Eastern Mediterranean Health Journal [2012; Vol.18, Issue 9]

Contents V olum e 18 N um ber 9 Septem ber 2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale Volume 18 / No. 9 September / Septembre 2012 9 ددع / شرع نماثلا دلجلما برمتبس / لوليأ Women across the generations, Morocco Globally 17.3 million people a year die of cardiovascular disease, almost 50% of whom are women. Children may also at increased risk of cardiovascular disease later in life, since in childhood they may be unable to control where and how to live a healthy heart life. Therefore, the focus of World Heart Day 2012 on 28 September is cardiovascular disease prevention among women and children. Letter from the Editor .............................................................................................................................................................. 901 Research articles In-hospital complications and 1-year outcome of acute coronary syndrome in patients with hypertension: findings from the 2nd Gulf Registry of Acute Cardiac Events ................................................................................. 902 Mapping family planning policy and programme best practices in the WHO Eastern Mediterranean Region: a step towards coordinated scale-up ..........................................................................................................911 HIV infection and related risk behaviours among female sex workers in greater Cairo, Egypt ............................ 920 Tehran dentists’ self-reported knowledge and attitudes towards HIV/AIDS and observed willingness to treat simulated HIV-positive patients .............................................................................................. 928 Knowledge, perceptions and practices towards medical ethics among physician residents of University of Alexandria hospitals, Egypt .................................................................................................................935 Study of adverse events of A/H1N1 vaccine among health care staff in selected provinces of Afghanistan, 2010 ................................................................................................................................................. 946 Estimation of malaria transmission intensity in Sennar state, central Sudan .......................................................... 951 Drug resistance pattern and outcome of treatment in recurrent episodes of tuberculosis ..................................957 Updating and validation of the socioeconomic status scale for health research in Egypt ................................... 962 Natural killer cell populations in Egyptians infected with hepatitis C virus ........................................................... 969 Lead exposure among the general population of Duhok governorate, Kurdistan region, Iraq ............................974 Consumption of iodized salt among households of Basra city, south Iraq ...........................................................980 Oral and pharyngeal cancers in Yemen: a retrospective study .............................................................................. 985 Cover 18-9.indd 1 9/6/2012 10:14:47 AM Subscriptions and Distribution Enquiries regarding subscriptions and distribution of the print edition of EMHJ should be addressed to: Printing and Marketing of Publications at: email: pam@emro.who.int; tel: (+202) 2276 5000; fax: (+202) 2670 2492 or 2670 2494 Permissions Requests for permission to reproduce or translate articles, whether for sale or non-commercial distribution should be addressed to EMHJ at: emhj@emro.who.int Correspondence Editor-in-chief EMHJ 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: khayat@emro.who.int/emhj@emro.who.int 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 health services; and for the exchange of ideas, con‑ cepts, 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 Col‑ laborating Centres and individuals within and outside the Region. 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 des ser‑vices de santé ainsi qu’à l’échange d’idées, de concepts, de données épidémiologiques, de résultats de recherches et d’autres informations, 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 collabora‑ teurs de l’OMS et personnes concernés au sein et hors de la Région. 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 is abstracted/indexed in the Index Medicus and MEDLINE (Medical Literature Analysis and Retrieval Systems on Line) and the ExtraMed‑Full text on CD‑ROM, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), CAB International, Lexis Nexis, Scopus and the Index Medicus for the WHO Eastern Mediterranean Region (IMEMR). ©World Health Organization 2012 All rights reserved 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 the World Health Organization 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 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 the World Health Organization in preference to others of a similar nature that are not mentioned. All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either express or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use. The named authors alone are responsible for the views expressed in this publication. ISSN 1020‑3397 Cover designed by Diana Tawadros Internal layout designed by Emad Marji and Diana Tawadros Printed by WHO Regional Office for the Eastern Mediterranean ميدقتل برنم ىهو .ةيلماعلا ةحصلا ةمظنمب طسوتلما قشرل ىميلقلإا بتكلما نع ردصت ىتلا ةيمسرلا ةلجلما ىه ةيئابولا تايطعلماو ميهافلماو ءارلآا لدابتلو ،اله جيوترلاو ةيحصلا تامدلخا فى ةديدلجا تاردابلماو تاسايسلا لك لىإ ةهجوم ىهو .طسوتلما قشر ميلقإب اهنم قلعتي ام ةصاخو ،تامولعلما نم كلذ يرغو ثاحبلأا جئاتنو زكارلماو ،ةينعلما ةيموكلحا يرغ تماظنلما اذكو ،ةيميلعتلا دهاعلما رئاسو ةيبطلا تايلكلاو ،ةيحصلا نهلما ءاضعأ .هجراخو ميلقلإا فى ةحصلاب ينمتهلما دارفلأاو ةيلماعلا ةحصلا ةمظنم عم ةنواعتلما طسوتلما قشرل ةيحصلا ةلجلما طسوتلما قشرل ةيلماعلا ةحصلا ةمظنلم ةيميلقلإا ةنجللا ءاضعأ نادلبلا ةيملاسلإا ناريإ ةيروهجم . ايبيل . سنوت . نيرحبلا . ناتسكاب . ةدحتلما ةيبرعلا تاراملإا . ناتسناغفأ . ندرلأا صرم . نانبل . تيوكلا . رطق . ينطسلف . نماُع . قارعلا . لاموصلا . نادوسلا . تيوبيج . نادوسلا بونج . نميلا . ةيروسلا ةيبرعلا ةيروهملجا ةيدوعسلا ةيبرعلا ةكلملما . برغلما 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 . South Sudan Sudan . Syrian Arab Republic . Tunisia . United Arab Emirates . Yemen 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 . Soudan du Sud . Tunisie . Yémen Cover 18-9.indd 2 9/6/2012 10:14:47 AM Contents La Revue de Santé de la Méditerranée orientale Eastern Mediterranean Health Journal Vol. 18 No. 9 9 ددع شرع نماثلا دلجلما•  2012  • Letter from the Editor ..............................................................................................................................................................................................................................................................................................................................901 Research articles In-hospital complications and 1-year outcome of acute coronary syndrome in patients with hypertension: findings from the 2nd Gulf Registry of Acute Cardiac Events W.M. Ali, K.F. Al Habib, A. Hersi, N. Asaad, K. Sulaiman, A. Al-Shiek-Ali and J. Al Suwaidi ...............................................................................................................................................................902 Mapping family planning policy and programme best practices in the WHO Eastern Mediterranean Region: a step towards coordinated scale-up P. Chikvaidze, H.H. Madi and R .K. Mahaini .........................................................................................................................................................................................................................................................................911 HIV infection and related risk behaviours among female sex workers in greater Cairo, Egypt I.A. Kabbash, I. Abdul-Rahman, Y.A. Shehata and A.A-R. Omar ...........................................................................................................................................................................................................................920 Tehran dentists’ self-reported knowledge and attitudes towards HIV/AIDS and observed willingness to treat simulated HIV-positive patients B. Khosravanifard, V. Rakhshan, M. Ghasemi, A. Pakdel, S. Baradaran-Eghbal, R . Sheikholeslami, T. Dadolahi-Sarab and H. Rakhshan ...........................................................928 Knowledge, perceptions and practices towards medical ethics among physician residents of University of Alexandria hospitals, Egypt A.M. Mohamed, M.A. Ghanem and A.A. Kassem .............................................................................................................................................................................................................................................................935 Study of adverse events of A/H1N1 vaccine among health care staff in selected provinces of Afghanistan, 2010 J.A. Mofleh, Z. Akbarian, N. Muserat, H. Yosofi, A. Alkozai and B. Noormal ...................................................................................................................................................................................................946 Estimation of malaria transmission intensity in Sennar state, central Sudan Z.A. Elmahdi, A.A. Nugud and I.M. Elhassan.......................................................................................................................................................................................................................................................................951 Drug resistance pattern and outcome of treatment in recurrent episodes of tuberculosis M. Marjani, P. Tabarsi, P. Baghaei, M. Shamaei, D. Mansouri, M.R. Masjedi and A.A. Velayati .........................................................................................................................................................957 Updating and validation of the socioeconomic status scale for health research in Egypt A. El-Gilany, A. El-Wehady and M. El-Wasify ........................................................................................................................................................................................................................................................................962 Natural killer cell populations in Egyptians infected with hepatitis C virus M. Rafik, G. Sidhom, R . Mamdouh, D. Ellebedy and M. Mohamed .......................................................................................................................................................................................................................969 Lead exposure among the general population of Duhok governorate, Kurdistan region, Iraq A.H. Al-Dosky, D.J. Al-Timimi and S.A. Al-Dabbag ...........................................................................................................................................................................................................................................................974 Consumption of iodized salt among households of Basra city, south Iraq S.M. Ebrahim and N.K. Muhammed...........................................................................................................................................................................................................................................................................................980 Oral and pharyngeal cancers in Yemen: a retrospective study E.S. Halboub, M.Abdulhuq and A. Al-Mandili ......................................................................................................................................................................................................................................................................985 M. Haytham Khayat MD, PhD, FRSPH, Editor-in-chief Editorial Board Ahmad Ezzat Abdou BSc, DPH, PhD, (Secretary) Naeema Al Gasseer MSc, PhD Ahmad Bassel Al-Yousfi PhD, PE, DEE Mohamed M. Ali BSc, MSc, PhD, DTMH Abdulla S. Assaedi MBBS, MPH Mounir Farag MD, DGS, DEmS, DHP Zuheir Hallaj MD, DPH, DrPH Nahla Khamis Ibrahim MD, MPH, Dr.PH (Epidemiology), DHPE (Medical Education) Mamunur Rahman Malik MBBS, Dip(Health Economics), MSc, MPhil Ezzedine Mohsni PhD Abdulaziz Saleh Dip(Hosp Pharm), Dip(Indus Pharm), PHD Kassem Sara MD, MAM Joanna Vogel MScHS, MScPH Mohamed Helmy Wahdan MD, DPH, PhD International Advisory Panel Dr S. Aboulazm. Professor of Orthodontics. Egypt Dr Abdul Rahman Al-Awadi BSc, MD, MPH, Honorary FRCM, Ireland Dr Law, Korea, Honorary  FRCS & P, Glasgow, FRCP, Edinbugh. Kuwait Dr Fariba Al-Darazi RN, MSc, PhD. Bahrain Dr M. Al-Nozha, MD, FRCP, FACC, FESC. Professor of Medicine and Consultant Cardiologist. Saudi Arabia Dr Ala’din Alwan MD, FRCP, FFPHM. Iraq Dr F. Azizi. Professor of Internal Medicine and Endocrinology. Islamic Republic of Iran Dr K. Bagchi BSc, MD, PhD. India Professor K. Dawson BA, MD, PhD, FRCP, FRACP, FRCPCH, DObst, RCOG. New Zealand Professor Kaussay Dellagi MD. Tunisia Dr R. Dybkaer MD. Denmark Dr M. Aziz El-Matri. Professor of Medicine. Tunisia Professor F. El-Sabban BSc, MS, PhD. United States of America Dr A.H. El-Shaarawi MSc (Stat), PhD (Stat). Canada Professor N. Fikri-Benbrahim PhD (Pub health) (SocSci). Morocco Professor A.T. Florence BSc (Pharm), PhD, DSc, FRSC, FRPharmS, FRSE. United Kingdom Professor Cheherezade M.K. Ghazi BS (Nursing), MS (Nursing), DPH, MPA. Egypt Professor M.A. Ghoneim MD, MD (Hons). Egypt Dr J.A. Hashmi DTM&H, FRCP. Pakistan Professor J. Jervell MD, PhD. Norway Professor G.J. Johnson MA, MD, BChir, FRCS (C), FRCOphth, DCEH. United Kingdom Dr M. Kassas. Emeritus Professor of Plant Ecology. Egypt Professor M.M. Legnain MBBS, MRCOG, FRCOG. Libyan Arab Jamahiriya Professor El-Sheikh Mahgoub DipBact, PhD, MD, FRCPath. Sudan Professor A.M.A. Mandil MSc (Paediatr), MPH, DrPH. Egypt Professor A.B. Miller MB, FRCP. Canada Professor S.S. Najjar MD. Lebanon Dr Abubaker A. Qirbi BSc, MD (Edin), FRCPC (Can), FRCP FRCPath (UK). Republic of Yemen Professor O.S.E. Rasslan MD, PhD. Egypt Professor W.A. Reinké MBA, PhD. United States of America Professor I.A. Sallam, MD, Dip High Surgery Cairo, Honorary FRCS, PhD (Glasgow),  LRCP, MRCS, FRCS (London), ECFMG. Egypt Dr C.Th.S. Sibinga FRCP (Edin), FRCPath. The Netherlands Mr Taoufik Zeribi Eng BSc, MSc. Tunisia Editors Fiona Curlet, Eva Abdin, Alison Bichard, Guy Penet Graphics Suhaib Al Asbahi, Hany Mahrous, Diana Tawadros Administration Nadia Abu-Saleh, Yasmine El Sakhawy المجلة الصحية لشرق المتوسط المجلد الثامن عشر العدد التاسع 109 رسالة من المحرر rotidE eht morf retteL تجاوبًا مع الشعار الذي اتخذته المنظمة عام 1102 لنفسها في اليوم العالمي للقلب، تحتفي المنظمة في الثامن والعشرين من أيلول/ سبتمبر من كل عام، وهو يدور حول الـَمْسَكن والقلب والصحة، ويركِّ ز على الوقاية من الأمراض القلبية والوعائية لدى النساء والأطفال. فعلى الرغم من أن عدد من يموت من النساء بسبب الأمراض القلبية الوعائية يساوي عدد الرجال الذين يموتون بسببها فإن هذه الأمراض لاتزال ُينظر إليها على أنها أمراض تخص الرجال، مما يؤدِّي إلى استهانة كثير من النساء بمخاطر هذه الأمراض. وفي هذا العدد من المجلة ورقة استندت إلى البيانات المستمّدة من السجل الخليجي الثاني للأحداث القلبية الحادة من حيث المضاعفات داخل المستشفيات وحصيلة سنة واحدة على حدوث المتلازمة التاجية الحادة لدى المرضى المصابين بارتفاع ضغط الدم. وقد وجد الباحثون أن ارتفاع ضغط الدم كان أكثر انتشارًا بين النساء منه بين الرجال، وأنه عامل منبئ مستقل بفشل القلب. ولو أنه لم تكن هناك اختلافات ُيْعَتدُّ بها بين الجنسين في معدل وفيات مرضى ارتفاع ضغط الدم. وتشير التقديرات إلى أن ما يزيد على مئتي مليون امرأة وفتاة في البلدان النامية؛ ممن لا يرغبن بالحمل، لا ُيتاح لهن الحصول على مانعات الحمل ولا الوصول إلى خدمات تنظيم الأسرة. علًما بأنه قد أصبح من الثابت أن تنظيم الأسرة لا يقتصر على إنقاذ الأرواح وعلى تحسين صحة النساء وأطفالهن، بل يتجاوز ذلك إلى تعزيز المجتمعات وتشجيع التنمية الاقتصادية. وفي تموز/يوليو 2102، ُعقدت قمة لندن حول تنظيم الأسرة، وكان هدفها “إحياء الالتزام العالمي بتنظيم الأسرة وبالحصول على مانعات الحمل باعتبار ذلك من الأولويات في التحوُّ ل في التنمية ومن الأولويات العالية المردود لقاء التكاليف؛ وتحسين الحصول على التزويد بمانعات الحمل وتوزيعها، وإزالة أو تقليص الحواجز التي َتُعوق تنظيم الأسرة”. وقد قامت منظمة الصحة العالمية التي شاركت في القمة، بإصدار العديد من الوثائق الموجزة حول السياسات المعنية بتنظيم الأسرة وأكَّ دت التزامها بتنظيم الأسرة وبتسريع وتيرة تقييم جودة موانع الحمل الجديدة والموجودة بالفعل، وتقديم الدعم للبلدان لإدماج خدمات تنظيم الأسرة في الرعاية الصحية الأساسية، وللدراسة المنهجية للأسباب التي تجعل عددًا كبيرًا من النساء غير قادرات على الحصول على مانعات الحمل عندما يكّن بأمسِّ الحاجة إليها. وفي هذا العدد ورقة تعرض تقييًما للممارسات والبرامج والسياسات المسندة بالبيِّنات في مجال تنظيم الأسرة لدى البلدان الأعضاء في الإقليم، باعتبارها من الخطوات التي تأخذ بنا للأمام في مسيرة الارتقاء المنسَّ ق بالخدمات. فالأجوبة التي وردت من ثمانية عشر بلدًا من البلدان الاثنين والعشرين في الإقليم أشارت إلى أن 49% منها قد أكَّ دت توافر ما لا يقل عن خمسة من أصل سبعة عناصر أساسية في المسح للبرامج الناجحة في مجال تنظيم الأسرة واستنتجت أن أفضل ممارسات الارتقاء بتنظيم الأسرة وتحسين صحة الأمهات والأطفال في الإقليم تتطلَّب تعاونًا وثيقًا بين جميع الأطراف المعنية. وبعُد، فإن الدورة التاسعة والخمسين لّلجنة الإقليمية في القاهرة سوف تعقد بإذن الله في الفترة من 1 إلى 4 تشرين الأول/أكتوبر 2102، ومن المواضيع التقنية التي ستناقش فيها: تعزيز النظم الصحية في بلدان الإقليم؛ والتزام الدول الأعضاء بالإعلان السياسي للجمعية العامة للأمم المتحدة حول الوقاية من الأمراض غير السارية ومكافحتها؛ وتلبية القدرات الوطنية الأساسية اللازمة لّلوائح الصحية الدولية في الوقت المحدَّ د من عام 4102.  esaesid ralucsavoidrac no gnisucof si rebmetpeS 82 no 2102 yaD traeH dlroW ,htlaeh ,traeh ,emoh fo emeht 1102 eht gniunitnoC a sa deviecrep llits si DVC tey dna nem sa DVC morf eid nemow ynam sa tsomlA .nerdlihc dna nemow gnoma noitneverp )DVC(  yrtsigeR fluG dn2 eht morf atad gnizilitu repap a si eussi siht ni dedulcnI .ksir rieht gnitamitserednu nemow ot gnidael ,esaesid s’nam htiw stneitap ni emordnys yranoroc etuca fo emoctuo raey-1 dna snoitacilpmoc latipsoh-ni ta dekool hcihw stnevE caidraC etucA fo -ciderp tnednepedni na saw ti taht dna nem naht nemow ni tnelaverp erom saw noisnetrepyh taht dnuof srohtua eTh .noisnetrepyh .stneitap evisnetrepyh eht ni ytilatrom ni secnereffid redneg tnacfiingis on ,revewoh ,erew ereTh .eruliaf traeh rof rotcaf evit -noc ssecca tonnac tnangerp teg ot tnaw ton od ohw seirtnuoc gnipoleved ni slrig dna nemow noillim 002 revo taht detamitse si tI nemow fo htlaeh eht evorpmi dna sevil evas ot ylno ton nwohs neeb sah gninnalp ylimaF .secivres gninnalp ylimaf dna sevitpecart  ylimaF no timmuS nodnoL eht ,2102 yluJ nI .tnempoleved cimonoce egaruocne dna seitinummoc nehtgnerts ot osla tub ,seibab dna -rofsnart dna evitceffe-tsoc a sa sevitpecartnoc ot ssecca dna gninnalp ylimaf ot stnemtimmoc labolg ezilativer“ ot dleh saw gninnalP ot sreirrab ecuder dna evomer dna ;seilppus evitpecartnoc fo noitubirtsid dna ssecca eht evorpmi ;ytiroirp tnempoleved lanoitam :ot tnemtimmoc sti demrffia dna sfeirb ycilop gninnalp ylimaf lareves deussi ,timmus eht dednetta hcihw ,OHW .”gninnalp ylimaf cisab otni secivres gninnalp ylimaf etargetni ot seirtnuoc troppus ;sevitpecartnoc ytilauq gnitsixe dna wen fo tnemssessa etarelecca detneserp repap A .ti deen yeht nehw noitpecartnoc niatbo tonnac llits nemow ynam os yhw enimaxe yllacitametsys dna ;erac htlaeh a sa noigeR eht fo setatS rebmeM ni gninnalp ylimaf ni secitcarp emmargorp dna ycilop tseb desab-ecnedive dessessa eussi siht ni  fo ytilibaliava eht demrfinoc hcihw fo %49 ,setatS rebmeM 22 fo 81 morf deviecer erew sesnopseR .pu-elacs detanidrooc sdrawot pets secitcarp tseb pu elacs ot taht dedulcnoc dna semmargorp gninnalp ylimaf lufsseccus fo stnemele laitnesse deyevrus 7 eht fo 5 tsael ta .sredlohekats lla gnoma noitaroballoc esolc seriuqer noigeR eht ni htlaeh dlihc dna lanretam evorpmi rehtruf dna gninnalp ylimaf ni  stcejbus lacinhcet eht gnomA .2102 rebotcO 4 ot 1 morf oriaC ni dleh eb lliw eettimmoC lanoigeR eht fo noisseS htnin-yftiF eTh -aralced lacitilop eht ot tnemtimmoc ’setatS rebmeM ;noigeR eht fo seirtnuoc ni gninehtgnerts smetsys htlaeh :era noissucsid rednu  4102 eht gniteem dna ;sesaesid elbacinummocnon fo lortnoc dna noitneverp eht no ylbmessA lareneG snoitaN detinU eht fo noit .snoitalugeR htlaeH lanoitanretnI eht rof seiticapac eroc lanoitan eht rof enildaed EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 902 In-hospital complications and 1-year outcome of acute coronary syndrome in patients with hypertension: findings from the 2nd Gulf Registry of Acute Cardiac Events W.M. Ali,1 K.F. Al Habib,2 A. Hersi,2 N. Asaad,1 K. Sulaiman,3 A. Al-Shiek-Ali 4 and J. Al Suwaidi 1 ABSTRACT Using data from the 2nd Gulf Registry of Acute Coronary Events (Gulf RACE-2) in 2008–09 we investigated the in-hospital complications and 1-year outcome of acute coronary syndrome (ACS) in patients with systemic hypertension from 6 Gulf countries. Of 7847 consecutive patients admitted with ACS, 3746 (47.7%) had hypertension. Hypertension was more prevalent in women, in Arabs than non-Arabs and in older age groups. Patients with hypertension were more likely than those without hypertension to present with dyspnoea and advanced Killip class. Among hypertensive patients, the mortality rate was higher only among those admitted with ST-elevation myocardial infarction. After adjustment for baseline variables, hypertension was an independent predictive factor for heart failure (OR = 1.31) and stroke (OR = 2.47). There were no significant differences in mortality in hypertensive ACS patients when stratified by sex, age or ethnicity. 1Department of Cardiology and Cardiovascular Surgery, Hamad General Hospital and Weill Cornell Medical School, Doha, Qatar (Correspondence to J. Al Suwaidi: jalsuwaidi @hotmail.com). 2King Fahad Cardiac Centre, King Khalid University Hospital College of Medicine, Riyadh, Saudi Arabia. 3Department of Cardiology, Royal Hospital, Muscat, Oman. 4Department of Cardiology, Sheikh Khalifa Medical City, Abu Dhabi, United Arab Emirates. Received: 03/06/11; accepted: 26/07/11 نم جئاتن :مدلا طغض عافترا ضىرم ىدل دحاو ٍماع ضيم دعب لئاصلحاو ىفشتسلما في ةدالحا ةيجاتلا ةمزلاتلما تافعاضم ةدالحا ةيبلقلا ثادحلأل نياثلا يجيللخا لجسلا يديوسلا مساج ،ليع خيشلا يولع ،نمايلس مظاك ،دعسلأا لاضن ،سيرلها دحمأ ،بيبلحا دلاخ ،ليع نسمح ديلو في تافعاضلما ءاصقتسلا 2009-2008 ينماعلل ةدالحا ةيبلقلا ثادحلأل نياثلا يجيللخا لجسلا في تايطعلما نم نوثحابلا دافتسا دقو :ةـصلالخا تلمشو .ةيجيلخ نادلب ةتس لىإ نومتني ةدالحا ةيجاتلا ةمزلاتلماب ضىرم ىدل يزاهلجا مدلا طغض عافترا نم دحاو ماع ضيم دعب لئاصلحاو ىفشتسلما .مدلا طغض عافتراب ًاباصم )%47.7( ًاضيرم 3746 مهنيب نم ناكو ،ةدالحا ةيجاتلا ةمزلاتلماب مهتباصلإ تايفشتسلما في ملهوبق مت ًاضيرم 7847 ةساردلا ناكو .رمعلاب ةمدقتلما تاعومجلما في برعلا يرغ ينب هنم برعلا ينبو ،لاجرلا ينب هنم ءاسنلا ينب ًاراشتنا رثكأ مدلا طغض عافترا نأ ينثحابلل حضتاو يرغ ىدل ماع ديزي لدعمبو ،Killip بيليك فينصت قفو مدقتم فنص عم سفنلا قيض مهيلع رهظي نأ مدلا طغض عافتراب ينباصلما ضىرلما لىع بلغي ةلضع في ءاشتحا ميهدل ناكو ملهوبق مت نم دنع طقف لىعأ تايفولا لدعم ناك ،مدلا طغض عافتراب ينباصلما ينب نمو .مدلا طغض عافتراب ينباصلما لماع مدلا طغض عافترا نأ اودجو ،ساسلأا طخ في تايرغتلما مهذخأب تايطعلما نوثحابلا ححص نأ دعبو .ST ةلصولا في عافترا عم قفارت بلقلا ينباصلما ىدل تايفولا لدعم ينب يربك قرف كانه نكي لمو .)OR = 2.47( ةيغامدلا ةتكسللو )OR = 1.31 ةيحجرلأا لدعم( يبلقلا لشفلل لقتسم ئبنم .ةينثلإاو رمعلاو سنلجا بسحب حئاشر لىإ مهميسقت دنع مدلا طغض عافتراو ةدالحا ةيجاتلا ةمزلاتلماب Complications hospitalières et issue à un an d'un syndrome coronarien aigu chez des malades hypertendus : résultats obtenus à partir du deuxième Registre des événements coronariens aigus du Golfe RÉSUMÉ À partir de données du deuxième Registre des événements coronariens aigus du Golfe (Gulf RACE-2), pour la période 2008 - 2009, nous avons enquêté sur les complications à l'hôpital et sur l'issue à un an d'un syndrome coronarien aigu chez des patients atteints d'hypertension artérielle systémique dans six pays du Golfe. Sur 7847 patients admis consécutivement pour un syndrome coronarien aigu, 3746 souffraient d'hypertension (47,7 %). L'hypertension était plus forterment prévalente chez les femmes, chez les patients arabes que chez les personnes d'un autre groupe ethnique et dans les tranches d'âge supérieur. Les patients hypertendus avaient une plus grande probabilité que les patients ayant une tension normale de souffrir de dyspnée et d'atteindre un stade élevé dans la classification de Killip. Chez les patients hypertendus, le taux de mortalité était plus élevé uniquement chez les patients admis pour un infarctus du myocarde avec sus-décalage du segment ST. Après ajustement des résultats en fonction des variables de référence, l'hypertension s'est révélé être un facteur prédictif indépendant de l'insuffisance cardiaque (OR = 1,31) et de l'accident vasculaire cérébral (OR = 2,47). Aucune différence significative n'a été observée dans la mortalité des patients atteints d'un syndrome coronarien aigu après stratification par sexe, par âge, ou par groupe ethnique. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 903 Introduction Hypertension is widely acknowledged to increase the risk of premature car- diovascular disease, including stroke, coronary artery disease (CAD), heart failure and peripheral vascular dis- ease [1–3]. In the INTERHEART study of patients from 52 countries, hypertension accounted for 18% of the population-attributable risk of a first myocardial infarction [2]. Fur- thermore, hypertension is the most common and most important risk factor for stroke and intracerebral haemorrhage and the incidence of these disease can be markedly re- duced by effective antihypertension therapy [4,5]. The global prevalence of hypertension is high. It is estimated that 26% of the world adult popula- tion (972 million) had hypertension in the year 2000 [6]. In the United States, 58 to 65 million people with hypertension were reported in the National Health and Nutrition Ex- amination Survey (NHANES) data from 1999–2000 [7]. The prevalence of hypertension was 28% and 44% in Canada and 6 European countries respectively [8]. We have recently reported a 49.4% prevalence of hypertension among pa- tients with acute coronary syndrome (ACS) using data from the 1st Gulf Registry of Acute Coronary Events (Gulf RACE) [9]. Hypertension was not independently associated with increased risk of in-hospital mortality except in patients with ST-elevation myocardial infarction (STEMI). In the present study, using data from the 2nd Gulf RACE registry (Gulf RACE- 2), we evaluated the in-hospital and 1-year outcomes of patients with ACS in relation to their hypertension status. We also evaluated the preva- lence of hypertension and outcomes of these patients according to age, sex and ethnicity. Methods Sample The data were collected from the Gulf RACE-2, which was a 9-month prospec- tive, multicentre study of 7847 consecu- tive patients with ACS from 6 adjacent Gulf countries (Bahrain, Saudi Arabia, Qatar, Oman, United Arab Emirates and Yemen). Patients diagnosed with ACS, including unstable angina and non-ST- and ST-elevation myocar- dial infarction (NSTEMI and STEMI respectively), were recruited from 65 hospitals. An on-site cardiac catheteriza- tion laboratory was available in 43% of the hospitals and a coronary care unit onsite in 71% of hospitals. There were no exclusion criteria and thus all patients with ACS were enrolled. Diagnosis of the different types of ACS and definitions of data variables were based on the American College of Cardiology clinical data standards [9]. The history of hypertension was obtained and recorded at the time of presentation. The study received ethical approval from the institutions’ ethics committees in all participating coun- tries. Data collection Details about the Gulf RACE methods have been published previously [10]. A case report form for each patient with suspected ACS was filled out on hospital admission by assigned physicians and/ or research assistants working in each hospital using standard definitions and was completed throughout the patient’s hospital stay. All forms were verified by a cardiologist then sent online to the principal coordinating centre, where the forms were further checked for mistakes before submission for final analysis. To avoid double-counting of multiple ad- missions to the registry, the patients’ national identification numbers and an assigned registry number were used. Thus, only the second ACS admission was included for each patient registered more than once. Patients were divided into hyper- tensive and non-hypertensive groups. Systemic hypertension was defined in the current registry as: either history of hypertension diagnosed and treated with medication, diet and/or exercise; or blood pressure > 140 mmHg systolic and/or diastolic pressure > 90 mmHg on at least 2 occasions. Any history of hypertension was obtained and record- ed at the time of presentation. For this study, patients were analysed according to the presence or absence of a diagno- sis of hypertension. Statistical analysis Patients’ characteristics were presented as proportions, mean and standard de- viation (SD) as appropriate. Whenever possible, rates were used to describe patient populations. The frequencies of categorical variables in the 2 groups (hy- pertensive and non-hypertensive) were compared using the chi-squared test. Continuous variables were compared using the 2-tailed Student t-test. Vari- ables influencing in-hospital mortality, cardiogenic shock and stroke were as- sessed with multiple logistic regressions after adjustment for confounders (i.e. age, sex, diabetes mellitus, hyperten- sion, dyslipidaemia, smoking, throm- bolytic therapy and khat use). Odds ratios (OR), 95% confidence interval (CI) and P-values were reported for significant predictors. A P-value < 0.05 was considered significant. All P-values were the results of 2-tailed tests. All data analyses were carried out using SPSS, version 14. Results Patients’ characteristics Gulf RACE-2 enrolled 7847 patients with ACS from 6 Gulf countries, 3746 (47.7%) of whom had hypertension. Patients with hypertension were EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 904 Table 1 Characteristics of hypertensive and normotensive patients with acute coronary syndrome Variable Hypertensive Normotensive P-value (n = 3746) (n = 4101) Demographic data Age [mean (SD)] (years) 59.7 (11.7) 54.3 (12.8) < 0.001 Male (No., %) 2239 59.8 3543 86.4 < 0.001 Arab ethnicity (No., %) 2868 76.6 2739 68.8 <0.001 Clinical history History of CAD (No., %) 1997 54.0 1229 30.0 < 0.001 History of heart failure (No., %) 382 10.0 140 3.0 < 0.001 Prior coronary revascularization (No., %) 708 19.0 275 6.7 < 0.001 Dyslipidaemia (No., %) 1847 49.0 742 18.0 < 0.001 Family history of CAD (No., %) 469 12.5 336 8.0 0.9 Diabetes mellitus (No., %) 2024 54.7 1089 26.9 < 0.001 Current smoking (No., %) 961 25.7 1839 44.8 < 0.001 History of renal failure (No., %) 277 7.4 39 1.0 < 0.001 History of stroke (No., %) 269 7.0 69 1.7 < 0.001 Prior aspirin use (No., %) 2133 58.5 1058 25.8 < 0.001 Prior β-blocker use (No., %) 1712 45.7 585 14.3 < 0.001 Prior ACE use (No., %) 1558 41.6 473 11.5 < 0.001 Clinical data at presentation Presentation within 12 h of event (No., %) 332 53 619 60.9 < 0.001 Heart rate [mean (SD)] (beats/min) 85 (20) 84 (20) 0.02 Systolic BP [mean (SD)] (mmHg) 144 (31) 128 (25) < 0.001 Diastolic BP [mean (SD)] (mmHg) 84 (19) 78 (16) < 0.001 BMI [mean (SD)] (kg/m2) 27.7 (6.0) 26.0 (4.8) < 0.001 Ischemic chest pain (No., %) 2993 74.2 3489 84.4 < 0.001 Dyspnoea (No., %) 574 14.2 226 5.5 < 0.001 Killip class > I (No., %) 1052 28.0 753 18.4 < 0.001 GRACE score (No., %) < 0.001 Low 2340 63.0 2963 73.6 Intermediate 1048 28.0 820 20.4 High 322 8.7 245 6.0 LV dysfunction (No., %) All cases 2124 73.0 2307 74.0 0.3 Severe cases 318 11.0 237 7.7 < 0.001 Coronary angiography (No., %) 1303 34.8 1254 30.6 < 0.001 Left main stem CAD 87 6.7 58 4.6 0.02 Single-vessel CAD 324 24.8 432 34.0 < 0.001 Two-vessel CAD 297 22.7 322 25.7 0.8 Three-vessel CAD 474 36.0 293 23.0 < 0.001 Laboratory findings [mean (SD)] First blood sugar (mg/dL) 11 (5.8) 9.7 (5.0) < 0.001 Fasting blood sugar (mg/dL) 7.5 (3.0) 7.0 (3.0) < 0.001 Peak CK-MB (ng/mL) 74 (127) 122 (177) < 0.001 First creatinine (µmol/L) 111 (91) 94 (59) < 0.001 Total cholesterol (mg/dL) 4.7 (1.8) 4.9 (1.4) < 0.001 HDL cholesterol (mg/dL) 1.0 (0.5) 1.0 (0.5) 0.5 LDL cholesterol (mg/dL) 3.0 (1.5) 3.2 (1.1) < 0.001 Fasting TG (mg/dL) 1.8 (1.1) 1.7 (1.1) 0.054 First haemoglobin (g/L) 13 (2) 14 (2) < 0.001 SD = standard deviation; CAD = coronary artery disease; ACE = angiotensin-converting enzyme; LV = left ventricular; BP = blood pressure; BMI = body mass index; GRACE = global registry of acute coronary events; CK-MB = creatine kinase MB-fraction; HDL = high-density lipoprotein; LDL = low-density lipoprotein; TG = triglycerides. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 905 significantly older and significantly more likely to be male than those without hypertension (Table 1). The clinical history showed that patients with hypertension were significantly more likely to have comorbidities, including higher rates of diabetes, dyslipidaemia, prior history of CAD, prior congestive heart failure, chronic renal insufficiency and stroke. How- ever, they were less likely to be current smokers (Table 1). On presentation to hospital, hyper- tensive patients with ACS had higher Killip class, body mass index (BMI), heart rate and systolic and diastolic blood pressure. Patients with hyperten- sion were more likely to present with dyspnoea and to seek medical advice early (within 12 hours of the event) compared with non-hypertensive pa- tients (Table 1). Baseline serum creatinine, fasting blood sugar and first blood sugar levels were higher in patients with hyperten- sion, whereas fasting serum triglycerides, low-density lipoprotein cholesterol and peak creatine kinase MB-fraction were higher in non-hypertensive patients. In hypertensive patients with ACS, NSTEMI was the most frequent di- agnosis (36.7%), followed by STEMI (33.0%) and unstable angina (30.0%). Among non-hypertensive patients, 56.6% had STEMI, followed by NSTEMI (24.0%) and unstable angina (19.0%) (Table 2). Overall coronary angiography was performed in 38.4% of hypertension patients and 30.6% of non- hypertensive patients. Hypertensive patients were more likely to have 3-ves- sel CAD and left main stem CAD, while non-hypertensive patients more com- monly had 2- and single-vessel CAD (Table 1). In-hospital and discharge treatment On admission, hypertensive patients were more likely to receive angioten- sin-converting enzyme inhibitors, angiotensin-receptor blockers, calcium- channel blockers and beta-blockers, but less likely to receive thrombolytics, unfractionated heparin and clopidogrel when compared with non-hypertensive patients. At discharge, hypertensive pa- tients were more likely to be prescribed calcium-channel blockers, angiotensin- receptor blockers and beta-blockers and less likely to receive clopidogrel, Table 2 Management of hypertensive and normotensive patients with acute coronary syndrome Variable Hypertensive Normotensive P-value No. % No. % Medication at admission Thrombolysis 540 46.9 1107 52.6 0.002 Aspirin 3673 98.0 4046 98.7 < 0.03 Clopidogrel 2802 74.8 3171 77.3 0.009 Heparin 1415 37.9 1824 44.5 < 0.001 Heparin (LMW) 1514 40.6 1434 35.0 < 0.001 Glycoprotein IIb/IIIa inhibitor 315 8.0 293 7.0 0.03 β-blocker 2878 76.8 2959 72.0 < 0.001 ACE inhibitor 2798 72.0 2848 69.4 0.01 Angiotensin receptor blocker 331 8.8 63 1.5 < 0.001 Primary percutaneous coronary intervention 562 22.0 574 20.0 0.1 Calcium-channel blocker 454 12.0 116 2.8 < 0.001 Medication at discharge Aspirin 3464 95.4 3819 95.5 0.8 Clopidogrel 2463 66.0 2845 69.5 0.01 Statin 3383 90.7 3764 91.9 0.05 β-blocker 3019 80.9 3188 77.9 0.001 ACE inhibitor 2633 70.6 2948 72.0 0.2 Angiotensin receptor blocker 408 10.9 128 3.0 < 0.001 Calcium-channel blocker 488 13.0 123 3.0 < 0.001 Diagnosis at discharge STEMI 1246 33.0 2326 56.6 < 0.001 NSTEMI 1374 36.7 993 24.0 Unstable angina 1128 30.0 789 19.0 ACE = angiotensin-converting enzyme; LMW = low molecular weight; STEMI = ST-elevation myocardial infarction; NSTEMI = non-ST elevation myocardial infarction. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 906 whereas prescription of aspirin, statin and angiotensin-converting enzyme inhibitors were comparable between the 2 groups (Table 2). Outcomes Compared with normotensive patients, hypertensive patients had significantly higher rates of in-hospital heart failure, recurrent myocardial ischaemia, atrial fibrillation and stroke (Table 3). Stroke was predominantly ischaemic in na- ture. Overall in-hospital, 1-month and 1-year mortality rates were not statisti- cally significant different between the 2 groups. However, hypertensive patients presenting with STEMI had higher rate of in-hospital mortality in addition to higher risk of heart failure, recurrent is- chaemia, re-infarction and stroke when compared with non-hypertensive pa- tients with STEMI. Among NSTEMI patients, however, hypertensive patients had a higher risk of heart failure only, while in patients with unstable angina, hypertension was associated with in- creased risk of heart failure and stroke when compared with non-hypertensive patients (Table 3). Bleeding complica- tions were comparable between the 2 groups. Multivariate analysis Age, female sex, hypertension, diabetes mellitus, BMI, smoking and being on thrombolytic drugs were adjusted as confounders in a multivariate analysis for predictors of in-hospital heart failure and stroke in patients with ACS. Hyper- tension, age, smoking, diabetes mellitus and BMI were independent risk factors for in-hospital heart failure (Table 4). Only hypertension was a significant risk factor for stroke, (OR = 2.47, 95% CI: 1.11–5.47, P = 0.02) (Table 4). Hypertension: prevalence and outcome by sex, age and ethnicity Hypertension was more prevalent among women with ACS (66.6%) than among men (38.7%) (P < 0.001). In the Table 3 Clinical outcomes of hypertensive and normotensive patients Variable Hypertensive Normotensive P-value No. % No. % All patients (n = 4036) (n =4129) In-hospital death 168 4.5 185 4.5 1.0 1-month death 266 8.0 297 8.2 0.7 1-year death 386 13.3 374 11.8 0.08 Heart failure 601 16.0 425 10.3 < 0.001 Recurrent ischaemia 613 16.4 601 14.7 0.04 Re-infarction 77 2.1 91 2.2 0.6 Cardiogenic shock 192 5.1 256 6.2 0.2 Major bleeding 24 0.6 22 0.5 0.5 Stroke 35 0.9 18 0.4 0.007 Haemorrhagic (9) (7) Thrombotic (20) (9) Unknown (6) (2) On ventilation 191 5.1 171 4.2 0.05 VF/VT 142 3.8 186 4.6 0.06 Atrial fibrillation 66 1.8 38 0.9 0.001 STEMI (n = 1244) (n = 2319) In-hospital death 105 8.4 141 6.1 0.008 Heart failure 229 18.4 275 11.8 < 0.001 Recurrent ischemia 278 22.3 376 16.2 < 0.001 Re-infarction 55 4.4 71 3.1 0.03 Cardiogenic shock 118 9.5 196 8.5 0.2 Stroke 21 1.7 16 0.7 0.005 On ventilation 97 7.8 119 5.1 0.001 VF/VT 69 5.7 109 4.8 0.2 NSTEMI (n =1374) (n = 993) In-hospital death 56 4.1 39 3.9 0.8 Heart failure 267 19.4 112 11.3 < 0.001 Recurrent ischemia 193 14.0 135 13.6 0.7 Re-infarction 19 1.4 19 1.9 0.3 Cardiogenic shock 68 5.0 50 5.1 0.9 Stroke 8 0.6 2 0.2 0.1 On ventilation 78 5.7 45 4.5 0.2 VF/VT 21 1.5 16 1.6 0.8 Unstable angina (n = 1128) (n =789) In-hospital death 7 0.6 5 0.6 0.9 Heart failure 105 9.3 38 4.8 < 0.001 Recurrent ischemia 142 12.6 90 11.4 0.4 Re-infarction 3 0.3 1 0.1 0.5 Cardiogenic shock 6 0.5 10 1.3 0.08 Stroke 6 0.5 0 0.0 0.04 On ventilation 16 1.4 7 0.9 0.2 VF/VT 3 0.3 5 0.6 0.2 STEMI = ST-elevation myocardial infarction; NSTEMI = non-ST elevation myocardial infarction; VF = ventricular fibrillation, VT = ventricular tachycardia. