Cousens S, Cox J, Campbell OMR .The influence of distance and level of care on delivery place in rural Zambia: a study of linked national data in a geographic information system. PLoS Med.2011;8(1): e1000394. https://doi.org/10.1371/journal. pmed.1000394 1414 Review EMHJ – Vol. 26 No. 11 – 2020 50. Bailey PE, Keyes EB, Parker C, Abdullah M, Kebede H, Freedman L. Using a GIS to model interventions to strengthen the emergency referral system for maternal and newborn health in Ethiopia. Int Fed Gynecol Obstet. 2011;115(3):300–9. https://doi. org/10.1016/j.ijgo.2011.09.004 51. Gjesfjeld CD, Jung J. How far? Using geographical information systems (GIS) to examine maternity care access for expectant mothers in a rural state. Soc Work Health Care. 2011;50(9):682–93. https://doi.org/10.1080/00981389.2011.575537 52. Grzybowski S, Stoll K, Kornelsen J. Distance matters: a population based study examining access to maternity services for rural women. BMC Health Serv Res. 2011;11:147. https://doi.org/10.1186/1472-6963-11-147 53. Fisher RP, Myers BA. Free and simple GIS as appropriate for health mapping in a low resource setting: a case study in eastern Indonesia. Int J Health Geogr. 2011;10:15. https://doi.org/10.1186/1476-072X-10-15 54. Bloch JR. Using geographical information systems to explore disparities in preterm birth rates among foreign-born and US-born black mothers. J Obstet Gynecol Neonatal Nurs. 2011;40(5):544–54. https://doi.org/10.1111/j.1552-6909.2011.01273.x 55. Målqvist M, Sohel N, Do TT, Eriksson L, Persson L. Distance decay in delivery care utilisation associated with neonatal mortali- ty. A case referent study in northern Viet Nam. BMC Public Health. 2010;10:762. https://doi.org/10.1186/1471-2458-10-762 56. Karlsen S, Say L, Souza JP, Hogue CJ, Calles DL, Gülmezoglu AM, et al. The relationship between maternal education and mortal- ity among women giving birth in health care institutions: analysis of the cross sectional WHO Global Survey on Maternal and Perinatal Health. BMC Public Health. 2011;11:606. https://doi.org/10.1186/1471-2458-11-606 57. Dummer TJB, Parker L. Hospital accessibility and infant death risk. Arch Dis Child. 2004;89(3):232–4. https://doi.org/10.1136/ adc.2003.030924 58. Heard NJ, Larsen U, Hozumi D. Investigating access to reproductive health services using GIS: proximity to services and the use of modern contraceptives in Malawi. Afr J Reprod Health. 2004;8(2):164–79. 59. McTavish S, Moore S, Harper S, Lynch J. National female literacy, individual socio-economic status, and maternal health care use in sub-Saharan Africa. Soc Sci Med. 2010;71(11):1958–63. https://doi.org/10.1016/j.socscimed.2010.09.007 60. Iyaniwura CA, Yussuf Q. Utilization of antenatal care and delivery services in Sagamu, south western Nigeria. Afr J Reprod Health. 2009;13(3):111–22. 61. Bishai D, Cohen RL, Alfonso YN, Adam T, Kuruvilla S, Schweitzer J. Factors contributing to child mortality reductions in 142 low- and middle-income countries between 1990 and 2010. Paper presented at the 2014 Annual Meeting of the Population Association of America, 1 May 2014, Boston, United States of America. 62. Caramani D, Eugster B. Which factors are necessary and/or sufficient for accelerated reduction of maternal and child mortality in low and middle-income countries? Results from a Boolean, qualitative comparative analysis (QCA). (Partnership for Mater- nal, Newborn & Child Health and partners, technical paper). Geneva: World Health Organization; 2014, (https://www.who.int/ pmnch/knowledge/publications/success_factors.pdf, accessed 12 July 2015). 63. Gabrysch S, Campbell OM. Still too far to walk: literature review of the determinants of delivery service use. BMC Pregnancy Childbirth. 2009;9:34. https://doi.org/10.1186/1471-2393-9-34 64. Koch E, Romero T, Romero CX, Akel C, Manriquez L, Paredes M, et al. Impact of education, income and chronic disease risk factors on mortality of adults: does “a pauper–rich paradox” exist in Latin American societies? Public Health. 2010;124(1):39–48. https://doi.org/10.1016/j.puhe.2009.11.008 65. Temmerman M, Verstraelen H, Martens G, Bekaert A. Delayed childbearing and maternal mortality. Eur J Obstet Gynecol Re- prod Biol. 2004;114(1):19–22. https://doi.org/10.1016/j.ejogrb.2003.09.019 66. Callaghan WM, Berg CJ. Pregnancy-related mortality among women aged 35 years and older, United States, 1991–1997. Obstet Gynecol. 2003;102 (5 Pt 1):1015–21. https://doi.org/10.1016/s0029-7844(03)00740-3 67. Berg CJ, MacKay AP, Qin C, Callaghan WM. Overview of maternal morbidity during hospitalization for labor and delivery in the United States, 1993–1997 and 2001–2005. Obstet Gynecol. 2009;113(5):1075–1081. https://doi.org/10.1097/AOG.0b013e3181a09fc0 68. Berg CJ, Callaghan WM, Syverson C, Henderson Z. Pregnancy-related mortality in the United States, 1998 to 2005. Obstet Gyne- col. 2010;116(6):1302–9. https://doi.org/10.1097/AOG.0b013e3181fdfb11 69. Conde-Agudelo A, Belizan JM, Lammers C. Maternal perinatal morbidity and mortality associated with adolescent pregnancy in Latin America: cross-sectional study. Am J Obstet Gynecol. 2004;192(2):342–9. https://doi.org/10.1016/j.ajog.2004.10.593 70. Patton GC, Coffey C, Sawyer SM, Viner RM, Haller DM, Bose K, et al. Global patterns of mortality in young people: a systematic analysis of population health data. Lancet. 2009(9693);374:881–92. https://doi.org/10.1016/S0140-6736(09)60741-8 71. Benova L, Cumming O, Campbell OMR. Systematic review and meta-analysis: association between water and sanitation envi- ronment and maternal mortality. Trop Med Int Health. 2014;19(4):368–87. https://doi.org/10.1111/tmi.12275 72. Yego F, D’Este C, Byles J, Williams JS, Nyongesa P. Risk factors for maternal mortality in a tertiary hospital in Kenya: a case control study. BMC Pregnancy Childbirth. 2014;14:38. https://doi.org/10.1186/1471-2393-14-38 73. Khan S, Wojdyla D, Say L, Gulmezoglu A, Van Look P. WHO analysis of cause of maternal death: a systematic review. Lancet. 2006;367(9516):1066–74. https://doi.org/10.1016/S0140-6736(06)68397-9 74. Ahmadian L, Cornet R, Kalkman C, de Keizer NF; NVA Working Group “Minimal Dataset for Preoperative Assessment”. Devel- opment of a national core dataset for preoperative assessment. Methods Inf Med. 2009;48(2):155–61. https://doi.org/10.3414/ ME9218 1415 EMHJ – Vol. 26 No. 11 – 2020Short research communication Reproductive and behavioural risk factors of low birth weight among newborns in Al Thawra Hospital, Sana’a, Yemen Idayu Idris,1 Manal Sheryan,1,2 Qistina Ghazali1 and Azmawati Nawi1 1Department of Community Health, Faculty of Medicine, UKM Medical Centre, Cheras, Kuala Lumpur, Malaysia (Correspondence to: A. Nawi: azma- wati@ppukm.ukm.edu.my). 2Al Thawra Hospital, Sana’a, Yemen. Abstract Background: Low birth weight can lead to infant death, especially during the first year of life. Aims: To assess risk factors related to low birth weight babies in Sana’a, Yemen. Methods: We conducted an unmatched case–control study of 252 women who came for delivery at Al Thawra Hospital, Sana’a, Yemen, between August and October 2016. Results: Significant risk factors for low birth weight were: birth interval < 2 years; history of pre-eclampsia during current pregnancy; preterm gestational age < 37 weeks; and khat chewing or smoking during pregnancy. After controlling for all the confounders, only birth interval < 2 years was significantly associated with low birth weight. Conclusion: Shorter birth interval is an important risk factor for low birth weight; therefore, improving maternal aware- ness of this should be emphasized during postnatal follow-up. Keywords: low birth weight, prematurity, pre-eclampsia, khat chewing, smoking. Citation: Idris I; Sheryan M; Ghazali Q; Nawi A. Reproductive and behavioural risk factors of low birth weight among newborns in Al Thawra Hospital, Sana’a, Yemen. East Mediterr Health J. 2020;26(11):1415-1419. https://doi.org/10.26719/emhj.20.061 Received: 14/05/19; accepted: 27/11/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Introduction Low birth weight (LBW) is defined as birth weight < 2.5 kg (1). Two main factors cause LBW: preterm deliv- ery (< 37 weeks’ gestation) and growth restriction, or a combination of both (2). LBW is related to maternal bi- ological, social and general health (3). Infants with LBW have almost 20 times greater risk of dying compared to normal-weight newborn infants (4). LBW is predic- tive of a newborn’s health and survival, and can lead to death during the first year of life. Unfortunately, even if an LBW infant survives, they may face life-threatening conditions throughout their life and be exposed to chron- ic health issues in later life (4). According to the World Health Organization (WHO), there are an estimated 25 million LBW infants born each year worldwide, which comprise 17% of all live births, and about 95% of them are born in developing countries (5). There is a high prevalence of LBW infants in Yemen, where 32% of all newborn infants have LBW (6). This study was conducted to assess the maternal sociodemographic, reproductive and lifestyle risk factors related to LBW in Sana’a City, Yemen. Methods Study design and sampling This was an unmatched case–control study conducted at Al Thawra Modern General Hospital, located in Sana’a City, Yemen during August–October 2016. This hospital is one of the biggest referral hospitals in Yemen. The cases and controls were selected according to the inclusion cri- teria, that is, women who gave birth to a single live infant and who were resident in Sana’a City for at least 1 year. Women who had given birth to newborns weighing < 2.5 kg were classified as cases, and women who had given birth to newborns weighing ≥ 2.5 kg were classified as controls. Exclusion criteria were newborns that weighed < 1.5 kg or > 4.0 kg, twins, stillbirths, and infants with a visible congenital anomaly. From the medical records, cases were selected by convenience sampling while con- trols were selected by simple random sampling. All new- borns were weighed within 1 hour after birth. Sample size was calculated using Open Epi version 3.01 using a formula one proportion. Based on a previous study in Su- dan, with a prevalence of LBW of 12.6% (7), the sample size calculated for this study was 252 mothers of newborns (126 cases and 126 controls). Data collection Data were collected from selected women through self-ad- ministered questionnaires and medical records. Ques- tionnaires were administered via face-to-face interview and questions were closed-ended in nature (Yes/No an- swers). The questionnaires and medical records consist- ed of women’s age, educational level, occupational status, birth parity, birth interval, history of previous abortion, history of pre-eclampsia during the current pregnancy, gestational age of the newborn, number of antenatal care visits, as well as khat chewing and smoking during the current pregnancy. Interviews were conducted by 2 well- trained female data collectors within the first 24 hours af- ter delivery, and medical record data were extracted after the interview. The researchers trained the data collectors 1416 EMHJ – Vol. 26 No. 11 – 2020Short research communication and supervised them during data collection and checked all the questionnaires to ensure accuracy. Ethical considerations The study was approved by the Medical Research and Ethics Committee of the University of Science and Tech- nology, Sana’a, Yemen. Before starting the interview, the participants were informed about the purpose of the study and its benefits. Verbal and signed consent were obtained from the participants before data collection. The respondents were assured that all the information would be confidential and only used for this study. Statistical analysis All the data were analysed by SPSS version 24.0. Qualita- tive variables were described by calculation of frequency, and quantitative variables were represented by mean and standard deviation. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to measure the risk. χ2 and Fisher’s exact tests were used to show significant asso- ciations between cases and controls, as well as associa- tion of various risk factors and LBW. We used multiple logistic regression to calculate the adjusted ORs (aORs) after controlling for all possible confounders, with the corresponding 95% CIs, for LBW concerning exposure of interest. Results A total of 126 cases and 126 controls were included in the study (Table 1). The mean age in the cases and controls was 25.9 (standard deviation; 7.06) and 25.2 (5.48) years, respectively. Most cases and controls were aged 20–29 years. In the case group, 67 (53.2%) women did not com- plete their primary education and 59 (46.8%) who did. In the control group, 58 (46%) women did not complete their primary education and 68 (54%) did. In terms of occupa- tion, 116 (92.1%) women in the case group were house- wives and 10 (7.9%) worked outside the home. In the control group, 113 (89.7%) women were housewives and 13 (10.3%) worked outside the home. There was no signif- icant association between cases and controls concerning sociodemographic and socioeconomic characteristics. Birth interval < 24 months, history of pre-eclampsia, preterm delivery, chewing khat, and smoking were significant risk factors for LBW (Tables 2 and 3). However, in multivariate analysis, the only significant risk factor for LBW was birth interval < 24 months (aOR = 2.24, 95% CI = 1.017–4.952, P = 0.045) after controlling for the effect of other predictors. Discussion LBW is a significant cause of morbidity and mortality among neonates and children (8). Infants’ gestational age has a vital role in determining BW. There is an in- creased risk of LBW for premature infants (< 37 weeks’ gestation). According to WHO, prematurity is the cause of LBW in about one third of LBW infants (8). The pres- ent study showed that LBW was significantly associated with preterm delivery, which was consistent with pre- vious studies (5,7,8). The reason for this might be that most of the fetal growth and weight gain is in the late period of pregnancy; thus, preterm infants receive few- er nutrients, which can lead to LBW. In this study, khat chewing during pregnancy was a significant risk factor for LBW, which was consistent with other studies (9,10). This might be due to the influence of khat on reducing maternal food intake, which affects proper weight gain during pregnancy and fetal growth. Smoking during pregnancy was a significant risk factor for LBW, which is supported by most studies worldwide (11,12). The harmful components of cigarettes or hubble-bubble get into the maternal circulation and then to the placenta, affecting fetal growth and weight. Recommendations can be made to reduce the prevalence of LBW. Screening should be conducted by healthcare professionals of pregnant women with high risks of delivering LBW infants, especially if the mothers have a birth interval < 24 months. Furthermore, awareness Table 1 Maternal sociodemographic characteristics of study participants Sociodemographic characteristics Cases, n = 126 Controls, n = 126 P* OR 95% CI Mean (SD) n % Mean (SD) n % Age (years) 25.92 (7.06) 25.26 (5.48) < 20 21 16.7 17 13.5 0.301 1.45 0.71–2.96 20–29 70 55.6 82 65.1 Ref ≥ 30 35 27.8 27 21.4 1.52 0.84–2.75 Educational level Did not complete basic education 67 53.2 58 46.0 0.257 1.33 0.81 –2.18 Completed basic education and more 59 46.8 68 54.0 Ref Occupational status Housewife 116 92.1 113 89.7 0.512 1.33 0.56–3.17 Work outside 10 7.9 13 10.3 Ref *2 Statistically significant at P < 0.05. CI = confidence interval; OR = odds ratio; SD = standard deviation. 1417 EMHJ – Vol. 26 No. 11 – 2020Short research communication raising and health education on how to carry on a healthy pregnancy should be focused individually. Improvement of lifestyle by all means, and avoiding smoking and chewing khat during pregnancy are essential. Pregnant women need access to suitable maternal health services, including antenatal care and nutritional counselling services. Conclusion Risk factors for LBW identified in this study can be re- duced and prevented by improving maternal health edu- cation, especially on pregnancy spacing. Funding: None Competing interests: None declared. Table 2 Maternal reproductive factors for low birth weight Reproductive factors Cases, n = 126 Controls, n = 126 P OR 95% CI n % n % Birth parity 1 57 45.2 50 39.7 0.655 1.19 0.60–2.36 2 23 18.3 24 19.0 Ref 3 or more 46 36.5 52 41.3 0.92 0.46–1.85 Birth interval < 24 months 33 47.8 20 26.7 0.009* 2.52 1.26–5.06 ≥ 24 months 36 52.2 55 73.3 Ref History of abortion Yes 27 21.4 23 18.3 0.527 1.22 0.66–2.27 No 99 78.6 103 81.7 Ref History of pre-eclampsia Yes 27 21.4 10 7.9 0.002* 3.16 1.46–6.86 No 99 78.6 116 92.1 Ref Gestational age Preterm (< 37 weeks) 38 30.2 1 .8 < 0.001* 53.98 7.27–400.53 Full term 88 69.8 125 99.2 Ref No. of antenatal care visits < 4 44 34.9 35 27.8 0.222 1.40 0.82–2.38 ≥ 4 82 65.1 91 72.2 Ref *χ2 Statistically significant at P < 0.05. CI = confidence interval; OR = odds ratio. Table 3 Lifestyle risk factors for low birth weight Lifestyle risk factors Cases Controls P OR 95% CI n % n % Khat chewing Yes 83 65.9 66 52.4 0.029* 1.76 1.06–2.92 No 43 34.1 60 47.6 Ref Smoking Yes 41 32.5 17 13.5 < 0.001* 3.09 1.64–5.82 No 85 67.5 109 86.5 Ref *χ2 Statistically significant at P < 0.05. CI = confidence interval; OR = odds ratio. 1418 EMHJ – Vol. 26 No. 11 – 2020Short research communication References 1. Low birthweight: country, regional and global estimates. UNICEF; 2004. (https://www.unicef.org/publications/index_24840. html, accessed 14 May 2020). 2. Bendhari ML, Haralkar SJ. Study of maternal risk factors for low birth weight neonates: a case–control study. Int J Med Sci Public Health. 2015; 4(7):987–90. http://dx.doi.org/10.5455/ijmsph.2015.20032015203 3. Deshpande Jayant D, Phalke DB, Bangal VB, D Peeyuusha BS. Maternal risk factors for low birth weight neonates: a hospital based case control study in rural area of Western Maharashtra, India. Natl J Commun Med. 2011 Oct–Dec; 2(3):394–8. 4. Child health USA 2011. Rockville, MD: US Department of Health and Human Services, Health Resources and Services Adminis- tration; 2011 (http://mchb.hrsa.gov/, accessed 14 May 2020). 5. Paramita S, Sharma N, Benjamin AI. Risk factors for low birth weight: a case control study in Ludhiana Punjab. Indian J Matern Child Health. 2009;11(1–4):9. 6. Riniker KS. Women’s health in Yemen: factors influencing maternal and infant health, fertility rates, the public health care system, education, and globalization. J Global Health Perspect. 2012. 7. Bener A, Salameh KM, Yousafzai MT, Saleh NM. Pattern of maternal complications and low birth weight: associated risk factors among highly endogamous women. ISRN Obstet Gynecol. 