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 907 spectrum of ACS, women with or with- out hypertension had significantly high- er in-hospital mortality (i.e. any death recorded during admission from time of arrival until discharge) and 1-month and 1-year mortality rates compared with men with or without hypertension (Table 5). The differences in mortal- ity between women and men among hypertensive patients were no longer statistically significant after adjustment for age, diabetes mellitus, renal failure, BMI, smoking and hyperlipidaemia. The prevalence of hypertension in- creased steadily with age, from 21.6% at age < 40 years to 59.0% at > 70 years. Mortality rates rose consistently with age among hypertensive and non- hypertensive patients with ACS, both in-hospital and during the follow-up period. When adjusted for sex and comorbidities, age was an independ- ent predictor for in-hospital mortality in hypertensive (OR = 1.04, 95% CI: 1.03–1.06, P < 0.001) and non-hyper- tensive patients (OR = 1.06, 95% CI: 1.05–1.07, P < 0.001). Hypertension was more prevalent in Middle-Eastern Arab ethnicity when compared to non-Arabs (51.2% versus 39.2%, P < 0.001) and more prevalent in female Arabs (66.8% versus 61.8%, P < 0.001). In-hospital, 1-month and 1-year mortality were higher in Arab hyper- tensive and non-hypertensive patients when compared with non-Arab ethnic- ity patients. In multivariate analysis, after adjustment for age, sex and comorbidi- ties, Arab ethnicity was an independent predictor of in-hospital mortality (OR = 1.75, 95% CI: 1.20–2.56, P = 0.004) but hypertension was not a predictor of in-hospital mortality (OR = 0.84, 95% CI: 0.65–1.10, P = 0.21). Discussion The data from the Gulf RACE-2 survey provides an insight into the prevalence and impact of hypertension in a broad Middle East population presenting with ACS. In Gulf RACE-1 we reported a high prevalence of hypertension in a Middle Eastern population presenting with ACS [9]. Hypertensive patients had a higher risk of heart failure than non-hypertensive patients. The mor- tality rate was also higher but only in hypertensive patients presenting with STEMI. The current study extends these observations using data from Gulf RACE-2, where the prevalence of hypertension was comparable to that reported from the first registry. Fur- thermore, the in-hospital mortality rate was comparable between the 2 groups. We also report for the first time that this comparable mortality rats extends up to 1 year follow-up with a trend of increased mortality among hyperten- sive patients. Patients with hyperten- sion had a higher risk of heart failure complications regardless of the ACS subtype. Hypertension patients also had higher risk of recurrent myocardial infarction and the development of atrial fibrillation. We also report for the first time that among hypertensive Middle- Eastern patients presenting with ACS hypertension was the only independent risk factor for stroke, which was primari- ly ischaemic in nature. Moreover, stroke risk appeared to be most prominent among STEMI patients. The literature on the effect of hyper- tension on short- and long-term mor- tality is conflicting; many studies have claimed a deleterious effect of hyperten- sion in ACS with an increasing risk of death [11–13] while others, including the current study and Gulf RACE-1, failed to show this [6–9,11–19]. Other investigators reported better outcomes in hypertensive patients presenting with ACS [20,21]. These conflicting results might be due to the complex impact of hypertension on ACS and the dif- ferences in patients’ baseline charac- teristics such as age, sex and ethnicity as well as therapy among these various studies. Hypertension is well known to be associated with other morbidities, such as diabetes mellitus, dyslipidaemia, obesity and previous CAD, that might collectively contribute to a deteriora- tion in prognosis [13,20,22]. Moreover, left ventricular hypertrophy, heart fail- ure and arrhythmia are known to be induced by hypertension [13,20,22]. However, some characteristics of hy- pertensive patients might attenuate the Table 4 Multivariate analysis for predictors of in-hospital heart failure and stroke in patients with acute coronary syndrome Variable In-hospital heart failure In-hospital stroke OR (95% CI) P-value OR (95% CI) P-value Age 1.03 (1.02–1.04) < 0.001 1.02 (0.99–1.05) 0.2 Sex 1.31 (0.97–1.76) 0.08 0.51 (0.14–1.85) 0.3 Hypertension 1.31 (1.05–1.65) 0.02 2.47 (1.11–5.47) 0.02 Dyslipidaemia 0.94 (0.73–1.22) 0.7 0.92 (0.39–1.07) 0.9 Diabetes mellitus 1.54 (1.22–1.94) < 0.001 1.42 (0.64–3.13) 0.3 Smoking 1.35 (1.05–1.72) 0.01 1.65 (0.68–3.98) 0.3 Thrombolytic therapy 0.58 (0.46–0.73) < 0.001 0.59 (0.27–1.30) 0.2 BMI 1.03 (1.01–1.05) 0.001 1.00 (0.93–2.17) 1.0 OR = odds ratio; CI = confidence interval; BMI = body mass index. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 908 adverse factors in ACS, including smaller infarction sizes [23], better coronary artery perfusion in the pres- ence of higher diastolic pressure, a well-developed collateral circulation in the setting of greater severity of coronary atherosclerosis and previous treatment with cardioprotective drugs [20]. This hypothesis of a bal- ance between exacerbating and attenuating factors was clearly evident in Gulf RACE-2. Hypertensive patients were older, more often female and were more likely to have diabetes, dyslipidaemia, higher BMI and prior CAD. On the other hand, hypertensive patients were more likely to be on aspirin and beta-blockers prior to admission. Hypertensive patients also were more likely to receive recommended medications at admission in- cluding beta-blockers, angiotensin converting enzyme inhibitors, angiotensin II receptor blockers. NSTEMI was the predominant presentation in hypertensive patients and peak creatine kinase MB-fraction was lower, which supports the smaller infarction size, while in non-hypertension, STEMI occurred in almost half of patients. The higher mortality rate of hypertensive patients with STEMI in our study is supported by many similar studies including our previous study in Gulf RACE-1 [9,23]. It seems that the deleterious effect of hyperten- sion became more evident when the infarction size was large, resulting in higher recurrent ischaemia, heart fail- ure and stroke, which in turn led in a higher risk of death. Several factors may explain the higher risk of heart failure among hypertensive patients presenting with ACS. Diastolic dysfunction, due to hypertrophic and interstitial changes in chronic hypertension, ventricular remodelling and greater activation of neurohormonal systems (both in the early post-infarction period and several months later) [24], plus the higher incidence of diabetes mellitus in hypertensive patients, each independently and collectively enhance the risk of heart failure [25]. Nevertheless, heart failure is a major predictor of mortality in ACS [26] and the higher mor- tality in our study was limited to hypertensive patients with STEMI only. Although severe left ventricular dysfunction (systolic dysfunction) was higher among hypertensive patients compared with non-hypertensive patients in our study group, the overall prevalence of left ventricular dysfunction was comparable between the 2 groups, suggesting that a significant proportion of heart failure in hypertension may be attributed to diastolic dysfunction. We found a higher incidence of stroke in hyper- tensive patients with ACS, which is unsurprising given that hypertension and acute myocardial infarction are independently associated with higher risk of stroke Ta bl e 5 M or ta lit y in h os pi ta l a nd a t 1 -m on th a nd 1 -y ea r f ol lo w -u p am on g hy pe rt en si ve a nd n or m ot en si ve p at ie nt s w it h ac ut e co ro na ry s yn dr om e by s ex , a ge a nd e th ni ci ty Va ri ab le In -h os pi ta l m or ta lit y 1- m on th m or ta lit y 1- ye ar m or ta lit y H yp er te ns iv e N or m ot en si ve H yp er te ns iv e N or m ot en si ve H yp er te ns iv e N or m ot en si ve N o. % N o. % N o. % N o. % N o. % N o. % Se x Fe m al e 63 /1 10 7 5. 7 48 /5 58 8. 6 99 /1 0 0 1 9. 9 64 /5 0 5 12 .7 14 7/ 88 1 16 .7 79 /4 49 17 .6 M al e 10 5/ 26 39 4. 0 13 7/ 35 43 3. 9 16 7/ 23 27 7.2 23 3/ 31 19 7.5 13 9/ 20 23 11 .8 29 5/ 27 14 10 .9 P- va lu e 0 .0 2 0 .0 0 8 < 0 .0 1 < 0 .0 0 1 < 0 .0 0 1 < 0 .0 0 1 A ge g ro up (y ea rs ) ≤ 40 4/ 14 5 2. 8 8/ 53 1 1.5 5/ 12 9 3. 9 21 /4 71 4. 5 6/ 10 6 5. 7 22 /4 11 5. 4 41 –7 0 11 3/ 29 68 3. 8 12 5/ 31 43 4. 0 18 4/ 26 33 6. 9 20 9/ 27 57 7.6 26 1/ 22 94 11 .4 26 1/ 23 98 10 .9 ≥ 71 51 /6 33 8. 1 52 /4 27 12 .2 77 /5 66 13 .6 67 /3 96 16 .9 11 9/ 50 4 23 .6 91 /3 54 27 .7 P- va lu e < 0 .0 0 1 < 0 .0 0 1 < 0 .0 0 1 < 0 .0 0 1 < 0 .0 0 1 < 0 .0 0 1 Et hn ic ity A ra b 14 4 5. 0 16 5 6. 0 23 4 9. 1 26 9 10 .8 34 2 15 .1 33 9 15 .3 N on -A ra b 24 2. 7 20 1.5 32 4. 2 28 2. 5 44 6. 8 35 3. 7 P- va lu e 0 .0 0 4 < 0 .0 0 1 < 0 .0 0 1 < 0 .0 0 1 < 0 .0 0 1 < 0 .0 0 1 طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 909 [27]. Ischaemic stroke was the main type in our registry, which means that atherosclerosis or embolic events are the likely causes, not thrombolytic ther- apy. Additionally we observed a higher prevalence of atrial fibrillation among hypertensive patients, which may be related to left atrial enlargement and the poor left ventricular function due to heart failure, both of which can make the heart the source of embolic stroke in hypertension patients with ACS. These findings emphasize the need for aware- ness of the risk of stroke by physicians treating hypertension in ACS patients. Finally, the prevalence of hyper- tension in women presenting with ACS was higher than in men. The high prevalence of hypertension among Middle Eastern women was also re- ported in Gulf RACE-1 (70% versus 43%) [28]. Furthermore, the over- all mortality rate was higher among women compared with men. This may be attributed to multiple factors including late and atypical presenta- tions, less medication prescribed and a higher incidence of heart failure in women than men, as was reported by other investigators [29]. Some limitations of the study can be noted. Our data were collected from an observational study and the fun- damental limitations of observational studies cannot be eliminated because of their non-randomized nature and unmeasured confounding factors. However, well-designed observational Table 6 Mortality among hypertensive and normotensive patients with acute coronary syndrome, unadjusted and adjusted to age and comorbidities by sex, age and ethnicity Variable OR (95% CI) Unadjusted Adjusted Sex In-hospital mortality 1.78 (1.42–2.23) 1.80 (1.42–2.85) 1-month mortality 1.54 (1.27–1.86) 1.57 (1.30–1.91) 1-year mortality 1.63 (1.38–1.93) 1.59 (1.34–1.89) Age In-hospital mortality 1.05 (1.04–1.06) 1.05 (1.04–1.06) 1-month mortality 1.04 (1.03–1.05) 1.05 (1.03–1.05) 1-year mortality 1.05 (1.04–1.06) 1.05 (1.04–1.06) Ethnicity In-hospital mortality 2.92 (2.13–4.0 0 2.94 (2.13–4.05) 1-month mortality 3.38 (2.58–4.41) 3.42 (2.61–4.51) 1-year mortality 3.41 (2.69–4.33) 3.40 (2.67–4.32) OR = odds ratio; CI = confidence interval. studies provide valid results and do not systematically overestimate the results compared with the results of randomized controlled trials. Finally, although long-term follow-up was re- ported, this was limited to mortality rates only and data about compliance with medical therapy in addition to the risk of other cardiovascular complica- tions including recurrent myocardial ischaemia, infarction and heart failure were lacking. Conclusions Hypertension is highly prevalent among Middle-Eastern patients presenting with ACS particularly in women, older age groups and among Middle-Eastern Arab ethnicity. Hypertension was in- dependently associated with increased risk of heart failure and stroke. In cases of STEMI, hypertension was associated with a higher risk of in-hospital mortal- ity and stroke. Acknowledgements Gulf RACE is a Gulf Heart Associa- tion (GHA) project and was financially supported by the GHA, Sanofi Aventis, and the College of Medicine Research Center at King Khalid University Hos- pital, King Saud University, Riyadh, Saudi Arabia. The sponsors had no role in study design, data collection, data analysis, writing of the report, or submis- sion of the manuscript. References 1. Psaty BM et al. Association between blood pressure level and the risk of myocardial infarction, stroke, and total mortality: the cardiovascular health study. Archives of Internal Medicine, 2001, 161:1183–1192. 2. Yusuf S et al., INTERHEART Study Investigators. Effect of poten- tially modifiable risk factors associated with myocardial infarc- tion in 52 countries (the INTERHEART study): case-control study. Lancet, 2004, 364:937–952. 3. Sesso HD et al. Systolic and diastolic blood pressure, pulse pres- sure, and mean arterial pressure as predictors of cardiovascular disease risk in Men. Hypertension, 2000, 36:801–807. 4. Staessen JA et al., The Systolic Hypertension in Europe (Syst-Eur) Trial Investigators. Randomised double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension. Lancet, 1997, 350:757–764. 5. Thrift AG et al.; Melbourne Risk Factor Study (MERFS) Group. Risk factors for cerebral hemorrhage in the era of well-controlled hypertension. Stroke, 1996, 27:2020–2025. 6. Kearney PM et al. Global burden of hypertension: analysis of worldwide data. Lancet, 2005, 365:217–223. 7. Fields LE et al. The burden of adult hypertension in the United States 1999 to 2000: a rising tide. Hypertension, 2004, 44:398–404. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 910 8. Wolf-Maier K et al. Hypertension prevalence and blood pres- sure levels in 6 European countries, Canada, and the United States. Journal of the American Medical Association, 2003, 289:2363–2369. 9. Ali WM et al. The prevalence and outcome of hypertension in patients with acute coronary syndrome in six Middle-Eastern countries. Blood pressure, 2010, 20(1):20–26. 10. Ali WM et al. Acute coronary syndrome and khat “herbal amphetamine” use: an observational report. Circulation, 2011, 124:2681 –2689. 11. Kang DG et al., Korea Acute Myocardial Infarction Registry Investigators. Clinical effects of hypertension on the mortality of patients with acute myocardial infarction. Journal of Korean Medical Science, 2009, 24:800–806. 12. Dumaine R et al. Thrombolysis in Myocardial Infarction (TIMI) Study Group. Association of a history of systemic hypertension with mortality, thrombotic, and bleeding complications following non-ST-segment elevation acute coronary syndrome. Journal of Clinical Hypertension, 2006, 8:315–322. 13. Thune JJ et al. Effect of antecedent hypertension and follow-up blood pressure on outcomes after high-risk myocardial infarc- tion. Hypertension, 2008, 51:48–54. 14. Willich SN et al. PIN Study Group. Cardiac risk factors, medica- tion, and recurrent clinical events after acute coronary disease; a prospective cohort study. European Heart Journal, 2001, 22:307–313. 15. Piegas LS et al., AFIRMAR Study Investigators. Risk factors for myocardial infarction in Brazil. American Heart Journal, 2003, 146:331–338. 16. Al Suwaidi J et al., GUSTO-IIb, GUSTO-III, PURSUIT, Glob- al Use of Strategies to Open Occluded Coronary Arteries. Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy, PARAGON-A Investiga- tors, Platelet IIb/IIIa Antagonism for the Reduction of Acute coronary syndrome events in a Global Organization Network. Prognostic implications of abnormalities in renal function in patients with acute coronary syndromes. Circulation, 2002, 106:974–980. 17. Majahalme SK et al. Comparison of patients with acute coro- nary syndrome with and without systemic hypertension. Ameri- can Journal of Cardiology, 2003, 92:258–263. 18. Boden WE. The post-myocardial infarction patient with hy- pertension. Long-term outcome. Postgraduate Medicine, 1992, 92:173–175, 179–180. 19. Ali I et al. Importance of a history of hypertension for the prognosis after acute myocardial infarction–for the Bucindolol Evaluation in Acute myocardial infarction Trial (BEAT) study group. Clinical Cardiology, 2004, 27:265–269. 20. Abrignani MG et al. In-hospital complications of acute myo- cardial infarction in hypertensive subjects. American Journal of Hypertension, 2005, 18:165–170. 21. Ivanuša M et al. The influence of hypertension on in-hospital outcome in patients with acute myocardial infarction. Medicin- ski Glasnik, 2009, 6:53–59. 22. Al Suwaidi J et al. Prevalence of the metabolic syndrome in patients with acute coronary syndrome in six middle eastern countries. Journal of Clinical Hypertension, 2010, 12:890–899. 23. Aursnes I, Landmark K. More non-Q-wave myocardial infarc- tions but similar infarct sizes in patients with hypertension. Blood Pressure, 2000, 9:98–103. 24. Richards AM et al. Antecedent hypertension and heart failure after myocardial infarction. Journal of the American College of Cardiology, 2002, 39:1182–1188. 25. Bener A et al. The effect of hypertension as a predictor of risk for congestive heart failure patients over a 10-year period in a newly developed country. Blood Pressure, 2004, 13:41–46. 26. Alsheikh-Ali AA et al., Gulf RACE Investigators. Heart failure in patients hospitalized with acute coronary syndromes: obser- vations from the Gulf Registry of Acute Coronary Events (Gulf RACE). European Journal of Heart Failure, 2009, 11:1135–1142. 27. Witt BJ et al. The incidence of stroke after myocardial infarc- tion: a meta-analysis. American Journal of Medicine, 2006, 119(4):354.e1–354.e9. 28. El-Menyar A et al. Comparison of men and women with acute coronary syndrome in six Middle Eastern countries. American Journal of Cardiology, 2009, 104:1018–1022. 29. Frazier CG et al., SYMPHONY and the Second SYMPHONY Investigators. Prevalence and management of hypertension in acute coronary syndrome patients varies by sex: observa- tions from the Sibrafiban versus aspirin to Yield Maximum Protection from ischemic Heart events postacute cOroNary sYndromes (SYMPHONY) randomized clinical trials. American Heart Journal, 2005, 150:1260–1267. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 911 Mapping family planning policy and programme best practices in the WHO Eastern Mediterranean Region: a step towards coordinated scale-up P. Chikvaidze,1 H.H. Madi 2 and R.K. Mahaini 3 ABSTRACT Family planning is recognized as among the 4 core components of safe motherhood. This survey aimed to map evidence-based best policy and programme practices in family planning in the Member States of the WHO Eastern Mediterranean Region. A self-administered, structured questionnaire was developed to survey different components of 7 essential elements of successful family planning programmes. Responses were received from the ministry of health in 18 out of 22 Member States. A total of 17 out of 18 responding countries (94%) (including 7 priority countries for Millennium Development Goals 4 and 5) confirmed the availability of at least 5 out of the 7 surveyed essential elements of successful family planning programmes. Documented available best practices in family planning suggest a need for close coordination and collaboration among stakeholders in scaling up these best practices, especially in priority countries, to improve maternal and child health in the Region. 1Women's Reproductive Health; 2Health Protection and Promotion Division; 2Family and Community Health Unit, World Health Organization, Regional Office for the Eastern Mediterranean, Cairo, Egypt (Correspondence to P. Chikvaidze: chikvaidzep@emro.who.int). Received: 28/07/10; accepted: 26/07/11 ءاقترلاا قيسنت وحن ةوطخ :طسوتلما قشر ميلقإ في ةسرلأا ميظنت جمانرب في تاسرمالماو تاسايسلا لضفأ ميسرت ينياهم ييرخ زمار ،ضيام ينسح ءافيه ،زديفكش اتاب ميسرت نوثحابلا هب ماق يذلا حسلما فدهتسيو .ةنومألما ةموملأا في ةيسيئرلا ةعبرلأا تانوكلما نم دحاو هنأ لىع ةسرلأا ميظنت لىإ رظنُي :ةـصلالخا نوثحابلا دعأ دقو .طسوتلما قشر ميلقإ في ءاضعلأا نادلبلا في ةسرلأا ميظنت في تانِّيبلاب ةدنسلما تاسايسلاو جمانبرلا تاسرامم لضفأ ينب طباورلا تاباجتسلاا نوثحابلا ملتساو .ةسرلأا ميظنت في ةحجانلا جمابرلل ةيساسلأا ةعبسلا صرانعلل ةفلتخلما تانوكلما حسلم ةددمح ةينب هلو ،ًايتاذ ذ ََّفنُي ًانايبتسا اهيف ماب( ،)%94( تباجأ يتلا ةشرع نيماثلا نادلبلا نم ًادلب شرع ةعبس دّكأ دقو .نيشرعلاو يننثلاا نادلبلا عوممج نم ًادلب 18 في ةحصلا ةرازو نم حاجنل ةيسيئر صرانع 7 لصأ نم 5 لىع ديزي ام رُفاَوَت ،)ةيفللأل ةيئمانلإا يمارلما نم سمالخاو عبارلا ْنينَيَم ْنرَملل ةبسنلاب ةيولولأا تاذ نادلبلا نم ةعبس ةحلصلما باحصأ ينب ينقيثو نواعتو قيسنت لىإ ةجالحا لىع ةسرلأا ميظنت في ةرفاوتلما ةقثولما تاسرمالما ّلدَتو .حسلما اهلمش يتلا ةسرلأا ميظنت جمارب .ميلقلإا في لافطلأاو تاهملأا ةحص ينستح لجأ نم ،ةيولولأا تاذ نادلبلا في مايسلاو ،هذه تاسرمالما لضفأب ءاقترلاا دوهج في Cartographie des meilleures pratiques dans la mise en oeuvre des politiques et des programmes de planification familiale dans la Région de la Méditerranée orientale : une étape vers un élargissement coordonné RÉSUMÉ La planification familiale est reconnue comme faisant partie des quatre éléments indispensables d'une maternité à moindre risque. La présente enquête visait à cartographier les meilleures pratiques reposant sur une base factuelle pour la mise en oeuvre des politiques et des programmes de planification familiale dans les États Membres de la Région OMS de la Méditerranée orientale. Un questionnaire structuré et destiné à être autoadministré a été élaboré afin d'évaluer les différentes composantes de sept éléments essentiels dans un programme de planification familiale efficace. Des réponses ont été reçues du ministère de la Santé de 18 États Membres sur 22. Au total, 17 pays sur 18 ayant répondu (94 %) (les sept pays prioritaires pour les objectifs 4 et 5 du Millénaire pour le développement compris) ont confirmé la disponibilité d'au moins 5 éléments essentiels étudiés sur les 7 requis pour un programme de planification familiale efficace. Les meilleures pratiques documentées et disponibles en matière de planification familiale suggèrent qu'une coordination et une collaboration étroites sont nécessaires entre les parties prenantes dans l'élargissement de la mise en oeuvre de ces meilleures pratiques, en particulier dans les pays prioritaires, afin d'améliorer la santé des mères et des enfants dans la Région. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 912 Introduction The World Health Organization (WHO) has estimated that in 2008 about 58 300 women and 510 000 newborns died in the countries of the Eastern Mediterranean Region (EMR) due to complications related to pregnancy and childbirth [1]. In order to achieve the United Nations Millennium Development Goals (MDG) 4A (reduce by two-thirds the under-5 mortality rate) and 5A (reduce by three-quarters the mater- nal mortality ratio), between 1990 and 2015, policy-makers in EMR countries need to pursue the most effective, evidence-based strategies in health service delivery for improving maternal and child health outcomes [2]. Family planning (FP) is well rec- ognized among 4 core components of safe motherhood, the other key health service strategies being: skilled attend- ance during pregnancy and childbirth; access to essential obstetric care; and postpartum care [3–5]. Promotion of FP to increase utilization of modern contraceptive methods, specifically in countries with high birth rates, has the potential to prevent up to 32% of all maternal deaths and almost 10% of childhood deaths, not to mention its indirect effects on promoting health through reducing poverty and hun- ger in overpopulated countries [6,7]. FP reduces mortality among women directly by avoiding the risk of death attributed to complications of child- bearing and birth. However, it also indirectly reduces maternal mortal- ity by shifting the risk associated with each pregnancy and birth away from high-risk maternal groups, e.g. among women who are too young (aged 15– 19 years [2]), older women (aged over 35 years [2]), multiparous women and women with existing health problems. In addition, achieving spacing of con- secutive births by 2 years has been es- timated to reduce the chance of dying in infancy or before the age of 5 years by up to 50% in developing countries [3,8,9]. In terms of financial benefits, it has been well demonstrated that investing in sexual and reproductive health, in particular FP, is cost-efficient for countries due to the substantial savings on costs related to health and social services and emotional distress to women, their family and society at large [10,11]. Although progress has been made towards improving FP services in many countries of the EMR, the prevalence of use of modern contra- ceptives remains low. This is especially pertinent to 8 EMR countries where maternal mortality levels remain unac- ceptably high (Afghanistan, Djibouti, Iraq, Morocco, Pakistan, Somalia, Su- dan and Yemen). These countries are classified as MDG priority countries due to poor performance in achiev- ing targets of goals 4A and 5A. In particular, Afghanistan, Pakistan and Sudan contribute up to 80% of the total annual maternal deaths in the EMR, while contraceptive prevalence rates for use of modern methods in these 3 countries are 17.4%, 21.7% and 5.7% respectively [1]. Achieving success in FP pro- grammes requires continued, coor- dinated and harmonized efforts from different national and international concerned stakeholders. Yet sig- nificant amount of time and resources could be saved by documenting, shar- ing, adopting and scaling up successful practices in FP services, that could be classified as best practices in the Region. In addition, documentation of best practices in FP would advise in- ternational donor and national efforts in evidence-based policy decision- making, programming and scaling- up. In the absence of any published literature on this subject in the Region we conducted this survey to identify and map evidence-based best policy and programme practices in FP in the countries of WHO EMR. Methods Setting The survey was conducted in the 22 Member States of the WHO EMR: 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 Emir- ates and Yemen. Survey instrument A self-administered structured ques- tionnaire was developed consisting of 77 questions that were grouped around 7 core categories widely recognized as essential components of successful FP programmes [12]. These were: • integrated FP services and mix of ser- vice delivery points; • staff training; • polices, regulations and guidelines ensuring the quality of FP services; • FP programmes targeted for special groups of population; • FP commodity security; • promotion of FP; and • FP programme planning, monitoring and evaluation. Process The survey instrument was pretested in 4 EMR countries and finalized ac- cording to the feedback received. To standardize the survey process, the questionnaire was delivered to national FP programme managers through WHO representative offices and departments of international af- fairs at the ministries of health in each country. A supporting covering letter explained the purpose of the survey, type of the information being collected, description of who could answer the questions, where to return completed questionnaires and the voluntary nature of the survey. Reminder letters were sent to encourage responses from طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 913 countries. Completed questionnaires were returned to the WHO Regional Office for the Eastern Mediterranean and examined for completeness. This process took place between May and August 2009. Analysis Descriptive statistics were used to ana- lyse the data. Results Responses were received from 18 out of 22 Member States of WHO EMR. Of the 18 countries who responded to the survey 17 (94%) confirmed the availability of at least 1 of the elements comprising 5 of the 7 essential com- ponents of successful FP programmes which we surveyed (Table 1). Integrated FP services and mix of service delivery points FP services were part of the basic health benefit package and were delivered at the primary health care level in 17 (94%) of the 18 respondent countries (Table 2). At primary health care level, contraceptives were provided to all women free of charge in over three- quarters of countries (78%). A range of different types of contraceptives were part of the essential drug list in 89% of countries. Counselling about FP was provided at antenatal care visits in 89% and postnatal care visits in 94%. While most countries provided actual FP methods postnatally (94%), few provided them as part of antenatal care (22%). FP counselling and methods were provided to parents as part of child care services in 61% of countries, vaccination services in 55% and sexu- ally transmitted infection/HIV services in 55%. Importantly, 7 of the MDG- priority countries in the region (Af- ghanistan, Iraq, Morocco, Pakistan, Somalia, Sudan and Yemen) reported the existence of many elements of best practices in terms of integrated FP services (Table 2) and other essential components of successful FP pro- grammes, as reported in the following sections (Tables 3–8). Staff training FP was less commonly part of the pre- service training programmes in medical universities (78% of countries) than through technical schools for midwives, nurses and lady health visitors (94%) Table 3). However, FP was part of in- service continuing medical education for physicians in 94% and in-service training for midwives, nurses and lady health visitors in 83% of countries. FP training guidelines and materials were reportedly evidence-based in 83% countries and were updated regularly in 72%. Polices, regulations and guidelines ensuring the quality of FP services A competency-based national quali- fication system that certified health workers to provide FP counselling and services was in place in only one- third of respondent countries (Table 4). To ensure the safety of FP services, the national regulations required up-to-date minimum standards for health facilities in 89%, medical equip- ment in 89%, medical commodities in 100% and infection prevention measures in 94% of countries. The national guidelines and protocols for FP counselling and service provision were reportedly evidence-based in 89% and regularly updated in 72% of countries. An effective and functional quality assurance system to ensure the quality of FP services provided was reported to be in place in half of countries. A supportive supervision system was in place to support service providers and improve their perfor- mance at primary health care level in 83% of countries and at secondary care level in 50%. FP programmes targeted to special groups of population Less than half of responding coun- tries had special FP programmes in place to meet the needs of certain vulnerable population groups: ado- lescents (39%), internally displaced populations or refugees in (50%), the poor (50%), periurban and slum populations (44%), males (22%) and persons with disabilities (22%) (Table 5). FP commodity security FP commodity security was reportedly ensured through: a well-functioning contraceptive logistics management information system (83%), data-based planning by the government (61%), effective supply chain management of all contraceptive commodities through- out the country (67%) and in certain geographic areas of the country (77%) (Table 6). Table 1 Number of essential components of successful family planning practices present (defined as at least 1 element of the component present) in the family planning programme of 18 countries of the Eastern Mediterranean Region Country No. of essential components present (n = 7) Afghanistan, Egypt, Iran (IR), Iraq, Jordan, Lebanon, Morocco, Pakistan, Qatar, Syrian Arab Republic, Yemen 7 Oman, Palestine, Saudi Arabia, Sudan 6 Bahrain, Somalia 5 United Arab Emirates 3 Djibouti, Kuwait, Libya, Tunisia n/r n/r = no response. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 914 Promotion of FP FP was actively promoted in accord- ing to the respondent EMR countries through: effective social marketing of FP methods (67%), via community education including wide distribution of quality education and information materials (77%) and through com- munity mobilization efforts (61%). Public–private partnerships for FP were implemented more commonly via community education activities (77%), but also via community outreach ser- vices (44%) and service provision pro- grammes (44%) countries (Table 7). FP programme planning, monitoring and evaluation Decisions on changes in FP pro- gramme were made based on evi- dence and the analysis of information obtained from health management information systems (61%), from FP programme evaluation (61%) and from special national and sub-national studies (77%) (Table 8). Evidence- based indicators were selected to monitor and evaluate FP programme 77% of countries. A similar proportion of countries (72%) regularly collected and analysed FP programme data us- ing health management information systems. Table 2 Integrated family planning (FP) services and mix of service delivery points present in all 18 respondent countries and the 7 priority countries of the Eastern Mediterranean Region, 2009 Item All countries Priority countriesa No. % Afg Irq Mor Pak Som Sud Yem FP services are part of the basic health benefit package and are delivered at the primary health care level 17 94 ü ü ü ü ü ü ü At primary health care level, contraceptives are provided to all women regardless of their ability to pay 14 78 ü ü ü ü ü ü ü A mix of different types of contraceptives are part of the country’s essential drug list 16 89 ü ü ü ü ü ü ü FP counselling is provided: At antenatal care visits 4 22 ü ü ü ü ü û ü At postnatal care visits 17 94 ü ü ü ü û ü ü FP methods are provided: At antenatal care visits 4 22 û û û ü û û ü At postnatal care visits 17 94 ü ü ü ü ü ü ü FP counselling and methods for parents are provided during: Child health care services 11 61 ü û ü ü û û ü Child vaccination services 10 55 ü û ü n/r û û ü STI/HIV services 10 55 ü ü û ü ü û ü FP counselling is provided by: General practitioner/ family doctor 16 89 û ü ü ü ü ü ü Obstetrician/ gynaecologist 17 94 ü ü ü ü ü ü ü Nurse 14 78 ü ü ü ü ü û ü Midwife 14 78 ü ü ü ü ü ü ü Community health worker 10 55 ü û n/r ü ü û ü Lady health visitor 11 61 ü û ü ü û ü ü FP methods are provided by: General practitioner/ family doctor 18 100 ü ü ü ü ü ü ü Obstetrician/ gynaecologist 18 100 ü ü ü ü ü ü ü Nurse 8 44 û û ü ü û û ü Midwife 12 67 ü û ü ü û ü ü Community health worker 8 44 ü û ü ü û û ü Lady health visitor 9 50 ü û ü ü û ü ü aAfghanistan, Iraq, Morocco, Pakistan, Somalia, Sudan and Yemen: classified as priority countries due to poor performance in achieving targets of millennium development goals 4A and 5A. ü= available; û= not available; n/r = no response received from the country. STI = sexually transmitted infection; HIV = human immunodeficiency virus. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 915 Discussion The major finding of this mapping sur- vey is that all 18 respondent countries in the EMR have confirmed the existence of the majority of key elements of best policy and programme practices in FP. More than 90% of them had already taken actions and most of them have gained corresponding benefits through: integrating FP services into basic health benefit package at primary health care level and other vertical programmes; including a mix of different contracep- tives into the national essential drug list; and by adopting evidence-based poli- cies, guidelines, standards and practices for both health provider education and Table 3 Staff training in family planning (FP) in all 18 respondent countries and the 7 priority countries of the Eastern Mediterranean Region, 2009 Item All countries Priority countriesa No. % Afg Irq Mor Pak Som Sud Yem FP is part of the pre-service training programmes: In medical universities 14 78 ü û ü ü û ü ü In technical schools for midwives, nurses and lady health visitors 17 94 ü ü ü ü ü ü ü FP is part of in-service training programmes: For continuing medical education for physicians 17 94 ü ü ü ü ü ü ü For midwives, nurses and lady health visitors 15 83 ü ü ü ü ü ü ü Training guidelines and materials are: Evidence-based 15 83 ü ü ü ü û ü ü Updated regularly 13 72 ü ü ü ü û ü ü aAfghanistan, Iraq, Morocco, Pakistan, Somalia, Sudan and Yemen: classified as priority countries due to poor performance in achieving targets of millennium development goals 4A and 5A. ü= available; û= not available. Table 4 Polices, regulations and guidelines ensuring the quality of family planning (FP) services in all 18 respondent countries and the 7 priority countries of the Eastern Mediterranean Region, 2009 Item All countries Priority countriesa No. % Afg Irq Mor Pak Som Sud Yem A competency-based national qualification system is in place that certifies health workers to provide quality FP counselling and services 6 33 ü ü û n/r û û ü To ensure the safety of FP services, national regulations set up-to-date minimum standards for: Health facilities 16 89 ü ü ü ü û ü ü Medical equipment 16 89 ü ü ü ü û ü ü Medical commodities 18 100 ü ü ü ü ü ü ü Infection prevention measures 17 94 ü ü ü ü ü ü ü National guidelines and protocols for FP counselling and service provision are: Evidence-based 16 89 ü ü ü ü û ü ü Regularly updated 13 72 ü ü ü ü û û ü An effective and functional quality assurance system is in place to ensure the quality of provided FP services 9 50 ü û ü ü û û ü Supportive supervision system is in place to support service providers and improve their performance at: Primary health care level 15 83 ü ü ü ü û ü ü Secondary health care level 9 50 ü ü ü ü û ü ü aAfghanistan, Iraq, Morocco, Pakistan, Somalia, Sudan and Yemen: classified as priority countries due to poor performance in achieving targets of millennium development goals 4A and 5A. ü= available; û= not available; n/r = no response received from the country. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 916 service provision. Reportedly, about 60% of responded countries had reli- able contraceptive logistics and sup- ply chain management systems and implemented community education and outreach activities for FP services, including services for internally dis- placed and refugee communities where applicable. The results also indicated cer- tain deficiencies in the availability of evidence-based FP programmes in 18 responded countries. For example, only one-third of respondent countries paid adequate attention to FP service qual- ity assurance through national certi- fication and accreditation systems. In addition, few countries reported having FP programmes specially designed to address the needs of vulnerable popu- lation groups as adolescents, persons with disabilities, men and periurban populations. There were several limitations to this survey. First, although the survey instru- ment was sent to ministries of health through WHO representative offices, we cannot be sure who completed the questionnaires and whether they were qualified to respond to the questions. Secondly, regardless of who completed the questionnaires, the survey was based on self-reporting and the results may therefore reflect over- or under- reporting of services. Thirdly, responses came from only one source, the coun- try’s ministry of health, and we are not sure survey respondents had consulted other concerned organizations, such as universities, international and local partner organizations and professional associations. Another limitation of this survey was that it did not allow an evalu- ation of the geographical distribution of reported FP best practices, as well as their quality and sustainability. The benefits of existing evidence- based practices do not always mate- rialize into scaled-up national level programmes in the EMR. For example, in our survey the 4 largest contributor countries to maternal and under-5-year- old mortality in EMR— Pakistan, Afghanistan, Sudan and Yemen—re- ported meeting a high percentage of the 77 best practices surveyed (84%, 94%, 65% and 96% respectively). However, Table 5 Special family planning (FP) programmes for vulnerable groups in all 18 respondent countries and the 7 priority countries of the Eastern Mediterranean Region, 2009 Item All countries Priority countriesa No. % Afg Irq Mor Pak Som Sud Yem Special programmes are in place designed to meet the needs of vulnerable target groups: Adolescents 7 39 û ü û n/r û û ü Internally displaced populations or refugees 9 50 ü û û ü û ü ü Poor 9 50 ü û ü ü û û ü Periurban and slum populations 8 44 ü û ü ü û û ü Males 4 22 ü û û n/r û û ü Persons with disabilities 4 22 ü û û n/r û û ü aAfghanistan, Iraq, Morocco, Pakistan, Somalia, Sudan and Yemen: classified as priority countries due to poor performance in achieving targets of millennium development goals 4A and 5A. ü= available; û= not available; n/r = no response received from the country. Table 6 Commodity security in family planning (FP) services in all 18 respondent countries and the 7 priority countries of the Eastern Mediterranean Region, 2009 Item All countries Priority countriesa No. % Afg Irq Mor Pak Som Sud Yem FP commodity security is ensured through: Well-functioning contraceptive logistics management information system 15 83 ü ü ü ü ü û ü Data-based planning by the government 11 61 ü û ü ü û û ü Effective supply chain management of all contraceptive commodities throughout the country 12 67 ü û ü ü û û ü Effective supply chain management of all contraceptive commodities in certain parts of the country 14 77 n/r ü n/r n/r û û n/r aAfghanistan, Iraq, Morocco, Pakistan, Somalia, Sudan and Yemen: classified as priority countries due to poor performance in achieving targets of millennium development goals 4A and 5A. ü= available; û= not available; n/r = no response received from the country. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 917 the contraceptive prevalence for use of modern methods in these countries is reported to be low (16%, 22%, 6% and 13% respectively) [1]. This calls into question the scale and quality of implementation of acknowledged best practices in these countries. For in- stance, FP services may be included in basic health benefit packages at primary health care level and integrated into other vertical programmes such as child care or sexually transmitted infections/ HIV prevention, but not actually ac- cessible to clients due to different rea- sons, such as deficiency in availability of qualified and/or motivated personnel, lack of contraceptive commodities or insufficient empowerment and social support to women. Similarly, although respondent countries widely acknowl- edged the existence of successful prac- tices in FP programmes, according to 2008 estimates, on average only 31.1% of married women in EMR were us- ing modern contraceptive methods to avoid unwanted pregnancy, which corresponds to a regional average total fertility rate as high as 4.2 children per woman of reproductive age (15 to 49 years) [13]. The results of our survey add to the scarce published literature on the availability of best FP policies and pro- grammes in the EMR. Some countries, such as Egypt, Jordan, Kuwait, Islamic Republic of Iran, Libya, Palestine and Tunisia, are champions both in imple- menting successful FP programmes and in achieving favourable indicators for contraceptive prevalence and ma- ternal and under-5 child mortality rates. However, such achievements will be difficult to observe in the MDG-priority countries unless existing constraints in terms of political commitment and bottlenecks in health service systems are addressed simultaneously. In their global assessment of the health sys- tem and policy environment for ma- ternal, newborn and children health, the Countdown working group on Table 7 Promotion of family planning (FP) services in all 18 respondent countries and the 7 priority countries of the Eastern Mediterranean Region, 2009 Item All countries Priority countriesa No. % Afg Irq Mor Pak Som Sud Yem FP is actively promoted through: Effective social marketing of FP methods 12 67 ü ü ü ü û ü ü Community education, including wide distribution of quality education and information materials 14 78 ü ü ü ü û û ü Community mobilization efforts 11 61 ü û ü ü ü û ü Public-private partnership is ensured in: Community education activities 14 78 ü ü ü ü û ü ü Community outreach services 8 44 ü ü n/r ü û û ü Service provision programmes 8 44 ü û n/r ü û ü ü aAfghanistan, Iraq, Morocco, Pakistan, Somalia, Sudan and Yemen: classified as priority countries due to poor performance in achieving targets of millennium development goals 4A and 5A. ü= available; û= not available; n/r = no response received from the country. Table 8 Programme monitoring and evaluation and planning of family planning (FP) services in all 18 respondent countries and the 7 priority countries of the Eastern Mediterranean Region, 2009 Item All countries Priority countriesa No. % Afg Irq Mor Pak Som Sud Yem Evidence-based indicators are selected to monitor and evaluate FP programme 14 78 ü ü ü ü û ü ü Health management information system regularly collects and analyses FP programme data 13 72 ü ü ü ü û ü ü Decisions on changes in FP programme are made based on the evidence and analysis of information obtained from: Health management information system 11 61 ü ü ü n/r û û ü FP programme evaluation 13 72 ü ü ü ü û û ü Special national and subnational studies 14 77 ü ü ü ü û ü ü aAfghanistan, Iraq, Morocco, Pakistan, Somalia, Sudan and Yemen: classified as priority countries due to poor performance in achieving targets of millennium development goals 4A and 5A. ü= available; û= not available; n/r = no response received from the country. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 918 policy and health systems suggested that the prerequisites for successful im- plementation and eventual scale-up of evidence-based practices are: strong national leadership and effective gov- ernance; adequate and sustainable fi- nancial resources; integration of desired services; and availability of qualified workforce [14]. In order to promote the diffusion of successful FP programmes during the past decade WHO and its partners have focused on identifying, documenting and categorizing successful policy and programme practices in FP. Recently 10 essential components of successful FP practices have been crystallized and shared widely [12]. In addition, the Im- plementing Best Practices consortium was established in 2003, sponsored by WHO, the United Nations Popula- tion Fund (UNFPA) and the United States Agency for International Devel- opment (USAID) and represented by more than 23 international agencies, to advance documentation and diffu- sion of knowledge about best practices to improve reproductive health out- comes [15]. Further, a special opera- tions research methodology called “A guide for fostering change to scale up effective health services” was developed and promoted by WHO and USAID to facilitate the process of scaling-up successful reproductive health practices in countries [16]. It is important to mention that suc- cessful experience from countries where relevant operations research activities have been applied to increase contra- ceptive uptake and/or reduce maternal mortality have reported certain com- mon factors that enabled success. These include: i) effective leadership and stewardship capacity of top national health authorities for coordination, collaborative planning and monitoring implementation of interventions; ii) a comprehensive approach in covering a range of key contributing factors, such as: policies, regulations and guidelines; health systems; health education and accreditation; health information and commodity security; iii) transparency and wide stakeholder involvement both from public and private sector; iv) continuous technical support and motivation of health organizations and personnel; and v) local capacity-build- ing, close monitoring and supportive su- pervision [17–22]. Increasing the use of modern FP methods has a wide range of potential benefits including: improved maternal and child health; enhanced empowerment of women by reducing the burden of excessive childbearing; reduction of poverty; contribution to environmental sustainability; and significant monetary savings in reduced costs of health care, social welfare and environmental sustainability [23]. Additional research is needed to fur- ther elucidate the scale and quality of the existing best FP practices identified in individual countries of the WHO EMR. Countries lagging in achieving MDG goals 4A and 5A should invest both in scaling-up identified FP best practices, as well as adapting to the local context practices that have proved successful elsewhere in the Region. Priority should be given to scaling-up: integration of quality and acceptable FP counselling and methods into primary and com- munity health care services; training and deployment of a sufficient skilled workforce; ensuring FP commodity security; and introducing performance motivation packages for qualified FP service providers. In conclusion, the results of the study enable us to recommend that identified best policy and programme practices in family planning should: advise planning and implementation of FP programmes in the Region; and facilitate closer col- laboration and coordination of efforts of stakeholders concerned to increase the prevalence of modern contracep- tive use in order to improve maternal and child health in the countries of the WHO EMR. References 1. The work of WHO in the Eastern Mediterranean Region. Annual report of the Regional Director. Cairo, World Health Organiza- tion, Regional Office for the Eastern Mediterranean, 2008. 2. Millennium Development Goals indicators. The official Unit- ed Nations site for the MDG indicators. Untied Nations [on- line] (http://millenniumindicators.un.org/unsd/mdg/Host. aspx?Content=Indicators/OfficialList.htm, accessed 1 July 2012). 3. Health benefits of family planning. Geneva, World Health Or- ganization, Unit of Family Planning and Population, Division of Reproductive Health, 1995. 4. Bullough C et al. Current strategies for the reduction of mater- nal mortality. International Journal of Obstetrics and Gynaecol- ogy, 2005, 112:1180–1188. 5. Bulatao RA, Ross JA. Which health services reduce maternal mortality? Evidence from ratings of maternal health services. Tropical Medicine and International Health, 2003, 8:710–721. 6. Cleland J et al. Family planning: the unfinished agenda. Lancet, 2006, 368:1810–1827. 7. Collumbien M, Gerressu M, Cleland J. Non-use and use of ineffective methods of contraception. In Ezzati M, et al., eds. Comparative quantification of health risks: global and regional burden of disease attributable to selected major risk factors. Geneva, World Health Organization, 2004:1255–1320 (NLM/ WA 105). 8. Ross JA, Blanc A. The contribution of family planning in reduc- ing maternal mortality. Paper presented at the International Conference on Family Planning Research and Best Practices, Uganda, 15–18 November, 2009:210–211. 9. Huang W, Ronsmans C. Systemic review of childbearing patterns and maternal mortality. Paper presented at the International Conference on Family Planning Research and Best Practices, Uganda, 15–18 November, 2009:213–214. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 919 10. Singh S et al. Adding it up: the benefits of investing in sexual and reproductive health care. Geneva, United National Population Fund/The Allan Guttmacher Institute, 2004. 11. Mavranezouli I. Health economics of contraception. Best Prac- tice and Research. Clinical Obstetrics and Gynaecology, 2009, 23:187–198. 12. Ten essential components of family planning programmes. Knowledge for health [online] (http://www.k4health.org/ topics/ten-elements-family-planning-success, accessed 15 July 2012). 13. Making pregnancy safer database. Cairo, World Health Organi- zation Regional Office for the Eastern Mediterranean [online] (http://www.emro.who.int/mps/pdf/statistics_2008.pdf, accessed 1 July 2012). 14. Countdown Working Group on Health Policy and Health Systems. Assessment of the health system and policy environ- ment as a critical complement to tracking intervention cover- age for maternal, newborn, and child health. Lancet, 2008, 371:1284–1293. 15. The Implementing Best Practices (IBP) Initiative. Implementing best practices in reproductive health [online] (http://www. ibpinitiative.org/, accessed 15 July 2012). 16. Implementing Best Practices Consortium. A guide for foster- ing change to scale up effective health services. Geneva, World Health Organization/United States Agency for International Development, 2007. 17. Simons R, Fajas P, Ghiron L. Scaling up health service delivery: from pilot innovations to policies and programmes. Geneva, World Health Organization, 2007 (NLM/WA 550). 18. Solomon M et al. Creating a compendium of best practices: con- tributions of the IBP team. Paper presented at the International Conference on Family Planning Research and Best Practices, Uganda, 15–18 November, 2009: 308. 19. Bukusi VA et al. The success story of implementing best practice in Kenya. Paper presented at the International Conference on Family Planning Research and Best Practices, Uganda, 15–18 November, 2009: 224-225. 20. Hoemeke L et al. Repositioning family planning in Rwanda: how a taboo topic became a priority number one, and a success story. Paper presented at the International Conference on Family Planning Research and Best Practices, Uganda, 15–18 Novem- ber, 2009:223–224. 21. Brambila C et al. Getting results used: evidence from reproduc- tive health programmatic research in Guatemala. Health Policy and Planning, 2007, 22:234–245. 22. Ruiz-Rodríguez M, Wirtz VJ, Nigenda G. Organizational ele- ments of health service related to a reduction in maternal mor- tality: the cases of Chile and Colombia. Health Policy, 2009, 90(2):149–155. 23. Cleland J. Contraception in historical and global perspective. Best Practice and Research. Clinical Obstetrics and Gynaecology, 2009, 23:165–176. Family planning: a global handbook for providers (2011 update) Family planning : a global handbook for providers (2011 update) is an essential resource for health-care professionals providing contraceptive methods. The handbook is the successor to The essentials of contraceptive technology, first published in 1997 by the Center for Communication Programs at Johns Hopkins Bloomberg School of Public Health. Experts from around the world have contributed to the development of the handbook, and many major international organizations and professional organizations working in family planning have endorsed and adopted this guidance. It is one of WHO’s “Family Planning Cornerstones”, a companion to the Medical eligibility criteria for contraceptive use, the Selected practice recommendations for contraceptive use and the Decision-making tool for family planning clients and providers. This publication is also available in Arabic and French. Further information about this and other WHO publications is available at: http://www.who.int/publications/en/ EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 920 HIV infection and related risk behaviours among female sex workers in greater Cairo, Egypt I.A. Kabbash,1 I. Abdul-Rahman,2 Y.A. Shehata 3 and A.A-R. Omar 1 ABSTRACT Although illegal in Egypt, prostitution exists. The prevalence of HIV infection among female sex workers (FSWs) in Cairo is not precisely known. This cross-sectional study investigated the high-risk behaviour for HIV infection and HIV prevalence among FSWs in greater Cairo. A total of 431 FSWs were interviewed about their sexual history with paid and unpaid partners, condom use and risky behaviour for HIV infection; all were tested for HIV. Use of alcohol and drugs was reported by 39.9% and 49.0% of the women respectively; 37.6% only used such substances while with a client. Male condoms were known by 72.6% but their use in the previous month was low (32.8%) and only 22.4% had used one with their last client. The main reasons for not using condoms were not thinking of it (40.6%) and client refusal (20.5%). All the women tested negative for HIV infection. The high-risk behaviour of many FSWs necessitates intervention programmes to reduce their risk of HIV infection. 1Department of Public Health and Community Medicine, Faculty of Medicine, Tanta University, Tanta, Egypt (Correspondence to I.A. Kabbash: iakabbash@yahoo.com, kabbash-ia@maktoob.com). 2National AIDS Control Programme, Ministry of Health and Population, Cairo, Egypt. 3Department of Public Health and Community Medicine, Faculty of Medicine, Menoufiya University, Menoufiya, Egypt. Received: 18/05/11; accepted: 21/08/11 صرم ،ىبركلا ةرهاقلا في اياغبلا ينب ابه ةلصلا تاذ تايكولسلاو زديلإا سويرفب ىودعلا رمع ميحرلا دبع ءماسأ ،ةتاحش اطع سراي ،نحمرلا دبع بايهإ ،شابك ليع ميهاربإ هذه صيقتستو .ةرهاقلا في اياغبلا ينب ةقدب زديلإا سويرفب ىودعلا راشتنا لدعم فرعي لاو .هرظيح نوناقلا نأ مغر ءاغبلل ةسرامم صرم في دجوت :ةـصلالخا ىرجأ ،ًاّيِغَب 431 ةساردلا تلمش دقو .ةرهاقلا في اياغبلا ينب زديلإا سويرف راشتنا لدعمو زديلإا سويرفب ىودعلل ةيلاع رطاخمب فوفحلما كولسلا ةساردلا ىودعلل رطاخلماب فوفحلما كولسلاو يركذلا لزاعلا مادختسا لوحو ،رجأ نودب وأ رجأ ءاقل نئابز عم نيهدل ةيسنلجا قباوسلا لوح تلاباقم نهعم نوثحابلا نأو )%49( تاردخلما وأ )%39.9( لوحكلا يطاعت نع اياغبلا تغلبأو .زديلإا سويرف فشكل تارابتخا نهلك نله نوثحابلا ىرجأ ماك ،زديلإا سويرفب قباسلا رهشلا للاخ هل نهمادختسا لدعم نأ لاإ ،نهنم %72.6 ىدل ًافورعم يركذلا لزاعلا ناكو .طقف نهنئابز دوجو ىدل نامدلإا داوم ينطاعتي نهنم %37.6 مدع وه يركذلا لزاعلا مادختسا مدعل سييئرلا ببسلا ناك دقو .نله نوبز رخآ لَبِق نم همادختسا لىإ نشرأ طقف نهنم %22.4 نأ ماك ،)%32.8( ًاضفخنم ناك تايبلس تارابتخلال نعضخ تيلالا ءاسنلا عيجم تناك دقو .)%19.3( يركذلا لزاعلا ةيهاركو ،)%20.5( همادختسلا نوبزلا ضفرو ،)%40.6( هب يركفتلا .زديلإا سويرفب ىودعلا رطامخ فيفختل لخدتلا جمارب دايجإ متيح اياغبلا نم يرثكلا ىدل ةيلاع رطاخمب فوفحلما كولسلا نأ نوثحابلا ىريو .زديلإا سويرفل Infection à VIH et comportements à risque associés chez les professionnelles du sexe dans le Grand Caire (Égypte) RÉSUMÉ Bien qu'illégale en Égypte, la prostitution existe. La prévalence de l'infection à VIH chez les professionnelles du sexe au Caire n'est pas connue précisément. La présente étude transversale a étudié les comportements associés à un risque important de contracter une infection à VIH et la prévalence de ce virus chez les professionnelles du sexe en activité dans le Grand Caire. Au total, 431 professionnelles du sexe ont été interrogées au sujet de leurs antécédents en matière de rapports sexuels avec des partenaires rémunérant ou non leurs services, d'utilisation du préservatif et de comportements à risque pour l'infection à VIH ; toutes les personnes interrogées ont été soumises à un test de dépistage du VIH. Parmi celles-ci, 39,9 % déclaraient consommer de l'alcool et 49,0 % de la drogue ; 37,6 % en consommaient uniquement avec leurs clients. Les préservatifs masculins étaient connus de 72,6 % d'entre elles mais elles étaient peu nombreuses (32,8 %) à en avoir utilisé au cours du mois précédent et seulement 22,4 % en avaient utilisé un avec leur dernier client. Les principales raisons de leur absence d'utilisation des préservatifs étaient l'oubli (40,6 %) et le refus du client (20,5 %). Toutes les professionnelles du sexe de l'étude ont obtenu des résultats négatifs dans le dépistage du VIH. Le comportement à haut risque de nombreuses professionnelles du sexe appelle des programmes d'intervention afin de réduire leur risque d'infection à VIH. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 921 Introduction Worldwide, commercial sex workers are considered one of the most at-risk groups, including for HIV infection [1]. Sex workers frequently have insufficient access to adequate health services, male and female condoms and other preven- tive services [2]. The extent of HIV infection among female sex workers (FSWs) in Greater Cairo, Egypt, is not precisely known due to the fact that few field studies have been undertaken and therefore a scarcity of data. Conduct- ing studies among female sex workers can be problematic because of the stigma related to sex work. That said, the prevalence of HIV infection in Greater Cairo still appears to be lower than that in other industrialized cit- ies [3]. Worldwide studies estimate that nearly 5 million people are being infected with HIV every year [4,5] and about half of those currently infected with HIV are females [5]. Rates of HIV infection are higher among those involved with sex work than in most other populations [2]. The number of FSWs has been increasing over the past 2 decades mainly for economic reasons [5–7]. Many studies indicate that the risk of getting infected with HIV is high among FSWs and their clients [8]. Paid sex remains an important fac- tor in many of the HIV epidemics in western, central and eastern Africa. It is estimated that 32% of new HIV infections in Ghana, 14% in Kenya and 10% in Uganda are linked to sex work (HIV infection among sex workers, their clients, or their other sex partners) [9]. In Egypt, commercial sex is illegal but it does exist. This study aimed to determine the prevalence of HIV infec- tion among FSWs in Greater Cairo, Egypt, and identify the HIV-related risk behaviours among them. Methods Setting and subjects The study was conducted in Greater Cairo which includes 3 Governo- rates: Cairo, Giza and Qualiobia. This cosmopolitan area accommodates a population of nearly 14 million people [10]. Places of gathering of FSWs were identified through focus groups discussions with FSWs and with El-Shehab nongovernmental organization (NGO) working with FSWs. The target population was FSWs representing different categories as identified by 3 focus group discussions that were conducted before starting the field work. Each focus group included 6–8 sex workers recruited by El-Shehab organization. The focus group discus- sions identified the different categories of FSWs and the main places they gath- ered. These were: Street FSWs; FSWs at bars and night clubs; Dancers; FSWs at brothels; FSWs at coffeehouses; FSWs at hotels. The sample size was based on the calculation: n = Zα0.95 × [p × (1–p)]/d 2, where p = prevalence of HIV infection among FSWs, estimated to be 0.3% and d = expected margin of error (3%). Thus the total sample size was estimated at 234. We increased this number to 431 FSWs to ensure adequate representa- tion of all categories. A list of places where sex workers tended to gather was prepared based on information gained from focus group discussions and the experience of field work done by El-Shehab. One place for each category was randomly selected (a street, night club/bar, hair dresser where dancer gather to prepare for work, a brothel and a coffee shop). Sex work- ers at hotels were difficult to access as we did not succeed in getting approval from hotel authorities to conduct field work there, so they had to be omitted from in the study. Data collection and tools This work was conducted during the period from August 2009 to end of April 2010. Data were collected by a question- naire which was adapted from the ques- tionnaire published by family health international [11]. It was translated into Arabic and some modifications were made to make it suitable for the situa- tion of sex work in Egypt. Modifications were done based on the experience of the research team and in consultation with experts working previously with this group from El-Shehab. The ques- tionnaire included questions covering the following: • Sexual practices of FSWs • HIV-related risk behaviour • Knowledge and practice of FSWs of male condoms. A pilot study was conducted with a sample of sex workers who were not included in the main study in order to test the questionnaire and ensure its feasibility for the study. Data were collected through inter- view of the participants by 6 trained former sex workers who worked with El- Shehab outreach programme for harm reduction for HIV infection among sex workers in different regions of greater Cairo. The data collectors were directed to the locations to collect data from all those available at the time of interview. The process was repeated and another group of locations was randomly cho- sen and visited until the desired sample size was obtained. For each category the suitable timing for visiting the place was identified to ensure cooperation of the study participants and avoid interfering with their work needs. A suitable place, which offered comfortable conditions and confidentiality, was chosen by the data collectors to conduct the inter- view. The rate of refusal by the target population was 3%. Financial incentives were given to the sex workers who par- ticipated in the study to compensate for EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 922 the time lost during the interview. This incentive helped to minimize the rate for refusal. Blood samples were also taken from each participant for HIV testing carried out using an ELISA method. The blood samples were taken by female workers trained in blood sampling who accom- panied the data collector teams dur- ing the field work. None of the FSWs refused to give a sample. Analysis The collected data were organized, tabulated and statistically analysed us- ing SPSS, version 17. Ethical considerations Data collection was anonymous. Con- fidentiality was ensured whereby no data were linked to names or specific persons. No pressure of any kind was used to recruit the study participants. The study objectives and contents of the questionnaire were explained to the participants before starting data collection. Witnessed (by the data col- lectors), informed verbal consent was taken before the interview and blood sampling. The collected data were only used for research purposes according to the objectives of the study and for the benefit of the participants. Results A total of 431 FSWs were included in the study. Their mean age was 29.38 [standard deviation (SD) 9.48] years and median age 27 years. Almost 61% had attended school with a mean duration of 10 years. The majority (65.9%) were ever or currently married during the study period. Over half of the participants (58.7%) first had sex in exchange of money or some other benefits between the ages of 15 and 25 years, with a mean age of 22.28 (SD 6.66) years. The mean duration of work- ing as a sex worker was 6.32 (SD 6.42) years. Monthly income from sex work was reported as < US$ 100 by 29.2% of the women while 57.5% reported a monthly income ranging between US$ 100 and 500 (Table 1). Nearly 40% admitted drinking alcohol, 7% daily and 32.9% occasion- ally. Drug use was reported by 49% of the women, which was mostly hashish (96.2%), and 37.6% of drug use was reported to be associated with sexual practices, i.e. they used drugs only when with a client. Injecting drug use was reported by 5.6% in the last year. The majority (61%) of the women reported freedom to choose clients while 8.1% never had this chance. Over one-third (35.3%) reported experiencing sexual violence , either the threat of violence to have sex or violence during sex, which was mostly physical as reported by 73.7%. Abnormal vaginal discharge in the last year was reported by 76.1% of the women while 15.5% reported suffer- ing of genital ulcers (Table 2). Having unpaid sexual partners was reported by 59.2% of the women; these were husbands (30.6%) or boyfriends (21.6%). Of the participants with un- paid partners, 38.4% reported that their partners had a spouse or other sexual partners. The number of times to have had sex with their unpaid partner in the month before data collection was 1–5 times for 48.6% of the women while 22.8% reported 10 or more times. Of the 255 women with unpaid partners, 29.4% reported abuse was used for sexual relations by the unpaid partner, which was mainly physical (74.7%). Use of drugs by unpaid partners was reported by 70.6% of the women, 10.6% of whom injected drugs (Table 3). Most of studied sex workers (72.6%) knew about male condoms. However, only 22.4% had used them with the last client and 32.8% had used them in the past month. Their use was at the sug- gestion of the sex worker, as reported by 51.5% of the women who used a condom with their last client, while 37.1% reported that the client asked for condom use. The main reasons for not using male condom were not thinking about it (40.6%), client refused (20.5%), Table 1 Sociodemographic characteristics of the female sex workers (n = 431) Characteristic Value Age (years) Mean (SD) 29.38 (9.48) Median 27.00 Attended school Yes [No. (%)] 262 (60.8) Mean (SD) no. of years at school 10.18 (3.24) Median no. of years at school 10 Married, now or previously [No. (%)] 283 (65.7) Mean (SD) age at first selling sex (years) 22.28 (6.66) Duration of sex work (years) Range 1–35 Mean (SD) 6.32 (6.42) Median 4.0 Monthly income (US$) [No. (%)] < 100 126 (29.2) 100– 158 (36.6) 200– 90 (20.9) 500+ 45 (10.4) Unidentified 12 (2.8) SD = standard deviation. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 923 do not like it (19.3%) and using another contraceptive method (13.5%). Con- dom use during every a sexual relation during last month was reported by only 7.7% while 67.2% reported never using condoms in the previous month (Table 4). Condom use with unpaid sexual partners was reported by 24.2% of the women which was at the recommenda- tion of the sex worker in 58.8% of cases and of the sexual partner in 29.4%. Use of condom was “never thought of it” by 37.7% of the women, while 17.8% had no reason, 17.3% did not like condoms and for 15.2% the unpaid sexual part- ner refused. Most sex workers (76.3%) reported never using a condom with unpaid sexual partners in the last month while only 4.0% used one every time (Table 5) The majority of the women identi- fied pharmacies as the place for con- dom purchase (74.3%). Almost all the women (97.9%) reported never using a condom more than once and 90.6% did not use lubricants with condoms. Con- dom tear was experienced by 12.8% of the FSWs; the majority of these women (61.8%) continued sexual relations with a new condom and 29.1% continued without a condom. Having sex during menses was reported by 29.7% of the women; 61.7% of those doing this did so without a condom. On the other hand, 35.1% reported ever using condoms while having sex during menstruation. Only 7.7% reported practising group sex and of these 21.2% reported using condoms while having group sex (Ta- ble 6). All the FSWs tested negative for HIV infection, giving a 0% prevalence of HIV infection. Discussion Our results show that the behaviours of FSWs in Cairo put them at risk of HIV infection. These include: alcohol intake, drug abuse, physical violence, inadequate use of protective measures, questionable self-efficacy in using con- doms, and having sex during menstrua- tion. Although alcohol intake among Egyptians is extremely low (less than 0.2% of males, and less than 0.01% of females) [12], alcohol intake was preva- lent among FSWs in this study; nearly 40% reported drinking alcohol. At the same time, drug use was reported by nearly a half of the FSWs, most com- monly smoking hashish (cannabis), which parallels the fact that hashish is the most commonly abused drug in Egypt [13]. Injecting drugs was rela- tively low (5.6%). This is comparable to findings of a study in London showing drug injection to be 13% among FSWs [14]. Most of these FSWs had com- mercial and non-commercial (unpaid) partners and some of these unpaid part- ners abused drugs by injection [14,15]. Nearly 60% of FSWs in our study were having sex with unpaid partners 70.6% of whom abused drugs and 10.6% were injecting drugs. This exposes FSWs to an additional HIV-risk, as drug intoxica- tion alters the users’ mental status and judgement, which, in turn, can increase the likelihood that they will engage in high-risk sexual behaviours. Violence against FSWs is wide- spread and it is committed by clients, controllers, managers of sex work es- tablishments and intimate partners [2]. Table 2 Distribution of sex workers in relation to some risky behaviour Variable No. (n = 431) % Alcohol intake in the last month Daily 30 7.0 At least once weekly 70 16.2 2–3 times monthly 72 16.7 None 259 60.1 Drug use Used drugsa 211 49.0 Injecting drugs in the last year 24 5.6 Use of drugs associated with sex 162 37.6 Freedom to choose clients Always 263 61.0 Most of the time 50 11.6 Sometimes 52 12.1 Rarely 31 7.2 Never 35 8.1 Experienced sexual violenceb Many times 26 6.0 Sometimes 28 6.5 Few times 98 22.7 Never 279 64.7 Type of violence (n = 152) Physical 112 73.7 Verbal 7 4.6 Both 33 21.7 Sexual health issues in the past 12 months Unusual vaginal discharge 328 76.1 Genital ulcers 67 15.5 a96.2% of reported drug use was cannabis. bEither the threat of violence to have sex or violence during sex. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 924 The same finding was obtained in our study where more than one-third of the studied sex workers reported hav- ing sex under the threat of abuse or abusive sex, which was mostly physical. Violence is associated with unpro- tected sex and increases the risk of HIV infection due to vaginal trauma and lacerations resulting from the use of force [2,15–18]. Sexually transmitted infections (STIs) are common among FSWs as shown by different studies [19–21]. However, the prevalence of such infec- tions differs from one study to another based on the circumstances of sex work organization and whether it is legal or illegal [14,15]. In the present study abnormal vaginal discharge suggestive of STIs was reported by the majority of the FSWs as well as genital ulcers, at a lower prevalence. Individuals who are infected with STIs are 5–10 times more likely than uninfected individuals to acquire or to transmit HIV through sexual contact. The breaking of the genital tract lining creates a portal of entry for HIV, and HIV infected indi- viduals with other STIs are more likely to shed HIV in their genital secretions [19]. Condoms are one of the most ef- fective methods to prevent sexual transmission of HIV infection and other STIs. They should be readily available for sex workers and their clients [2]. Studies among FSWs show different rates of condom use in different coun- tries. Condoms were always used by 93.4% of FSWs in Santiago, Chile [22], 44% in Calcutta, India [1] and 98% in London, UK [14]. Many reasons were given for not using a male condom in- cluding never having thought of it, client refusal, dislike of condoms and using an- other contraceptive method. Condom use among the women in our study was relatively lower with their unpaid partners. FSWs did not perceive the risk of HIV infection from their intimate unpaid partners in spite of our finding that these partners usually abused drugs and had other sexual partners. The same finding of low condom use with unpaid partners has been documented in other similar studies. [14,20,23]. Among a minority of the sex workers in our study reporting use of a condom with the last client, self-efficacy of condom use was questionable among those who experienced condom tear as almost Table 3 Distribution of sex workers by sexual practices with unpaid partners Variable No. % Unpaid sexual partners during last 6 months (n = 431) Husband 132 30.6 Boyfriend 93 21.6 Multiple boyfriends 28 6.5 Girlfriend 2 0.5 None 176 40.8 Unpaid partner has another sexual partner (spouse or other) (n = 255) 98 38.4 Average daily number of sexual relations last month with unpaid partner (n = 255) 0 12 4.7 1– 124 48.6 5– 61 23.9 10– 31 12.2 > 15 27 10.6 Experienced sexual violence a (n = 255) Many times 17 6.7 Sometimes 20 7.8 Few times 38 14.9 Never 180 70.6 Type of violence (n = 75) Physical 56 74.7 Verbal 3 4.0 Both 16 21.3 Unpaid partner’s drug use Uses drugs (n = 255) 180 70.6 Injects drugs (n = 180) 19 10.6 aEither the threat of violence to have sex or violence during sex. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 925 Table 4 Distribution of sex workers in relation to their experience with male condom during sex with clients Variable No. % Ever heard about male condom (n = 431) 313 72.6 Use of condoms with clients Used a condom in the past month (n = 313) 102 32.6 Used a condom with last client (n = 313) 70 22.4 Person recommending male condom use (n = 70) Sex worker 36 51.5 Client 26 37.1 Both 8 11.4 Reasons for not using male condom (n = 243)a Unavailable 11 4.5 It is expensive 2 0.8 Client refusal 50 20.5 Don’t like it 47 19.3 Using other contraceptive methods 33 13.5 Not necessary to use it 11 4.5 Never thought of it 99 40.6 Don’t know 13 5.3 Other reasons 31 12.7 Frequency of male condom use in the last month with clients (n = 311) Every time 24 7.7 Almost all the time 28 9.0 Sometimes 50 16.1 Never 209 67.2 aNot mutually exclusive. Table 5 Distribution of sex workers in relation to their experience with male condom during sex with unpaid sexual partners Variable No. % Ever used a condom with unpaid sexual partner (n = 252) 61 24.2 Used a condom during the last sexual act with unpaid sexual partner (n = 253) 34 13.4 Person recommending condom use (n = 34) Sex worker 20 58.8 Sexual partner 10 29.4 Both 4 11.8 Reason for not using condom (n = 191)a Condom unavailable 2 1.0 Expensive 0 0.0 Partner refused 29 15.2 Don’t like it 33 17.3 Using other contraceptive methods 23 12.0 Not necessary to use it 9 4.7 Did not consider using it 72 37.7 Don’t know 34 17.8 Other reasons 24 12.6 Frequency of male condom use during last month ith unpaid sexual partner (n = 253) Every time 10 4.0 Almost all the time 16 6.3 Sometimes 34 13.4 Never 193 76.3 aNot mutually exclusive. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 926 of condom use, may be explained by the low prevalence of HIV infection in FSWs’ communities and clients. In a study in India, which covered differ- ent districts, the prevalence of HIV infection showed a large inter-district variation ranging from 2% to 38% which correlated with the prevalence of infection among clients in the differ- ent districts [15]. The zero prevalence of HIV infec- tion among our participants indicates a low prevalence of HIV infection in spite of the observed risky behaviours for infection. This indicates that the potential HIV epidemic may be in the first stages among this most at-risk population and it could be a golden opportunity for intervention pro- grammes to be implemented to limit its spread among sex workers and their clients. Limitations of the study There are two noteworthy limitations regarding the present study. The first concerns the nature of this research; this study focused on a very sensitive subject, with much stigma attached to it, which has not previously been studied in Egypt. Clearly, this represents a chal- lenging task for research. As such, the research was limited in scope and aimed mainly to generate knowledge for fur- ther research. The second limitation has to do with the generalizability of the findings beyond the cases studied. Give the very specific group of participants the extent to which the results can be generalized to a wider population is limited. Conclusion FSWs in Greater Cairo are exposed to different factors that put them at high risk of HIV infection. These factors include: alcohol intake, drug abuse, physical violence, inadequate use of protective measures, question- able self-efficacy in using condoms, Table 6 Distribution of studied female sex workers in relation to their knowledge about male condoms Variable No. % Places where condoms can be obtained (n = 424) Pharmacies 315 74.3 Supermarkets 1 0.2 Clinics 4 0.9 Hospitals 14 3.3 Bars/hotels 3 0.7 From a friend 15 3.5 Nongovernmental organizations 23 5.4 Don’t know 100 23.6 Have used a condom more than once (n = 429) Yes 9 2.1 No 420 97.9 Have used a lubricant with condoms (n = 427) Yes 40 9.4 No 387 90.6 Have experienced condom tear (n = 429) Yes 55 12.8 No 374 87.2 Response to condom tear (n = 55) Stopped sexual act 5 9.1 Continued after using a new condom 34 61.8 Continued without condom 16 29.1 Had sex during menses (n = 431) Yes 128 29.7 No 303 70.3 Used condom for sex during menses (n = 128) Yes 45 35.1 No 79 61.7 No answer 4 3.1 Group sex a Have had group sex (n = 431) 33 7.7 Used condom during group sex (n = 33) 7 21.2 aHaving sex with more than 1 partner at the same time. one-third continued sexual relations without a new condom. Similar findings were reported in a study conducted in Singapore [21]. An additional risk factor for HIV infection, as shown by a study on FSWs in Thailand, is having sex during menses [24]. With the shedding of the uterine mucosal lining, numerous blood vessels are open to receive the virus if the male partner is infected with HIV [24]. Sex during menses was reported by almost one-third of the FSWs in our study and almost two- thirds of them did so without using condoms. The prevalence of HIV infection among FSWs in Greater Cairo was 0%. A similar finding was found among FSWs in Santiago, Chile, which was attributed to high compliance of con- dom use among them [22]. In our study, however, this low prevalence of HIV infection, despite a low rate طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 927 and an additional risk of having sex during menstruation. These risk fac- tors urgently require more focused HIV-related health research, and highly specific long-term intervention programmes. Acknowledgements The authors express their thanks and appreciation to Dr Mervat El-Gue- neidy, Professor of Neuropsychiatric Nursing at the University of Alexandria References 1. Sarkar K et al. Epidemiology of HIV infection among brothel- based sex workers in Kolkata, India. Journal of Health, Popula- tion, and Nutrition, 2005, 23:231–235. 2. UNAIDS guidance note on HIV and sex work. Geneva, Joint United Nations Programme on AIDS (UNAIDS), 2009–2012. 3. AIDS epidemic update, 2009. Geneva, Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Or- ganization, 2009. 4. Abu-Raddad L et al. Characterizing the HIV/AIDS epidemic in the Middle East and North Africa: time for strategic action. Wash- ington DC, World Bank, 2010 (http://www-wds.worldbank. org/external/default/WDSContentServer/WDSP/IB/2010/ 06/04/000333038_20100604011533/Rendered/PDF/5488 90PUB0EPI11C10Dislosed061312010.pdf, accessed 12 August 2012). 5. AIDS epidemic update 2005. Geneva, Joint United Nations Programme on HIV/AIDS and World Health Organization, 2005. 6. Buckingham RW, Webb N. Condom use among the female sex worker population in Thailand. International Journal of STD & AIDS, 2004, 15:210–211. 7. 100% Condom Use Programme in entertainment establishments. Manila, World Health Organization Regional Office for the Western Pacific, 2000. 8. Wolffers I, van Beelen N, Brussa L. Violence, repression and other health threats. Sex workers at risk. Research for sex work, 2001, 4:1–32 9. UNAIDS Report on the global AIDS epidemic 2010. Geneva, Joint United Nations Program on HIV/AIDS (UNAIDS), 2010. 10. Rohaa SM, Hafez Y. Monitoring urbanization growth in Cairo City. Journal of Engineering and Applied Sciences, 2002, 49:667– 679. 11. Amon J et al. Behavioral surveillance surveys: Guideline for re- peated behavioral surveys in populations at risk for HIV. Durham, NC, Family Health International, 2000:210–230. 12. World Health Organization. WHO global status report on alcohol 2004. Eastern Mediterranean for Region. Country profiles: Egypt (http://www.who.int/substance_abuse/publi- cations/en/egypt.pdf, accessed 10 July 2012). 13. Illicit drug market in Greater Cairo. Cairo, United Nations Office for Drug Control and Crime Prevention, Regional Office for the Middle East and North Africa, 2001 (Technical report on drugs and crime in North Africa and the Middle East). 14. Ward HD et al. Prostitution and risk of HIV: female prostitutes in London. British medical journal, 1993, 307:356–358. 15. Ramesh BM et al. Determination of HIV prevalence among female sex workers in four south Indian states analysis of cross-sectional surveys in twenty-three districts. AIDS (London, England), 2008, 22(Suppl. 5):S35–S44. 16. Mahfoud Z et al. HIV/AIDS among female sex workers, inject- ing drug users and men who have sex with men in Lebanon: re- sults of the first biobehavioral surveys. AIDS (London, England), 2010, 24(Suppl. 2):S45–S54. 17. Shannon K, Csete J. Violence, condom negotiation, and HIV/ STI risk among sex workers. Journal of the American Medical As- sociation, 2010, 304(5):573–574. 18. Renzaho A, Pallotta-Chiarolli M. Commercial sex work, surviv- al sex, sexual violence and HIV/AIDS prevention in Arumeru District, Arusha Region of Tanzania. Open Tropical Medicine Journal, 2009, 2:27–38. 19. Da Ros CT, da Silva Schmitt C. Global epidemiology of sexu- ally transmitted diseases. Asian Journal of Andrology, 2008, 10:110–114. 20. Barrientos JE et al. Prevalencia del VIH, conocimientos sobre el SIDA, y uso del condón en trabajadoras sexuales de Santiago, Chile [HIV prevalence, AIDS knowledge, and condom use among female sex workers in Santiago, Chile]. Cadernos de saúde pública, 2007, 23:1777–1784. 21. Vall-Mayans M et al. Sexually transmitted Chlamydia trachoma- tis, Neisseria gonorrhoeae, and HIV-1 infections in two at-risk populations in Barcelona: female street prostitutes and STI clinic attendees. International Journal of Infectious Diseases, 2007, 11:115–122. 22. Bruckova M et al. Short report: HIV infection among com- mercial sex workers and injecting drug users in the Czech Republic. American Journal of Tropical Medicine and Hygiene, 2006, 75:1017–1020. 23. Wee S et al. Determinants of inconsistent condom use with female sex workers among men attending the STD clinic in Singapore. Sexually Transmitted Infections, 2004, 80:310–314. 24. Kilmarx PH et al. HIV-1 seroconversion in a prospective study of female sex workers in northern Thailand. AIDS (London, England), 1998, 12:1889–1898. for her help. We thank all the women who participated in our study and those who helped by referring us to others. We also thank El-Shehab or- ganization for its help in conducting this study. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 928 Tehran dentists’ self-reported knowledge and attitudes towards HIV/AIDS and observed willingness to treat simulated HIV-positive patients B. Khosravanifard,1 V. Rakhshan,2 M. Ghasemi,3 A. Pakdel,4 S. Baradaran-Eghbal,5 R. Sheikholeslami,5 T. Dadolahi- Sarab 5 and H. Rakhshan 5 ABSTRACT Dentists’ self-reported attitudes towards patients with HIV/AIDS might not reflect their actual behaviour. In this study 2 observers posed as HIV-positive patients and directly evaluated the behaviour of 300 dentists in Tehran, Islamic Republic of Iran. Two months later another observer interviewed the same dentists at their offices regarding AIDS-related knowledge and self-reported attitudes. Only 14.9% of dentists agreed to treat the simulated HIV-positive patients, 78.5% referred and 6.6% rejected them. Older age, longer work experience, graduation from a non-Iranian university and not having additional degrees were significantly related to adverse behaviours. Mean scores were 8.3 (SD 9.7) (out of 18) for knowledge and 17.5 (SD 7.1) (out of 39) for attitude. There were no significant correlations between dentists’ knowledge and attitude or between knowledge and behaviour, but there was a weak positive correlation between self-reported attitude and observed behaviour 1Department of Orthodontics; 2Department of Dental Anatomy and Morphology; 3Department of Periodontics; 4Department of Paediatric Dentistry; Dental Branch, Islamic Azad University, Tehran, Islamic Republic of Iran. 5Private Practice, Tehran, Islamic Republic of Iran (Correspondence to Vahid Rakhshan: vahid.rakhshan@gmail.com). Received: 24/03/11; accepted: 16/05/11 ظوحللما بيحترلاو ،نارهط في هسويرفب ىودعلاو زديلإا لوح مهسفنأ نع نانسلأا ءابطأ اهنع غلبي يتلا فقاولماو فراعلما زديلإا سويرفل ينيبايجإ ضىرلم ةيكامح تلااح ةلجاعمب ناشخر ديحم ،باراس يلهاداد هاجمات ،يملاسلإا خيش اضر ،لابقا ناردارب مايش ،لدكاب هفطاع ،يمساق دوممح ،ناشخر ديحو ،درف نياوسرخ مانبه فيو .ةيلعفلا متهايكولس عم هسويرفب ىودعلاو زديلإا ضىرم نم مهفقاوم لوح مهسفنأ نع نانسلأا ءابطأ اهمدقي يتلا تاغلابلا ىشماتت لا دق :ةـصلالخا .ةيملاسلإا ناريإ ةيروهجم ،نارهط في نانسأ بيبط 300 كولسل شرابم مييقت عم ،زديلإا سويرفل نايبايجإ مانهأ لىع يننثا ينبقارم ضرع مت ةساردلا هذه ىودعلاو زديلإا نع مهسفنأب اهنع نوغلبي يتلا مهفقاومو مهفراعم لوح متهادايع في متهاذ ءابطلأل ىرخأ ةلباقم ثلاث بقارم ىرجأ نيرهش ضيم دعبو مهنم %78.5 ماق مث ،زديلإا سويرفل ةيبايجلإا نوكايح نيذلا ضىرلما ةلجاعم لىع اوقفاو دق نانسلأا ءابطأ نم %14.9 نأ ينثحابلل حضتاو .هسويرفب ،لوطأ ةترفل ةيلمعلا ةبرلخاو ،رمعلاب مدقتلا يهف ةرئاضلا تايكولسلاب هب دَتْعُي ًاطابترا طبترت يتلا لماوعلا امأ .مهتلجاعم مهنم %6.6 ضفرو مهليوحتب غلب دقو ،فراعملل )18 لممج نم( )9.7 يرايعم فارحنا( 8.3 يطسولا زَرَلحا غلب دقو .ةيفاضإ تاداهش باستكا مدعو ،ةيناريإ يرغ ةعماج نم جرختلاو ينب لاو ،نانسلأا ءابطأ فقاومو فراعم ينب ابه ُّدَتْعُي تاطُبارت دهاشُت لمو ،)39 لممج نم( )7.1 يرايعم فارحنا( 17.5 فقاوملل ةبسنلاب يطسولا َزَرلحا .ةبقارلماب ظوحللما كولسلا ينبو فقاولما نع تياذلا غلابلإا ينب دهوش يذلا فيعضلا بيايجلإا طباترلا ءانثتساب ،مهكولسو مهفراعم Connaissances et attitudes autodéclarées des dentistes à Téhéran par rapport au VIH/sida et disposition observée à soigner les patients se présentant comme positifs pour le VIH RÉSUMÉ Il est possible que l'attitude autodéclarée des dentistes vis-à-vis des patients atteints du VIH/sida ne corresponde pas à leur comportement en situation réelle. Dans le cadre de cette étude, deux observateurs se sont fait passer pour des patients séropositifs pour le VIH et ont directement évalué le comportement de 300 dentistes de Téhéran (République islamique d'Iran). Deux mois plus tard, un autre observateur a interrogé les mêmes dentistes dans leur cabinet au sujet de leurs connaissances et de leur attitude autodéclarées à propos du sida. Seuls 14,9 % des dentistes ont consenti à traiter les patients ayant déclaré une séropositivité fictive, 78,5 % les ont orientés vers d'autres confrères et 6,6 % ont refusé de les soigner. Un âge plus avancé, une expérience professionnelle plus longue, des études suivies dans une université non iranienne et le fait de ne pas avoir de diplômes supplémentaires étaient des facteurs significativement associés à des comportements de refus. Les scores moyens étaient de 8,3 (E.T. 9,7) (sur 18) pour les connaissances et de 17,5 (E.T. 7,1) (sur 39) pour l'attitude. Aucune corrélation significative n'a été observée entre les connaissances et les attitudes des dentistes ni entre les connaissances et le comportement, mais une faible corrélation positive a été retrouvée entre l'attitude autodéclarée et le comportement observé. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 929 Introduction Dental professionals are at a high risk for cross-infection with HIV [1] and may therefore avoid treating HIV-positive dental patients [2]. This has serious pub- lic health consequences. For instance, in 60% to 70% of HIV-positive individuals oral symptoms of HIV infection are the first signs of the syndrome appearing and these can be used as appropriate diagnostic criteria in the detection of AIDS [3]. Additionally, healthy asymp- tomatic HIV-positive dental patients may conceal their condition in order to receive appropriate treatment, which may have serious consequences such as increased risk of cross-contamination [4]. The attitudes of medical health professionals towards treating patients living with HIV/AIDS seem to have changed over time and there appear to be a wide range of attitudes across different cultures [5–20]. There is also evidence that legal measures alone can- not change health professionals’ atti- tudes towards HIV patients. Improved awareness of dentistry personnel with respect to HIV carriers and the likeli- hood of infection might play a positive role in improving dentists’ attitude and confidence towards treating patients with HIV [5,15]. However, previous studies worldwide of dentists’ attitudes towards treating HIV/AIDS-positive patients were based only on dentists’ or patients’ declarations, which might not necessarily be in line with clini- cians' actual behaviours, and no studies have directly evaluated actual behav- iours/reactions (practices) of dentists during visits by HIV/AIDS-positive patients. Furthermore, no similar stud- ies have been performed on a sample of Tehran dentists. We aimed to evalu- ate Tehran dentists’ practices directly using simulated patients [21] and then assess their AIDS-related knowledge and their declared attitudes towards treating AIDS patients in face-to-face interviews. This was the first study of its kind and it was hoped that the infor- mation would be of value for alerting the authorities to allocate educational or infection control resources more effectively and to avoid discrimination against HIV patients. Methods This was a prospective, cross-sectional study carried out during 2008–09. Sample To obtain a test power of 0.85, and based on a pilot study, an estimated 300 dentists were required to partici- pate. A list of all dentists in Tehran was obtained from the Iranian Medical Council. Using simple random sam- pling from the alphabetic list of about 6100 Tehran dentists, and to allow for dropouts, a total of 411 dentists were randomly selected and visited. The final analysis was based on the 300 dentists (230 general practitioners and 70 specialists) who participated in both phases of the study. Data collection Observed behaviour with simulated patients Two female dental students aged 22 and 23 years old were recruited as ob- servers/simulated patients. An appoint- ment for an oral examination session was made at each selected dentist by visiting or calling the dentist’s office (each observer evaluated half the se- lected number of dentists). In the first session, just before the oral examination and/or when the patient’s profile was being recorded, the observer claimed to the dentist or the assistant, both orally and in written form, to be bloodborne HIV-positive. The observer also en- sured that the dentist was fully aware of the condition by telling him/her directly as well before any examina- tions. None of the clinicians suspected the observers were healthy researchers. The dentist’s reaction was documented and the behaviour was classified as: accepting (accepted the patient and scheduled the next session) or non- accepting (refused to admit the patient but referred to a specialist centre or rejected the patient without referring to other colleagues). Thirty dentists were randomly se- lected to be assessed again, in terms of their reactions toward AIDS patients, by another observer about 1 month later. The reliability of dentists’ behav- iours was determined as 89% (Cohen κ = 0.89, P = 0.001). Interviews using knowledge and atti- tude questionnaires After 2 months, a third observer called and visited the same dentists and asked to interview them face-to-face about their knowledge and attitudes using a structured questionnaire. The ques- tionnaire was specially designed for this study by a panel of experts. The follow- ing items were documented: demo- graphic data (age, sex); office location, based on income status categorized by Tehran municipality (Tehran south, west and east districts versus central and north district); office type (private office or general clinic with different de- partments); work experience; degrees obtained (Masters degree, degrees other than in the field of dentistry); country of graduation (Iranian versus foreign university); work experience; university staff member; continuing education courses (number of courses attended and time elapsed since last course). The knowledge section of the questionnaire included 18 questions regarding diagnostic criteria and man- agement of patients with HIV/AIDS (scored 1 or 0) [22–24]. The questions included the signs and symptoms of AIDS in dental patients, diagnosis of HIV and the latent period, methods of transmission, procedures for steriliza- tion and avoidance of transmission, etc. The scores were stratified into 3 groups by a panel consisting of 8 specialists in EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 930 dentistry and a statistician: good knowl- edge (score 15–18); average knowl- edge (score 10–15); poor knowledge (score < 10). The attitudes section of the ques- tionnaire included 13 questions with 4 Likert-scale answers which represented the extent of dentists’ willingness to participate in treating AIDS patients (scored from 3 to 0, i.e. certainly dis- agree, disagree, agree, certainly agree). The questions covered society’s respect for and beliefs about AIDS patients, their own and other dentists’ willing- ness and duty to treat AID patients, concerns about treating such patients, costs of treatment, etc. Attitudes were categorized into 3 groups: negative at- titude (score < 20); average attitude (score 20–30); positive attitude (score 30–39). Ethical considerations The study was given ethical approval by the Research Committee of the Islamic Azad University, as well as the Medical Council of Iran. No personal identification or occupational data (i.e. names, phone numbers, exact ad- dresses and license numbers) were collected. Each questionnaire included a consent form, in which the first phase (direct observation) was described to the dentists and they were assured that their personal information had not been collected. Statistical analysis Descriptive statistics were calculated. In the behaviour section, dentists were classified into accepting or non- accepting groups. In the knowledge and attitude sections, the good and average groups were merged. The frequency distribution of each potentially corre- lating factor was calculated across the sample, then in order to dichotomize quantitative variables (e.g. dentist’s age) to binominal ones, the distribution of each quantitative factor was divided into groups of greater and smaller than the rounded mean value of that factor. Odds ratios (OR) and 95% confidence intervals (CI) were calculated. Cor- relations between the variables and behaviour, knowledge and attitude were assessed using the chi-squared test. Spearman correlations (rho) were done using ranked scores for knowledge (good =2, average/poor = 1) and for attitudes (positive/average = 2, negative = 1). The level of significance was set at 0.05. Results Background characteristics Of the 411 dentists visited a total of 111 were unavailable or refused to par- ticipate in the second phase leaving 300 dentists for the analysis. The mean age of the dentists was 46.5 (SD 19.1) years (range 27–64 years) and 71.0% of them were males; 86.3% worked in private offices, 68.3% worked in the centre or north of the city. Their mean work experience was 14.7 (SD 21.8) years (range 1–31); 7.0% had university teaching experi- ence, 80.0% had graduated from Iranian universities, and 4.3% had other degrees as well as dentistry qualifications. The average years since the last continuing education course was 1.9 (SD 1.5) years (range 0–3). Observed behaviour Of the 300 dentists visited, 44 (14.7%) accepted the simulated HIV-positive patient for treatment, while 256 (85.3%, 95% CI: 80.9%–88.9%) refused. Of the non-accepting dentists, 236 (78.6%) referred the patient elsewhere for treat- ment, while 20 (6.7%) refused even to refer. Age, work experience, the coun- try of graduation (Islamic Republic of Iran versus abroad) and having other degrees were significantly related to the dentists’ practice (Table 1). Older dentists were significantly more unwill- ing to treat AIDS patients (OR = 3.12, 95% CI: 1.08–10.0), and those who had studied at foreign universities were also less willing, which was borderl- ine significant (OR = 2.8, 95% CI: 0.96–8.16). In contrast, dentists with < 10 years work experience (OR = 4.96, 95% CI: 2.30–10.7) and dentists with additional degrees other than dentistry (OR = 3.97, 95% CI: 1.24–12.8) were significantly more likely to agree to treat AIDS patients. Participants with experience of teaching at universities were also more willing, but not sig- nificantly so, to treat the simulated pa- tients (OR = 2.54, 95% CI: 0.93–6.94) (Table 1). Self-reported knowledge In the knowledge section, 160 (53.3%) of the 300 dental professionals who agreed to be interviewed scored poor (95% CI: 47.7%–58.9%), 72 (24%) average and 68 (22.7%) good. The mean knowledge score was 8.3 (SD 9.7) (range 0–17), out of a maximum score of 18. In the attitude section, 125 (41.7%) of participants scored negative (95% CI: 36.2%–47.3%), 174 (58%) average and only 18 (0.6%) positive. Dentists’ knowledge was significant- ly associated with the time since the last continuing education course (Table 1). Dentists not admitted to continu- ing education courses during the last 2 years were slightly but not significantly more likely to score better in knowl- edge (OR = 1.68, 95% CI = 0.93–3.02). Faculty members were 2.4 times more likely to show better knowledge statuses but this was not statistically significant (OR = 2.43, 95% CI: 0.95–6.20) (P = 0.06) (Table 1). Self-reported attitudes The mean attitude score was 17.5 (SD 7.1) (range 5–33) out of 33. Dentists with good attitudes were significantly more likely to have graduated from an Iranian university than abroad (OR = 3.33, 95% CI: 1.84–6.01) (Table 1). Faculty members were more likely to show good attitudes (OR = 2.41, 95% CI: 0.86–6.78), although this was not statistically significant. Those with work طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 931 Ta bl e 1 D en ti st s’ k no w le dg e, a tt itu de a nd p ra ct ic es to w ar ds p at ie nt s w it h H IV a cc or di ng to d em og ra ph ic a nd w or k- re la te d va ri ab le s Va ri ab le Be ha vi ou r a ss es sm en t a K no w le dg e sc or es b A tt itu de s co re sc To ta l A cc ep ti ng (n = 4 4) N on -a cc ep ti ng (n = 2 56 ) χ2 P- va lu e ≥ 10 (n = 1 40 ) < 10 (n = 1 60 ) χ2 P- va lu e ≥ 20 (n = 17 5) < 20 (n = 1 25 ) χ2 P- va lu e N o. N o. N o. N o. N o. N o. N o. Se x Fe m al e 87 15 72 0 .6 5 0 .4 2 37 50 0 .8 4 0 .3 6 52 35 0 .0 3 0 .8 6 M al e 21 3 29 18 4 10 3 11 0 12 5 88 A ge (y ea rs ) < 40 15 0 32 11 8 10 .7 0 .0 0 1 70 80 0 .0 0 1 1.0 0 87 63 0 .0 1 0 .9 1 ≥ 40 15 0 12 13 8 70 80 88 62 W or k ex pe ri en ce (y ea rs ) < 10 36 14 22 19 .2 0 .0 0 1 20 16 1.3 0 0 .2 5 32 4 15 .7 0 0 .0 0 1 ≥ 10 26 4 30 23 4 12 0 14 4 14 3 12 1 La st c on ti nu in g ed uc at io n < 2 ye ar s 24 4 38 20 6 0 .8 6 0 .3 5 10 8 13 6 3. 0 4 0 .0 8 14 1 10 3 0 .16 0 .6 9 ≥ 2 ye ar s 56 6 50 32 24 34 22 M Sc d eg re e Ye s 70 7 63 1.5 9 0 .2 1 37 33 1.4 1 0 .2 4 43 27 0 .5 3 0 .4 7 N o 23 0 37 19 3 10 3 12 7 13 0 10 0 U ni ve rs ity st af f Ye s 21 6 15 3. 49 0 .0 6 14 7 3. 63 0 .0 6 16 5 2. 96 0 .0 9 N o 27 9 38 24 1 12 6 15 3 15 9 12 0 C ou nt ry o f g ra du at io n Ira n (IR ) 24 0 40 20 0 3. 84 0 .0 5 11 1 12 9 0 .0 8 0 .7 7 15 4 86 16 .8 0 0 .0 0 1 O th er 60 4 56 29 31 21 39 O ffi ce ty pe G en er al 41 7 34 0 .2 2 0 .6 4 20 21 0 .0 9 0 .7 7 24 17 0 .0 0 1 0 .9 8 Pr iv at e 25 9 37 22 2 12 0 13 9 15 1 10 8 W or k di st ri ct N or th /c en tr al 20 5 31 17 4 0 .11 0 .7 4 58 10 5 1.1 6 0 .2 8 12 6 79 2. 61 0 .11 W es t/ ea st /s ou th 95 13 82 82 55 49 46 O th er d eg re es Ye s 13 5 8 6. 15 0 .0 1 8 5 1.2 1 0 .2 7 7 6 0 .11 0 .7 4 N o 28 7 39 24 8 13 2 15 5 16 8 11 9 a A cc ep te d pa tie nt fo r t re at m en t o r r ef us ed to a cc ep t p at ie nt (r ef er re d/ no t r ef er re d) ; b Kn ow le dg e sc or es : m in . t o m ax . 0 –1 8; c A tti tu de sc or es : m in . t o m ax . 0 –3 9. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 932 experience < 10 years were 6.8 times more likely to admit AIDS patients (OR = 6.77, 95% CI: 2.33–19.7). Dentists working in north and central Tehran districts had a tendency towards bet- ter attitudes (OR = 1.50, 95% CI = 0.92–2.45). Correlations between knowledge, attitude and practice The Spearman correlation coefficient did not indicate any significant cor- relations between dentists’ knowledge and attitude (rho = 0.102, P = 0.077) or between knowledge and behaviour (rho = 0.008, P = 0.367). However, a signifi- cant (but weak) positive correlation was found between attitude and observed behaviour (rho = 0.379, P = 0.02). Discussion In the present study, a great number of dentists in Tehran failed to score appropriately in knowledge, attitude and practices toward patients with HIV. Dentists’ age, work experience, their country of graduation, and having degrees other than dental degrees sig- nificantly correlated with their observed practice in accepting or refusing to treat simulated patients. Our results of direct observations showing that 14.9% of dentists agreed to treat the simulate patients were very different from the results of Hazelkorn, who reported a 99% rate of acceptance to treat homosexual dental patients in 1989 in the United States [21]. How- ever, that study only assessed dentists’ behaviours when facing a dental pa- tient with a higher risk of having AIDS, not an HIV-positive patient. Besides, the extent of the AIDS epidemic and the negative attitudes towards it may have been different in that time and locality. Our results concerning den- tists’ knowledge about HIV/AIDS were generally poorer than in other published studies [5,8–10]. The differ- ence in subjects’ knowledge in various investigations worldwide might be at- tributed to the effect of education via the media (e.g. television, radio), which vary across countries, and the younger ages evaluated in those studies. The literature indicates that younger den- tists’ attitudes and knowledge might be better [6,9,13]. However, comparison between the results from other settings is difficult as the studies have not used uniform methods/questionnaires to evaluate participants’ attitudes or knowledge. The attitudes of medical health professionals (dentists, physi- cians and other health care staff) to- wards treatment of AIDS patients seem to have changed over time and there might be cultural differences in atti- tudes too. The findings in the literature vary widely, ranging from positive at- titudes or appropriate knowledge—in Japan, 1997 [5], Mexico, 1998 [6], UK, 2005 [7], Nigeria, 2009 [8], Islamic Republic of Iran, 2009 [9] and Turkey, 2010 [10]—to relatively negative at- titudes or poor knowledge observed in populations selected from USA, 1989 to 2006 [11–15], Japan, 2006 [16], Islamic Republic of Iran, 2006 [17], Nigeria, 2007 [18], UK, 2008 [19], and the Netherlands, 2008 [20]. Even allowing for difference in the questionnaires and methods, the rate of positive attitudes towards delivering oral health services to AIDS patients in the present study was consider- ably lower than in many other studies [5–7,9–11,16,17], similar to certain studies [7,16] and greater than the remainder [6,15,17]. This might be due to a combination of factors includ- ing sample differences such as mean ages, population sizes of the cities in which the studies were conducted (e.g. Tehran versus Shiraz [17] versus Mex- ico city [6]), differences in the level of education (e.g. dentists [6,7,17], dental students [10,14], dental residents [12], medical students [10,14] or dental fac- ulty [13]), cultural differences among different countries (e.g. attitudes to- wards having sexual relationships with various partners in different cultures), frequency of AIDS-related education via the media [5] (e.g. in the Islamic Republic of Iran versus in Japan with a 94% rate of AIDS-oriented educa- tion via television) and the year of the study (e.g. in the early years of AIDS epidemic [21] versus 2008–09 in the present study). There were also meth- odological differences between these studies, which used mass surveys such as interviewing with mail or during conferences [7], interviews at dental offices (the present study) and direct observations [21]. A negative correlation was found between dentist’s age and his/her observed reaction to an HIV-positive patient. However, data derived from the questionnaires failed to reveal any significant correlation between age and dentists' knowledge and attitudes. Therefore, in spite of their self-reported attitudes and knowledge, which were not significantly different from younger dentists, older dentists might be less financially interested in difficult cases, have higher workloads or be more disinclined to perform standard and appropriate infection control proto- cols. Other questionnaire-based stud- ies agree with this finding, showing non-significant effects of dentists’ age and work experience [8,13,14,17] and a significant effect of dentists’ sex on self-reported attitude [8,12,13,25]; however, there are some controversies over the effect of age [7,10] and sex [13,14]. The same reasoning might ac- count for the lower tendency of more experienced dentists (who might have busier offices and/or be older) to treat HIV-positive patients. Possession of degrees other than dentistry was asso- ciated with more favourable behaviour among clinicians. Further studies are needed to understand more about the reasons for this. In the present study, unlike the age factor, the country of graduation (Islamic Republic of Iran طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 933 or abroad) significantly affected cli- nicians’ practices, both self-reported and observed. However, despite their poorer behaviour and attitudes, den- tists who studied abroad had appro- priate knowledge comparable to the homeland-graduated dentists. Hence, their poorer attitude and behaviour scores might be rooted in differences in the prevalence of AIDS in certain countries (e.g. Philippines or Russia) which might foster greater concern about treating HIV patients among dentists graduated from those coun- tries. In contrast to research conducted in Japan [5], we found no significant cor- relation between dentists’ knowledge with their attitudes and behaviours. Knowledge may raise practitioners’ confidence to ignore negative messages regarding patients with HIV/AIDS; however, knowing how to manage and treat such patients is not the only factor a clinician may consider when deciding whether to treat a patient. Unlike many other countries, the main carriers of HIV in the Islamic Republic of Iran are intravenous drug users. This may give rise to concerns that the HIV-positive patient is financially and ethically impov- erished, leading to further problems for the dentist, which might also influence his/her decision to treat. Furthermore, elementary standard precautions might be absent in private dental practices [17], which is a risk when dealing with AIDS patients. The infection control protocols needed for AIDS patients raise the cost of treatment, which might create a disincentive to treat them for some clinicians. Dentists should be taught that universal precautions should be used with all patients, since dentists and patients themselves will not always be aware of who is HIV-positive. Therefore, although knowledge is nec- essary to improve attitudes [15], it may not be sufficient. The weak association found between the results of direct ob- servation and dentists’ claimed attitudes in this study highlights the low reliability of self-reporting and the need for direct assessment methods in such surveys. The face-to-face interview and direct survey approaches used in this study seem to be capable of obtaining more reliable results compared with designs based on mass data collection methods. However, this design is limited by factors such as being much slower to complete as well as the presence of observers’ subjective bias, although the latter was reduced by determining clear-cut be- haviour categories. The observers might experience difficulties in pretending to be HIV-positive or face inappropriate behaviours of some dental personnel after revealing their HIV-positive status. Indeed several dentists expressed their anger over the observers and some used impolite language. Some dental profes- sionals suspected the third observer was a government investigator or tax collector, which was a factor in their re- fusal to participate and the subsequent loss of some data collected in the first phase. Another criticism of the study design is that during the first phase the dentists were deceived, which might give rise to ethical concerns. However, their personal data were not collected, and signed consent forms were taken from them in the second phase, after thoroughly describing the study. It might have been better to recruit trained actors as simulated patients [21] or ac- tual HIV-positive patients. However, the students successfully convinced all the dentists they were AIDS patients. Another limitation of all of the studies in the field is the lack of standard question- naires, which reduces the comparability of the results. In this study, simple ran- dom sampling was used due to its con- venience; a stratified random sampling method would be more representative of the population. Conclusions Declared attitudes of dentists towards AIDS patients seemed to differ from their actual behaviours. A high propor- tion of Tehran dentists were unwilling to provide oral health care to AIDS patients, suggesting that there should be better legal provisions for AIDS patients facing discrimination [15]. Educational programmes are needed to improve attitudes of dentistry personnel—es- pecially those who are older, more experienced and foreign-graduated— towards patients with HIV/AIDS and improve knowledge about the risks of transmission [5,10]. References 1. Greenberg M, Glick M. Burket’s oral medicine: diagnosis and treatment. New York, BC Decker, 2003. 2. Seacat JD, Litt MD, Daniels AS. Dental students treating pa- tients living with HIV/AIDS: the influence of attitudes and HIV knowledge. Journal of Dental Education, 2009, 73:437–444. 3. Lewis DA et al. Description and evaluation of an education and communication skills training course in HIV and AIDS for dental consultants. European Journal of Dental Education, 2000, 4:65–70. 4. Bharat S, Aggleton P, Tyrer P. India: HIV and AIDS-related dis- crimination, stigmatization and denial. UNAIDS best practice col- lection key material. Geneva, Joint United Nations Programme on HIV/AIDS, 2001 (UNAIDS/0146 E). 5. Kitaura H et al. Knowledge and attitudes of Japanese dental health care workers towards HIV-related disease. Journal of Dentistry, 1997, 25:279–283. 6. Irigoyen M, Zepeda M, López-Cámara V. Factors associated with Mexico City dentists' willingness to treat AIDS/HIV-posi- tive patients. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics, 1998, 86:169–174. 7. Crossley ML. An investigation of dentists’ knowledge, attitudes and practices towards HIV+ and patients with other blood- EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 934 borne viruses in South Cheshire, UK. British Dental Journal, 2004, 196:749–754. 8. Uti OG et al. Infection control knowledge and practices related to HIV among Nigerian dentists. Journal of Infection in Develop- ing Countries, 2009, 3:604–610. 9. Sadeghi M, Hakimi H. Iranian dental students’ knowledge of and attitudes towards HIV/AIDS patients. Journal of Dental Education, 2009, 73:740–745. 10. Turhan O et al. Knowledge, attitudes and behaviour of stu- dents from a medicine faculty, dentistry faculty, and medical technology Vocational Training School toward HIV/AIDS. In- ternational Journal of Occupational Medicine and Environmental Health, 2010, 23:153–160. 11. Wallack JJ. AIDS anxiety among health care professionals. Hos- pital and Community Psychiatry, 1989, 40:507–510. 12. Cohen LA et al. Attitudes of advanced dental education stu- dents toward individuals with AIDS. Journal of Dental Educa- tion, 2005, 69:896–900. 13. Cohen LA, Romberg E, Grace E. Revisiting the attitudes of dental faculty toward individuals with AIDS. Journal of Dental Education, 2001, 65:249–252. 14. Cohen LA, Romberg E, Grace E. A revisitation of dental stu- dent’s attitudes toward individuals with AIDS. Journal of Dental Education, 2000, 64:298–301. 15. Sears B, Ho D. HIV discrimination in health care services in Los Angeles County: the results of three testing studies. Los Angeles, Williams Institute, 2006. 16. Inada K et al. Factors related to the attitude of general dental practitioners toward treatment of HIV carriers. Journal of Den- tal Health, Tokyo, 2006, 56:240. 17. Askarian M, Mirzaei K, McLaws ML. Attitudes, beliefs, and infection control practices of Iranian dentists associated with HIV-positive patients. American Journal of Infection Control, 2006, 34:530–533. 18. Obuekwe O et al. Knowledge and Attitude of clinical level dental students concerning HIV/AIDS. Nigerian Dental Journal, 2007, 15:3–8. 19. Elford J et al. HIV-related discrimination reported by people living with HIV in London, UK. AIDS and Behavior, 2008, 12:255–264. 20. Stutterheim S, Bos A, Schaalma H. HIV-related stigma in the Netherlands. Maastricht, Netherlands, AIDS Fonds and Univer- sity of Maastricht, 2008. 21. Hazelkorn HM. The reaction of dentists to members of groups at risk of AIDS. Journal of the American Dental Association, 1989, 119:611–619. 22. Andreoli T et al. Andreoli and Carpenter's Cecil essentials of medi- cine. Philadelphia, WB Saunders, 2007. 23. Fauci A et al., eds. Harrison's principles of internal medicine. New York, McGraw-Hill, 2008. 24. Little J et al. Dental management of the medically compromised patient. St Louis, Missouri, Mosby, 2007. 25. McQuistan MR et al. Dentists’ comfort in treating underserved populations after participating in community-based clinical experiences as a student. Journal of Dental Education, 2008, 72:422–430. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 935 Knowledge, perceptions and practices towards medical ethics among physician residents of University of Alexandria hospitals, Egypt A.M. Mohamed,1 M.A. Ghanem 2 and A.A. Kassem 3 ABSTRACT This cross-sectional study was conducted to assess the knowledge, perceptions and practices towards medical ethics of physician residents at university hospitals in Alexandria, Egypt. A self-administered structured questionnaire was used for knowledge and perceptions and a checklist for observations of doctor –patient interactions in the outpatient setting. Only 18.0% of the 128 participating residents had obtained their knowledge from their medical education and 29.9% were dissatisfied with the roles played by the ethics committee. Most of the residents had satisfactory knowledge and 60.2% had satisfactory perceptions regarding ethical issues. The lowest perception score was in the domain of disclosing medical errors. Only 48.0% of the residents were compliant with the principles of medical ethics in practice and 52.0% of patients were dissatisfied with their treating physicians. The study identified areas of unsatisfactory knowledge and practices towards ethical issues so as to devise means to sensitize residents to these issues and train them appropriately. 1Department of Community Medicine; 2Department of Forensic Medicine and Toxicology; 3Department of Family Medicine, Faculty of Medicine, University of Alexandria, Alexandria, Egypt (Correspondence to A.M. Mohamed: aida_mohey@yahoo.com). Received: 27/03/11; accepted: 24/07/11 صرم ،ةيردنكسلإا ةعماج تايفشتسم في ينميقلما ءابطلأا ينب ةيبطلا تايقلاخلأا لوح تاسرمالماو كرادلماو فراعلما مساق دحمأ ،مناغ اهم ،دممح ةدياع تايفشتسم في ينميقلما ءابطلأا ينب ةيبطلا تايقلاخلأا لوح تاسرمالماو كرادلماو فراعلما مييقتل ةضرعتسلما ةساردلا هذه نوثحابلا ىرجأ :ةـصلالخا لعافتلا لوح تاظحلاملل ةيدّقفت ةمئاق اومدختسا ماك ،كرادلماو فراعلما لىع فرعتلل ًايتاذ ذ َّفنُي ًانايبتسا اومدختساو .