2012; 2012:540495. http://dx.doi.org/10.5402/2012/540495 PMID:22991672 Facteurs de risque reproductifs et comportementaux liés au faible poids de naissance chez les nouveau-nés de l’hôpital Al Thawra de Sanaa au Yémen Résumé Contexte : Un faible poids de naissance peut entraîner la mort du nourrisson, en particulier pendant la première année de vie. Objectifs : La présente étude visait à évaluer les facteurs de risque liés au faible poids de naissance des enfants nés à Sanaa au Yémen. Méthodes : Nous avons mené une étude cas-témoins non appariés auprès de 252 femmes venues accoucher à l’hôpital Al Thawra de Sanaa, au Yémen, entre août et octobre 2016. Résultats : Les facteurs de risque significatifs du faible poids de naissance étaient un intervalle entre deux naissances inférieur à deux ans, des antécédents de pré-éclampsie pendant la grossesse concernée, l’âge gestationnel prématuré inférieur à 37 semaines et la consommation de khat ou le tabagisme pendant la grossesse. Après contrôle de tous les facteurs de confusion, seul un intervalle entre deux naissances inférieur à deux ans était significativement associé à un faible poids de naissance. Conclusion : La réduction de l’intervalle entre les naissances représente un facteur de risque important de faible poids de naissance ; par conséquent, il convient de mettre l’accent sur une meilleure sensibilisation des mères sur ce point à l’occasion du suivi postnatal. نميلاب ءاعنص في ةروثلا ىفشتسم في ديلاولما نزو ضافخنلا ةيكولسلاو ةيباجنلإا رطلخا لماوع يوان تياوامزأ ،ليازغ انيتسق ،نايشر لانم ،سيردإ وياديإ ةصلالخا .مهرماعأ نم لىولأا ةنسلا للاخ ةصاخو ،ع َّضُرلا ةافو لىإ ةدلاولا دنع نزولا ضافخنا يدؤي نأ نكمي :ةيفللخا .نميلاب ءاعنص في ةدلاولا دنع نزولا ضافخناب ينباصلما لافطلأاب ةقلعتلما رطلخا لماوع مييقت لىإ ةساردلا هذه تفده :فادهلأا ةترفلا في ،نميلاب ءاعنص في ةروثلا ىفشتسم في ةدلاولل نضرح ةأرما 252 هعوممج الم اهعون نم ةديرف دهاوشو تلااح ةسارد انيرجأ :ثحبلا قرط .2016 لولأا نيشرت/ربوتكأو بآ/سطسغأ ينب خيرات دوجوو ؛ْينَماع نم لقأ لىإ ةدلاولا تارم ينب ةلصافلا ةترفلا ِّندت ةدلاولا دنع نزولا ضافخناب ةباصلإل ةمهلما رطلخا لماوع ينب نِم :جئاتنلا دعبو .لملحا ءانثأ ينخدتلا وأ تاقلا غضمو ،ًاعوبسأ 37 نع لقي يذلا ركبلما ليْمَلحا رمعلاو ،ليالحا لملحا ءانثأ لملحا ممست تامدقلم قباس نزولا ضافخناو ْينَماع نم لقأ لىإ ةدلاولا تارم ينب ةلصافلا ةترفلا ِّندت ينب ىوس مهم طابترا كانه نكي لم ،كابرلإا لماوع عيجم لىع ةرطيسلا .ةدلاولا دنع تاهملأا يعو زيزعت لىع ديكأتلا يغبني ،اذلو ؛ةدلاولا دنع نزولا صقنل ًماهم رطخ لماع ةدلاولا تارم ينب ةلصافلا ةترفلا صرِق دَعُي :تاجاتنتسلاا .ةدلاولا دعب ام ةعباتم ءانثأ رملأا اذبه 1419 EMHJ – Vol. 26 No. 11 – 2020Short research communication 8. Sutan R, Mohtar M, Mahat AN, Tamil AM. Determinant of low birth weight infants: a matched case control study. Open J Prev Med. 2014;4:91–9. 9. Demelash H, Motbainor A, Nigatu D, Gashaw K, Melese A. Risk factors for low birth weight in Bale zone hospitals, South-East Ethiopia: a case–control study. BMC pregnancy and childbirth. 2015 Oct 13;15:264. http://dx.doi.org/10.1186/s12884-015-0677-y PMID:26463177 10. Abdel-Aleem MA. Khat chewing during pregnancy: an insight on an ancient problem impact of chewing khat on maternal and fetal outcome among Yemeni pregnant women. J Gynaecol Neonat Biol. 2015; http://dx/doi.org/10.15436/2380-5595.15.004 11. Zheng W, Suzuki K, Tanaka T, Kohama M, Yamagata Z, Okinawa Child Health Study Group. Association between maternal smoking during pregnancy and low birthweight: effects by maternal age. PLoS One. 2016 Jan 21;11(1):e0146241. http://dx.doi. org/10.1371/journal.pone.0146241 PMID:26795494 12. Stojanović M, Bojanić V, Musović D, Milosević Z, Stojanović D, Visujić A et al. Maternal smoking during pregnancy and soci- oeconomic factors as predictors of low birth weight in term pregnancies in Niš. Vojnosanit Pregled. 2010 Feb;67(2):145–50 (in Serbian) http://dx.doi.org/10.2298/vsp1002145s PMID:20337097 1420 EMHJ – Vol. 26 No. 11 – 2020Short research communication Barriers to the use of dental services by children in Lebanon and association with parental perception of oral health care Ingrid Karam,1 Miran A. Jaffa2 and Joseph Ghafari1 1Division of Orthodontics and Dentofacial Orthopedics, Department of Otorhinolaryngology, Head and Neck Surgery, American University of Beirut Medical Center, Beirut, Lebanon. 2Epidemiology and Population Health Department, Faculty of Health Sciences, American University of Beirut, Beirut, Lebanon. (Correspondence to: Miran Jaffa: ms148@aub.edu.lb). Abstract Background: Oral health is important to general health but use of dental services varies considerably, particularly for children. Aims: We aimed to determine factors associated with parents’ use of dental services for their children in Lebanon, and their perception of dental care relative to medical care. Methods: A convenience sample of public and private schools in Beirut was selected between January and May 2013. Parents of children in grades 2–6 (aged 7–12 years) were invited to complete a questionnaire covering socioeconomic characteristics and use of dental services. Logistic regression analysis was used to assess the relationship between use of dental services, and parents’ socioeconomic characteristics and awareness and perceptions of dental services. Results: The parents of 316 children returned the questionnaire. Most children (72.8%) had been taken to the dentist in the past year, mainly for emergency care. Most parents (78.2%) considered dental care as important as or more important than medical care, and 89.9% were willing to contribute to dental insurance. Use of dental services was significantly associated with: older age of the parent (odds ratio, OR = 1.04; 95% confidence interval, CI: 1.02–1.06); awareness of dental care centres offering affordable treatment (OR = 3.18; 95% CI: 1.52–6.68); and children being in private schools (OR = 2.00, 95% CI: 1.08– 3.95). It was negatively associated with > 4 children in the family compared with 1 child (OR = 0.18; 95% CI: 0.04–0.81). Conclusion: Barriers to dental care for children were mostly related to economic factors. Keywords: dental care for children, oral health care, dental insurance, Lebanon Citation: Karam I; Jaffa MA; Ghafari J. Barriers to the use of dental services by children in Lebanon and association with parental perception of oral health care. East Mediterr Health J. 2020;26(11):1420–1424. https://doi.org/10.26719/emhj.20.079 Received: 02/05/19; accepted: 03/11/19 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo) Introduction Poor oral health, which is related to excessive sugar in- take and low levels of fluoride exposure (1,2), affects 60–90% of schoolchildren globally (3) and the most so- cially disadvantaged populations (4), particularly given the high cost of treatment in low- and middle-income countries (5,6). The use of dental services has been widely investigated in relation to sociodemographic variables. Utilization is lower in younger people (21–35 years) and people of lower socioeconomic status (7). On the other hand, the use of dental services in children is greater when parents have a higher education (4,7,8) and have a better perception about oral health, including preven- tive and comprehensive care rather than only emergency treatment (e.g. toothache) (4,9). In addition, utilization of dental care is greater when third-party dental coverage is available – uninsured people are twice as likely to ne- glect dental care as insured people (10). Low utilization of dental services was associated with treatment expenses in Lebanon (11), where only about 20% of the population benefit from dental insurance through military and civil servant programmes (12). In addition, oral health is globally the most common unmet health care need (13). This fact raises the question about people’s perception of oral care in relation to medical care. Although a statistically significant association has been reported between perceived general and dental health (14), the relationship between such perception and use of dental services is not well known. We hypothesized that the perception of the importance of oral care in relation to medical care may be associated with use of dental services, and that utilization among Lebanese schoolchildren would reflect the oral care perception of parents and related practices. Private and public schools are fairly equally distributed in Beirut – 52.8% and 47.2% of pupils attend private and public institutions, respectively (15). They therefore represent a good model for socioeconomic background. The aims of our study were to: (i) identify the factors associated with access to dental care in schoolchildren from different socioeconomic backgrounds and with the barriers preventing parents from using needed dental services for their children; and (ii) assess parental perception of dental care in relation to medical care. Methods Study design and sample This was a cross-sectional study of children in grades 2 to 6 (aged 7–12 years) attending five private and two public 1421 EMHJ – Vol. 26 No. 11 – 2020Short research communication schools in Beirut that agreed to allow children and their parents to participate in the study. The sample was a con- venience sample in the first phase (January–May 2013) of a multiphase study on the children’s oral health. In this phase, the children were given a consent form and a questionnaire to take home to their parents for them to sign and complete within one week. Parents who re- turned the signed consent form and the completed ques- tionnaire were included in the study. Power analysis showed that with a sample size of 316 responses, an effect size of 0.07 (between a small (0.02) and the medium (0.15) effect size) and a probability level of 0.05, the statistical power is 0.8, with 20 predictors. Data collected The questionnaire included standard demographic and socioeconomic questions and questions on the use of dental services. It did not include any scale that required validation. We did a pilot study on a random sample of 10 parents to ensure that the questions were clear to the participants. We excluded these parents from the study to avoid introducing bias. The recorded categorical var- iables were: · Outcome variable: use of dental services – parents had taken their children to the dentist in the past year (yes/no) and reasons for