صرم ،ةيردنكسلإا ةعماج فراعلما لىع اولصح دق طقف ةساردلا في ينهماسلما 128 ـلا ينميقلما نم %18 نأ ينثحابلل َّينبتو .ةيجرالخا تادايعلا في ضيرلماو بيبطلا ينب لدابتلما لىع ثعبت فراعم ينميقلما مظعم ىدل ناك دقو .تايقلاخلأا ةنلج ايهّدؤت يتلا راودلأاب ينضار اونوكي لم مهنم %29.9 نأو ،مله يبطلا ميلعتلا نم نع حاصفلإا لامج في يه كرادلما مّلس لىع ةجرد ىندأ تناكو .تايقلاخلأا اياضق لوح ضىرلا لىع ثعبت كرادم مهنم %60.2 ىدل ناكو ،ضىرلا نع ينضار يرغ ضىرلما نم %52 ناك ماك ،ةسرمالماب اهقيبطت ءانثأ ةيبطلا تايقلاخلأا ئدابمب ينمزتلم طقف ينميقلما نم %48 ناك دقو .ةيبطلا ءاطخلأا ةدافتسلاا نكميو ،تايقلاخلأا اياضقب ضىرلا اهيف تاسرمالماو فراعلما ثعبت لا يتلا تلااجلما لىع ةساردلا تفرعت دقو .منهولجاعي نيذلا ءابطلأا .مئلام ٍوحن لىع اهيلع مهبيردتو اياضقلا هذه لوح ينميقلما ءابطلأا ىدل يعولا ىوتسمب ءاقترلاا فدهتست تايجيتاترسا ةغايصل ةساردلا جئاتن نم Connaissances, perceptions et pratiques en matière d'éthique médicale des internes des centres hospitaliers universitaires d'Alexandrie (Égypte) RÉSUMÉ La présente étude transversale a été menée afin d'évaluer les connaissances, les perceptions et les pratiques en matière d'éthique médicale des internes exerçant dans des centres hospitaliers universitaires d'Alexandrie (Égypte). Un questionnaire autoadministré et structuré a été utilisé pour évaluer les connaissances et les perceptions des répondants. Une liste de contrôle des observations réalisées à partir des interactions entre le médecin et le patient en consultation externe a aussi été utilisée. Seuls 18,0 % des 128 internes participant avaient acquis leurs connaissances pendant leurs études de médecine et 29,9 % n'étaient pas satisfaits du rôle joué par le comité d'éthique. La plupart des internes avaient des connaissances satisfaisantes et 60,2 % une perception satisfaisante des questions éthiques. Le score le plus faible a été obtenu pour la perception de la divulgation des erreurs médicales. Seuls 48,0 % des internes respectaient les principes de l'éthique médicale dans la pratique, alors que 52,0 % des patients étaient mécontents de leur médecin traitant. L'étude a permis d'identifier des domaines où les connaissances et les pratiques en matière d'éthique étaitent insuffisantes. Les résultats pourront être exploités pour concevoir des moyens permettant de sensibiliser les internes à ces questions et de les former convenablement. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 936 Introduction Medical ethics has been included in the training curriculum of health profes- sionals in many countries and there has been a growth in the number of ethics specialists and ethics committees. Yet complaints from the public appear to be proliferating [1]. This may be a reflection of an increased public aware- ness as well as poor practices within the health care sector. Traditional medical training offers little help in resolving the ethical dilemmas encountered by health care professionals. It has been argued that very few physicians are exposed to training in this important area of medi- cal practice, yet on qualifying health care professionals are expected to know about ethical practice when applying their skills [1]. In the faculty of medicine in Al- exandria, Egypt, formal teaching of medical ethics at the undergraduate and postgraduate levels is done as part of medical jurisprudence. One problem is that medical students do not have much interest in learning about medical ethics as they consider other subjects to be more impor- tant for qualifying for examinations. Moreover, professional conduct and etiquette are learnt by observation rather than through theory. Yet stu- dents may not find role models for ethical practice among their teachers. Senior doctors’ practices influence the behaviour and attitudes of junior doctors. The teaching environment and methods are also not favourable for the teaching of ethics [2]. To our knowledge, there have been no previous surveys of postgraduate clinical trainees affiliated to the Univer- sity of Alexandria faculty of medicine. This study was conducted to reveal knowledge and perception of medi- cal ethics among physician residents in University of Alexandria hospitals and to assess the degree to which they adhere to ethical practices during out- patient consultations. Methods Setting and sample A descriptive cross-sectional survey was conducted at the 3 university hospitals of Alexandria from August 2009 to Sep- tember 2010. All physician residents of the faculty of medicine at University of Alexandria, working in 16 clinical de- partments, were invited to participate in this survey (n = 255, according to the hospital information system, 2008). A total of 128 completed and submitted the questionnaires. Observations were done on 100 residents (28 refused to be observed). A sample of patients attending 10 of the 16 outpatient clinics were invited to participate in the survey. A multistage random sampling technique was used to select patients. With the assump- tion that patients’ perception of good ethical practices were 50% and using significance level of 0.05% a sample size of 100 patients was estimated. The outpatient registration record for such clinics was used to enrol patients for the study using systematic random sam- pling. All patients agreed to participate in the study. Data collection Tools Three tools were developed: • An anonymous self-administered questionnaire was devised for this study based on a review of the lit- erature. It collected data about the demographic characteristics of physi- cians (sex, occupational category and postgraduate year), frequency of ethi- cal problems encountered in practice, sources of knowledge about medical ethics, who they preferred to consult about ethical problems when these arise and knowledge of ethics commit- tees and their roles. Another section included a range of statements de- signed (after a thorough review of the literature) to identify respondents’ knowledge about medical ethics, per- ceptions towards physician–patient ethics, physician–colleague relation- ships and disclosure of medical errors. • An observation checklist assessed how far the physicians adhered to specified principles of medical ethics: informed consent, privacy, confiden- tiality and collaboration of patients in the process of decision-making for treatment. • An interview tool obtained infor- mation about the demographic characteristics of patients and their satisfaction with adherence of physi- cians to the principles of medical eth- ics: physicians’ practices of informed consent, privacy and confidentiality and the way that physicians discussed treatment with them. All these tools were pilot-tested on 15 residents and 15 patients. The main obstacle encountered was the poor cooperation of residents in returning the questionnaire. This was due to high workload and shift work patterns. The response rate was 77.6% and this did not differ significantly by department of affiliation. A random subset of 30 residents also completed the survey to assess the reliability of the research tools. Intra-class correlation coefficients were used to compute internal reliabil- ity estimates. Evaluations of reliability of tools were done using Cronbach alpha. Scoring Variables were processed as following: • The knowledge about medical eth- ics scale comprised 13 statements (scored 1 = know and 0 = don’t know; negative questions were re- verse scored) (α = 0.86). The total knowledge score was calculated as a percentage of the maximum possible score. • The perceptions towards physi- cian–patients ethics scale comprised 13 statements concerning ethical conduct, autonomy, paternalism, confidentiality, informing patients about wrong-doing and informing relatives about the patient’s condi- tion, informed consent, treating طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 937 non-compliant or violent patients, religious beliefs influencing treat- ment and abortion and euthanasia. The perceptions towards disclosing medical errors scale comprised 4 statements. The perception scale of physician–colleague relationships comprised 6 statements (all scored as 1 = agree and 0 = disagree). Com- posite perception scores was calcu- lated by summing the 23 perception statements scores, negative questions were reverse scored (α = 0.90). The total perception score was calculated as a percentage of the maximum pos- sible score. • The compliance scale comprised 15 statements (scored 1 = done and 0 = not done) (α = 0.82). The total compliance score was calculated as a percentage of the maximum possible score. • Patients’ satisfaction scale comprised 9 statements (scored 1 = satisfied and 0 = dissatisfied) (α = 0.83). The total satisfaction score was calculated as a percentage of the maximum possible score. Satisfactory knowledge, perception and compliance of residents to ethical practices and adequate patients’ satisfac- tion were defined as percentage scores above the median percentage score. Ethical considerations The ethics committee of the faculty of medicine approved the study. Informed consent was taken from physicians and patients before starting data collection, after a full explanation of the purpose and aims of the study. Questionnaires were distributed by the investigator herself with a covering letter indicating the purpose of the study, confidentiality procedures and faculty review approval. Informed consent was taken from phy- sicians before starting collecting data, after a full explanation of the purpose and aims of the study. Participants’ anonymity was preserved. Data about responses and views of physicians and patients as well as the performance of each resident remained undisclosed by the researcher. Data processing and analysis After data collection, the raw data was coded and scored and a coding instruc- tion manual was prepared. Data were fed to the computer and statistical anal- ysis was performed using SPSS, version 18.0. The significance of the results was judged at the 5% level of significance. Results Physicians’ characteristics The characteristics of the 128 resi- dent physicians who completed the questionnaires have been described in a previous paper [3]. Two-fifths of residents (40.6%) had learned about medical ethics from their own experi- ence at work, 27.3% had read about it and 18.0% had their knowledge from their medical education; only 14.1% gained their knowledge from exposure to a training course on the subject. Of the total residents approached, 98.0% admitted facing ethical problems and the rest (2.0%) denied it. More than one-third (37.0%) stated that they faced them daily. One-third (33.0%) of physicians consulted their immediate supervisor when they encountered an ethical problem, 24.0% went to the eth- ics committee, 19.0% asked advice from a colleague and the rest either consulted a professor (15.0%) or the head of the department (9.0%). Physicians’ knowledge about ethics committees Only 4 physicians (3.1%) did not know about the existence of the ethics com- mittee in the faculty of medicine. All residents felt that among the roles of this committee was to approve and guide re- search. Many residents (72.6%) claimed that it had a role in teaching medical ethics and 8.1% that it conducted medi- cal ethics conferences. Asked about its role, 58.9% thought it was to ensure standard ethical practices of health care personnel, 24.2% to advise health care personnel about ethical problems and 12.1% to advise administrators. A few residents stated that the committee re- solved conflicts between professionals (2.4%) and between professionals and patients’ relatives (1.6%). Apart of those who denied its existence (3.1%), 29.9% were dissatisfied with its role. Physicians’ knowledge about medical ethics and rights of patients All residents except 1 believed that medical ethics was an essential subject for physicians and 86.7% were able to define it. A majority (89.8%) agreed that rapport can be established between physician and patient in medical prac- tice. Most residents (91.4%) could mention at least 1 of the 4 patients’ rights: privacy (89.8%), obtaining an informed consent (87.5%), verac- ity (85.9%) and beneficence (84.4%). More than two-thirds of them (68.7%) stated non-maleficence and less than two-thirds mentioned autonomy as a right. Justice and confidentiality were mentioned each by 59.4%. Less than half the physicians (45.3%) were of the opinion that disclo- sure of medical reports is a good idea. The majority (97.7%) agreed that the patient has the right for a second medi- cal opinion and 74.0% felt it is good for to patients learn about their own disor- ders through the Internet and/or books. The majority of residents (96.9%) agreed that palliative care is good and that medical treatment should rely heavily on drugs (88.3%). When asked whether as a routine physicians should describe and/or explain aspects of therapy there was agreement about ex- plaining methods of drug use (99.2%) and effects of the drug on the patient’s illness (93.0%), while fewer physicians would explain side-effects (79.7%) or names of drugs (52.3%) (Table 1). The overall knowledge score ranged from 30.7%–92.3% with a median and EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 938 interquartile range of 80.8% (IQR 9.8%). Over two-thirds of the residents (69.5%) had satisfactory knowledge (total knowledge scores greater than median score). No significant knowl- edge differences were noted between residents by sex (P = 0.729), depart- ment affiliation (P = 0.258) or post- graduate year (P = 0.253) (data not shown). Physicians’ perceptions about different domains of medical ethics The majority of residents disagreed with several statements about the phy- sician–patient relationship: avoiding legal action (95.3%), inability to apply it in modern care (90.6%), assisting patients who wished to die (97.7%), re- fusing to treat violent patients (88.3%) and refusing abortions (85.2%). The majority of physicians (96.1%) however, agreed that children should never be treated without the consent of their parents/guardians except in an emergency and that patients should always be told if something is wrong (83.6%). Physicians mostly agreed that confidential information can only be disclosed if the patient gives explicit consent or if expressly provided for in the law (77.3%) but fewer agreed to tell close relatives about a patient’s condi- tion (50.0%) or that patients should be told to find another doctor if they are refused treatment due to beliefs (60.2%) (Table 2). When asked about the disclosure of medical errors 32.0% would only do it in cases of death or disability while 64.1% would act in cases of only minor medi- cal harm. Fear of malpractice was the main reason for non-disclosure of medi- cal errors (75.0%) but 34.4% agreed that patients’ forgiveness was an important motive for disclosure (Table 2). Concerning physician–colleague relationships 70.3% agreed that physi- cians should treat colleagues respect- fully and work cooperatively, 54.7% that they should resolve conflicts with colleagues and 48.4% that they should cooperate with others in the care of patients. All the residents condemned stealing other colleagues’ patients and only 23.4% agreed that physicians should be paid to procure the referral of a patient. However, 60.9% did not ap- prove reporting colleagues’ misconduct (Table 2). The overall total perception score was 74.8% (IQR 6.4%) and 60.2% of residents had a satisfactory level of per- ception. Of the 3 domains, the highest scores were for the domain of physi- cian–colleague relationship (62.6% had a satisfactory level) and the lowest for disclosing medical errors (40.7% satis- factory), while 59.6% scored satisfactory for physician–patient relationships. No significant differences were observed as regards level of perception and sex (P = 0.749) or department of affilia- tion (P = 0.295). However, residents enrolled in 2nd and 3rd years (70.5% and 66.7%) were more likely to have a satisfactory perceptions compared with those enrolled in first (41.2%) and 4th year of postgraduate (48.4%) (data not shown). Table 1 Physicians’ knowledge about medical ethics and patients’ rights (n = 128) Knowledge statementa Know Don’t know No. % No. % Medical ethics is an essential subject for physicians 127 99.2 1 0.8 Briefly define the term “medical ethics” 111 86.7 17 13.3 Rights of patients that should be acknowledgedb 117 91.4 11 8.6 Disclosure of medical reports is a good idea 58 45.3 70 54.7 Rapport can be established between a physician and the patient in medical practice 115 89.8 13 10.2 It is good for patients to know about their own disorders via the Internet and/or books 94 74.0 33 26.0 Patients have the right to a second medical opinion 125 97.7 3 2.3 Palliative care is good 124 96.9 4 3.1 Medical treatment should rely heavily on drugs 113 88.3 15 11.7 Physicians as a routine should describe and/or explain: Names of drugs 67 52.3 61 47.7 Methods of drug use 127 99.2 1 0.8 Effects of drugs 119 93.0 9 7.0 Side-effects of drugs 102 79.7 26 20.3 Total – – – – The median (interquartile range) was 80.8% (9.8). aStatements are mutually exclusive; bResident was considered knowledgeable if he/she mentioned at least 4 (out of 8) of patient’ rights: privacy, informed consent, veracity, beneficence, non-maleficience, autonomy, justice and confidentiality. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 939 Table 2 Physicians’ perceptions towards medical ethics (n = 128) Perception domain/perception statement Agree Disagree Median (IQR) score No. % No. % % Physician–patient relationship 77.3 (8.9) Ethical conduct is important only to avoid legal action 6 4.7 122 95.3 Patient’s wishes must always be adhered to 74 57.8 54 42.2 Physician should do what is best irrespective of patient’s opinion 46 35.9 82 64.1 Patient should always be told if something is wrong 107 83.6 21 16.4 Confidential information can only be disclosed if the patient gives explicit consent or if expressly provided for in the law 99 77.3 29 22.7 Confidentiality cannot be applied in modern care and should be abandoned 12 9.4 116 90.6 Close relatives must always be told about a patient’s condition 64 50.0 64 50.0 Patients need to consent only for operations but not for tests or medications 23 18.0 105 82.0 Children should never be treated without the consent of their parents or guardians (except in an emergency) 123 96.1 5 3.9 Doctors and nurses should refuse to treat patients who behave violently 15 11.7 113 88.3 Patients who refuse treatment due to beliefs should be instructed to find another doctor 77 60.2 51 39.8 Health care worker cannot refuse to do an abortion if the law allows it to be performed 19 14.8 109 85.2 Patient who wishes to die should be assisted in doing so 3 2.3 125 97.7 Disclosing medical errors 52.7 (6.9) Would disclose medical error resulting in minor medical harm (prolonged treatment/discomfort or extra cost) 82 64.1 46 35.9 Would disclose medical error only if it caused major harm (disability or death) 41 32.0 87 68.0 Fear of malpractice is an important reason why doctors do not disclose medical errors 96 75.0 32 25.0 Patients’ forgiveness plays an important role in motivating doctors to disclose medical errors 44 34.4 84 65.6 Physician-colleague relationship 69.4 (8.9) Treat colleagues respectfully and work cooperatively to maximize patient care 90 70.3 38 29.7 Pay or receive fee to procure referral of a patient to a colleague 30 23.4 98 76.6 Steal patients from colleagues 0 0.0 128 100.0 Report colleagues’ misconduct 50 39.1 78 60.9 Apply the main ethical principles relating to cooperation with others in patient care 62 48.4 66 51.6 Resolve conflicts with other health care providers 70 54.7 58 45.3 Total – – – – 74.8 (6.4) IQR = interquartile range. Physicians’ ethical practices Observations of physicians’ compli- ance with ethical practices showed that only 48.0% of residents overall were compliant with the principles of medical ethics in practice. All residents took in- formed consent and complied with the principle of not harming the patient), 94.0% of them showed respect towards patients’ needs and 93.0% accepted the patients’ request not to be examined by medical students (Table 3). Privacy during examination was ensured by 90.0%; however, only 76.0% made sure EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 940 that no third party was present. The majority of residents (89.0%) complied with the principles of confidentiality. The lowest median score was for patient management [62.7% (IQR 3.2%)]. Only 40.0% of residents in- formed their patients about the cost of treatment, 53.0% shared the solution of their problem with patients or fully informed them about side-effects of treatment (54.0%); however, 98.0% fully informed their patients as regards the cause of disease. Only 62.0% of residents made a full examination to reach their diagnosis. Significant sex differences were noted as 61.8% of woman physicians were compliant with the principles of medical ethics compared with 31.2% of men (P = 0.003) (data not shown). Residents in psychiatry (80.0%) and paediatrics (80.0%) were significantly more likely to comply with medical ethics than those in internal medicine (50.0%), family medicine (40.0%) or other departments (38.3%) (P = 0.003). No significant differences however, were observed as regards compliance and postgraduate year (P = 0.539). Patients’ characteristics Of the 100 patients approached, 22.0% were aged < 20 years and 43.0% were > 40 years, 56.0% were male, 48.0% them were currently married, 31.0% did not hold any educational certificates and 53.0% were residing in rural areas. The majority of patients were seeking cura- tive care (81.0%) and the rest attended for preventive or rehabilitative care. Patients’ satisfaction with services and environment Only 48.0% of patients were satisfied with their treating physician. About Table 3 Physicians’ observed compliance to ethical practices during physician–patient interactions (n = 100) Compliance practice Done Not done Median (IQR) score No. % No. % % Informed consenta 100.0 (0.0) Took informed consent from patient (before history- taking, before examination or before exposing any body part) 100 100.0 0 0.0 Patient right not to be harmed 100.0 (0.0) Did not prescribed medication that was not indicated (e.g. only for research) 100 100.0 0 0.0 Did not order investigations that were not indicated (e.g. only for research) 100 100.0 0 0.0 Respect and dignity 82.6 (5.1) Showed respect, dignity, responsiveness and attention to patient’s health needs 94 94.0 6 6.0 Accepted patient’s request not to be examined by medical students 93 93.0 7 7.0 Privacy 74.5 (3.6) Used a private room/screen for examination 90 90.0 10 10.0 Ensured nobody present other than medical team 67 67.0 33 33.0 Confidentiality 72.0 (6.2) Did not share information without patient’s consent 89 89.0 11 11.0 Case management 62.7 (3.2) Gave patient the right to refuse treatment and change health care provider 3 3.0 97 97.0 Gave patient the opportunity to share in decision about treatment 53 53.0 47 47.0 Fully informed patient about cause(s) of illness 98 98.0 2 2.0 Fully informed patient about side-effects of treatment 54 54.0 46 46.0 Fully informed patient about cost of drugs 40 40.0 60 60.0 Made full examination of patient to reach diagnosis 62 62.0 38 38.0 Informed patient about follow-up visits 77 77.0 23 23.0 Total – – – – 70.5 (5.6) aResident was considered compliant if he/she practised any of the 3 items of informed consent. IQR = interquartile range. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 941 two-thirds (67.0%) were satisfied with the doctor’s practice in obtaining an informed consent and 62.0% with hav- ing privacy during medical consulta- tion. Only 49.0% were satisfied with the doctor’s practice of confidentiality. Just over half (52.0%) claimed that they were satisfied with the doctor’s way of communicating with them (Table 4). Patients’ views of physicians’ performance Of the patients 61.0% claimed that the physician discussed the causes of their illness, 37.0% complications of the disease, 60.0% treatment modal- ity and 68.0% drugs and their possible side-effects. Only 42.0% were informed about the cost of treatment (Table 4). Patients were significantly more likely to be satisfied with their ethical treatment if they were aged < 20 years (P =0.017), of lower educational level (P = 0.003) and receiving preventive or rehabilitative care (P = 0.004) (Table 5). No significant differences were observed as regards the satisfaction of patients according to their sex (P = 0.091), marital status (P = 0.395) or residence (P = 0.072). Significant dif- ferences were also observed as regards frequency of visits; patients attending 2+ times were more likely to be dissatis- fied (83.7%) than those on the first visit (28.1%) (P = 0.002). Discussion In order to construct an effective eth- ics teaching curriculum for physician residents, the first step is to determine their current basic knowledge, percep- tions and practices related to ethical issues and patients’ rights. Ethical problems were common among physi- cian residents in our study. The major- ity of residents (98.0%) encountered ethical issues during their practice and 37.0% of them claimed that they faced them on a daily basis. This indicates the importance of preparing residents to deal with ethical dilemmas and providing them with guidance, sup- port and supervision. Only 33.0% of respondents preferred to consult their immediate supervisor about ethical issues and senior staff, and the head of the department was less frequently consulted. This finding is consistent with the commonly preferred option to settle the matter at the department level rather than taking it to the higher levels [4]. The end result will be that senior staff, who should act as mentors, will not be aware of ethical problems facing their junior staff, either because they have no time or due to administrative and communication barriers. Similar results were reported by Nakao et al. in Japan [5]. Our results showed that senior staff in Alexandria need to be more involved in the ethical problems that do arise and need to be prepared for guidance and handling of ethical problems. It is interesting to note that few re- spondents (18.0%) had obtained their knowledge of ethics from their medi- cal education and 40.6% had learned about medical ethics from their own experience at work. This shows that the curricular training regarding ethics is either inadequate or ineffective. Also, if residents feel that their main source of knowledge of health care ethics is during experience at work, job experi- ence should be used to reinforce ethical knowledge and practice. Ethics committees are the most prominent formal institutional mecha- nism for considering and resolving ethical dilemmas in medicine. Despite that, hospital ethics committees are largely untested, unproven and un- known entities [6]. The present study indicated that the majority of physi- cians acknowledged the presence of an ethics committee in the faculty. Despite this, only a few of them stated that they approached the committee for advice. Awareness about the ethi- cal committee in this study disagrees with another study regarding physi- cians’ perceptions of a hospital ethics Table 4 Patients’ satisfaction with physicians’ ethical practices (n = 100) Satisfaction item Satisfied Dissatisfied No. % No. % Physician’s practices in general Manner in communicating 52 52.0 48 48.0 Ensuring informed consent 67 67.0 33 33.0 Ensuring privacy 62 62.0 38 38.0 Ensuring confidentiality 49 49.0 51 51.0 Physician’s practices in informing and discussing Causes of illness 61 61.0 39 39.0 Complications of illness 37 37.0 63 63.0 Treatment modalities 60 60.0 40 40.0 Drugs and its possible side effects 68 68.0 32 32.0 Cost of treatment 42 42.0 58 58.0 The median (interquartile range) was 63.9% (7.3). EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 942 committee from the United States, in which large number of residents ex- pressed dismay at the “invisibility” of the ethics committee [7]. Of our resi- dents, 29.9% were dissatisfied with the role played by the present committee. Therefore, the role of the medical ethics committee in University of Alexandria hospitals should be strengthened to in- clude involvement not only in research but also in education and advice about ethical issues. The present study revealed that only 48.0% of physician residents were com- pliant with all the principles of ethics. Their levels of knowledge and percep- tions were also lower than expected. Poor compliance might therefore be attributed to unsatisfactory levels of knowledge and perceptions about the subject, or it may be a reflection of the relatively high non-response rate to the questionnaire. While a majority of our residents were aware of the common ethical issues and patient’s rights, they had unsatisfactory levels of knowledge about certain ethical issues. This result suggests that medical ethics education in Alexandria should be strengthened in topics where knowledge levels were low. Medical education should also address changing attitudes and percep- tions of residents. Previous research suggested that positive attitudes of residents toward ethics preparation improved beneficial outcomes of edu- cational innovations [8]. Such findings support the belief that assessing train- ees’ attitudes, views and preferences is important in developing curricular approaches attuned to their concerns and experiences [9]. This information may help in the creation of curricular content and methods that are more acceptable to trainees [10,11]. In ac- cordance, we found that the majority of residents disagreed with several state- ments related to physician–patient relationships. However, the strong negative perceptions of the residents about certain issues may reflect differ- ences in the intensity of training. In a study on attitudes towards patient autonomy, residents in the United Kingdom showed a lower commit- ment to patient autonomy than did any of the United States groups [12]. The fact that 42.2% of residents in our study did not feel that the patient’s wishes should be adhered to at all times shows the lack of knowledge about the funda- mental principles of medical ethics. Table 5 Patients’ satisfaction with physicians’ ethical practices by patients’ sociodemographic variables (n = 100) Variable Total (n = 100) Satisfied ( n = 48) Dissatisfied (n = 52) χ2 P-value No. % No. % Age group (years) < 20 22 15 68.2 7 31.8 6.72 0.01720– < 40 35 17 48.6 18 51.4 40+ 43 16 27.2 27 62.8 Sex Male 56 34 60.7 22 39.3 4.34 0.091 Female 44 14 31.8 30 68.2 Marital status Married 48 28 58.3 20 41.7 0.96 0.395 Unmarried 52 20 38.5 32 61.5 Education level Secondary education or higher 60 12 20.0 48 80.0 0.003a Less than secondary education/no formal education 40 36 90.0 4 10.0 Residence Urban 47 17 36.2 30 63.8 4.77 0.072 Rural 53 31 58.5 22 41.5 Type of care Preventive 10 9 90.0 1 10.0 6.93b 0.004Curative 81 31 38.3 50 61.7 Rehabilitative 9 8 88.9 1 11.1 Frequency of visits First 57 41 71.9 16 28.1 8.73 0.002 Second or more 43 7 16.3 36 83.7 aFisher exact test; bMonte Carlo test. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 943 Critical steps toward improving the safety of the health care system include ensuring that the system is aware of its errors so that effective remedies can be applied, and enhancing the trust- worthiness of the health care system for patients by disclosing errors that are meaningful to them [13]. In Fein et al.’s study all residents believed that errors should be disclosed. Important influences on whether disclosure would occur fell into 4 categories: provider, pa- tient, error and institutional culture fac- tors [14]. The present study addressed some of the above-mentioned factors. The majority of residents agreed that fear of malpractice was the main reason for non-disclosure of medical errors, while patients’ forgiveness was another important motivator for disclosure. Further qualitative research is needed to apply the previous grounded model of error disclosure in order to deline- ate areas for interventions to increase disclosure as a step toward improving patient safety. The present work described resi- dents’ perceptions about how physicians should behave towards one another. The obligation to report incompetence, impairment or misconduct of one’s colleagues is emphasized in codes of medical ethics [15], but the majority of our respondents did not approve of reporting it. However, the application of this principle is not easy. A physician may be reluctant to report a colleague’s misbehaviour because of friendship or sympathy or for fear of hostility on the part of the accused and possibly other colleagues [16]. Our observations of the actual prac- tice of medical ethics by the residents showed that the majority of residents showed respect towards their patients. This is widely regarded as the most im- portant principle of medical ethics [17]. The burden of patients at outpatient clinics of public hospitals in Egypt often makes it impossible for the doctors to follow the full protocol of privacy and patients’ problems are often discussed in front of all present in the room. Such a practice may prevent the patient from revealing his/her complete history and symptoms [18]. A study from a public hospital in Karachi reported that the number of patients complaining of lack of privacy was greater than in more de- veloped countries [19]. Nevertheless, the present work showed that privacy was mostly ensured during consulta- tions. Similarly, the principle of confi- dentiality (information privacy) was also adequately practised. This was in accordance with a study in Canada that showed many family physicians fully understood their obligations towards patient confidentiality [20]. On the other hand, adherence to case management ethical principles in medical practice was inadequate in Al- exandria teaching hospitals. One of rea- sons behind such a practice is that the cultural trends in Egypt still tend to ac- cept the paternalistic model of medical care. This is in line with the Arab culture as a whole, where decision-making is often left purely to the doctors or other family members of the patient. A study from Karachi in Pakistan reflect similar practices, in which patients are willing to accept what doctors choose for them, while doctors are satisfied with their role as decision-makers [21]. While the situ- ation in the US was not much different until the 1960s [22], current medical practice attaches great importance to the concepts of informed consent and shared decision-making. This differs substantially from the situation in Al- exandria [23] and experts have called it a “cultural artefact”, in that reliance on this concept is not universal [24]. Even in the US, there is often a clash between these ethical standards and the moral intuitions of many physicians [25]. It is noteworthy that there were other reasons for inadequate ethical practices in our setting. Although in- novative ethical curricula have been shown to improve the confidence and practices of doctors with regards to medical ethics [26], medical education in Alexandria does not include bioethics training as a major component of the medical curriculum [27]. Lack of ap- plied ethics training has also been noted in other countries such as Germany [28] and even the US, which has always championed the cause of bioethics [29]. The lack of education in ethics means that trainees can only learn from the practices of their consultants, most of whom belong to the era when a pater- nalistic approach towards patients was in vogue. This leads to a vicious cycle where every subsequent generation of doctors believes in paternalism. Even doctors who favour practices such as informed consent often abandon these practices in the belief that most of their patients are uneducated and would not be able to decide what is best for them. It is true, though, that patients often do not want to take decisions and allow doctors to decide for them. Further- more, the lack of accountability and legal recourse means that doctors who do not respect patient ethics are never taken to task in Egypt [23]. The present work compared whether the patients’ satisfaction and views of these ethical practices matched correctly with the observed assessment of residents’ performance. Less than half of patients were satisfied with their physicians. In some instances, patients’ were satisfied even when the physician’s performance was judged to be unethical by the investigator. This finding can be explained by patients’ misunderstand- ings of their ethical rights [30]. Public health programmes are needed to make patients aware of their legal rights to informed consent, confidentiality and privacy. The discordance observed may be related to characteristics of these patients as the study indicated that dis- satisfied patients were those who were older, of higher education, obtained cu- rative care and made frequent visits. Pa- tients can under- or overestimate their ethical rights and hence their opinion may not necessarily be in line with cur- rent standards [1]. In other instances, EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 944 patients’ views about the physician’s performance were similar to what was actually observed. For example dissat- isfaction with instructions related to treatment reflects a distrust of doctors. Education needs to be directed at help- ing doctors gain the trust of patients [31]. Significant sex differences were observed between residents in the practice of medical ethics, with women being more compliant with ethical practices than men. This pattern rep- licates previous work [32]. Reasons for sex differences in adherence to ethical practices, and how these differences may influence patient care, remain uncertain [33]. Moazam proposed that women tend to adhere more to standards of morality and that their ideas of “goodness” depend on pleas- ing and helping others [34]. These re- sults are an invitation for further study. Residents in psychiatry and paediatrics were more likely to comply with medi- cal ethics standards than those in other departments. The practice of psychia- try and that of paediatrics, perhaps more than other medical specialties, place a heavy emphasis on dealing with people who may not always feel the need for expert assistance or may not be competent to understand their own needs. Ethics training demands careful References 1. Muaz F. Family, patient and physician in medical decision making. Hastings Center Report, 2000, 6:28–37. 2. McCullough LB, Chervenak FA. Informed consent. Clinics in Perinatology, 2007, 34:275–285i. 3. Mohamed AM, Ghanem MA, Kassem AA. Problems and perceived needs for medical ethics education of resident phy- sicians in Alexandria teaching hospitals, Egypt. Eastern Mediter- ranean Health Journal, 2012, 18(8):827–835. 4. Mallardi V. Le origini del consenso informato [The origin of informed consent]. Acta Otorhinolaryngologia Italica, 2005, 25:312–327. 5. Nakao HX, Chishaki A, Obayashi M. Awareness of ethical issues by nursing professionals at a general local hospital in Japan. Fukuoka Igaku Zasshi, 2008, 99:175–183. 6. Denham MJ, Foster A, Tyrrell DA. Work of a district ethical committee. British Medical Journal, 1979, 2:1042–1045. 7. Hern HG Jr. Ethics and human values committee survey: (AMI Denver Hospitals: Saint Luke’s, Presbyterian Denver, Presby- terian Aurora: Summer 1989). A study of physician attitudes and perceptions of a hospital ethics committee. HEC Forum, 1990, 2:105–125. 8. Perkins HS, Geppert CM, Hazuda HP. Challenges in teaching ethics in medical schools. American Journal of the Medical Sci- ences, 2000, 319:273–278. 