taking them to the dentist (routine check-up and prevention, emergency moti- vated by pain, or orthodontics). · Sociodemographic characteristics – sex of the re- sponding parent, marital status (married/separated/ divorced), respondent parent’s educational level (read and write, primary school, middle and secondary school, college/university), number of children in the family (1,2,3,4, > 4). · Socioeconomic characteristics: family monthly in- come (< 1 , ≥ 1 million Lebanese pounds (US$ 1 = 1500 Lebanese pounds]), children’s school (public or pri- vate), family medical/dental insurance (yes/no). · Respondent parent’s awareness of dental care centres offering affordable treatment (yes/no). · Respondent parent’s perception of their children’s oral health (less/equally/more important than their general health). · Respondent parent’s willingness to use dental servic- es more for their children if they were covered by or paid premiums towards dental insurance (yes/no). Statistical analysis We used simple and multivariable logistic regression analyses to evaluate the association between use of den- tal services and the independent variables. Variables that were statistically significant (P < 0.05) in the simple logis- tic regression analysis were entered in the multivariable analysis. Given that observations from the same school tend to be correlated, we used a logistic regression anal- ysis with a cluster effect at the school level. This method incorporates within-school intracluster correlation in the computation of standard errors. Data are presented as unadjusted and adjusted odds ratios (OR) and 95% con- fidence intervals (CI). Stata, version 12.1 was used for the analyses. Ethical considerations The multiphase study was approved by the Institutional Review Board of the American University of Beirut. All participating parents signed an informed consent form. Results The parents of 316 children in grades 2 to 6 (aged 7–12 years) returned the questionnaire and signed the in- formed consent form. Age was recorded for 273 (86.4%) parents: mean age and standard deviation (SD) was 38.30 (SD 6.40) years, range 16–57 years. The mean age of the children was 9.5 (SD 1.5) years. The proportion of completed questionnaires was about the same for public (n = 151, 47.8%) and private (n = 165, 52.2%) schools. Most parents (n = 230; 72.8%) had taken their children to the dentist at least once in the past year, 61.3% of whom (n = 141/230) had made more than one visit. The reasons for the visits were: decay and acute pain (n = 206; 89.6%), regular check-up (n = 153; 66.5%) and appearance of teeth (n = 131; 57.0%). Most parents (n = 284; 89.9%) were willing to invest in dental health and increase their utilization of dental services if costs were covered by dental insurance. We found statistically significant associations between use of dental services and eight variables in the simple regression analysis (Table 1): respondent’s age, respondent’s educational level, number of children in family, monthly family income, school type, medical insurance, perception of oral and general health, and awareness of the presence of dental care centres. Sex (P = 0.481) and marital status (P = 0.705) were not significantly associated with use of dental services and were not included in the multivariable analysis. In the multivariable analysis, four of these variables remained significantly associated with use of dental services – use of dental services increased with the older age of the respondent parent, fewer children in the family (> 4 children was significant in the unadjusted and adjusted analysis), awareness of dental care centres offering affordable treatments, and children being in private schools (Table 1). Discussion Many of our results corroborate the findings of previ- ous studies. For example, the most common reason for dental visits was emergency care (acute pain and decay) rather than preventive care and orthodontics (16), and use of dental services was associated with economic status, educational level of the parent, and monthly fam- ily income (4,7–17). However, our main results relate to the effect on behaviour of awareness of affordable den- tal centres and the perception that dental care is equal- ly important as or more important than medical care. 1422 EMHJ – Vol. 26 No. 11 – 2020Short research communication The fact that almost twice as many of the parents who had medical insurance used dental services for their children compared with those without medical insurance shows that insurance coverage increases utilization, potential- ly improving oral health and preventive care visits. This trend has been shown in low-income Americans with a dramatic increase (from 33% to 80%) in dental service uti- lization when they have medical insurance (18). Further- more, most parents of children in both public and private schools (89.8%) were willing to invest in dental health, which shows the importance of implementing dental in- surance within the medical insurance programmes. Health issues are concerned with life and death or quality of life. Most dental and many medical problems are not life-threatening but they can affect quality of life considerably. Yet, oral health care and general health care are considered independent entities. Why should a line be drawn between a headache and a toothache, whether mild or debilitating? The roots of a seemingly artificial split may lie in education (separate medical and dental schools) and insurance (separate for medical and dental needs) (19). Policy-makers should rethink this categorization and consider integration based on the principle that oral health is part of total health. The financial burdens of dental services may be addressed by giving greater coverage for preventive care and progressive copayment for different treatment options of more demanding procedures. Our study has some limitations, mainly the potential response bias and the convenience nature of the sample (selection was based only on agreement of schools and parents to participate in the study). More research is warranted on perception and awareness of dental needs and care with a larger sample from different geographic areas. Such research should evaluate the psychological disposition to seek preventive dental care, and ways to encourage everyone to have regular check-ups. Table 1 Simple and multivariable logistic regressions for dental services utilization adjusted for intracluster correlation Variable Values Unadjusted OR (95% CI) P Adjusted ORa (95% CI) P Mean (SD) Respondents’ age (years) 38.30 (6.40) 1.04 (1.02–1.07) < 0.001 1.04 (1.02–1.06) < 0.001 No. (%) Educational level Read and write/Primary (Ref) 74 (23.8) Intermediate/Secondary 94 (30.2) 0.93 (0.34–2.53) 0.901 1.02 (0.33–3.10) 0.967 College/University 143 (46.0) 1.86 (1.14–3.03) 0.013 1.32 (0.70–2.47) 0.385 Children in the family 1 (Ref) 12 (3.8) 2 102 (32.3) 0.34 (0.07–1.63) 0.180 0.55 (0.17–1.80) 0.328 3 113 (35.8) 0.22 (0.03–1.31) 0.097 0.33 (0.08–1.28) 0.111 4 58 (18.4) 0.18 (0.02–1.16) 0.072 0.24 (0.05–1.00) 0.050 > 4 31 (9.8) 0.18 (0.03–0.85) 0.031 0.18 (0.04–0.81) 0.025 Monthly family income (Lebanese poundsb) < 1 000 000 (Ref) 134 (44.8) ≥ 1 000 000 165 (55.2) 2.28 (1.42–3.66) < 0.001 1.63 (0.89–2.98) 0.108 School type Public (Ref) 151 (47.8) Private 165 (52.2) 2.69 (1.71–4.23) < 0.001 2.00 (1.08–3.95) 0.027 Have medical insurance No (Ref) 100 (33.0) Yes 203 (67.0) 1.99 (1.35–2.93) < 0.001 0.94 (0.54–1.65) 0.84 Perception of oral health compared with general health Less important (Ref) 52 (17.4) As important/more important 247 (82.6) 2.13 (1.01–4.52) 0.047 1.02 (0.01–1.65) 0.905 Awareness of dental care centres offering affordable services No (Ref) 157 (52.9) Yes 140 (47.1) 1.99 (1.25–3.19) 0.004 3.18 (1.52–6.68) 0.002 OR: odds ratio; CI: confidence interval; SD: standard deviation; Ref: reference category. aAdjusted for all other variables in the multivariable analysis. bUS$ 1 = 1500 Lebanese pounds. Percentages (%) are computed out of the total responses for each question 1423 EMHJ – Vol. 26 No. 11 – 2020Short research communication In conclusion, the use of dental services reflected the disparity in economic and educational levels of the parents. Even the poorer respondents would be prepared to share in the cost of dental insurance coverage, regardless of the presence or absence of medical insurance. This finding indicates that people perceive oral health as part of total health and suggests that the present system of separate coverage for medical and dental needs should be reconsidered. Funding: None. Competing interests: None declared. Recours aux services dentaires chez les enfants au Liban : obstacles et association avec la perception parentale des soins de santé bucco-dentaire Résumé Contexte : La santé bucco-dentaire est importante pour la santé générale, mais le recours aux services dentaires varie considérablement, en particulier chez les enfants. Objectifs : La présente étude visait à déterminer les facteurs associés au recours par les parents aux services dentaires nécessaires pour leurs enfants au Liban, et leur perception de ces soins par rapport aux soins médicaux. Méthodes : Un échantillon de commodité d’écoles publiques et privées de Beyrouth a été sélectionné entre janvier et mai 2013. Les parents d’enfants scolarisés en école primaire (âgés de 7 à 12 ans) ont été invités à remplir un questionnaire dans lequel ils devaient indiquer les caractéristiques socio-économiques et le recours aux services dentaires. L’analyse de régression logistique a été utilisée pour évaluer la relation entre le recours aux services dentaires et les caractéristiques socio-économiques des parents ainsi que la sensibilisation à ces services et la perception de ceux-ci. Résultats : Les parents de 316 enfants ont renvoyé le questionnaire. La plupart des enfants (72,8 %) avaient été emmenés chez le dentiste au cours de l’année précédente, principalement pour des soins d’urgence. La plupart des parents (78,2 %) considéraient que les soins dentaires étaient aussi importants ou plus importants que les soins médicaux, et 89,8 % étaient disposés à contribuer à l’assurance dentaire. Le recours aux services dentaires était associé de maniere significative à : un âge plus avancé du parent (odds ratio, OR = 1,04 ; intervalle de confiance à 95 %, IC : 1,02‐1,06) ; la connaissance de centres de soins dentaires offrant un traitement abordable (OR = 3,18 ; IC à 95 % : 1,52‐6,68) ; et à la scolarisation des enfants dans des écoles privées (OR = 2,00, IC à 95 % : 1,08‐3,95). Il y avait une corrélation négative dans les familles de plus de quatre enfants comparativement aux foyers à enfant unique (OR = 0,18 ; IC à 95 % : 0,04‐0,81). Conclusion : Les obstacles aux soins dentaires pour les enfants étaient principalement liés à des facteurs économiques. لماوعلاو مفلل ةيحصلا ةياعرلا ةيهمأ كاردإ نود لوتح يتلا قئاوعلا :نانبل في نانسلأا بط تامدلخ لافطلأا مادختسا كلذب ةطبترلما يرفغ فزوج ،افاج نارم ،مرك ديرجنإ ةصلالخا .لافطلأل ةبسنلاب ماّيس لا ،ًايربك ًاتوافت توافتي نانسلأا بط تامدخ مادختسا نكلو ،ةماعلا ةحصلل ةيهمأب مفلا ةحص ىظتح :ةيفللخا ةياعر لىإ متهرظنو ،نانبل في ملهافطلأ ةمزلالا نانسلأا بط تامدلخ ءابلآا مادختساب ةطبترلما لماوعلا ديدتح لىإ ةساردلا هذه تفده :فادهلأا .ةيبطلا ةياعرلل ةبسنلاب نانسلأا ءابلآ ةوعد ته ِّجُوو .2013 رايأ/ويامو نياثلا نوناك/رياني ينب ةترفلا في تويرب في ةصالخاو ةماعلا سرادلما نم ةيئاوشع ةنيع تيرتخا :ثحبلا قرط ،ةيداصتقلاا ةيعماتجلاا صئاصلخا لوانتي م َّظنُم نايبتسا لماكتسلا )ةنس 12-7 ينب مهرماعأ حواترت نيذلا( 6-2 فوفصلا في لافطلأا تاهمأو ةيعماتجلاا صئاصلخاو ،نانسلأا بط تامدخ مادختسا ينب ةقلاعلا مييقتل يتسجوللا رادحنلاا ليلتح مِدخُتساو .نانسلأا بط تامدخ مادختساو .انهأشب تاروصتلاو نانسلأا بط تامدخب يعولاو ،ةيداصتقلاا يقلتل ماّيس لا ،ضيالما ماعلا في نانسلأا بيبط لىإ )%72.7( لافطلأا مظعم بهذ دقو .هلماكتسا دعب نايبتسلاا ًلافط 316 ءابآ م َّلس :جئاتنلا مهدادعتسا نع مهنم %89.8 برعأو ،رثكأ وأ اهسفن ةيبطلا ةياعرلا ةيهمأب ىظتح نانسلأا ةياعر نأ )%78.2( ءابلآا مظعم برتعاو .ةئراطلا ةياعرلا :ليي ماب ًاظوحلم ًايبايجإ ًاطابترا نانسلأا بط تامدخ مادختسا طبترا ،تايرغتلما ددعتلما ليلحتلا فيو .نانسلأا لىع ينمأتلا تامدخ في ةكراشملل روسيم ًاجلاع م ِّدقت يتلا نانسلأا ةياعر زكارم ةفرعمو ،)1.06–1.02 :%95 ةقث لصافب 1.04 :ةيحجرلأا ةبسن( نيدلاولا دحأ رمُع م ُّدقت لصافب ،2.00 = ةيحجرلأا ةبسن( ةصالخا سرادلماب ينقحتللما لافطلأاو ،)6.68–1.52 :%95 ةقث لصافب 3.18 = ةيحجرلأا ةبسن( ةفلكتلا 1424 EMHJ – Vol. 26 No. 11 – 2020Short research communication References 1. Inadequate or excess fluoride: a major public health concern. Geneva: World Health Organization; 2010 (https://apps.who.int/ iris/bitstream/handle/10665/329484/WHO-CED-PHE-EPE-19.4.5-eng.pdf?ua=1, accessed 8 May 2020). 2. Guideline: sugars intake for adults and children. Geneva: World Health Organization; 2015. (https://apps.who.int/iris/bitstream/ handle/10665/149782/9789241549028_eng.pdf?sequence=1&isAllowed=y, accessed 8 May 2020). 3. Petersen PE, Ogawa H. Prevention of dental caries through the use of fluoride – the WHO approach. Community Dent Health. 2016;33(2):66–8. https://doi.org/10.1922/CDH_Petersen03 4. Kelly SE, Binkley CJ, Neace WP, Gale BS. Barriers to care-seeking for children’s oral health among low-income caregivers. Am J Public Health. 2005;95(8):1345–51. https://doi.org/10.2105/AJPH.2004.045286 5. Nunn JH. The burden of oral ill health for children. Arch Dis Child. 2006;91(3):251–3. https://doi.org/10.1136/adc.2005.077016 6. Shi L, Stevens GD. Disparities in access to care and satisfaction among US children: the roles of race/ethnicity and poverty sta- tus. Public Health Rep. 2005;120(4):433–8. https://doi.org/10.1177/003335490512000410 7. Wamala S, Merlo J, Bostrom G. Inequity in access to dental care services explains current socioeconomic disparities in oral health: the Swedish national surveys of public health 2004–2005. J Epidemiol Community Health. 2006;60(12):1027–33. https:// doi.org/10.1136/jech.2006.046896 8. Bell JF, Huebner CE, Reed SC. Oral health need and access to dental services: evidence from the National Survey of Children’s Health, 2007. Matern Child Health J. 2012;16(Suppl 1):S27–34. https://doi.org/10.1007/s10995-012-0992-0 9. Cohen LA, Bonito AJ, Eicheldinger C, Manski RJ, Macek MD, Edwards RR, et al. Behavioral and socioeconomic correlates of dental problem experience and patterns of health care-seeking. J Am Dent Assoc. 2011;142(2):137–49. https://doi.org/10.14219/jada. archive.2011.0056 10. Roberts-Thomson KF, Slade GD. Factors associated with infrequent dental attendance in the Australian population. Aust Dent J. 2008; 53:358–62. https://doi.org/10.1111/j.1834-7819.2008.00079.x 11. Ammar W. Health system and reform in Lebanon. J Med Liban. 2001;5:72–4. 12. Ammar W, Awar M. What does the World Health Report 2000 bring to Lebanon? J Med Liban, 2001;49(3):123–5. 13. Petersen PE. Oral health. In: Heggenhougen K, Quah S, editors. International encyclopedia of public health. Cambridge (MA): Academic Press; 2008. 14. Richmond S, Chestnutt I, Shennan J, Brown R. The relationship of medical and dental factors to perceived general and dental health. Community Dent Oral Epidemiol. 2007;35(2):89–97. https://doi.org/10.1111/j.1600-0528.2007.00296.x 15. Yaacoub N, Badre L. Education in Lebanon. Beirut: Central Administration of Statistics; 2012 (http://www.cas.gov.lb/images/ PDFs/SIF/CAS_Education_In_Lebanon_SIF3.pdf, accessed 19 May 2020). 16. Petersen PE. World Health Organization global policy for improvement of oral health- World Health Assembly 2007. Int Dent J. 2008;58(3):115–21. https://doi.org/10.1111/j.1875-595x.2008.tb00185.x 17. Liu J, Probst JC, Martin AB, Wang JY, Salinas CF. Disparities in dental insurance coverage and dental care among us children: the national survey of children’s health. Pediatr. 2007;119(Suppl 1):S12–21. https://doi.org/10.1542/peds.2006-2089D 18. Okada LM, Wan TT. Factors associated with increased dental care utilization in five urban, low-income areas. Am J Public Health. 1979;69(10):1001–9. https://doi.org/10.2105/ajph.69.10.1001 19. Ghafari JG. “Medically necessary” orthodontic care: challenges and applications. Semin Orthod. 2016; 22(3):167–76. https://doi. org/10.1053/j.sodo.2016.05.001 ةبسن( دحاو لفطب ًةنراقم ةسرلأا في لافطأ 4 نم رثكأ دوجوب ًايبلس ًاطابترا نانسلأا بط تامدخ مادختسا طبتراو .)3.95–1.08 :%95 ةقث .)0.81–0.04 :%95 ةقث لصافب 0.18 = ةيحجرلأا .ةيداصتقا لماوعب بلاغلا في لافطلأل نانسلأا ةياعر ضترعت يتلا قئاوعلا طبترت :تاجاتنتسلاا 1425 Report EMHJ – Vol. 26 No. 11 – 2020 Profil épidémiologique d’une intoxication au méthanol, El Hajeb (Maroc) Sanah Essayagh,1 Mariama Bahalou,2* Meriem Essayagh3 et Touria Essayagh4* 1Laboratoire Agroalimentaire et Santé, Faculté des Sciences et Techniques, Université Hassan 1er, Settat (Maroc). 2Délégation de la Santé, Meknès (Maroc). 3Faculté de Médecine et de Pharmacie, Université Mohammed V, Rabat (Maroc). 4Laboratoire Sciences et Technologies de la Santé, Institut Supérieur des Sciences de la Santé, Université Hassan 1er, Settat (Maroc) (Correspondance à adresser à : essayagh.toula@gmail.com (TE). *Ces auteurs ont apporté la même contribution au travail. Résumé Contexte : l’intoxication au méthanol est un problème pertinent dans les pays en développement. Nous signalons une intoxication au méthanol qui s’est produite le 22 mai 2017 à El Hajeb (Maroc). Objectifs : décrire l’ampleur de l’intoxication, déterminer sa source et instaurer les mesures préventives nécessaires. Méthodes : nous avons mené une enquête transversale. Un questionnaire normalisé comprenant des données socio-économiques, les symptômes cliniques et l’heure de consommation a été administré en face à face aux cas. Des prélèvements biologiques ont été effectués pour analyse toxicologique et physico-chimique. Les données ont été saisies et analysées sur Epi Info version 7. Résultats : au total, 26 cas ont été colligés avec un âge moyen de 39,7 (écart type [ET] 11,1) ans et un sex ratio homme/ femme de 5,5. Tous les cas intoxiqués étaient de faible niveau socio-économique. La durée moyenne de latence entre consommation et apparition des symptômes était de 1,5 (ET 1) jours. Les symptômes signalés étaient les faibles troubles de la conscience chez 14 cas (53,8 %), les douleurs abdominales chez 10 cas (38,5 %), les céphalées chez neuf cas (34,6 %), les vomissements chez huit cas (30,8 %) et le coma chez sept cas (27,1 %). La létalité a été de 65 % et quatre cas ont développé une cécité. Les résultats de laboratoire ont confirmé la présence de méthanol dans le sang avec des valeurs supérieures à 0,6 g/L. La dose de méthanol dans la bouteille incriminée était de 217 g/L. Conclusion : la sensibilisation de la population au danger du méthanol est importante. Une sensibilisation des professionnels de santé aux signes cliniques et à la conduite à tenir face à une intoxication au méthanol est nécessaire. Citation: Essayagh S; Bahalou M; Essayagh M; Essayagh T. Profil épidémiologique d’une intoxication au méthanol, El Hajeb (Maroc). East Mediterr Health J. 2020;26(11):1425–1429. https://doi.org/10.26719/2020.26.11.1425 Mots-clés : méthanol, intoxication, El Hajeb Reçu : 01/10/19 ; accepté : 16/12/19 © Organisation mondiale de la Santé 2020 Certains droits réservés. La présente publication est disponible sous la licence Creative Commons Attribution – Pas d’utilisation commerciale – Partage dans les mêmes conditions 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons. org/licenses/by-nc-sa/3.0/igo ). Introduction