9. Lazarus CJ et al. The program for professional values and eth- ics in medical education. Teaching and Learning in Medicine, 2000, 12:208–211. 10. Christakis DA, Feudtner C. Ethics in a short white coat: the ethical dilemmas that medical students confront. Academic Medicine, 1993, 68:249–254 11. Hébert PC, Meslin EM, Dunn EV. Measuring the ethical sensiE- tivity of medical students: a study at the University of Toronto. Journal of Medical Ethics, 1992, 18:142–147. 12. Fulford KW, Yates A, Hope T. Ethics and the GMC core cur- riculum: a survey of resources in UK medical schools. Journal of Medical Ethics, 1997, 23:82–87. 13. Boothman RC et al. A better approach to medical malpractice claims? The University of Michigan experience. Journal of Health and Life Sciences Law, 2009, 2:125–159. 14. Fein S et al. A conceptual model for disclosure of medical er- rors. Advances in Patient Safety, 2005, 2:135–146. 15. Williams JR. Medical ethics manual, 2nd ed. Ferney-Voltaire Cedex, France, World Medical Association, 2009. 16. Del Carmen MG, Joffe S. Informed consent for medical treat- ment and research: a review. Oncologist, 2005, 10:636–641. 17. Beauchamp TL, Childress JF. Principles of biomedical ethics. New York, Oxford University Press, 2001:174–175. 18. Moskop JC et al. From Hippocrates to HIPAA: privacy and confidentiality in emergency medicine–part I: conceptual, moral, and legal foundations. Annals of Emergency Medicine, 2005, 45:53–59. 19. Shiraz B et al. Medical ethics in surgical wards: knowledge, atti- tude and practice of surgical team members in Karachi. Indian Journal of Medical Ethics, 2005, 2:94–96. 20. Geiderman JM, Moskop JC, Derse AR. Privacy and confis- dentiality in emergency medicine: obligations and chal- lenges. Emergency Medicine Clinics of North America, 2006, 24:633–656. 21. Jafarey A. Informed consent: views from Karachi. Eastern Medi- terranean Health Journal, 2006, 12(Suppl. 1):S50–S55. 22. Oken D. What to tell cancer patients. A study of medical attitudes. Journal of the American Medical Association, 1961, 175:1120–1128. 23. Ali WF. Patient’s consent to treatment with reference to the de- velopment of medical ethics in Alexandria teaching hospitals. Wiadomosci Lekarskie, 2007, 60:198–200. 24. Ruhnke GW et al. Ethical decision making and patient au- tonomy: a comparison of physicians and patients in Japan and the United States. Chest, 2000, 118:1172–1182. 25. Boisaubin EV. Observations of physician, patient and family perceptions of informed consent in Houston, Texas. Journal of Medicine and Philosophy, 2004, 29:225–236. 26. Mahmoud K. Informed consent and medical ethics. Annals of King Edward Medical College, 2005, 11:247–249. 27. Ali WF. Patient’s consent to treatment with reference to the de- velopment of medical ethics in Alexandria teaching hospitals. Wiadomosci Lekarskie, 2007, 60:198–200. attention to this particular challenge [35]. The present survey had some limitations. It relied on self-reported knowledge and perceptions and it in- volved a sample only of residents at a single teaching institution. This limits the generalization of results to other settings. The low response rate to the questionnaire may also have created a sampling bias. Nevertheless, the study highlights the need to identify physician residents who have unsatisfactory levels of knowledge, perceptions and practice towards ethical issues, to devise means to sensitize them to these issues and to train them appropriately. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 945 28. Moehring B, Schildmann J, Vollmann J. End-of-life decisions: a comparative survey on (teaching) experiences, views, and ethico-legal knowledge of final-year medical students in Germany and France. Journal of Palliative Medicine, 2011, 14(11):1206 –1210. 29. Mattick K, Bligh J. Undergraduate ethics teaching: revisiting the Consensus Statement. Medical Education, 2006, 40:329–332 30. Amin FM et al. An audit of information provided during Preop- erative informed consent. Pakistan Journal of Medical Sciences, 2006, 22:10–13. 31. Imam SZ et al. Patients’ satisfaction and opinions of their experiences during admission in a tertiary care hospital in Pakistan—a cross sectional study. BMC Health Services Research, 2007, 7:161. 32. Cleary PD et al. Patients evaluate their hospital care: a national survey. Health Affairs (Project Hope), 1991, 10:254–267. 33. Jafarey AM, Farooqui A. Informed consent in the Pakistani milieu: the physician’s perspective. Journal of Medical Ethics, 2005, 31:93–96. 34. Moazam F. Family, patient and physician in medical decision making: a Pakistani perspective. Hastings Center Report, 2000, 6:28–37. 35. International ethical guidelines for biomedical research involving human subjects. Geneva, Council of International Organiza- tions for Medical Science, 2002. 9th Global Summit of National Ethics Committees An increasing number of countries have created national ethics committees (NECs) to provide advice about ethical issues related to health to their executive and legislative branches as well as to the general public. The Global Summit of the National Ethics Committees is a biennial world meeting that has been held since 1996 and which: • Provides a unique platform for exchange of information about on-going work of the NECs. • Gives the opportunity for open debate, focusing on specific issues such as the protection of human participants in health research, stem cell research, end of life choices, etc. • Facilitates access to working documents prepared and circulated previous to the meeting. • Contributes to update information regarding the status and functions of national bodies. The 9th Global Summit of National Ethics Committees is being hosted by Tunisia in Carthage from 26 to 28 September 2012. This is the first time the Summit will take place in Africa. Further information about the Summit is available at: http://www.9gsnec.tunisia2012.rns.tn/congre_en/ EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 946 Study of adverse events of A/H1N1 vaccine among health care staff in selected provinces of Afghanistan, 2010 J.A. Mofleh,1 Z. Akbarian,2 N. Muserat,3 H. Yosofi,4 A. Alkozai 5 and B. Noormal 6 ABSTRACT At the end of March 2010 an A/H1N1 vaccination campaign was conducted in Afghanistan using donated vaccines. However, no surveillance system for detection of adverse events following immunization was in place. We report a cross-sectional, descriptive survey in 4 provinces of Afghanistan to assess the rate of adverse events among health care staff immunized with A/H1N1 monovalent vaccine 4 weeks after vaccination. Using random sampling proportionate to size, 350 staff (mean age 36 years, range 16–65 years) were surveyed using a questionnaire. The highest self-reported rates of adverse events were pain at the injection site (53%), fever in the first 3 days after immunization (40%), body pain (39%), tiredness (33%), swelling at the injection site (29%) and redness at the injection site (28%). More females than males suffered adverse reactions and the rates varied across different provinces, ranging from 79% in Balkh to 23% in Kabul. 1Eastern Mediterranean Public Health Network; 2Disease Early Warning System (DEWS) Office, Western Region; 3DEWS Office Central Region; 4DEWS Office Northern Region; 5DEWS Office Eastern Region; 6Afghan Public Health Institute, Ministry of Public Health, Kabul, Afghanistan (Correspondence to J. Mofleh: jmofleh@yahoo.com). Received: 11/07/11; accepted: 02/11/11 2010 ناتسناغفأ نم ةاقتنم تايلاو في ةيحصلا ةياعرلا في ينلماعلا ينب A/H1N1 ازنولفنلإا حاقلل ةرئاضلا ثادحلأا ةسارد لمرون يرشب ،يزوكلا ليميإ ،يفسوي ليعنيسح ،تسرم ديون ،نايبركا فيرظ ،حلفم دحمأ داوج ؛ةحنام فارطأ اهتم َّدق يتلا تاحاقللا مادختساب ناتسناغفأ في A/H1N1 ازنولفنلإا دض حيقلت ةلحم تيرجأ 2010 سرام/راذآ رهش ةيانه في :ةـصلالخا يرجأ حسلم ةيفصو ةضرعتسم ةسارد نوثحابلا مدقيو .هل ةيلاتلا ةرئاضلا ثادحلأا فشكل د ُّصترلل ماظن يأ عينمتلا دعب ذيفنتلا عضوم عضوي لمو ؤفاكتلا ديحولا حاقللاب عينمتلا اوقلت نيذلا ةيحصلا ةياعرلا في ينلماعلا ينب ةرئاضلا ثادحلأا لدعم مييقت فدهتسا ،ناتسناغفأ في تايلاو عبرأ في تلمشو ،ةساردلا مجح عم ًايئاوشع بسانتت تانيع نوثحابلا مدختسا دقو .حيقلتلا لىع عيباسأ ةعبرأ ضيم دعب كلذو ،A/H1N1 ازنولفنلإل داضلما ةرئاضلا ثادحلأل تلادعلما لىعأ تناك دقو .نايبتسا مادختساب يّرحتلا متو ،)ًاماع 65 لىإ 16 لاجلماو ،ًاماع 36 يطسولا رمعلا( ينلماعلا نم 350 بعتلاو ،)%39( ةيدسلجا ملالآاو ،)%40( عينمتلا تبقعأ يتلا لىولأا ةثلاثلا مايلأا في ىملحاو ،)%53( نقلحا عضوم في لملأا وه اهنع غلابلإا مت يتلا تفلتخاو ةرئاضلا تلاعافتلا نم لاجرلا ىناع امم رثكأ ءاسنلا تناع دقو .)%28( نقلحا عضوم رارحماو ،)%29( نقلحا عضوم مروتو ،)%33( .لباك في %23و خلب في %79 ينب ًةحِوا َرترُم ،ىرخأو ةيلاو ينب تلادعلما Étude des manifestations indésirables suite à la vaccination contre la grippe A (H1N1) chez les agents de santé dans des provinces afghanes sélectionnées sur l'année 2010 RÉSUMÉ Fin mars 2010, une campagne de vaccination contre la grippe A (H1N1) a été menée en Afghanistan avec des vaccins faisant l'objet de dons. Toutefois, aucun système de surveillance pour détecter les manifestations postvaccinales indésirables n'était en place. Nous transmettons les résultats d'une enquête descriptive transversale, menée dans quatre provinces de l'Afghanistan pour évaluer le taux de manifestations postvaccinales indésirables chez les agents de santé vaccinés avec le vaccin monovalent contre la grippe A (H1N1) quatre semaines après l'injection. Sélectionnés par échantillonnage aléatoire proportionnel, 350 agents de santé (âge moyen: 36 ans, fourchettes : 16–65 ans) ont participé à l'enquête en répondant à un questionnaire. Les manifestations postvaccinales indésirables les plus fréquemment déclarées étaient une douleur au niveau du site d'injection (53 %), de la fièvre dans les trois premiers jours suivant la vaccination (40 %), des courbatures (39 %), de la fatigue (33 %), un gonflement (29 %) ou une rougeur (28 %) au niveau du site d'injection . Les femmes étaient plus nombreuses que les hommes à être affectées par des réactions indésirables et les taux de réaction variaient entre les provinces, allant de 79 % dans la province de Balkh à 23 % dans la province de Kaboul. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 947 Introduction Emergence of the A/H1N1/2009 virus in early 2009 was the trigger for the first pandemic influenza this century [1]. The virus spread to several countries within weeks, which reduced the pos- sibility of rapid containment to zero, and the World Health Organization (WHO) declared pandemic influenza on 11 June 2009 [2]. By the time the virus was detected it was widespread in Mexico, parts of the United States and Canada, hence the rapid containment of the disease was practically impossi- ble, and other recommendations of the International Health Regulations, e.g. pharmaceutical and not pharmaceutical interventions, were implemented. On the principle of access to health and treatment, after declaration of phase 5, WHO started deploying 3 million doses of the antiviral drug osletamivir to Mexico and to 71 pre-identified low-in- come countries, of which 50 000 doses of the medicine reached Afghanistan by October 2009. Influenza A/H1N1 monovalent vaccine was licensed in September 2009 and was on the market by the end October 2009 [3,4]. Use of this vaccine was considered to be safe by producers, the Centers for Disease Control and WHO [5–9]. A total of 78 million doses of pandemic influenza vaccine were deployed to 77 countries in 2009 and 2010. In February 2010 Afghanistan received 0.5 million doses of Glaxo- SmithKline pandemic influenza vac- cine to protect the health workforce, pregnant women, young children and people with chronic illnesses. The Ministry of Public Health in Afghani- stan decided to vaccinate health care staff first and planned to receive more vaccine for other high-risk groups lat- er. Influenza vaccination was not part of the routine regime of vaccination in the country and therefore no sur- veillance system for detection of ad- verse events following immunization (AEFI) with influenza vaccine was in place. Also as the vaccines reached Afghanistan at the peak of pandemic influenza, the vaccination campaign was conducted immediately to ensure the integrity of the health system in Afghanistan and prevent a collapse in the system. This paper reports a survey to assess the rate of self-reported AEFI among a sample of health care staff im- munized with A/H1N1 monovalent vaccine in 4 provinces of Afghanistan in 2010. Methods Study design This was a descriptive, cross-sectional study with a simple random sampling proportionate to size method. Afghani- stan received the donation of vaccines from WHO in February 2010 and the vaccination campaign was conducted at the end of March 2010. This study was conducted at the end of April 2010, just over 4 weeks after administration of the vaccines. Study setting and sample The study was conducted in the 4 major provinces of Kabul, Nangarhar, Balkh and Herat. Health facilities in the selected provinces were the sampling frame and the subjects were health care workers who received H1N1 vaccina- tion. Health care facilities were ran- domly selected from the list of health facilities that received the vaccine. Study candidates were also selected randomly from the list of the health care workers receiving vaccination in the respective health facilities. A total of 27 100 people received A/H1N1 monovalent vaccine in these 4 prov- inces of Afghanistan. OpenEpi was used to draw a sample of 417 from the pool of subjects. Data collection A questionnaire was developed, field tested and applied to all study candi- dates by a team of 11 qualified trained surveyors who conducted the inter- views and filled the questionnaires. Oral consent from participants was obtained prior to interview. Reports of local AEFI were all subjective and based on the respondents’ own reports. Reports of systemic AEFI were verified where possible by medical doctors and were registered. Analysis The data were entered to Epi Info database and analysed using Epi Info and Microsoft. Excel statistical pack- ages. The rates of adverse reactions or adverse events were compared with the manufacturer’s reported rate of adverse events following immunization with the influenza monovalent vaccine [10]. The manufacturer categorized the AEFI as: common (> 1/100 but < 1/10 people), uncommon (> 1 /1000 but < 1/100), rare (1/10 000 but < 1/ 1000) and very rare (< 1/10 000). Results Background characteristics After excluding forms with inaccurate or missing data 360 of the selected candidates were included in the study: 91 (25%) females and 279 (75%) males; 86% were married. The mean age was 36 years, range 16–65 years. There were 137 in Kabul (103 males/34 females), 87 in Nangarhar (64 males/14 females), 66 in Balkh (48 males/18 females) and 80 in Herat (55 males/25 females). The dis- tribution by educational level showed 44 (12%) staff with no education (i.e. ancillary staff), 23 (6%) primacy edu- cation, 13 (4%) secondary education, 39 (11%) high school graduates, 142 (39%) paramedical (12 –15 years of education) and 100 (28%) university graduates. Adverse events reported Table 1 summarizes the rate of each ad- verse event. Pain at the site of injection EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 948 was the most common AEFI, reported by 53% of study participants. Other local reactions—swelling, redness and hardness at the injection site—were reported by 29%, 28% and 24% of re- spondents respectively. Some of the study participants (13%) experienced all the classic signs of local inflammation (heat, redness, swelling and pain). The second most common AEFI was a systemic one, fever in the first 3 days after immunization, reported by 40% of respondents. Some of the respondents (9%) had all the signs of local inflammation plus fever and 5 (1%) reported all the injection site effects together with fever, body pain and shivering. Overall, 10% of the study participants reported influenza-like ill- ness in the first 7 days after the influenza vaccination. Body pain was the third most com- mon complain, experienced by 39% of all those who received vaccine: 49% among females versus 36% among males. Other common complaints re- ported by respondents were: tiredness (33%), muscle pain (26%), headache (21%), muscle weakness (16%), chills (13%) and fainting (9%). Blurred vision, neuritis, bleeding, vasculitis and convulsions were rarely reported (0.3%, 0.3%, 0.6%, 0.8% and 0.8% re- spectively). No infections at the site of injec- tion, purulent discharge from the site of injection, anaphylactic shock or deaths were recorded after influenza immuni- zation. Adverse events reported by demographic characteristics Feelings of pain at the injection site were experienced by more females (62%) than male respondents (50%). Fever was also more common among females Table 1 Rates of adverse events following immunization of a sample of health staff in Afghanistan with A/H1N1 vaccine (n = 360) Adverse event Females (n = 91) Males (n = 270) Both sexes (n = 361) No. % No. % No. % Pain at injection site 56 62.2 136 50.4 192 53.2 Body pain 44 48.9 96 35.6 140 38.9 Fever 44 48.9 100 37.0 144 40.0 Tiredness 33 36.7 87 32.2 120 33.3 Swelling at injection site 33 36.7 71 26.3 104 28.9 Hardness at injection site 29 32.2 57 21.7 86 23.9 Redness at injection site 27 30.0 72 26.7 99 27.5 Muscle pain 27 30.0 67 24.8 94 26.1 Chills 24 26.7 24 8.9 48 13.3 Headache 23 25.4 52 19.3 75 20.8 Muscle weakness 14 15.6 42 15.6 56 15.6 Shivering 12 13.3 15 5.6 27 7.5 Sweating 10 11.0 20 7.4 30 8.3 Fainting 9 10.0 24 8.9 33 9.2 Coryza 9 10.0 17 6.3 26 7.2 Hypotension 9 10.0 15 6.5 24 6.7 Numbness 6 6.7 12 4.4 18 5.0 Tingling 6 6.7 9 3.3 15 4.2 Insomnia 5 5.6 17 6.3 22 6.1 Tinnitus 5 5.6 9 3.3 14 3.9 Neuralgia 4 4.4 6 2.2 10 2.8 Generalized rash 4 4.4 4 1.5 8 2.2 Blurred vision 2 2.2 8 3.0 10 2.8 Hypertension 2 2.2 1 0.4 3 0.8 Bruising 2 2.2 1 0.4 3 0.8 Convulsion 1 0.0 2 0.7 3 0.8 Bleeding 1 0.0 1 0.4 2 0.6 Vasculitis 0 0.0 3 1.1 3 0.8 طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 949 (54%) than males (37%). The rate of fever was 63% in Nangarhar, 56% in Herat, 49% in Balkh and 16% in Kabul. Pain at the injection site differed across residents of different provinces, ranging from 79% in Balkh to 23% in Kabul. The highest rate of pain among females was reported in Nangarhar province (93%), followed by Herat (84%), Balkh (72%) and Kabul (26%). Experience of AEFI also varied by edu- cational level. Pain was reported more in those with only had primary edu- cation (83%), followed by university graduates (58%), high-school gradu- ates (55%), no education (ancillary staff) (54%), paramedics (46%) and secondary education (39%). Discussion The study showed that the highest rate of AEFI was a local one—pain at the injection site—in 53% of partici- pants. A systemic AEFI—fever—was reported by 40%. All the classic signs of local inflammation (heat, redness, swelling and pain) were reported by 13% of subjects. The findings of this study agree with a study conducted in Australia, in which the highest rate of adverse events were systemic (53.8%) and injection site events (56.3%) [11]. However, comparing the results of this study with similar studies conducted in the United States of America and Canada prior to the licensing of the A/H1N1 vaccine [5] (Table 2) show much higher rates of pain, redness and swelling at the injection site in our study and for all adverse events in this study except for headache. The find- ings were also not replicated in another study conducted in Canada in 2010 [12] and a multi-centre double-blind randomized trail conducted in China in 2009 in which the authors reported a local reaction of 5.3% and pain of 4.6% among a sample size of over 11400 individuals who received the first dose of the influenza vaccine [13]. However, a study in Canada reported that the incidence of AEFI was higher in those who received higher doses of the AH1N1 vaccine compared with those who received lower doses of the vaccine and this may be one of the reasons for a higher number of adverse effects in Afghanistan. There are many factors that can contribute to higher rates of adverse effects following immunization: a higher dose than recommended may be administered, (e.g. > 15 µg); poor training of injectors may increase the risk of AEFI (e.g. shallow or deep injections may cause pain, redness and swelling of the injection site); and knowledge of participants about adverse events may lead to bias in an- swering. Among many difficulties that Afghanistan faced in this the first ever influenza vaccination campaign in the country were limited trained vaccina- tors, limited capacity for the logistic of vaccines which were not included in the routine immunization schedule, willingness of the vaccine recipients and attitudes of people toward the vaccine and vaccine safety. This study had some limitations which should be noted. A cross-sec- tional study was conducted, while a prospective, cohort study would have been better. The results were a snapshot which was taken 4 weeks after vaccina- tion. No objective measures were in- cluded in the study; all responses were based on the participants’ self-reports and we could not clinically verify their responses because it was 4 weeks after the administration of the vaccine. As most of the reactions were mild and the study population was health care workers they were either self-treated or treated by colleagues without proper registration of the adverse event. No AEFI surveillance system for influenza vaccination was available in Afghani- stan to compare our data with. The study was conducted only 4 weeks after implementation of the influenza vac- cination campaign and therefore no delayed reactions/adverse effects were recorded. Nevertheless, this was the first time that a major influenza vaccina- tion campaign was conducted at the country level and it provides useful baseline data for future research. Im- plementation of an H1N1 vaccination campaign needs in-depth planning, logistics and training to reduce associ- ated adverse effects and ensure injec- tion safety. A proper AEFI surveillance system should be implemented to capture all events associated with the immunization. Table 2 Comparison of the rates of adverse events in the current study in Afghanistan with data from the United States of America (USA) and Canada prior to the licensing of the A/H1N1 vaccine [5] Adverse event Rate of adverse event (%) USA Canada Afghanistan (current study) Pain 24 21 53 Fever 11 1 40 Fatigue 17 10 33 Redness 11 14 28 Swelling 10 6 29 Myalgia 13 11 26 Headache 28 10 21 Chills 5 3 13 EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 950 References 1. Influenza-like illness in the United States and Mexico. Briefing note, 24 April 2009. Geneva, World Health Organization, 2009 (http://www.who.int/csr/don/2009_04_24/en/index.html, accessed 31 July 2012). 2. DG statement following the meeting of the emergency committee. Briefing note, 11 June 2009. Geneva, World Health Organiza- tion, 2009 (http://www.who.int/csr/disease/swineflu/4th_ meeting_ihr/en, accessed 31 July 2012). 3. Use of influenza A (H1N1) 2009 monovalent vaccine. Recom- mendations of the Advisory Committee on Immunization Practices (ACIP), 2009. Morbidity and Mortality Weekly Report, 2009, 58(early release):1–8. 4. Pandemic influenza vaccines: current status. Pandemic (H1N1) 2009. Briefing note 11, 24 September 2009. Geneva, World Health Organization, 2009 (http://www.who.int/csr/disease/ swineflu/notes/pandemic_influenza_vaccines_20090924/ en/index.html, accessed 31 July 2012). 5. Safety of pandemic influenza A(H1N1) 2009 vaccines. Extract from report of GACVS meeting of 8–9 December 2010, pub- lished in the WHO Weekly Epidemiological Record on 28 Janu- ary 2011. Geneva, World Health Organization, 2011 (http:// www.who.int/vaccine_safety/topics/influenza/pandemic/ h1n1_safety_assessing/Dec_2010/en/index.html, accessed 31 July 2012). 6. Safety of pandemic A (H1N1) influenza vaccines. Extract from report of GACVS meeting of 3–4 December 2009, published in the WHO Weekly Epidemiological Record on 29 January 2010. Geneva, World Health Organization, 2010 (http:// www.who.int/vaccine_safety/topics/influenza/pandemic/ h1n1_safety_assessing/Dec_2009/en/index.html, accessed 31 July 2012). 7. Safety of pandemic vaccines. Pandemic (H1N1) 2009. Briefing note 6, 6 August 2009. Geneva, World Health Organization, 2009 (http://www.who.int/csr/disease/swineflu/notes/ h1n1_safety_vaccines_20090805/en/index.html, accessed 31 July 2012). 8. Pandemic influenza vaccines: current status, Pandemic (H1N1) 2009. Briefing note 11, 24 September 2009 Geneva, World Health Organization, 2009 (http://www.who.int/csr/disease/ swineflu/notes/pandemic_influenza_vaccines_20090924/ en/index.html, accessed 31 July 2012). 9. Safety of pandemic (H1N1) 2009 vaccines. Briefing note, 30 Octo- ber 2009. Geneva, World Health Organization, 2009 (http:// www.who.int/csr/disease/swineflu/frequently_asked_ques- tions/vaccine_preparedness/safety_approval/en/index. html, accessed 31 July 2012). 10. Highlights of prescribing information. Quebec City, GlaxoS- mithKline, 2010 (http://www.fda.gov/downloads/Biologics- BloodVaccines/Vaccines/ApprovedProducts/UCM190377. pdf, accessed 8 August 2012). 11. Greenberg ME et al. Response to a monovalent 2009 influ- enza A (H1N1) vaccine. New England Journal of Medicine, 2009, 361:2405–2413. 12. Summary of Manitobans vaccinated for pandemic H1N1 influ- enza. Surveillance report, 7 February 2010. Winnipeg, Manitoba, Province of Manitoba Health Department, 2010 (http://www. gov.mb.ca/health/publichealth/surveillance/h1n1/stats4. html#2, accessed 31 July 2012). 13. Liang XF JZ et al. Safety and immunogenicity of 2009 pan- demic influenza A H1N1 vaccines in China: a multicentre, double-blind, randomised, placebo-controlled trial. Lancet, 2010, 375:56–66. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 951 Estimation of malaria transmission intensity in Sennar state, central Sudan Z.A. Elmahdi,1 A.A. Nugud 2 and I.M. Elhassan 3 ABSTRACT Understanding the behaviour of malaria vectors is crucial for planning mosquito control programmes. The aim of this study was to estimate the malaria transmission intensity in 2 different ecological zones in a highly endemic malaria area of Sennar state in central Sudan over the main transmission period. Species confirmation by PCR indicated that Anopheles arabiensis was the only malaria vector in the study area, with high anthropophilic behaviour (84.9% human-feeding). ELISA studies showed Plasmodium falciparum sporozoite rates rose from 1.8% to 4.5% and the average entomological inoculation rates rose from 2.4 to 4.2 infectious bites per person per night in September (the beginning) to November (the end) of the 3-month transmission season. The proportion of malaria-positive slides ranged from 50.1% to 57.0%. The proportion of human-blood positive mosquitoes was significantly higher in the irrigated area (El Booster) compared with the non-irrigated area (Rahal). 1Institute of Endemic Diseases, University of Khartoum, Khartoum, Sudan (Correspondence to I.M. Elhassan: ibrahimelhassan@iend.org). 2National Health Laboratory, Federal Ministry of Health, Khartoum, Sudan. 3Faculty of Medicine, University of Jazan, Jazan, Saudi Arabia. Received: 20/01/11; accepted: 21/06/11 نادوسلا طسو ،راَّنس ةيلاو في ايرلالما ةياسر ةدش ريدقت نسلحا دممح ميهاربإ ،دقن ديملحا دبع ،يدهلما بنيز ةدش ريدقت لىإ ةساردلا هذه فدتهو .ضوعبلا ةحفاكم جمابرل طيطختلا في ةيهملأا غلاب ًارمأ ايرلاملل لقانلا ضوعبلا تايكولس مهف دَعُي :ةـصلالخا لىع كلذو ،نادوسلا طسو في رانس ةيلاو في ايرلاملل عفترم نطوت نم نياعت ةقطنم نمض ةيئيبلا تماسلا ثيح نم نافلتتخ ينتقطنم في ايرلالما ةياسر ديحولا لقانلا يه ةيبرعلا ةليفونلأا نأ زارميلوبلل ليسلسلا لعافتلا مادختساب ضوعبلا عاونأ نم دكأتلا حضوأ دقو .ةيسيئرلا ةياسرلا ةترف ىدم يعانلما زتملماب ةساردلا تحضوأو .)%84.9( ةعفترم ةجردب شربلا لىع ىّذغتت نأ لّضفت ةيكولسلا ةيحانلا نم انهأو ،ةساردلا ةقطنم في ايرلاملل نم تعفترا دق تاشرحلل حيقلتلا تلادعم نأو ،%4.5 لىإ لصتل %1.8 نم تعفترا دق ةيلجنلما تاروصملل ةيغوبلا تاناويلحا تلادعم نأ )ازيلإ( ةياسرلا لصف في برمفون/نياثلا نيشرت رهشب ًاءاهتناو برمتبس/لوليأ رهش نم ًاءدب كلذو ،4.2 لىإ لصتل ةليل لكل صخش لكل ةيدْعُم تاَعْسَل 2.4 مدل ةيبايجلإا تاضوعبلل ةيوئلما ةبسنلا تناك دقو .%57.0و %50.1 ينب تحواترف ايرلاملل ةيبايجلإا حئاشرلل ةيوئلما ةبسنلا امأ .رهشأ ةثلاثل دتمي يذلا .)لاحرلا( ةيورلما يرغ قطانلما في اهنم )ترسوبلا( ة َّيِوْرَلما قطانلما في هب ُّدَتْعُي رادقمب لىعأ شربلا Estimation de l'intensité de la transmission du paludisme dans l'État de Sennar (centre du Soudan) RÉSUMÉ La compréhension du comportement des vecteurs du paludisme est cruciale pour la planification des programmes de lutte contre ces moustiques. L'objectif de la présente étude était d'estimer, pendant la principale période de transmission, l'intensité de la transmission du paludisme dans deux zones écologiquement différentes de l'état de Sennar, au centre du Soudan, où le paludisme est fortement endémique. La confirmation de l'espèce par la méthode PCR a indiqué qu'Anopheles arabiensis était le seul vecteur du paludisme présent dans la zone étudiée, et qu'il avait un comportement hautement anthropophile (alimentation d'origine humaine à 84,9 % ). La méthode ELISA a révélé que les taux sporozoïtiques de Plasmodium falciparum avaient augmenté de 1,8 % à 4,5 % et les taux d'inoculation entomologique moyens de 2,4 à 4,2 piqûres infectieuses par personne et par nuit entre septembre, le début de la période, et novembre, c'est-à-dire la fin de la saison de transmission qui dure trois mois. La proportion de lames positives pour le paludisme allait de 50,1 % à 57,0 %. La proportion de moustiques positifs pour le sang humain était significativement plus elevée dans la zone irriguée (El Booster) que dans la zone non irriguée (Rahal). EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 952 Introduction Malaria is a major health problem in Sudan. With the exception of a small area at the Sudan–Egypt border, the whole population is considered to be at risk of infection. More than 95% of malaria cases in Sudan are due to Plas- modium falciparum [1] and mosquitoes of the Anopheles gambiae complex Giles are the main vectors [2–6]. As malaria transmission in Africa is characterized by complex species heterogeneity, un- derstanding the factors that facilitate transmission are very important in tar- geted malaria control and eradication programmes. The sporozoite infection rate (SR) of the mosquito population is a power- ful and sensitive measure for describing the epidemiology of malaria in a particu- lar area. It is also used to determine the entomological inoculation rate (EIR), to establish both vector identity and dif- ferences in transmission intensity over space and time and for the assessment of malaria control programmes. The EIR estimates the level of exposure to P. falciparum-infected mosquitoes. It is the most common measure for assessing malaria endemicity and transmission intensity and is particularly useful when estimating the effect of human–vector contact [7]. SR and EIR are largely de- pendent on environmental conditions and would be expected to vary seasonal- ly. Studies have compared transmission intensity in different ecological zones and have reported significant variations in the transmission intensity of malaria between rural and urban settings within the same locality and even between villages separated by short distances [8]. Development of enzyme-linked immunosorbent assays (ELISA) for the detection of circumsporozoite protein provide an effective tool for determina- tion of the infection rates in mosquitoes and the consequent incrimination of malaria vector(s) [9] and for identify- ing the blood meals of vectors [10]. Understanding the feeding preferences of a given malaria vector has paramount epidemiological importance in selec- tion of the appropriate vector control measure(s). In the present study, we used poly- merase chain reaction (PCR) for the identification of the vector responsible for malaria transmission and ELISA to determine the SR and blood-meal sources of the malaria vectors in an area of central Sudan characterized by highly seasonal malaria transmission. Methods Study area The study was conducted in Sennar state which lies in a rich savannah region between latitude 12.5°–14.7° N and longitude 32.9°–35.4° S in central east- ern Sudan. It shares borders with Gezira state in the north, White Nile and Up- per Nile states in the west, Gadarif state in the east and Blue Nile state and Ethio- pia in the south. The summer extends from March to May, with average daily temperatures of 32–40° C and relative humidity of 25%. The rainy season starts early in June and continues until Sep- tember. Winter starts in October with average daily temperatures of 20–25 °C [Sudan Meteorological Services, 2005, unpublished]. The population is close to 2 million. The study area is characterized by stable and highly seasonal malaria transmis- sion. For the purpose of this study the following sites were selected: • Rahal area (33°55' E, 13°10' N); lies in central-east Sennar and it is an ur- ban area. Most of the people work as employees in different organizations or have their own jobs. The houses are generally well-constructed. Around 90% of the people, especially children, use impregnated or non-impregnated bednets. The main mosquito breed- ing sites are rain pools. • El Booster area in western Sennar (33°36' E, 13°32' N). It lies in the middle of cotton and sugarcane agri- cultural schemes. Most of the people are working in farming in cotton and sugar schemes. The houses are poorly constructed and the use of bednets is not common. The breeding sites are mainly rain pools and different types of drains. Ethical approval for this project was obtained from the national ethics committee and the ethics committee of the Institute of Endemic Diseases, University of Khartoum. Mosquito sampling and processing Mosquito collections Entomological surveys were conducted during the main malaria transmission months between September and No- vember 2008. Mosquito collections were made once per month from each site from 7 randomly selected rooms by using the pyrethrum spray catch method. Mosquitoes were sorted ac- cording to their feeding status as unfed, fresh fed, half-gravid and gravid and then identified to species based on morphological characteristics [11]. Specimens resembling An. gambiae were individually preserved in labelled tubes containing silica gel and stored at room temperature until further processing. Species identifications using PCR Females belonging to the An. gambiae complex were identified to species using the PCR technique described by Scott et al. [12]. A small segment from a leg of a mosquito was placed directly into reaction mixture containing species- specific primers, dNTPs, buffer and polymerase. The resultant amplicon was detected on 2% agarose gel stained with ethidium bromide. Detection of sporozoite infections For detection of sporozoite infections, circumsporozoite ELISA was carried out as previously described [9,13]. In brief, the head and thorax of each mos- quito were separated from the abdo- men and tested for the presence of P. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 953 falciparum circumsporozoite antigen. Mosquitoes were ground in 50 µL of boiled casein containing (Igepal CA 630) and the final volume was brought to 250 µL with blocking buffer; 50 µL of the triturate was used in sporozoite enzyme-linked immunosorbent assays. Positive reactions were determined visually and spectrophotometrically. All ELISA chemicals were obtained from Sigma and the reagents (MAPs and peroxidases) were provided by the US Centers for Disease Control and Prevention and Kirkegaard and Perry Laboratories Inc. Identification of blood-meal sources All blood-fed An. arabiensis were tested against anti-human antigens using the ELISA method of Bier et al. [14]. Mos- quito abdomens were ground in 50 µL grinding buffer and completed the total volume to 450 µL; 50 µL of mos- quito homogenate was used for blood meal detection. Positive samples were determined visually and spectrophoto- metrically. Calculation of SR and EIR SR was calculated by dividing the number of mosquitoes positive for P. falciparum by the number of tested mosquitoes. The EIR, expressed as the number of infectious bites per person per unit time, was derived as a product of the SR and the human biting rate [15]. The monthly EIRs were derived by multiplying the daily EIR [infectious bites per person per night (ib/p/n)] by 30 days. Parasitological data Parasitological data on the prevalence of malaria in the area were obtained from the malaria health authority’s reports (State Ministry of Health, cen- tral Sudan). These data were used to correlate entomological parameters with the prevalence of malaria in the study area. Statistical analysis Data entry was performed using the Microsoft Excel program. SPSS, ver- sion 11.5 software was used for data analysis. Frequency distributions were calculated. Student t-test was used to evaluate whether there was a significant difference between measures from the 2 areas. The level of P < 0.05 was used to indicate statistical significance. Results Vector identification A total of 970 anopheline mosquitoes were collected. Morphological iden- tification and subsequent application of species-specific PCR analysis con- firmed that An. arabiensis was the only member of the An. gambiae complex in the study area. Out of 412 mosquitoes (randomly selected) 370 (89.8%) were identified as An. arabiensis. Sporozoite infection rate All An. arabiensis were tested for the presence of P. falciparum sporozoite and 30 specimens were found to be posi- tive. The mean positivity rates over the 3-month period were 4.2% (SD 2.0%) and 2.3% (SD 0.8%) in El Booster and Rahal areas respectively (P > 0.05). The SR values increased from September through to November in each area (Table 1) and both areas combined (Figure 1). Blood-meal sources A total of 360 An. arabiensis collected from the study sites were tested for human blood and 307 mosquitoes (85.3%) were positive for human blood. The proportion of mosquitoes that were human-blood positive was significantly higher in El Booster (86.9%) than in Rahal (82.9%) (P = 0.01). Entomological inoculation rate The mean EIR over the 3-month transmission season was significantly higher in El Booster [5.6 (SD 2.0) ib/p/n] than Rahal [2.1 (SD 0.5) ib/p/n ] (P = 0.01). The EIR in the 2 areas combined increased from September through to November (Figure 1). The highest EIR values were recorded in November in each area: 6.8 and 2.5 ib/p/n for El Booster and Rahal respectively. Table 1 Monthly sporozoite and entomological inoculation rates in the 2 study areas of Sennar state, central Sudan Measure/area Month Mean (SD) Sep. Oct. Nov. Sporozoite rate (%) El Booster 2.1 4.4 6.0 4.2 (2.0) Rahal 1.4 2.5 3.0 2.3 (0.8)a Entomological inoculation rate (infectious bites/person/night) El Booster 3.3 6.6 6.8 5.6 (2.0) Rahal 1.5 2.3 2.5 2.1 (0.5)b aP > 0.05; bP = 0.01 between the 2 areas. SD = standard deviation EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 954 Relationship between parasitological and entomological parameters The proportion of malaria-positive slides rose from 50.1% in September through to 57.0% in November. High positive correlations were found between the proportion of malaria-positive slides and both SR (0.88) and EIR (0.90) (Figure 1). Discussion Information concerning the species composition, biology and behaviour of malaria vectors is crucial for under- standing the vector’s epidemiological role in malaria transmission and there- fore for designing and planning control programmes. Previously, vector identifi- cation in Sudan was conducted through morphological studies which indicated that the principal vector of malaria in the country was An. gambiae s.l. The subse- quent genetic studies (crossing experi- ments) revealed that the only member of An. gambie found in central Sudan was An. gambiae species B [16,17]. This was followed by cytotaxonomy stud- ies which showed that An. arabiensis was the only malaria vector in central Sudan [5]. The non-extracted DNA protocol of Scott et al. has been shown to be sensitive [12], easy and suitable for routine work as large number of samples can be identified in a single day [18]. In the present study, we used the PCR technique in order to confirm the species identification carried out earlier in Sudan. Our results confirmed that An. arabiensis was the only member of An. gambiae complex present in the study areas. In this study we found that the SR ranged from 1.4% to 6.0%. The highest rates were reported during the peak of the transmission season (November) in both sites, when the conditions are less favourable for breeding and the vector population is mostly composed of older mosquitoes [19]. The infectiv- ity rates in both study sites were found to be within the normal range reported for members of An. gambiae complex [20]. Increases in SR were observed from the beginning of the transmis- sion season (September) to the peak of the rainy season (November) in both areas. A relatively broad range of SRs have been recorded in other African countries [21–23]. In central Sudan, no sporozoites were detected in all dissec- tions carried out in the Blue Nile health project area which covered the Gezira irrigation in the period between 1981 and 1989. However, Elsafi reported a SR of 0.45% in Gezira area of central Su- dan in 1992 [unpublished data], while in the Gadaref area of eastern Sudan, a SR of 1.44% was reported during the malaria transmission season of 1995 [6]. The relatively high SRs recorded in the present study could be explained the high anthropophilic feeding behaviour of An. arabiensis in the absence of other animals. In addition, all samples of the current study were collected during the malaria transmission season. Malaria transmission intensity is best expressed as the EIR, which di- rectly reflects the exposure of humans to pathogenic Plasmodium spp. parasites [24]. It has been observed that EIR val- ues in Africa vary widely both spatially and temporally from 0 to 884 infectious bites per person per year [25], with rural areas experiencing higher intensities of transmission than urban areas [24]. In the present study we compared 2 sites representing rural and irrigated areas in order to get more accurate informa- tion on variations in the transmission of malaria in different environmental and ecological situations of this region. Ir- rigation schemes provide ideal breeding 1.8 3.4 4.