L’alcool à brûler contient des concentrations variables de méthanol. Ce dernier est utilisé dans les antigels des automobiles et les solvants (1). Il est aussi utilisé comme substitut de l’alcool éthylique dans plusieurs boissons alcoolisées frelatées dans la classe de faible niveau socio- économique dans les pays en développement, surtout en cas de prohibition de l’alcool. Le méthanol peut être ingéré de manière accidentelle ou avec intention suicidaire, entraînant ainsi des intoxications sous forme isolée ou collective. À l’intérieur du corps, il est métabolisé en acide formique qui, sans traitement, entraîne la cécité ou la mort. Les études montrent des létalités allant de 76 % à 89 % (2,3). Le méthanol peut également affecter le tissu cérébral. L’évolution de l’intoxication au méthanol est conditionnée par le degré d’acidose et le délai entre l’exposition et l’instauration du traitement spécifique (1). Les signes cliniques de l’intoxication au méthanol sont non spécifiques, aboutissant à un diagnostic et un traitement tardifs, et à une surmortalité. Le traitement se base soit sur l’administration d’éthanol pour obtenir une concentration constante d’éthanol d’environ 1g/L (4,5), soit sur l’hémodialyse pour épuration (5,6). Dans le cadre d’une alerte, l’équipe de surveillance épidémiologique d’El Hajeb a informé le 22 mai 2017, à 10 h 30, l’équipe de Meknès du transfert aux urgences du Centre hospitalier préfectoral (CHP) de Meknès de quatre cas groupés présentant des troubles visuels, des douleurs abdominales et des vomissements. Une enquête épidémiologique pour vérification des cas a été effectuée afin de confirmer l’existence d’une intoxication collective, de décrire son ampleur, de déterminer sa source et d’instaurer les mesures préventives. Méthodes Site de l’enquête Pays d'Afrique du nord, le Maroc a une population de 33,8 millions d’habitants. Il se compose de 12 régions, dont celle de Fès-Meknès. Cette région est composée de neuf provinces, dont El Hajeb, qui regroupe 16 communes et compte une population d’environ 35 282 habitants (7). Elle dispose d’un hôpital provincial qui, en présence de cas sévères, transfère les patients à l’hôpital provincial de Meknès ou au centre hospitalier régional (CHU) de Fès. 1426 Report EMHJ – Vol. 26 No. 11 – 2020 Schéma d’étude et définition de cas Nous avons mené une enquête transversale à visée descriptive. L’enquête s’est déroulée durant la période allant 22 au 29 mai 2017. Elle a concerné Meknès- El Hajeb-Fès. Un cas a été défini comme toute personne habitant El Hajeb, ayant consommé une boisson alcoolisée et ayant présenté, entre le 17 et le 23 mai 2017, au moins l’un des signes suivants : troubles visuels, céphalées, vomissements, nausées, douleurs abdominales, altération de l’état général, troubles de la conscience ou coma. Les cas ont été recensés à partir du registre des urgences de l’hôpital d’El Hajeb. Une recherche active des cas a aussi été menée auprès de la population d’El Hajeb, en concertation avec les autorités locales. Un interrogatoire avec les malades survivants ou leurs proches a été réalisé. Le suivi des cas a été effectué auprès des hôpitaux d’El Hajeb et de Meknès, et des services de réanimation et d’ophtalmologie du CHU de Fès. Épidémiologie descriptive Un questionnaire a été administré en face à face pour recueillir des informations sur : 1) les données socio- démographiques ; 2) les signes cliniques ; 3) la date de consommation de méthanol et d’apparition des symptômes ; 4) le lieu de procuration de l’alcool ; 5) l’évolution du malade ; et 6) les complications. Le fournisseur agréé et les cas ont été interrogés sur la procédure de préparation du méthanol. La police judiciaire d’El Hajeb a procédé à la saisie du lot de méthanol pour stopper l’épidémie et assurer l’analyse toxicologique et physico-chimique. Les données ont été analysées sur Epi Info version 7. Enquête toxicologique et physico-chimique Des prélèvements biologiques ont été effectués et envoyés pour analyse au Centre Antipoison et de Pharmacovigilance. Le méthanol et l’éthanol ont été dosés par chromatographie en phase gazeuse (gas chromatography, GC) avec un détecteur à ionisation de flamme (flame ionisation detector, FID) couplée à un échantillonneur automatique d’extraction en espace de tête (headspace, HS) (GC-FID-HS). La recherche des drogues a été réalisée par des méthodes immuno- chromatographiques. Considérations éthiques Tous les participants avaient donné leur consentement informel verbal. Les tests de laboratoire ont été réalisés à la suite d’un diagnostic de routine. Résultats Épidémiologie descriptive Durant l’épidémie, 26 cas ont été identifiés. L’âge moyen était de 39,7 (ET 11,1) ans, avec des extrêmes allant de 28 à 58 ans et un sex ratio homme/femme de 5,5. Tous les patients étaient de faible niveau socio-économique, sans domicile fixe pour six cas (23,1 %) (Tableau 1). La durée moyenne de latence entre consommation et apparition des symptômes était de 1,5 (ET 1) jour. Les cas ont commencé à se produire le 20 mai pour atteindre le maximum le 21 mai avec 11 cas. Le dernier cas a été enregistré le 23 mai à 18 heures (Figure 1). Les symptômes les plus enregistrés étaient les suivants : faibles troubles de la conscience chez 14 cas (53,8 %), douleurs abdominales chez 10 cas (38,5 %), céphalées chez neuf cas (34,6 %), vomissements chez huit cas (30,8 %) et coma chez sept cas (27,1 %). Dix-sept personnes sont décédées, soit une létalité de 65 %, et quatre Tableau 1 Caractéristiques des cas d’intoxication au méthanol, El Hajeb (Maroc), mai 2017 Caractéristiques Total (n = 26) Nbre (%) Âge moyen (ans) 39,7 (ET 1,1) Sexe Masculin 22 (84,6) Féminin 4 (15,4) Situation matrimoniale Non marié 21 (80,8) Marié 5 (19,2) Revenu mensuel (US$) < 150 26 (100,0) ≥ 150 0 (0,0) Type de domicile Sans domicile fixe 6 (23,1) Avec domicile fixe 20 (76,9) Consommation d’alcool local Oui 20 (76,9) Non 6 (23,1) Toxicomanie Oui 18 (69,2) Non 8 (30,8) ET : écart type. Figure 1 Courbe épidémique de l’intoxication au méthanol, El Hajeb (Maroc), 22 mai 2017 13 12 11 10 9 8 7 6 5 4 3 2 1 20 21 22 23 Début des symptômes Arrivée du lot consommé Saisie du lot consommé N om br e de ca s 1427 Report EMHJ – Vol. 26 No. 11 – 2020 ont développé une cécité, soit 15,4 % (Tableau 2). Vingt cas ont déclaré avoir consommé de l’alcool frelaté (alcool préparé localement), qui provenait du même fournisseur agréé d’El Hajeb. Enquête toxicologique et physico-chimique Les résultats des trois prélèvements biologiques des patients ont révélé la présence de méthanol dans le sang avec des valeurs de 0,7 g/L, 1,35 g/L et 1,94 g/L. Le niveau d’éthanol dans le sang était inférieur à 0,1 g/L. Les prélèvements d’urine ont révélé la consommation de cannabis chez un malade (59 mg/mL). Les flacons d’alcool toxique saisis par la police ont montré des concentrations de méthanol de 217 g/L, 7 g/L et 6 g/L. Discussion L’intoxication au méthanol est un problème dans les pays en développement. Au Maroc, une intoxication au méthanol a eu lieu en 1996 avec 76 cas, dont sept décès et quatre cas de cécité (8-11). La majorité des cas d’intoxication au méthanol résultent d’une ingestion/ inhalation ou d’une absorption transcutanée (12). La dose létale est de 1 mL/kg (13). La dose toxique de méthanol dépend de l’individu et du traitement reçu. Une concentration de méthanol dans le sang supérieure à 500 mg/L est associée à des toxicités graves alors qu’une concentration supérieure à 1500-2000 mg/L entraîne la mort chez les patients non traités (14). Dans notre étude, les trois patients pour lesquels des prélèvements biologiques ont pu être effectués et chez qui les valeurs du méthanol dans le sang étaient de 0,7 g/L, 1,35 g/L et 1,94 g/L sont décédés. Ceci pourrait expliquer la létalité élevée dans notre étude. Dans l’enquête, tous les cas étaient de faible niveau socio-économique, ce qui est compatible avec la littérature (15). Dans notre étude, on a constaté une prédominance masculine. Ceci pourrait être expliqué par le contexte marocain où seule une minorité de femmes sont consommatrices de méthanol, ce qui est semblable à l’étude de Sanaei-Zadeh (16). Les signes cliniques et la consommation d’alcool à brûler suggèrent une intoxication au méthanol. Cela a été confirmé lors de l’analyse des lots d’alcool frelaté. La répartition des cas par moment d’apparition des symptômes suggère une source commune ponctuelle d’intoxication. En effet, lors de l’enquête, le lot de méthanol incriminé a été celui du 19 mai. Le cas index a été un consommateur du 20 mai. Ce dernier avait présenté une douleur abdominale vers 20 heures, pour laquelle il a été traité. Le 21 mai, à 2 h 30, il est retourné à l’hôpital pour aggravation de cas, où il décède. La létalité de 65 % dans notre série est proche des valeurs observées dans la littérature qui oscillent entre 76 % et 89 % (2,3). La littérature montre que, lors de l’ingestion de méthanol, l’intoxication passe par plusieurs phases : la phase de latence comprise entre 9-24 heures en moyenne et liée à la dose ingérée ; la phase de symptomatologie initiale de l’intoxication qui est peu spécifique et qui est marquée par des troubles neurologiques, des signes digestifs et une polyurie associée à une polydipsie ; et la phase d’état où le méthanol ingéré est oxydé au niveau hépatique en formaldéhyde et en acide formique par des enzymes. L’accumulation de l’acide formique est responsable de la toxicité et sa concentration est en lien avec la morbidité, la mortalité et la profondeur de l’acidose métabolique observée dans l’intoxication au méthanol (17). Cet intervalle libre de plusieurs heures entre l’ingestion du méthanol et l’apparition des signes cliniques explique le retard dans la prise en charge des cas et la létalité élevée. À cela pourrait s’ajouter le manque de connaissances de la part des professionnels de santé sur la conduite à tenir face à une intoxication au méthanol. Limites de l’étude Notre étude a connu certaines limites, à savoir, la non-réalisation des prélèvements biologiques pour l’ensemble des patients, d’où la non-collecte des données en relation avec la dose de méthanol dans le sang, et la non-collecte des données sur la quantité d’alcool ingérée. Conclusion L’enquête a révélé une intoxication collective au méthanol. Une sensibilisation du grand public au danger du méthanol est importante. Une sensibilisation des professionnels de santé aux signes cliniques et à la conduite à tenir face à une intoxication au méthanol s’avère nécessaire. Financement : aucun. Conflits d'intérêts : aucun déclaré. Tableau 2 Répartition des symptômes de l’intoxication au méthanol, El Hajeb (Maroc), mai 2017 Symptômes Fréquence (%) (n = 26) Faibles troubles de la conscience 14 (53,8) Douleurs abdominales 10 (38,5) Céphalées 9 (34,6) Vomissements 8 (30,8) Coma 7 (27,1) Polypnée 6 (23,1) Troubles visuels 5 (19,2) Cécité 4 (15,4) Irritabilité 4 (15,4) 1428 Report EMHJ – Vol. 26 No. 11 – 2020 Epidemiological profile of methanol poisoning, El Hajeb, Morocco Abstract Background: Methanol poisoning is of particular importance in low and middle-income countries. We reported on a methanol poisoning incident that occurred 22 May 2017, in El Hajeb (Morocco). Aims: This study aimed to describe the extent of the intoxication, determine its source and implement the necessary preventative measures. Methods: We conducted a cross-sectional survey. A standardized questionnaire including socio-economic data, clinical symptoms and time of use was administered face-to-face to cases of methanol poisoning. Biological samples were taken for toxicological analysis. Data were entered and analyzed on Epi Info version 7. Results: Twenty-six cases of methanol poisoning were surveyed with a mean age of 39.7 (SD 11.1) years and a male/female sex ratio of 5.5. All intoxicated cases were of low socioeconomic status. The mean latency period between use and symp- tom onset was 1.5 (SD 1) days. Reported symptoms were mildly altered consciousness in 14 cases (53.8%), abdominal pain in 10 cases (38.5%), headache in 9 cases (34.6%), vomiting in 8 cases (30.8%) and coma in 7 cases (27.1%). Mortality was 65% and 4 cases developed blindness. Laboratory results confirmed the presence of methanol in the blood with values greater than 0.6 g/L. The dose of methanol in the associated bottle was 217 g/L. Conclusion: Public awareness of the dangers of methanol intoxication is important. Health professionals need to be aware of the clinical signs and what to do in the event of methanol poisoning. Références 1. Methanol. Properties, production, uses, & poisoning. Encyclopaedia Britannica (s.d). (https://www.britannica.com/science/ methanol, consulté le 20 mars 2020). 2. Zobnine IV, Liubimov BM, Malyh AF,TretyakovAB, Teterina IP, Pazukov EA, et al. Intoxication collective par le méthanol à Irkoutsk en décembre 2016. Toxicol Anal Clin. 2017 May;29(2S):S77–S78. https://doi.org/10.1016/j.toxac.2017.03.118 3. Liu JJ, Daya MR, Carrasquillo O, Kales NS. Prognostic factors in patients with methanol poisoning. J Toxico Clin Toxicol. 1998;36(3):175–81. https://doi.org/10.3109/15563659809028937 برغلما ،بجالحا ةنيدم ،لوناثيلماب ممستلل يئابولا مسترلما غياصلا ةيروت ،غياصلا ميرم ،ولاحب ةميرم ،غياصلا انس ةصلالخا /ويام 22 خيراتب تثدح لوناثيلماب ممست ةلاح نع انغلبأ دقو .لخدلا ةطسوتلماو ةضفخنلما نادلبلا في ةصاخ ًةلكشم لوناثيلماب ممستلا لثمي :ةيفللخا .ةيبرغلما بجالحا ةنيدم في 2017 رايأ .ةمزلالا ةيئاقولا يربادتلا ذيفنتو ،هردصم ديدتحو ،ممستلا ةجرد فصو لىإ ةساردلا هذه تفده :فادهلأا ،ةيداصتقلااو ةيعماتجلاا تانايبلا لىع لمتشاو لوناثيلماب ممست تلااح عم هجول ًاهجو ًانايبتسا انيرجأ ماك .ًايعطقم ًاحسم انيرجأ :ثحبلا قرط Epi-Info-version (7) جمانرب انمدختساو .ابه ةيمسلا ليلحتل ةيجولويب تانيع تعُجو .لوحكلا لوانت تقوو ،ةيريسرلا ضارعلأاو .اهليلتحو تانايبلا لاخدلإ لىإ روكذلا ةبسن تغلبو .)11.1 يرايعلما فارحنلاا( ًاماع 39.7 تلاالحا رماعأ طسوتم غلبو ،لوناثيلماب ممست ةلاح 26 حسلما لمش :جئاتنلا ةيادبو لوحكلا يطاعت ينب نومكلا ةترف طسوتم غلبو .ضفخنم يداصتقاو يعماتجا عضو تاذ ممستلا تلااح عيج تناكو .5.5 ثانلإا نطبلا في ملاآو ،)%53.8( ةلاح 14 في يعولا في فيفخ يرغت :يه اهنع غلبُلما ضارعلأا رثكأ تناكو .)1 يرايعلما فارحنلاا( ًاموي 1.5 ضارعلأا ةتاملإا لدعم غلبو .)%27.1( تلااح 7 في ةبوبيغو ،)%30.8( تلااح 8 في ءيقو ،)%34.6( تلااح 9 في عادصو ،)%38.5( تلااح 10 في ةعرج تغلبو .ترل/مارج 0.6 نع ديزت تايمكب مدلا في لوناثيم دوجو ةيبرتخلما جئاتنلا تدكأو .ىمعلاب تلااح 4 تبيصأ ماك ،تلاالحا نم %65 .ترل/مارج 217 رضرلا في تببست يتلا ةروراقلا في لوناثيلما بجاولا تاءارجلإابو ،ةيريسرلا تاملاعلاب ينيحصلا ينينهلما ةيعوت بيجو .لوناثيلماب ممستلا راطخأب ماعلا يعولا ةدايز مهلما نم :تاجاتنتسلاا .لوناثيلماب ممستلا ةلاح في اهذاتخا 1429 Report EMHJ – Vol. 26 No. 11 – 2020 4. Jacobsen D, Jansen H, Wiik‐Larsen E, Bredesen JE, Halvorsen S. Studies on methanol poisoning, Acta Med Scand. 1982;212(1- 2):5–10. https://doi.org/10.1111/j.0954-6820.1982.tb03160.x 5. Palatnick W, Redman LW, Sitar DS, Tenenbein M. Methanol half-life during ethanol administration: implications for management of methanol poisoning. Ann Emerg Med. 1995 Aug;26(2):202–7. https://doi.org/10.1016/S0196-0644(95)70152-4 6. McCoy HG, Cipolle RJ, Ehlers SM, Sawchuk RJ, Zaske DE. Severe methanol poisoning: application of a pharmacokinetic model for ethanol therapy and hemodialysis. Am J Med. 1979 Nov;67(5):804–7. https://doi.org/10.1016/0002-9343(79)90738-1 7. Recensement général de la population et de l’habitat 2014 : Population légale du Maroc. Rabat: Haut-Commissariat au Plan; 2018 (https://www.hcp.ma/downloads/RGPH-2014_t17441.html, consulté le 23 mars 2020). 8. Rhalem N, Jalal G, Soulaymani R. Intoxication par le méthanol. 1996 (https://docplayer.fr/27698320-Intoxication-par-le- methanol-n-rhalem-gh-jalal-r-soulaymani.html). 9. Sutton TL, Foster RL, Liner SR. Acute methanol ingestion. Pediatr Emerg Care. 2002 Oct;18(5):360–3. https://doi. org/10.1097/00006565-200210000-00009 10. Jacobsen D, McMartin KE. Methanol and ethylene glycol poisonings. Mechanism of toxicity,clinical course, diagnosis and treatment. Med Toxicol. 1986 Sep-Oct;1(5):309–34. https://doi.org/10.1007/bf03259846 11. Treichel JL, Henry MM, Skumatz CM, Eells JT, Burke JM. Formate, the toxic metabolite of methanol, in cultured ocular cells. Neurotoxicology. 2003 Dec;24(6):825–34. https://doi.org/10.1016/S0161-813X(03)00059-7 12. Kurtas O, Imre KY, Ozer E, Can M, Birincioglu I, Butun C, et al. The evaluation of deaths due to methyl alcohol intoxication. Biomed Res. 2017;28(8):3680–7. 13. Théfenne H, Turc J, Carmoi T, Gardet V, Renard C. Intoxication aiguë au méthanol : réflexion à partir d’un cas. Ann Biol Clin. 2005;63(5):556–60. (https://docplayer.fr/57683568-Abc-intoxication-aigue-au-methanol-reflexion-a-partir-d-un-cas-pratique- quotidienne.html). 14. United Nations Environment Programme (UNEP)/International Labour Organization (ILO)/World Health Organization (WHO). International Programme on Chemical Safety (IPCS). Methanol. Geneva: World Health Organization; 1997 (Environmental Health Criteria Series, No 196). 15. Paasma R, Hovda KE, Tikkerberi A, Jacobsen D. Methanol mass poisoning in Estonia: outbreak in 154 patients. Clin Toxicol(Phila). 2007;45(2):152–7. https://doi.org/10.1080/15563650600956329. 16. H. Sanaei-Zadeh H, S.K. Esfeh SK, N. Zamani N, F. Jamshidi F, S. Shadnia S. Hyperglycemia is a strong prognostic factor of lethality in methanol poisoning, J Med Toxicol. 2011 Sep;7(3):189–4. https://doi.org/10.1007/s13181-011-0142-x. 17. Francis ST, Nair JR, Shiji PV, Mohamed S, Geetha P, Sasidharan PK. A case series of acute methanol poisoning from Northern Kerala. Emergency Med. 2016;6(2): 312. https://doi.org/10.4172/2165-7548.1000312. WHO events addressing public health priorities 1430 EMHJ – Vol. 26 No. 11 – 2020 Enteric and diarrhoeal diseases surveillance, prevention and control in the Eastern Mediterranean Region1 Citation: Enteric and diarrhoeal diseases surveillance, prevention and control in the Eastern Mediterranean Region. East Mediterr Health J. 2019;26(11):1430–1431. https://doi.org/10.26719/2020.26.11.1430 Copyright © World Health Organization (WHO) 2020. Open Access. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO license (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). 