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Sep. Oct. Nov. % o f m o sq u it o es 2.4 3.3 4.2 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Sep. Oct. Nov. In fe ct io u s b it es p er p er so n p er n ig h t 50.1 50.2 57.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 Sep. Oct. Nov. % o f s lid es Figure 1 Monthly sporozoite rate, entomological inoculation rate and proportion of malaria-positive slides in the 2 study areas of Sennar state, central Sudan (combined data) (a) Sporozoite rate (b) Entomological inoculation rate (c) Malaria-positive slides rate طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 955 sites for malaria vectors in Africa [26]. However, no simple association has been found between irrigation and the degree of exposure to malaria parasites (EIR) as measured using classical entomological methods [27,28]. Our results indicated that exposure to ma- laria infection was significantly higher in the study population of the irrigated area (El Booster) compared with the non-irrigated area (Rahal). This result is in agreement with a previously study carried out in Ghana [28]. A number of studies have reported that the high EIR values and therefore high risk of exposure to infective bites usually occur during the wet seasons in which 75% of the annual EIR is re- corded [24]. Our results showed that the overall EIR ranged between 1.5–6.9 ib/p/n. These values were comparable to other EIR values reported from dif- ferent African countries,; for example, in Kenya Shililu et al. reported a range of 0–7 ib/p/ts [29]. Furthermore, ranges of 3–5 ib/p/ts and 5–12 ib/p/ts were reported from Gambia and Burundi respectively [30,31]. Nonetheless the highest ranges of EIR (122–702 ib/p/ ts) have been recorded from Tanzania [26,32]. We noted an increase in the risk of malaria infection from the beginning of the malaria transmission season (Sep- tember) through to the end of the ma- laria season (November), as indicated by the increases in SR and EIR values. A significant positive correlation was found between the SR, EIR and the proportions of malaria-positive slides reported during the study months. Identification of blood-meal sourc- es for mosquitoes is an important fac- tor in the study of malaria transmission and vectorial capacity [33]. Also feed- ing behaviour is of great importance with respect to the planning of vector control programmes [34]. Our overall results of blood-meal analysis showed that the main source of blood for the majority (almost 85%) of An. arabiensis was human. Generally, An. arabiensis is regarded as being more exophilic, ex- ophagic and zoophilic than An. gambiae s.s., but although it is a highly efficient vector of malaria [34], it appears to be an east–west behavioural cline. For An. arabiensis populations in West Africa, the proportion of blood meals from hu- mans has been reported as 80%–100% indoors, with most feeding and resting occurring indoors [35]. By contrast, in East Africa, a greater proportion of An. arabiensis feed on cattle and rest out- doors. In Tanzania, for example, White et al. reported human blood indices of 61% and 7% for samples collected in- doors and outdoors, respectively [36]. These behavioural differences may be linked to observed genetic variability [37] and the opportunistic behaviour of An. arabiensis [38]. However, the high rate of feeding on humans in this study showed that humans were indeed the preferred host, and An. arabiensis is highly anthropophilic. These results confirmed a previously published report from Ethiopia [39]. In the current study the proportion of human-blood positive mosquitoes was significantly higher in El Booster compared with Rahal. This could be explained by the fact that inhabitants of Rahal area tend to keep cattle and other grazing animals around their set- tlements and these animals may have a zooprophylactic effect. In this study, the high percentage of indoor collected An. arabiensis females from the 2 areas were found to be fed on humans. Thus, the use of long-lasting insecticidal nets and residual spraying are the most suit- able measures for malaria control in central Sudan. Conclusions This study confirmed that An. arabien- sis is the only member of An. gambiae complex present in Elbooster and Rahal areas of Sennar state. Malaria in central Sudan is highly seasonal with the peak during the known transmission season occurring in November. Acknowledgements The technical assistance of staff of the Malaria Training Centre, Sennar state, is greatly acknowledged. Dr Robert Wirtz and Dr Melissa Avery, Department of Medical Entomology, Centre of Dis- ease Control and Prevention, Atlanta, United States of America, are greatly acknowledged for providing ELISA reagents. Dr M. Balkew, University of Addis Ababa, is thanked for his techni- cal advice and valuable comments. This study obtained financial assistance from the MIM/TDR research project (ID No. TDA 40048A). References 1. Elhassan IM et al. High proportion of subclinical Plasmodium falciparum infections in an area of seasonal and unstable ma- laria in Sudan. American Journal of Tropical Medicine and Hy- giene, 1995, 53:78–83. 2. Zahar AR. Vector bionomics in the epidemiology and control of malaria. Part 1. Section Ш, (D) East Africa. Geneva, World Health Organization, 1985 (WHO/DOCVBC/85.3). 3. Lewis DJ. Some mosquitoes of the Blue Nile valley in Republic of the Sudan. Bulletin of Entomological Research, 1958, 49:133–155. 4. Haridi AM. Partial exophily of Anopheles gambiae species B in the Khashm Elgirba area in eastern Sudan. Bulletin of the World Health Organization, 1972, 46:39–46. 5. Petrarca V et al. Cytogenetics of the Anopheles gambiae com- plex in Sudan, with special reference to An. arabiensis: rela- tionships with East and West African populations. Medical and Veterinary Entomology, 2000, 14:149–164. 6. Karrar AA. Seasonality of malaria transmission in Assar and Daraweesh villages (Gadaref Province, Gadaref, State, Eastern EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 956 Sudan [MSc thesis]. Khartoum, Sudan, Department of Zoology, University of Khartoum, 1996. 7. Kelly-Hope LA, McKenzie FE. The multiplicity of malaria trans- mission: a review of entomological inoculation rate meas- urements and methods across sub-Saharan Africa. Malaria Journal, 2009, 8:19. 8. Robert V et al. Malaria transmission in urban sub-Saharan Af- rica. American Journal of Tropical Medicine and Hygiene, 2003, 68:169–176. 9. Wirtz RA et al. Field evaluation of enzyme-linked immunosorb- ent assays for Plasmodium falciparum and Plasmodium vivax sporozoites in mosquitoes (Diptera: Culicidae) from Papua New Guinea. Journal of Medical Entomology, 1987, 24:433–437. 10. Voller A et al. Enzyme-linked immunosorbent assay. In: Rose NR, Fredman H, eds. Manual of clinical immunology, 2nd ed. Washington DC, American Society of Microbiology, 1980:359– 371. 11. Gillies MT et al. The Anophelinae of Africa south of Sahara (Ethiopian zoogeografical region). Johannesburg, South African Institute of Medical Research, 1968 (Publication No. 5). 12. Scott JA, Brogdon WG, Collins FH. Identification of single specimens of the Anopheles gambiae complex by the polymer- ase chain reaction. American Journal of Tropical Medicine and Hygiene, 1993, 49:520–529. 13. Zavala F et al. Monoclonal antibodies to circumsporozoite proteins identify the species of malaria parasite in infected mosquitoes. Nature, 1982, 299:737–738. 14. Beier JC et al. Bloodmeal identification by direct enzyme- linked immunosorbent assay (ELISA), tested on Anopheles (Diptera: Culicidae) in Kenya. Journal of Medical Entomology, 1988, 25:9–16. 15. MacDonald G. The epidemiology and control of malaria. Ox- ford, Oxford University Press, 1957. 16. Lewis DJ. Some mosquitos of the Sudan. Bulletin of Entomologi- cal Research, 1956, 47:723–735. 17. Petrarca V et al. Dati preliminary sul complesso Anopheles gam- biae in Sudan. Parassitologia, 1986, 28:304–306. 18. Van Rensburg AJ et al. The polymerase chain reaction method as a tool for identifying members of the Anopheles gambiae complex (Diptera:Culicidae) in northeastern Tanzania. Journal of the American Mosquito Control Association, 1996, 12:271–274. 19. Beier JC. Malaria parasite development in mosquitoes. Annals Review of Entomology, 1998, 43:519–543. 20. Magesa SM et al. Trial of pyrethroid impregnated bednets in an area of Tanzania holoendemic for malaria. Part 2. Effects on the malaria vector population. Acta Tropica, 1991, 49:97–108. 21. Joshi GP, Service MW, Pradhan GD. A survey of species A and B of the Anopheles gambiae Giles complex in the Kisumu area of Kenya prior to insecticidal spraying with OMS-43 (feni- trothion). Annals of Tropical Medicine and Parasitology, 1975, 69:91–104. 22. Fettene M et al. Behaviour of Anopheles arabiensis and Anoph- eles quadriannulatus sp. B mosquitoes and malaria transmission and Southwestern Ethiopia. Africa Entomology, 2004, 12:83–87. 23. Okbaldet YB. Biology and transmission potential of malaria vector moasquitoes in Elabered sub-zone, Eritrea [MSc thesis]. Bloemfontein, South Africa, Department of Zoology and Ento- mology, Faculty of Natural and Agricultural Sciences, Univer- sity of Free State, 2001. 24. Hay SI et al. Annual Plasmodium falciparum entomological inoculation rates (EIR) across Africa: literature survey, Internet access and review. Transactions of the Royal Society of Tropical Medicine and Hygiene, 2000, 94:113–127. 25. Mbogo CN et al. Relationships between Plasmodium falcipa- rum transmission by vector populations and the incidence of severe disease at nine sites on the Kenyan coast. American Journal of Tropical Medicine and Hygiene, 1995, 52:201–206. 26. Temu EA et al. The role of four anopheline species (Diptera: Culicidae) in malaria transmission in coastal Tanzania. Transac- tions of the Royal Society of Tropical Medicine and Hygiene, 1998, 92:152–158. 27. Ijumba JN et al. Irrigated crop production is associated with less malaria than traditional agricultural practices in Tanzania. Transactions of the Royal Society of Tropical Medicine and Hy- giene, 2002, 96:476–480. 28. Appawu M et al. Malaria transmission dynamics at a site in northern Ghana proposed for testing malaria vaccines. Tropical Medicine and International Health, 2004, 9:164–170. 29. Shililu JI et al. Seasonal density, SRs and entomological inocula- tion rates of Anopheles gambiae and Anopheles funestus in high- altitude sugarcane growing zones in Western Kenya. Topical Medicine and International Health, 1998, 3(9):706–710. 30. Lindsay SW et al. A malaria control trial using insecticide-treat- ed bed nets and targeted chemoprophylaxis in a rural area of The Gambia, west Africa. 7. Impact of permethrin-impregnated bed nets on malaria vectors. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1993, 87(Suppl. 2):45–51. 31. Van Bortel W et al. Deltamethrin-impregnated bednets as an operational tool for malaria control in a hyper-endemic region of Burundi: impact on vector population and malaria morbid- ity. Tropical Medicine & International Health, 1996, 1:824–835. 32. Shiff CJ et al. Malaria infection potential of anopheline mos- quitoes sampled by light trapping indoors in coastal Tanzanian villages. Medical and Veterinary Entomology, 1995, 9:256–262. 33. Hewitt S, Rowland M. Control of zoophilic malaria vectors by applying pyrethroid insecticides to cattle. Tropical Medicine and International Health, 1999, 4:481–486. 34. White GB. Anopheles gambiae complex and disease transmis- sion in Africa. Transactions of the Royal Society of Tropical Medi- cine and Hygiene, 1974, 68:278–301. 35. Molineaux L, Gramiccia G. The Garki project. Geneva, World Health Organization, 1980:109–115. 36. White GB. The Anopheles gambiae complex and malaria trans- mission around Kisumu, Kenya. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1972, 66:572–581. 37. Coluzzi M et al. Chromosomal differentiation and adaptation to human environments in the Anopheles gambiae complex. Transactions of the Royal Society of Tropical Medicine and Hy- giene, 1979, 73:483–497. 38. Gillies MT, Coetzee M. A supplement to the anopheline of Africa South of the Sahara. Johannesburg, South African Institute of Medical Research, 1987 (Publication No. 55). 39. Tirados I et al. Blood-feeding behaviour of the malarial mos- quito Anopheles arabiensis: implications for vector control. Medical and Veterinary Entomology, 2006, 20:425–437. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 957 Drug resistance pattern and outcome of treatment in recurrent episodes of tuberculosis M. Marjani,1 P. Baghaei,1 P. Tabarsi,1 M. Shamaei,1 D. Mansouri,1 M.R. Masjedi 1 and A.A. Velayati 1 ABSTRACT Patterns of drug resistance in recurrent cases of tuberculosis may be different than in those without a history of treatment. In this retrospective study, the drug resistance pattern and outcome of treatment with DOTS category I (CAT I) regimen was compared in 63 recurrent cases and 872 new cases of pulmonary tuberculosis from April 2003 to January 2008 at the National Research Institute of Tuberculosis and Lung Disease in Tehran, Islamic Republic of Iran. Resistance to isoniazid and ethambutol was significantly more common in recurrent cases, but there were no differences in rates of resistance to rifampin, pyrazinamide, streptomycin or the rate of multi-drug resistant strains. Resistance to streptomycin was the most common. No significant differences in treatment outcome and deaths were found between the 2 groups. Due to the low frequency of multi-drug resistance in the recurrent cases, a CAT I regimen may be suitable for empirical therapy before drug sensitivity results become available. 1Clinical Tuberculosis and Epidemiology Research Centre, National Research Institute of Tuberculosis and Lung Disease, Masih Daneshvari Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Islamic Republic of Iran (Correspondence to M. Marjani: marjani@nritld.ac.ir). Received: 20/10/10; accepted: 08/01/11 لسلل ةرركتلما تماجلها في ةلجاعلما لئاصحو ةيودلأل ةمواقلما طمانأ یتیلاو برکا یلع ،یدجسم اضر دممح ،یروصنم دوواد ،ییماش دوعسم ،یسبرط مایب ،ییاقب هناورب ،نياجرم دیمج نوثحابلا نراق ،ةیداعتسلاا ةساردلا هذه فيو ةلجاعم قباوس مضت يتلا كلت نع لسلل ةدواعلما تلاالحا في ةیودلأل ةمواقلما طمانأ فلتتخ دق :ةـصلالخا تلاالحا نم 63 ىدل )ستود( شرابلما فاشرلإا تتح دملأا ةيرصقلا ةلجاعلما ةیجیتاترسا نم I ةئفلا ماظنب ةلجاعلما لئاصحو ةیودلأل ةمواقلما طمن لوح ثوحبلل ينطولا دهعلما في ،2008 ریانی/نياثلا نوناکو 2003 لیربأ/ناسین ينب ةترفلا في يوئرلا لسلل ةدیدلجا تلاالحا نم 872و ةرركتلما في ابه ُّدَتْعُی ةجردب ًاعویش رثکأ تناک لوتوبماثیلإلو دیزاینوزیلإل ةمواقلما نأ حضتاو ،ةیملاسلإا ناریإ ةیروهجم في ،نارهط في ةئرلا ضارمأو لسلا ةیودلأ ةمواقلما يرارذلا لدعم في لاو ،ينسیموتبترسلاو دیمانیزايربلاو ينبمافیرلل ةمواقلما تلادعم في قرف كانه نكی لم ينح في ،ةرركتلما تلاالحا ًةجیتنو .ينتعومجلما ينب تایفولاو ةلجاعلما لئاصح في ابه ُّدتعُی قراوف كانه نكت لمو ،ًاعویش رثکلأا يه ينسیموتبترسلل ةمواقلما تناکو .ةددعتم ةمئلام نوكت دق شرابلما فاشرلإا تتح دملأا ةيرصقلا ةلجاعلما ةیجیتاترسا نم I ةئفلا نإف ،ةددعتم ةیودلأ ةمواقملل ضفخنلما )راركتلل( رُتاوتلل .ةیودلأل سسحتلا جئاتن رفاوتت نأ لبق ةیبیرجتلا ةلجاعملل Profils de pharmacorésistance et résultats du traitement des épisodes récurrents de tuberculose RÉSUMÉ Les profils de pharmacorésistance dans les cas récurrents de tuberculose peuvent être différents de ceux qui n'ont pas d'antécédents thérapeutiques. Dans la présente étude rétrospective, les profils de pharmacorésistance et les résultats du traitement de brève durée sous surveillance directe (DOTS) de catégorie I ont été comparés dans 63 cas récurrents et 872 nouveaux cas de tuberculose pulmonaire entre avril 2003 et janvier 2008 à l’Institut national de recherche sur la tuberculose et les maladies pulmonaires de Téhéran (République islamique d'Iran). La résistance à l'isoniazide et à l'éthambutol était nettement plus fréquente dans les cas récurrents, mais aucune différence n'a été observée dans les taux de résistance à la rifampine, la pyrazinamide, la streptomycine, ni dans le taux des souches multirésistantes. La résistance à la streptomycine était la plus fréquente. Aucune différence significative dans les résultats thérapeutiques et les décès n'a été observée entre les deux groupes. En raison de la faible fréquence de polypharmacorésistance parmi les cas récurrents, le traitement de catégorie I peut convenir en tant que traitement empirique en attendant que les résultats des tests de sensibilité aux médicaments deviennent disponibles. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 958 Introduction With 8.8 million new cases and the deaths of 2 million people reported an- nually worldwide, tuberculosis (TB) is still an important health problem [1,2]. Studies have shown that a previous his- tory of TB treatment, especially irregular drug therapy, is an important risk factor for resistance to anti-TB drugs [3,4]. In most communities, recurrent cases make up the largest proportion of previ- ously treated patients [5]. The pattern of anti-TB drug resistance in this group may be different from that in defaulters and patients with treatment failure. Ide- ally, drug susceptibility testing should be obtained from all previously treated TB patients. In settings where drug sus- ceptibility results are not routinely avail- able and if country-specific data show low or medium levels of multi-drug resistant TB (MDR), TB cases with a first episode of relapse may receive the DOTS category II regimen (CAT II). This regimen consists of isoniazid (INH), rifampin (RIF), pyrazinamide (PZA), ethambutol (ETB) and strep- tomycin (STR) for 2 months, followed by the first 4 drugs without STR for 1 month and INH, RIF and ETB for the remaining 5 months [6]. The CAT II regimen is more costly and has more adverse effects than the DOTS category I regimen (CAT I) which consists of 2 months of INH, RIF, PZA and ETB and 4 months of INH and RIF [7]. The incidence of TB in the Islamic Republic of Iran was estimated as 22 per 100 000 people in 2005; 1.9% of confirmed TB patients were recurrent TB cases according to a World Health Organization (WHO) report [2]. Re- sistance to RIF and concomitant resist- ance to INH and RIF have the most effect on the outcome of treatment. Therefore information about drug resistance prevalence is necessary for more effective treatment of previously treated cases and for national strategic planning [5]. The aim of the present study was to identify the drug resistance patterns among recurrent and new TB cases and assess the efficacy of the CAT I regimen in these groups. Methods Study setting This retrospective study was conducted at the National Research Institute of Tu- berculosis and Lung Disease (NRITLD), Tehran, Islamic Republic of Iran, from April 2003 to January 2008. NRITLD is a specialized centre for TB and lung disease and has a bidirectional relation- ship with the national TB programme. New cases diagnosed at NRITLD are referred to the national TB programme for supervision of their treatment and the national TB programme refers complex TB cases to NRITLD. This research was reviewed and approved by the ethics committee of the centre. Sample Over the study period 63 cases of recur- rent pulmonary TB and 872 cases of new pulmonary TB with mycobacte- riologic confirmation were diagnosed. All of them were recruited to the study. Recurrent cases were defined as TB patients with previous history of TB who were cured or completed treat- ment based on WHO definitions. New cases were defined as any pulmonary TB patient with no history of treatment or a history of anti-TB therapy for less than 1 month [6]. All patients were over 14 years of age. Data collection Sputum smear and culture testing were performed for all patients. The tests were done at the WHO-approved National Mycobacteriology Reference Labora- tory. The laboratory is supervised by the Swedish Institute for Infectious Disease Control and the Research Institute of Tu- berculosis of the Japan Anti-Tuberculosis Association. Anti-TB drug susceptibility tests (proportional method) were per- formed for culture-positive specimens. Both groups were treated with standard CAT I regimen (see earlier), except for MDR cases or patients with severe adverse drug reactions in which modifications to their drug regimen was necessary. Treatment was initiated as inpatients or outpatients and continued by the national TB programme under the DOTS strategy. Demographic information (age, sex, nationality and residency), smoking and drug misuse status, drug resistance pattern and outcome of treatment were obtained for both groups. Successful treatment consisted of cure (patient had negative sputum smear at the end of treatment) and treatment completed (patient completed treatment but did not meet the criteria for cure). Treat- ment failure was defined according to WHO guidelines [6]. Data analysis For statistical analysis, all data gathered were entered into SPSS, version 15. The association between qualitative vari- ables was evaluated by the chi-squared test or Fisher exact test. Student t-test and Mann–Whitney tests were used for quantitative variables with normal dis- tribution and non-distributed variables respectively. A P-value < 0.05 was con- sidered statistically significant. Adjusted Mantel–Haenszel test was utilized to exclude covariates. Results Background characteristics of the groups The study sample included 63 (6.7%) recurrent cases and 872 new cases of pulmonary TB from April 2003 to January 2008. The background char- acteristics, outcome of treatment and drug resistance pattern of both groups are summarized in Table 1. Males were 61.9% of the recurrent cases and 49.4% of new cases, although this dif- ference was not significant. There was no significant difference in the mean طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 959 age of recurrent cases [49.9 (SD 20.1) range 15–87 years] compared with new cases [51.8 (SD 21.4) range 14–90 years]. Among new cases 78.0% were Iranian and 22.0% were other nation- alities, primarily Afghan. In the recurrent group 79.4% were Iranian. Smoking and drug misuse were more common in recurrent cases (46.0% and 36.5% respectively) than new cases (26.6% and 15.4% respectively) (P = 0.001). TB-free duration In the recurrent group the mean TB- free duration between the 2 episodes of disease was 99 months (median 36 months, range 6–540 months). Over half (53.4%) of the recurring cases occurred during the first 3 years after the previous episode. There was no relationship between the first TB-free period and drug resistance pattern or outcome of treatment (P = 0.29 and P = 0.53 respectively): mean time be- tween 2 episodes of TB was 10.56 years for recurrent cases without any drug resistance and 6.97 years for recurrent cases with any drug resistance, while for outcome of treatment it was 9.54 and 3.71 years for successfully treated and unsuccessfully treated respectively. Drug susceptibility Drug susceptibility testing was per- formed on 46 culture-positive recurrent cases [16 (25.4%) had negative cultures and data were missing for 1 case]. In the newly occurring group 95 patients had negative cultures and among the culture-positive cases drug susceptibil- ity testing was done for 508 cases (data missing for the remaining cases). Resistance to first-line anti-TB drugs was found in strains cultured from 22/46 (47.8%) recurrent cases and this was a significantly higher proportion than among newly occurring cases (136/508, 26.8%) (P = 0.002). Although resistance to INH and ETB were more common in cultures from re- current cases than new cases (P = 0.022 and P = 0.012 respectively), there was no statistical difference in resistance to RIF, PZA and STR. In both groups resistance to STR was the highest among all drugs (28.3% in recurrent and 20.3% in new cases). Two MDR strains were cultured in recurrent cases and 10 in new cases (4.3% and 2% respectively) (P = 0.262). In recurrent cases the CAT I regi- men was modified only in 2 MDR cases and another 6 cases due to severe ad- verse drug reactions, especially drug hepatitis. Outcome of treatment Excluding two MDR cases, 87.5% of the recurrent cases were treated success- fully with the CAT I regimen. Table 1 Demographic profile and outcome of treatment of new and recurrent tuberculosis cases Variable New cases Recurrent cases P-value No. % No. % (n = 872) (n = 63) Sex Male 431 49.4 39 61.9 NS Female 441 50.6 24 38.1 Age [mean (SD)] 51.8 (21.4) 49.9 (20.1) NS Nationality Iranian 680 78.0 50 79.4 NS Other 192 22.0 13 20.6 Smoking Yes 232 26.6 29 46.0 0.001 No 640 73.4 34 54.0 0.001 Drug abuse Yes 134 15.4 23 36.5 NS No 738 84.6 40 63.5 Alcohol use Yes 74 8.5 9 14.3 NS No 798 91.5 54 85.7 Successful treatmenta (n = 683) (n = 58) Yes 621 90.9 51 87.9 NS Death Yes 56 8.2 7 12.1 NS aSuccessful treatment consisted of cure and treatment completed. SD = standard deviation; NS = not significant. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 960 There was no significant difference between the new and recurrent groups in the rates of successful treatment (90.9% and 87.9% respectively) or in the mortal- ity rate (8.2% and 12.1% respectively). Using the Mantel–Haenszel test to ad- just for the effect of smoking and drug misuse showed a higher rate of success- ful treatment in recurrent cases than new cases (P = 0.007, odds ratio = 1.107, 95% confidence interval = 1.078–1.137), but no significant difference in the death rate. Discussion Our study compared drug resistance patterns and the efficacy of the CAT I regimen between new and recurrent cases of TB. Resistance to INH and ETB was more common among recurrent cases but there was no difference in rates of resistance to RIF, PZA and STR and in the rate of MDR strains. Resistance to STR was high in our setting (over 20% in both groups). A total of 87.5% of non-MDR recurrent cases were treated successfully with the CAT I regimen. Rates of successful treatment and death were similar between the 2 groups. This study had some limitations. First of all, our centre is a national refer- ral centre for TB patients in the Islamic Republic of Iran, so further prospective studies in the wider community are necessary. Lack of information about HIV status was the second limitation, although the prevalence of coinfection with HIV and TB was low in the coun- try in the period of study (2.2% in new cases of TB) [8]. The most important limitation was the difficulty of following most patients for a long period after completion of treatment due to address changes or emigration. In low TB incidence communities, the most frequent cause of recurrence is true relapse or treatment failure [9,10], whereas in high TB incidence populations reinfection is the most common mecha- nism for recurrence of disease [11,12]. Therefore, efficacy of the CAT II regimen in recurrent cases may be dependent on the frequency of drug-resistant strains in the community, the primary mechanisms of recurrent disease and the efficiency of the national TB programmes [5]. There are no randomized clinical trials to sup- port STR consumption in the CAT II regimen or the superiority of CAT II over CAT I among relapse cases. Resistance to RIF may occur with use of first-line anti-TB drugs in a sub- population with a high frequency of INH resistance and lead to the genera- tion of MDR-TB strains [13]. However, resistance to RIF or both RIF and INH have the greatest effect on the outcome of treatment [14]. National TB programmers need data about country-specific drug re- sistance patterns to inform decisions on each country’s standard treatment regimens for defined patient groups [6]. Many studies have shown that resistant strains are more frequent in TB patients with a history of treatment than in new TB cases [10] but that the rate of this resistance varies among communities. In addition, in most studies, recurrent cases were not analysed separately from other previously treated patients. In a study in Malawi 81% of recurrent cases were sensitive to first-line anti-TB drugs [15]. However, there was resistance to at least 1 anti-TB drug among 85.9% and 84.4% of TB cases with a history of treat- ment in Tashkent (Uzbekistan) and Baku (Azerbaijan) respectively [16]. In another study performed from 2003 to 2004 in our centre, 78% of previously treated patients (recurrent and others) had any resistance to first-line anti-TB drugs [17]. In the present study, we found resistance to at least 1 first-line anti-TB drug in 47.8% of recurrent cases. Drug resistance surveys often show that TB patients relapsing have a me- dium or low likelihood of harbouring MDR stains; such patients can receive a Table 2 Drug resistance pattern in bacterial strains isolated from culture-positive new and recurrent tuberculosis cases Variable New cases (n = 508) Recurrent cases (n = 46) P-value No. % No. % Any drug resistance Yes 136 26.8 22 47.8 0.002 No 372 73.2 24 52.2 Any resistance to: Isoniazid 69 13.6 12 26.1 0.022 Rifampin 23 4.5 4 8.7 NS Pyrazinamide 11 2.2 3 6.5 NS Ethambutol 17 3.3 5 10.9 0.012 Streptomycin 103 20.3 13 28.3 NS Multi-drug resistance Yes 10 2.0 2 4.3 NS NS = not significant. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 961 re-treatment regimen of first-line drugs [6]. Previous studies showed different results for the efficacy of the CAT II regi- men. In 2003 Salaniponi et al. demon- strated a 65% success rate with the CAT II regimen in Malawi but the frequency of STR resistance was low in their set- ting and 81% of strains were sensitive to first-line anti-TB drugs. However, the mortality rate was as high as 25% [16]. In similar studies, success rates were es- timated as 74.8% and 64.9% in Morocco and Turkey respectively [14,18]. In our study, recurrent cases (ex- cluding the 2 MDR cases) were treated with the CAT I regimen, with an 87.5% success rate. In the recurrent patients who were fully sensitive to first-line anti-TB drugs the success rate reached 95.5%. This was significantly higher than that reported by Espinal et al., although their analysis did not separate the pa- tients with recurrence from the patients with treatment failure [19]. We cannot justify the efficiency of the CAT II regi- men in our patients as it was not used, but the CAT II regimen does not have any additional benefits in cases with sensitive strains and, on the basis of the drug resistance patterns, it is probably suboptimal in patients harbouring re- sistant strains. In a study in Thailand, MDR-TB was found in 80% of treatment failed cases but the frequency of MDR was as low as 8% in recurrent cases [20]. In our study, we found only 2 MDR-TB cases among recurrent cases (4.6%) and there was no statistical difference between the proportion of new and recurrent cases in the frequency of RIF resistance or of MDR strains. The high frequency of STR resistance in our setting is notable and it might be related to use of this drug in the treatment of brucellosis, which is an endemic dis- ease in the Islamic Republic of Iran [21]. In summary, our study showed a need to perform drug sensitivity test- ing for recurrent TB cases. Moreover resistance to STR is common in our set- ting. The CAT I regimen was effective for treatment of non-MDR recurrent cases. Due to the low frequency of MDR among this group, the CAT I regimen may be suitable for empirical therapy before drug sensitivity test results be- come available. Further studies in differ- ent communities, with consideration of long-term outcomes of treated patients, are necessary to investigate the best ap- proach to the management of patients with recurrent TB. References 1. Corbett EL et al. The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Archives of Internal Medicine, 2003, 163:1009–1021. 2. Global tuberculosis control: surveillance, planning, financing. WHO report 2007. Geneva, World Health Organization, 2007. 3. Kritski AL et al. Retreatment tuberculosis cases. Factors associ- ated with drug resistance and adverse outcomes. Chest, 1997, 111:1162–1167. 4. Espinal MA et al. Determinants of drug-resistant tuberculosis: analysis of 11 countries. International Journal of Tuberculosis and Lung Disease, 2001, 5:887–893. 5. Zignol M et al. Patients with previously treated tuberculosis no longer neglected. Clinical Infectious Diseases, 2007, 44:61–64. 6. Treatment of tuberculosis: guidelines, 4th ed. Geneva, World Health Organization, 2009 (WHO/HTM/TB/2009.420). 7. Harries AD et al. Recurrent tuberculosis: definitions and treat- ment regimens. International Journal of Tuberculosis and Lung Disease, 1999, 3:851–854. 8. Global tuberculosis control: surveillance, planning, financing. WHO report 2008. Geneva, World Health Organization, 2008 (WHO/HTM/TB/2008.393). 9. Jasmer RM et al. Tuberculosis Trials Consortium. Recurrent tu- berculosis in the United States and Canada: relapse or reinfec- tion? American Journal of Respiratory and Critical Care Medicine, 2004, 170:1360–1366. 10. El Sahly HM et al. Recurrent tuberculosis in Houston, Texas: a population-based study. International Journal of Tuberculosis and Lung Disease, 2004, 8:333–340. 11. Shamputa IC et al. Endogenous reactivation and true treat- ment failure as causes of recurrent tuberculosis in a high inci- dence setting with a low HIV infection. Tropical Medicine and International Health, 2007, 12:700–708. 12. van Rie A et al. Exogenous reinfection as a cause of recurrent tuberculosis after curative treatment. New England Journal of Medicine, 1999, 341:1174–1179. 13. Yoshiyama T et al. Development of acquired drug resistance in recurrent tuberculosis patients with various previous treat- ment outcomes. International Journal of Tuberculosis and Lung Disease, 2004, 8:31–38. 14. Sevim T et al. Treatment outcome of relapse and defaulter pulmonary tuberculosis patients. International Journal of Tu- berculosis and Lung Disease, 2002, 6:320–325. 15. Salaniponi FM et al. Characteristics, management and out- come of patients with recurrent tuberculosis under routine programme conditions in Malawi. International Journal of Tuberculosis and Lung Disease, 2003, 7:948–952. 16. Anti-tuberculosis drug resistance in the world: 4th global report. The WHO/IUATLD global project on anti-tuberculosis drug resistance surveillance 2002–2007. Geneva, World Health Or- ganization, 2008 (WHO/HTM/TB/2008.394). 17. Mirsaeidi MS et al. Trends of drug resistant mycobacterium tuberculosis in a tertiary tuberculosis center in Iran. Saudi Medical Journal, 2007, 28:181–187. 18. Ottmani SE et al. Results of cohort analysis by category of tu- berculosis retreatment cases in Morocco from 1996 to 2003. International Journal of Tuberculosis and Lung Disease, 2006, 10:1367–1372. 19. Espinal MA et al. Standard short-course chemotherapy for drug-resistant tuberculosis: treatment outcomes in 6 coun- tries. Journal of the American Medical Association, 2000, 283:2537–2545. 20. Quy HTW et al. Drug resistance among failure and relapse cases of tuberculosis: is the standard re-treatment regimen adequate? International Journal of Tuberculosis and Lung Dis- ease, 2003, 7:631–636. 21. Pappas G et al. Brucellosis. New England Journal of Medicine, 2005, 352:2325–2336. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 962 Updating and validation of the socioeconomic status scale for health research in Egypt A. El-Gilany,1 A. El-Wehady 1 and M. El-Wasify 2 ABSTRACT This study aimed to update and re-validate the scoring system of Fahmy and El-Sherbini for measurement of socioeconomic status in health research in Egypt. The new socioeconomic status scale has 7 domains with a total score of 84. Intra-and inter-observer variability and the internal consistency of the scale were assessed. A linear regression model was performed to determine the relative importance of each domain to the total score. Kappa coefficient was used to measure the agreement between the socioeconomic levels of the new and the old scales. There was a strong correlation between most of the 7 domains of the scale. Cronbach α for the scale was 0.66. The education domain contributed to 0.898 of variation in total score. There was a moderate agreement (κ = 0.76) and strong positive significant correlation (r = 0.93) between the socioeconomic levels and scores of both scales. We conclude that the new socioeconomic status scale is valid and reliable. 1Department of Public Health; 2Department of Psychiatry, College of Medicine, University of Mansoura, Mansoura, Egypt (Correspondence to A. El-Gilany: ahgilany@gmail.com; ahgilany@hotmail.co.uk). Received: 03/04/11; accepted 02/11/11 هتحص نم ققحتلاو صرم في ةيحصلا ثوحبلا في مدختسلما يداصتقلااو يعماتجلاا عضولا سايق ملس ثيدتح يفيصولا دممح ،يداهولا لداع ،نيلايلجا يدالها دبع مدختسي يذلا يداصتقلااو ،يعماتجلاا عضولل "ينيبشرلاو يمهف" ماظن سايق ملس لوح تامولعلما ثيدتح ةساردلا هذه فدهتست :ةـصلالخا زارحلأا لياجمإ غلبيو تلاامج ةعبس ديدلجا يداصتقلااو يعماتجلاا عضولا سايق ملس مضيو .