1 This summary is extracted from the Summary report on the Meeting on enteric and diarrhoeal diseases surveillance, prevention and control with a focus on cholera, typhoid and rotavirus in the Eastern Mediterranean Region, Cairo, Egypt, 2–5 March 2020 (https://applications.emro.who.int/docs/ WHOEMEPI359E-eng.pdf?ua=1). Introduction Estimates suggest that the incidence of diarrhoeal dis- eases in 2015 in the WHO Eastern Mediterranean Region exceeded 300 million episodes, with children under five years of age accounting for 53% of cases (1). Over 103 692 deaths from diarrhoea occurred in the Region in 2015, with a mortality rate of 16.0 per 100 000 (1). The majori- ty of these deaths (63.3%) occurred in children aged un- der 5 years (65 670 deaths); WHO estimates that in 2017 there were 11–21 million cases of typhoid fever, leading to 128 000–161 000 deaths (1). Adopting parallel approaches to managing the separate agents of diarrhoeal diseases does not constitute an efficient approach to prevention and control. Within the WHO Regional Office for the Eastern Mediterranean, the Vaccine Preventable Diseases/Polio Transition unit of the Department of Universal Health Coverage/ Communicable Diseases and the Infectious Hazard Management unit of the WHO Health Emergencies programme have initiated an integrated approach to addressing diarrhoeal diseases under one programme that encompasses all interventions. As part of this initiative, the WHO Regional Office held a meeting in Cairo, Egypt, from 2 to 5 March 2020 on enteric and diarrhoeal diseases surveillance, prevention and control with a focus on cholera, typhoid and rotavirus in the Eastern Mediterranean Region (2). The objectives of the meeting were to: · understand the epidemiology and overall burden of diarrhoeal diseases, including rotavirus, cholera and typhoid fever, in the Eastern Mediterranean Region; · share lessons learned from global paediatric diarrhoe- al surveillance and rotavirus surveillance progress in relation to rotavirus vaccine introduction; · share global updates on the rotavirus vaccine availa- ble and in the pipeline; · discuss the current cholera and typhoid situation in the Region and challenges to managing current out- breaks, including cross-border coordination mecha- nisms; · share experiences from countries on the use of oral cholera vaccine, including in prevention and control of cholera in the Region; · share experiences from countries on the use of ty- phoid conjugate vaccine, including for prevention and control of extensively drug-resistant salmonella typhoid (XDR S. Typhi); and · develop an overall approach for the prevention and control of diarrhoeal diseases in the Region. Summary of discussions Discussions demonstrated the interest of countries in improving their surveillance systems for diarrhoeal diseases, based on other countries’ experience. Improv- ing rotavirus vaccine coverage was seen as crucial for decreasing diarrhoeal disease morbidity and mortality among children under 5 years of age. It was also agreed that surveillance, outbreak preparedness and response, coupled with WASH preventive measures, need to be combined to manage and control cholera outbreaks in the Region. Participants proposed recommendations based on the knowledge and experiences shared by countries and the support offered by partners. The recommenda- tions point to the need for robust integrated approaches to all aspects of surveillance, prevention and control of diarrhoeal diseases in the Region. The recommendations addressed several areas including surveillance, outbreak preparedness and response, case management, laborato- ry enhancement, and building on existing systems and mechanisms to optimize health benefits and make use of existing resources. Recommendations To WHO · Surveying countries in the Region to assess diarrhoe- al disease surveillance capacity and data needs, in or- der to guide WHO and partners on the establishment of a comprehensive diarrhoeal diseases surveillance system; · establishing rotavirus surveillance in countries where it does not exist; · establishing a rotavirus regional reference laboratory; · enhancing capacities in the Region to track XDR-ty- phoid in coordination with work on antimicrobial resistance in S. Typhi; WHO events addressing public health priorities 1431 EMHJ – Vol. 26 No. 11 – 2020 · improving understanding of the burden of typhoid fever in countries of the Region and increase aware- ness and advocacy for the prevention and control of typhoid fever (and associated antimicrobial resist- ance) as a public health priority in the Region; · developing or updating national cholera control plans aligned to the Ending Cholera global 2030 roadmap; and · coordinating with relevant institutions to conduct modelling and forecasting studies for cholera and other waterborne diseases. To Member States · Strengthening capacity to follow diagnostic proce- dures and case management of typhoid and cholera as per WHO recommendations; · adopting a multisectoral approach, including collab- oration between different ministries, in prevention and control of cholera outbreaks; · aligning the national cholera control plan to the End- ing Cholera global roadmap; and · enhancing community awareness and engagement, as an integral component of prevention and control of diarrhoeal diseases. References 1. World Health Organization. Immunization, vaccines and biologicals – typhoid. Geneva: World Health Organization; 2019 (https://www.who.int/immunization/diseases/typhoid/en/). 2. World Health Organization Regional Office for the Eastern Mediterranean (WHO/EMRO). Meeting on enteric and diarrhoe- al diseases surveillance, prevention and control with a focus on cholera, typhoid and rotavirus in the Eastern Mediterranean Region, Cairo, Egypt, 2–5 March 2020. Cairo: WHO/EMRO; 2020 (https://applications.emro.who.int/docs/WHOEMEPI359E-eng. pdf?ua=1). Eastern Mediterranean Health Journal IS the official health journal published by the Eastern Mediterranean Regional Office of the World Health Organization. It is a forum for the presentation and promotion of new policies and initiatives in public health and health services; and for the exchange of ideas, concepts, epidemiological data, research findings and other information, with special reference to the Eastern Mediterranean Region. It addresses all members of the health profession, medical and other health educational institutes, interested NGOs, WHO Collaborating Centres and individuals within and outside the Region. طسوتلما قشرل ةيحصلا ةلجلما ةماعلا ةحصلا في ةديدلجا تاردابلماو تاسايسلا ميدقتل برنم ىهو .ةيلماعلا ةحصلا ةمظنمب طسوتلما قشرل ىميلقلإا بتكلما نع ردصت ىتلا ةيمسرلا ةلجلما ىه قشر ميلقإب اهنم قلعتي ام ةصاخو ،تامولعلما نم كلذ يرغو ثاحبلأا جئاتنو ةيئابولا تايطعلماو ميهافلماو ءارلآا لدابتلو ،اله جيوترلاو ةيحصلا تامدلخاو ةمظنم عم ةنواعتلما زكارلماو ،ةينعلما ةيموكلحا يرغ تماظنلما اذكو ،ةيميلعتلا دهاعلما رئاسو ةيبطلا تايلكلاو ،ةيحصلا نهلما ءاضعأ لك لىإ ةهجوم ىهو .طسوتلما .هجراخو ميلقلإا فى ةحصلاب ينمتهلما دارفلأاو ةيلماعلا ةحصلا La Revue de Santé de la Méditerranée Orientale EST une revue de santé officielle publiée par le Bureau régional de l’Organisation mondiale de la Santé pour la Méditerranée orientale. Elle offre une tribune pour la présentation et la promotion de nouvelles politiques et initiatives dans le domaine de la santé publique et des services de santé ainsi qu’à l’échange d’idées, de concepts, de données épidémiologiques, de résultats de recherches et d’autres informa- tions, se rapportant plus particulièrement à la Région de la Méditerranée orientale. Elle s’adresse à tous les professionnels de la santé, aux membres des instituts médicaux et autres instituts de formation médico-sanitaire, aux ONG, Centres collaborateurs de l’OMS et personnes concernés au sein et hors de la Région. Correspondence Editor-in-chief Eastern Mediterranean Health Journal WHO Regional Office for the Eastern Mediterranean P.O. Box 7608 Nasr City, Cairo 11371 Egypt Tel: (+202) 2276 5000 Fax: (+202) 2670 2492/(+202) 2670 2494 Email: emrgoemhj@who.int Members of the WHO Regional Committee for the Eastern Mediterranean Afghanistan . Bahrain . Djibouti . Egypt . Islamic Republic of Iran . Iraq . Jordan . Kuwait . Lebanon Libya . Morocco . Oman . Pakistan . Palestine . Qatar . Saudi Arabia . Somalia . Sudan . Syrian Arab Republic Tunisia . United Arab Emirates . Yemen طسوتلما قشرل ةيلماعلا ةحصلا ةمظنلم ةيميلقلإا ةنجللا ءاضعأ نادلبلا ةيملاسلإا ناريإ ةيروهجم . ايبيل . سنوت . نيرحبلا . ناتسكاب . ةدحتلما ةيبرعلا تاراملإا . ناتسناغفأ . ندرلأا برغلما . صرم . نانبل . تيوكلا . رطق . ينطسلف . نماُع . قارعلا . لاموصلا . نادوسلا . تيوبيج . ةيروسلا ةيبرعلا ةيروهملجا نميلا . ةيدوعسلا ةيبرعلا ةكلملما Membres du Comité régional de l’OMS pour la Méditerranée orientale Afghanistan . Arabie saoudite . Bahreïn . Djibouti . Égypte . Émirats arabes unis . République islamique d’Iran Iraq . Libye . Jordanie . Koweït . Liban . Maroc . Oman . Pakistan . Palestine . Qatar . République arabe syrienne Somalie . Soudan . Tunisie . Yémen Subscriptions and Permissions Publications of the World Health Organization can be obtained from Knowledge Sharing and Production, World Health Organization, Regional Office for the Eastern Mediterranean, PO Box 7608, Nasr City, Cairo 11371, Egypt (tel: +202 2670 2535, fax: +202 2670 2492; email: emrgoksp@who.int). Requests for permission to reproduce, in part or in whole, or to translate publications of WHO Regional Office for the Eastern Mediterranean – whether for sale or for noncommercial distribution – should be addressed to WHO Regional Office for the Eastern Mediterranean, at the above address; email: emrgoegp@who.int. Cover photo: © WHO / Budi Chandra EMHJ is a trilingual, peer reviewed, open access journal and the full contents are freely available at its website: http://www/emro.who.int/emhj.htm EMHJ information for authors is available at its website: http://www.emro.who.int/emh-journal/authors/ EMHJ is abstracted/indexed in the Index Medicus and MEDLINE (Medical Literature Analysis and Retrieval Systems on Line), ISI Web of knowledge, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), Embase, Lexis Nexis, Scopus and the Index Medicus for the WHO Eastern Mediterranean Region (IMEMR). © World Health Organization (WHO) 2020. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO licence (https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Disclaimer The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. The authors alone are responsible for the views expressed in this publication and they do not necessarily represent the views, decisions or policies of the institutions with which they are affiliated. If authors are staff members of the World Health Organization, the authors alone are responsible for the views expressed in this publication and do not necessarily represent the decisions, policy or views of the World Health Organization. ISSN 1020-3397 Cover 26-7.indd 4,6 23/07/2020 01:02