هتحص نم ققحتلاو صرم في ةيحصلا ثوحبلا في فرعتلل يطلخا فّوحتلل جذومن قِّبُطو ةَدِح لىع بقارم لك ىدل فلاتخلااو ينبقارلما ينب فلاتخلاا نوثحابلا مَّيق دقو .84 هيف )سايقلا تاجرد( في ةيعماتجلااو ةيداصتقلاا تايوتسلما ينب قفاوتلا سايقل )اباك( لماعلما نوثحابلا مدختساو .زارحلأا لممج نمض لامج لكل ةيبسنلا ةيهملأا لىع سايقل( افلأ خابنورك ةميق تغلب دقو .سايقلا ملسل ةعبسلا تلااجلما مظعم ينب يوق طبارت دوجو ينثحابلل حضتاو .ديدلجاو ميدقلا سايقلا ْيَمَّلُس طسوتم قفاوت كانه ناكو سايقلا جئاتن لممج في توافتلا نم 0.898 في يميلعتلا لاجلما مهاس ماك .0.66 سايقلا ملسل )ليخادلا فاصتلاا ةيقوثوم جتنتساو .ينسايقلما لاك في زارحلأا ينبو ةيعماتجلااو ةيداصتقلاا تايوتسلما ينب )r = 0.93( ًايئاصحإ هب دَتْعُي يوق بيايجإ طُبارتو )0.76 =اباك( .ةيقوثولمابو ةحصلاب عتمتي ديدلجا يداصتقلااو يعماتجلاا عضولا سايق ملس نأ نوثحابلا Mise à jour et validation d'une échelle du statut socioéconomique pour la recherche en santé en Égypte RÉSUMÉ La présente étude visait à mettre à jour et à revalider le système d'évaluation de Fahmy et El-Sherbini relatif à la mesure du statut socioéconomique dans la recherche en santé en Égypte. La nouvelle échelle du statut socioéconomique compte sept domaines pour un score total de 84. La variabilité inter- et intra-observateurs et la cohérence interne de l'échelle ont été évaluées. Un modèle de régression linéaire a été mis en oeuvre pour déterminer l'importance relative de chaque domaine par rapport au score total. Le coefficient Kappa a été utilisé pour mesurer la concordance entre les niveaux socioéconomiques de la nouvelle échelle et de l'ancienne. Une forte corrélation a été observée entre la plupart des sept domaines de l'échelle. Le coefficient α de Cronbach pour l'échelle était de 0,66. Le domaine portant sur le niveau d'études contribuait pour 0,898 dans la variation du score total. Une concordance modérée (κ = 0,76) et une forte corrélation positive (r = 0,93) ont été observées entre les statuts socioéconomiques et les résultats des deux échelles. Nous en avons conclu que la nouvelle échelle du statut socioéconomique était valable et fiable. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 963 Introduction The terms socioeconomic status (SES), socioeconomic position and social class are widely used in health research [1]. Assessment of SES is an important aspect in community-based health research as this is a major determinant of health and nutritional status as well as of mortal- ity and morbidity [2–4]. SES is usually measured by 3 variables: education, oc- cupation and income [4–10]. Occupa- tional categories based on prestige, skills, social influence, and/or power have been the primary basis for SES classification in western European countries [1]. While SES has been seen as an inherited factor, in modern society it may migrate on the basis of factors such as occupation, edu- cation, income, type of housing, material possessions, etc. [10]. In Egypt there have been only 2 recorded attempts to develop scales to determine SES. The scale of Fahmy and El-Sherbini, which was originally published in 1983 [5], is the most com- monly used in health research, while the second scale of El-Shakhs [6] has been developed for use in educational research. However, social transforma- tions and the fast growing economy of the country may have reduced the validity of these scales for measuring SES over the decades. Furthermore, monetary inflation and the consequent fall in the value of the currency make the economic criteria of the scale less relevant. We suggested the need for an economic revision of this scale in order to account for the devaluation of the Egyptian pound and other changes over time. This study therefore aimed to up- date and re-validate the scoring system of Fahmy and El-Sherbini for measure- ment of SES in health research [5]. Methods Sample The target population was a con- venience sample of 100 households distributed proportionally between both urban and rural areas (40% and 60%, respectively). Ten (10) clusters, each cluster with 10 households, were selected from 6 villages and 4 sectors of Mansoura city in Egypt. The villages were selected to represent small and large villages, as well as those with and without rural health services. The urban sectors were chosen to represent differ- ent social strata of the community. The first home in each cluster was randomly chosen. Empty homes were replaced by their neighbours. In each household we interviewed one adult family member (aged 18+ years)who volunteered to participate in the study. A total of 96 agreed to partici- pate in the study and were interviewed; 6 questionnaires were excluded from analysis (3 due to incomplete data, 2 because the family member interviewed at the first visit was not available at the second visit and 1 due to inability to complete the retest interview as the family was unavailable at their home). Thus 90 questionnaires were analysed (response rate of 90%). Study tool The old scoring system of Fahmy and El-Sherbini [5] was updated by the authors, based on a literature review and extensive discussion with a jury of 10 experts (3 psychologists, 3 soci- ologists and 4 public health specialists). The relative weight of each item and its allocation to different domains were defined by the relevant jury. In case of disagreement between juries, the most frequently cited weight and domain allocation of each item was chosen. The final scale included 7 domains with a total score of 84, with a higher score indicating better SES: Education and cultural, Occupation, Family, Family possessions, Economic, Home sanita- tion, Health care (Box 1). The original scale of Fahmy and El-Sherbini included items with no domains and there was more than one model for different health problems [5]. Furthermore, the old scale was not tested for its reliability and validity. It measured the woman’s (and man’s) education and occupation as 1 item and we judged this to be inappropriate as education does not necessarily cor- respond to occupation. The updated scale included all the variables of the previous one. In ad- dition new items and domains were included, e.g. access to health informa- tion, more items in the family, family possessions and economic domains, and the health care domain. The vari- ables were grouped into 7 domains and the weight of each item was up- dated according to the opinion of the jury experts. Both the old and the new scales were developed in English, the official language of scientific writing in the medicine and health fields in Egypt. Data collection With the help of the local health facili- ties of the chosen areas, 2 researchers (A. El-W. and M. El-W.) contacted the target families and agreed a day and time for home visits. At the visit the re- searchers introduced themselves to the head of the household and obtained verbal approval for his/her participa- tion in the study. To test the reliability of the new scale (final version) it was applied to the 90 households during 3 assessments. To assess inter-observer variability the first 2 assessments were performed consecutively on the same visit by the 2 researchers at an interval of 15 to 20 minutes, with either re- searcher 1 or researcher 2 applying the first assessment, thus preventing ha- bituation bias. To assess intra-observer reliability the third assessment was applied after 10 days by researcher 1. Personal data of the interviewed family member were collected during the first visit. The tool was completed by the 2 interviewers at the 2 occasions for all subjects to ensure standardization of the wording of the questions and statements. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 964 Name of head of family: Address: Definition of the family: It includes nuclear or joint family. Married couple with unmarried children or without children. Head of the family will be either husband/wife. Dependent father/mother/brother/sister does not become member of the family unless he/she is earning and one kitchen with pooled income is managed by him/her. This scale includes 7 domains with a total score of 84 Socioeconomic level: to be classified into very low, low, middle and high levels depending on the quartiles of the score calculated. NB In case of death or retirement of husband or wife, record the education and occupation before death or retirement Education and cultural domain (for both husband & wife) (score = 30) Highest level of education Husband Wife Illiterate 0 0 Read & write 2 2 Primary 4 4 Preparatory 6 6 Secondary (general & technical of 3 or 5 years) 8 8 Intermediate (2 years) institutes 10 10 University graduate 12 12 Postgraduate degree 14 14 Access to health information (1 each for the following items): Printed materials, e.g. books, posters, booklets, etc.; Audiovisual message on television &/or radio Occupation domain (for both husband & wife) (score = 10) Occupation Husband Wife Non-working/house wife 0 0 Unskilled manual worker 1 1 Skilled manual worker/farmer 2 2 Trades/business 3 3 Semi-professional/clerk 4 4 Professional 5 5 Family possessions domain (score = 12: 1 each for the presence of items given below) Refrigerator – Radio – Television – Washing machine – Telephone/ mobile phone – Car – Agricultural land – Non-agricultural land for housing – Shop or animal shed – Other house (beside the house in which the family is living) – Animals/poultry – Computer/ Internet Family domain (score = 10) Residence: Urban slum = 0; Rural = 1; Urban = 2 Number of family members (parents, children & all de- pendents): < 5 members = 2; ≥ 5 members = 1 Number of earning family members: 1 member = 1; 2 members = 2; ≥ 3 members = 3 Education of children (aged ≥ 5 years, whether free or private education): All children going or ever gone to school/university = 3; ≥ 50% going or ever gone to school/university = 2; < 50% going or ever gone to school/university = 1; None go/gone to school/univer- sity/not applicable = 0 Home sanitation domain (score = 12) Services (1 each for the presence of the following items): Pure wa- ter supply – Electricity – Natural gas – Sewerage system – Munici- pal collection of solid wastes – Flush latrine – Air conditioning Type of house: Owned, ≥ 4 rooms = 4; Owned, < 4 rooms = 3; Rented, ≥ 4 rooms = 2; Rented, < 4 rooms = 1; No place to reside = 0 Crowing index: (number of family members divided by number of rooms): ≤ 1 person per room = 1 = 1; > 1 person per room = 0 Economic domain (score = 5) Income from all sources: In debt = 0; 1 Just meet routine expenses = 1; Meet routine expenses and emergencies = 2; Able to save/invest money = 3 Family receives governmental support: Yes = 1; No = 0 Family pays tax: Yes = 1; No = 0 Health care domain (score = 5) Usual source of health care: Private health facilities = 5; Health in- surance = 4; Free governmental health service = 3; More than one of the above sources = 2; Traditional healer/self-care = 1 Box 1 Scoring of scale for measuring family socioeconomic status (SES) for health research in Egypt طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 965 Ethical considerations The objectives and content of the ques- tionnaire as well as the confidentially of data collected were discussed with an adult member of the household. Par- ticipation was on a voluntary basis after giving verbal consent. Households were included if the same adult member was available for retest on a mutually agreed day. At the time of starting this research there was no research ethics committee in our institution. Data analysis Data were analysed using SPSS, version 16. Descriptive statistical analysis was used for personal data and the total score. The correlation between differ- ent domains of the scale as well as their inter-and intra-observer reliability were determined by Pearson correlation coefficient. The internal consistency of the scale was assessed by Cronbach α. Multivariate stepwise linear regression analysis was performed to detect the relative contribution of each domain to the total score. R2 and added R2 were calculated. The new scale was catego- rized into 4 socioeconomic levels ac- cording to the 3 quartiles. Agreement between the different levels of SES of the new and the old scales was assessed by kappa coefficient of agreement. A P- value ≤ 0.05 was considered statistically significant. Results The respondents were 47 women (52.2%), 36 men (40.0%), 4 female youths (4.4%) and 3 male youths (3.3%). Their age ranged from 18 to 65 years with a mean of 35.5 (standard deviation 12.0) years. Their other socio- economic features are shown in Table 1. Table 2 shows that the total mean SES scores were nearly equal in the test, retest and second rater rounds (48.2, 48.0 and 47.9 respectively). Table 3 lists the correlation coef- ficients between the 7 domains and the total SES score. There was no significant correlation between the family domain and the educational, family posses- sions, economic and home sanitation domains or the economic and health care domains. The other correlation coefficients were either moderately or highly significant. Within each domain, both the test- retest and inter-rater correlation coef- ficients were very strong, ranging from 0.84 to 1.00 (Table 4). The different domains showed a moderate internal consistency (Cronbach α = 0.66) (not shown in the tables). Table 5 shows the multiple linear regression of the SES scale. The most important predictor was the education and cultural domain (R2 = 0.898), fol- lowed by the occupation domain (R2 = 0.042). Both the home sanitation and health care domains had the lowest con- tributions to the SES score (R2 = 0.005 and 0.002 respectively). Table 6 shows that there was a mod- erate coefficient of agreement between Table 1 Socioeconomic characteristics of the interviewed household members (n = 90) Variable No. % Age (years) < 25 13 14.4 25 –< 50 60 66.6 ≥ 50 17 18.9 Sex Male 39 43.3 Female 51 56.7 Residence Rural 52 57.8 Urban, slum 17 18.9 Urban, non-slum 21 23.3 Education Illiterate 21 23.3 Below secondary 17 18.9 Secondary 28 31.1 Above secondary 24 26.7 Occupation Housewife 26 28.9 Farmer/ manual worker 15 16.7 Professional/ semiprofessional 29 32.2 Othera 20 22.2 aTrades, business, students, not working and retired. Table 2 Descriptive statistics of scores on the proposed socioeconomic status scale (total score range 0–84) Variable Test scores Retest scores Second rater scores Min.–max. 19–78 18–78 19–78 Mean (SD) 48.2 (14.8) 48.0 (14.9) 47.9 (14.6) 1st quartile 36.8 36.0 37.0 2nd quartile (median) 46.5 46.0 47.0 3rd quartile 58.3 60.3 58.3 SD = standard deviation. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 966 the levels of the new and the old SES scores (κ = 0.76) and there was a strong positive significant correlation between both scores (r = 0.93). Discussion A number of classifications and scales for SES have been devised and used in different countries and settings; each has its own strengths and inadequacies and no single scale is suited to every sociocultural setting around the world. However, the general consensus is that occupation, income and education are the key variables that determine the SES of a person [8]. The original scale that we updated in this study [5] measures income in monetary terms (e.g. monthly in- come per capita). This is often difficult to ascertain. There are difficulties in obtaining income because people may be reluctant to talk about their income openly [10]. Instead we asked about the total family income in non- monetary terms. Also an indirect meas- ure of income was included in other domains, covering family possessions and ownership of agricultural land and non-agricultural land for housing, own- ership of other houses, etc. All these contribute to the family income. In rural Egypt the whole family—men, women and children—may be work- ing. Likewise, child labour is widely prevalent in urban-based enterprises, especially in the informal sector [11]. Because of these considerations, we used the number of earning members of the family, whatever the type and seasonality of the occupation. Evidence from the literature in- dicates that while standard measures of education and income are cor- related, these correlations are gener- ally not strong enough to justify using education as a proxy for income (or vice versa). Earnings can vary con- siderably among people of similar educational level, particularly across different social groups [1]. Both in- come and education can influence the etiology of many health outcomes, in part through pathways involving material resources. Education can also affect a range of non-economic social Table 3 Correlation coefficients between different domains and total socioeconomic status (SES) scale Domain Domain Education & cultural Occupation Family Family possessions Economic Home sanitation Health care Occupation 0.66*** – – – – – – Family 0.20 0.23* – – – – – Family possessions 0.56*** 0.38*** 0.09 – – – – Economic 0.50*** 0.38*** 0.17 0.62*** – – – Home sanitation 0.47*** 0.34*** 0.15 0.61*** 0.49*** – – Health care 0.49*** 0.27** 0.30** 0.24* 0.18 0.37*** – Total SES 0.95*** 0.75*** 0.32** 0.69*** 0.62*** 0.62*** 0.55*** *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. Table 4 Test–retest and inter-rater correlation coefficients of different domains and total socioeconomic status (SES) scale Domain Test–retest Inter-rater Education & cultural 0.94*** 1.00*** Occupation 0.98*** 0.96*** Family 0.94*** 0.92*** Family possessions 0.95*** 0.93*** Economic 0.84*** 0.92*** Home sanitation 0.96*** 0.89*** Health care 0.92*** 0.91*** Total SES 0.93*** 0.99*** ***P ≤ 0.001. Table 5 Best model obtained by multiple linear regression for prediction of socioeconomic status Domain β Added R2 t-value Education & cultural 0.59 0.898 93.9*** Occupation 0.11 0.042 22.5*** Family 0.20 0.027 40.0*** Family possessions 0.10 0.014 25.0*** Economic 0.14 0.012 25.0*** Home sanitation 0.09 0.005 20.5*** Health care 0.06 0.002 12.2*** Constant = 1.4; model F = 10073.1***; model R2 = 0.999. ***P ≤ 0.001. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 967 characteristics (e.g. health-related knowledge, literacy, problem-solving skills, prestige, influence over others and one’s own life) with important health effects [12–16]. Income is not a proxy for wealth. The total accumu- lated economic resources or wealth could be at least as important for health. Wealth can buffer the effects of temporarily low income due to un- employment or illness and can reflect power or influence over others [1,17]. In the absence of a gold stand- ard or reference scale, we compared the revised scale with the original scale of Fahmy and El-Sherbini [5], which is most commonly used in health research in Egypt. There was a strong positive correlation between both scales. Furthermore, there was moderate agreement between the socioeconomic levels of the 2 scales. This relatively low agreement can be explained by the different cutoff points in each scale. In the old scale the cutoff points were based on arbi- trary points (85%, 75% and 50% of the total score) [5]. However, in the new scale the cutoff points were based on statistical parameters (1st, 2nd and 3rd quartiles). The relative increase in the percentage of people of high social level in the new scale could be a reflection of better opportunities for higher education in Egypt since the time when Fahmy and El-Sherbini’s scale was developed. Multiple linear regression of the SES scale revealed that the most im- portant predictor was the education domain (R2 = 0.898), followed by the occupation domain. Both home sani- tation and health care domains had the lowest contribution to SES. This raises the possibility of using a shorter version of the scale as a proxy measure of SES. This short version could be useful for clinical and hospital-based research. However, a short form would need further validation before its ap- plication. Busy clinicians considering SES as confounder may ask about 3 or 4 domains, e.g. education, occupation, family and family possessions. For re- searchers seeking a causal association of SES with specific outcomes, it could be more appropriate to use the full version of the new SES scale. What is new in the proposed scale? SES scores were classified into very low, low, middle and high levels, depending on the quartiles of the calculated score rather than a fixed point. This could be more practical for application in dif- ferent settings and populations with varying socioeconomic backgrounds. Education and occupation were dealt with as separate domains. In recent decades the association between educa- tion and occupation has diminished. Income was expressed in perceived terms, independent of the actually mon- etary salary or income. This overcomes the problem of the devaluation of the Egyptian currency. Also family wealth or possessions, as an indicator of family economic status, was included in the scale. Indirect measures of income were also included, e.g. receiving government financial support or paying taxes. The Table 6 Agreement between the socioeconomic status (SES) scores of the updated scale and Fahmy and El-Sherbini scale [5] in a sample of 90 people SES level of proposed scale SES level of Fahmy & El-Sherbeni scale Total Very low Low Middle High No. % No. % No. % No. % No. % Very low 18 – 2 – 0 – 0 – 20 22.2 Low 4 – 19 – 3 – 0 – 26 28.9 Middle 0 – 2 – 16 – 1 – 19 21.1 High 0 – 0 – 4 – 21 – 25 27.8 Total 22 24.4 23 25.6 23 25.6 22 24.4 90 100.0 % agreement = 82.2%; κ coefficient = 0.76; correlation coefficient between the 2 scores, r = 0.93***. ***P ≤ 0.001. content of the new scale was validated by a jury of experts in this field and its items and domains were tested for reli- ability by Cronbach alpha as well as inter and intra-rater reliability. Finally, the contribution of each component to the total score was assesses by linear regression model. The study had some limitations which should be noted. The sample was selected on a convenience ba- sis which means that the results may not be generalizable to other popula- tions. The cross-sectional design of this study made it impossible to assess fluctuations in income over time and their impact on the SES scale. It is diffi- cult to calculate the SES of individuals living alone (single member family), such as retired elderly living alone and dependent on income contrib- uted by their children. The situation is even more complicated in extended families with more than more nuclear families. This new scale should be tested in different community and clinical set- tings in Egypt. Also further research is needed to evaluate the predictive validity of the scale in predicting morbidity (both communicable and noncommunicable), mortality and health care utilization, among others. It will be useful to translate the SES scale into Arabic language to be used for self-evaluation of SES. Both the Arabic and the short versions should be adequately tested. Application of this new scale in different Arab cultures is strongly recommended EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 968 References 1. Braveman PA et al. Socioeconomic status in health research: one size does not fit all. Journal of the American Medical Asso- ciation, 2005, 294:2879–2888. 2. Ghosh A, Ghosh T. Modification of Kuppuswamy’s socioeco- nomic status scale in context to Nepal. Indian Pediatrics, 2009, 46:1104–1105. 3. Agarwal AK. Social classification: the need to update in the present scenario. Indian Journal of Community Medicine, 2008, 33:50–51. 4. Aggarwal OP et al. A new instrument (scale) for measuring the socioeconomic status of a family: preliminary study. Indian Journal of Community Medicine, 2005, 30:111–114. 5. Fahmy S, El-Sherbini AF. Determining simple parameters for social classifications for health research. Bulletin of the High Institute of Public Health, 1983, 13:95–108. 6. El-Shakhs AE. A scale of socioeconomic status of the family. The scale manual. 2nd ed. Cairo, Anglo-Egyptian Stationery, 1995. 7. Collins AL, Goldman N. Perceived social position and health in older adults. Princeton, New Jersey, Office of Population Research, University of Princeton, 2006 (Working Paper No. 2006–08). 8. Gupta P, Ghai OP. Textbook of preventive and social medi- cine, 2nd ed. New Delhi, CBS Publishers and Distributors, 2007:626. 9. Kulkarni AP, Baride JP. Textbook of community medicine. Mum- bai, Vora Medical Publications, 1998:28. 10. Park K. Park’s textbook of preventive and social medicine, 18th ed. Jabalpur, India, Banarsi Das Bhanot, 2005:506. 11. Datt G, Jolliffe D, Sharma M. A profile of poverty in Egypt: 1997. Washington DC, International Food Policy Research Institute, 1998 (FCND Discussion Paper No. 49). 12. Ross CE, Mirowsky J. Refining the association between educa- tion and health: the effects of quantity, credential, and selec- tivity. Demography, 1999, 36:445–460. 13. Winch C. The economic aims of education. Journal of Philoso- phy of Education, 2002, 36:101–117. 14. Mirowsky J, Ross CE. Education, personal control, lifestyle and health: a human capital hypothesis. Research on Aging, 1998, 20:415–449. 15. Reynolds JR, Ross CE. Social stratification and health: educa- tion’s benefit beyond economic status and social origins. Social Problems, 1998, 45:221–247. 16. Ross CE, Van Willigen M. Education and the subjective quality of life. Journal of Health and Social Behavior, 1997, 38:275–297. 17. Brenner MH. Health, productivity and the economic environ- ment: dynamic role of socioeconomic status. In: Green GM, Baker F, eds. Work, health and productivity. New York, Oxford University Press, 1991:241. Note from the Editor We wish to draw the kind attention of our potential authors to the importance of applying the editorial requirements of the EMHJ when preparing their manuscripts for submission for publication. These provisions can be seen in the Guidelines for Authors, which are published in the first issue of each volume and are available online at: http://www.emro.who.int/ emh-journal/authors/. We regret that we are unable to accept papers that do not conform to the editorial requirements. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 969 Natural killer cell populations in Egyptians infected with hepatitis C virus M. Rafik,1 G. Sidhom,2 R. Mamdouh,1 D. Ellebedy 2 and M. Mohamed 2 ABSTRACT Natural killer (NK) cells are key players in the immune response to viruses. This study examined the effect of hepatitis C virus (HCV) on the frequency of NK cells and their subsets in individuals with different clinical outcomes; 20 positive for anti-HCV and HCV-RNA (chronic hepatitis C), 20 positive for anti-HCV but negative for HCV-RNA (spontaneously resolved) and 20 healthy controls free of HCV. There was a significant reduction in the frequency of total NK cells in the chronic group compared to the control (P = 0.001) or resolved (P = 0.01) groups. The percentage of CD56(bright) cells was significantly higher than the control group (P = 0.04). While the percentages of CD56 (dim) cells and their CD16 expression were lower in the chronic group, this was not statistically significant. The frequency of CD3+CD56– T cells was significantly lower in both the chronic and resolved groups compared to the control group (P = 0.04). Our results confirm a potential role of NK cells and the different subsets in the pathogenesis of chronic HCV infection. 1Clinical Pathology Department, Faculty of Medicine, Ain Shams University, Cairo, Egypt. 2Clinical and Chemical Pathology Department, Medical Division, National Research Center, Cairo, Egypt (Correspondence to G. Sidhom: gloriasidhom@yahoo.com). Received: 20/04/10; accepted: 07/06/10 "سي" سيويرفلا دبكلا باهتلاب ينباصلما ينيصرلما ىدل ةيعيبطلا ةكتافلا ايلالخا تاعوممج دممح ىم ،ىديبللا ايلاد ،حودمم اشر ،مهديس ايرولج ،قيفر ىنم "سي" سيويرفلا دبكلا باهتلا رثأ ةساردلا هذه ىّرحتتو .تاسويرفلل ةيعانلما ةباجتسلاا في ًايسيئر ًارود ةيعيبطلا ةكتافلا ايلالخا يدؤت :ةـصلالخا ًايبايجإ ًاصخش 20 ةساردلا تلمشو ؛ةفلتمخ ةيريسر لئاصحب ينباصلما صاخشلأا ىدل ةيعرفلا اتهاعوممج لىعو ةيعيبطلا ةكتافلا ايلالخا راركت لىع تاداضلم ًايبايجإ ًاصخش 20و )"سي" سويرفلاب نمزلما دبكلا باهتلا( "سي" دبكلا باهتلا سويرف انرلو "سي" دبكلا باهتلا سويرف تاداضلم سويرف نم ينلالخا ءاحصلأا دهاوشلا نم 20و ،)يئاقلت ءافش( سي دبكلا باهتلا سويرف انرل نويبلس مهنكلو "سي" دبكلا باهتلا سويرف دهاوشلا ةعومجمب ةنراقم يننمزلما ضىرلما ةعوممج ىدل ةيعيبطلا ةكتافلا ايلالخا لممج رتاوت في ظوحلم صقن كانه ناكو "سي" دبكلا باهتلا ردقب لىعأ ناك )ةقشرلما( CD56 ايلاخلل ةيوئلما ةبسنلا نأ ًاضيأ حضتاو .) P = 0.01( "سي" دبكلا باهتلا نم اوفش نيذلا ةعومجمب وأ ) P = 0.001( يننمزلما ضىرلما ةعوممج ىدل ضفخأ CD16 اهيربعتو )ةتماقلا( CD56 ايلاخلل ةيوئلما بسنلا نأو ،دهاوشلا ةعوممج ىدل وه امم ) P = 0.04( هب دتعي نيذلاو يننمزلما ضىرلما ْيَتعوممج نم لك ىدل هب ُّدَتْعُي رادقمب ضفخأ CD3 +CD56 – T ايلالخا راركت ناك ماك .يئاصحإ دادتعا كلذل نوكي نأ نود ايلاخلل لمتحلما رودلا د ِّكؤت اهيلع اولصح يتلا جئاتنلا نأ نوثحابلا ىريو .) P = 0.04( ةدهاشلا ةعومجلما عم ةنراقلماب "سي" دبكلا باهتلا نم اوفش ."سي" دبكلا باهتلا سويرفب ىودعلا ضارمإ في ةفلتخلما ةيعرفلا اتهاعوممجو ةيعيبطلا ةكتافلا Populations de cellules tueuses naturelles chez des Égyptiens infectés par le virus de l'hépatite C RÉSUMÉ Les cellules tueuses naturelles jouent un rôle majeur dans la réponse immunitaire aux virus. La présente étude a examiné l'effet du virus de l'hépatite C (VHC) sur la fréquence des cellules tueuses naturelles et de leurs sous- groupes chez des personnes présentant des issues cliniques différentes : 20 patients positifs pour les anticorps anti- VHC et pour son ARN (hépatite C chronique) ; 20 personnes positives pour les anticorps anti-VHC mais négatives pour son ARN (résolution spontanée) ; et 20 témoins en bonne santé ne présentant pas d'infection par le virus de l'hépatite C. Une réduction importante de la fréquence des cellules tueuses naturelles totales dans le groupe des patients porteurs d'une infection chronique a été observée par rapport au groupe des témoins (P = 0,001) ou au groupe des patients dont l'infection a connu une résolution spontanée (P = 0,01). Le pourcentage des cellules CD56 (bright) était significativement supérieur (P = 0,04) à celui du groupe témoin. Si le pourcentage des cellules CD56 (dim) et de leur expression de CD16 était inférieur dans le groupe de patients infectés chroniquement, la différence n'était pas statistiquement significative. La fréquence des cellules T CD3+CD56 était très inférieure à la fois dans le groupe de patients infectés chroniquement et dans le groupe ayant bénéficié d'une résolution spontanée par rapport au groupe témoin (P = 0,04). Nos résultats confirment le rôle potentiel des cellules tueuses naturelles et de ses différents sous-groupes dans la pathogénèse de l'infection chronique par le virus de l'hépatite C. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 970 Introduction Hepatitis C virus (HCV) infection has the ability to establish chronic infection in approximately 75%–80% of cases [1–4]. The elimination or persistence of the infection depends on a balance between the immune response and the virus replication rate [5] and play a cru- cial role in the pathogenesis of chronic hepatitis C [6]. Natural killer (NK) cells are potent antiviral cells which eliminate the vi- rus through direct killing and cytokine production, and are the key players in immune responses to viruses [7]. Like many other viruses, HCV has evolved strategies to evade detection and elimi- nation by NK cells and these have been proposed as a mechanism of virus per- sistence and chronic infection [8]. According to CD56 expression, NK cells are divided into 2 subsets, dim and bright [9]. Dim NK cells are mainly cytolytic and comprise more than 90% of NK cells in peripheral blood, whereas bright NK cells are immune-regulatory mainly through cytokine production, and display homing receptors required for migration to 2ry lymph nodes [10,11]. CD56 together with natural T (NT) cells comprise about 5%–15% of the peripheral T cell pool and up to 50% of T cells within the liver microenviron- ment, which is the primary site of HCV replication [12]. These cells have been reported to be impaired in the blood and/or liver of chronic HCV patients [12–16]. HCV may evade the immune re- sponse either by inhibition of NK cell activation receptor genes or by decreas- ing the activity of these cells by reducing their number and function in chroni- cally infected individuals [17,18]. Also, the inheritance of particular inhibitory killer immunoglobulin-like receptors genes, which control NK cell activity, may play a role in predisposing to chro- nicity [19]. Recently, altered NK cell frequency has been suggested to contribute to the impaired cellular immune response and virus persistence [8]. Some T cells are known to express NK cells receptors (NKRs) and mediate functions of both T cells and NK cells [13]. Studies sug- gest that HCV inhibits receptor genes in the activation of NK cells [20], and the phenotypes and/or functional activities of various populations of NKR+ T cells have been reported to be impaired in chronic hepatitis C patients [16,21]. In order to study the possible role that NK cells might play in the patho- genesis of the different clinical outcomes of HCV infection in our population, we aimed in this study to assess the impact of HCV on the frequency and subset distribution of circulating NK cells, in- cluding CD56 dim and bright popula- tions and CD56 + NT cells in different clinical outcomes of HCV infection. Methods Participants The study included 60 participants: 40 health care workers (HCWs), who were a part of a screening programme for HCV infection among HCWs at the internal medicine hospital of Ain- Shams University, and 20 age- and sex- matched healthy volunteers recruited from the same hospital, who were negative for both HCV antibody and HCV RNA, who served as a control group. Exclusion criteria for all groups included interferon or ribavirin therapy and dialysis. Written informed consent was obtained from all the participants and the study was approved by the ethi- cal committee of the National Research Centre in Cairo, Egypt. Methodology All participants were screened for HCV antibodies (HCV-Ab) by a 3rd generation enzyme-linked immuno- sorbent assay (Diasorin SPA, Italy) and HCV-RNA by qualitative real-time polymerase chain reaction (Stratagene Mx3000P). The percentage of total circulating NK cells, NK subsets (in- cluding both CD56 dim and bright, with their CD16 expression), total T cells and CD56+NT cells was charac- terized by flow cytometry using anti CD3, CD56 and CD16 monoclonal antibodies. For flow cytometric analysis, 3 mL whole blood were withdrawn from each participant on ethylenediamine- tetraacetic acid (EDTA), and layered over Ficoll-Hypaque (density 1.077 g/L) (Amersham Biosciences) and centrifuged at 2000× g for 20 min at 24 °C. The mononuclear layer was care- fully separated, washed 3 times with phosphate buffer saline (PBS), and the cell pellet was resuspended in 1 mL FACS buffer (500 mL PBS + 5 mL pooled human serum + 5 mL fetal bo- vine serum + 1 mL sodium azide). Then, the mononuclear layer was stained with CD16-FITC, CD56-PE and CD3-PE- Cy5 monoclonal antibodies (Sigma), according to the manufacturer’s instruc- tions. Briefly, 5 µL of each monoclonal antibody were added to 100 µL of the resuspended cells and incubated for 20 min at 4 °C in the dark. Then 1 mL of FACS buffer was added followed by centrifugation for 5 min at 1500× g. Finally, 500 µL of supernatant were discarded and samples were ready for processing on the flow cytometer. Flow cytometric analysis was done using the EPICS-XL Coulter (Coulter Diagnostics, Hialeah, Florida, United States of America), with standard filter setting using 488 nm wavelength laser excitation. The following parameters were used: linear forward scatter, side scatter (log), log fluorescence 1 for detection of CD16-FITC, log fluores- cence 2 for detection of CD56-PE and log fluorescence 4 for CD3-PE-Cy5. Statistical analysis Data were analysed using SPSS, version 15.0. Qualitative data are presented as frequency and percentage, while طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 971 quantitative data are presented as mean, standard deviation (SD) and range. The chi-squared test was used to compare qualitative variables, and the Student t-test was used to compare quantitative variables. A P-value < 0.05 was considered statistically significant; P < 0.01 was considered highly signifi- cant. Results According to their HCV-Abs and HCV-RNA results, our patients were divided into a “spontaneously resolved patient group” which included 20 pa- tients who were positive for anti-HCV and negative for HCV-RNA, and a “chronic hepatitis C patient group” included 20 patients who were positive for both anti-HCV and HCV-RNA. The mean percentage of the cell popula- tions in the different groups are shown and compared in Table 1. On comparing the percentage of different cell populations in the differ- ent groups, the percentage of the total circulating NK cell population was significantly lower in chronic hepatitis C patients [mean 3.3 (SD 1.9)] com- pared to either the spontaneously re- solved group [9.3 (SD 6.3)] (P = 0.01) or the healthy control group [9.5 (SD 4.5)] (P = 0.001) (Table 1). As regards NK subsets, the CD56 (bright) NK cell population was sig- nificantly higher in chronic hepatitis C patients than in the healthy control group [mean 12.2 (SD 7.1) vs 5.9 (SD 5.5), P = 0.04]. In the spontaneously resolved group the CD56 (bright) NK cell population was also higher but this was not statistically significant [mean 12.2 (SD 7.1) vs 10.6 (SD 9.40), P = 0.4]. Although the percentage of CD56 (dim) NK cells was lower in chronic hepatitis C patients than in the control group, this was not was statistically sig- nificant (P = 0.05) (Table 1). CD16 expression was lower on CD56 (dim) NK cells of chronic hepatitis C patients than healthy controls or spontaneously resolved patients but not significantly so (P = 0.13 and P = 0.2 respectively). The percentage of non-CD56 expressing T cells (CD3+CD56–T) was significantly lower in both chronic hepatitis C patients and spontaneously resolved patients compared to healthy controls (P = 0.04). The percentage of CD56+ NT cells showed no statistically significant difference between the differ- ent groups. Discussion In our study, we examined the frequency of circulating NK cell populations (total and subsets) as well as total T cells and CD56+NT cells in peripheral blood of both chronic hepatitis C patients and spontaneous resolvers compared to a control group of healthy volunteers using flow cytometry. Our results revealed that the total circulating NK cells were highly signifi- cantly different in the different studied groups. A significant reduction in the frequency of circulating total NK cells in chronic hepatitis C patients compared to either healthy controls or sponta- neous resolvers was found. Total NK cell frequency and/or cytotoxic activity have been reported to be low in chroni- cally infected patients in several studies [8,22]. As regards NK cell subsets, the per- centage of CD56(bright) NK cells was significantly higher in chronic hepati- tis C patients than in the resolved or control groups, while the percentage of CD56(dim) NK cells was lower in chronic hepatitis C patients than in the other groups. This concurs with sev- eral studies [8,16,23,24]; Morishima et al. found that the percentage of CD56(dim) cytolytic NK cells was 30% lower in HCV-infected patients compared with uninfected subjects and this was paralleled by a decrease in circulating NK cells [16]. Meier et al. suggested that the deficit in interleukin 15 (IL-15) might be responsible for Table 1 Frequency (%) of the different cell populations in different hepatitis C virus (HCV) infection outcomes Cell population Controls Chronic hepatitis Ca Spontaneously resolvedb P-value (control vs chronic) P-value (control vs resolved) P-value (chronic vs resolved) Mean (SD) Mean (SD) Mean (SD) Total T cells 64.1 (10.4) 50.2 (16.9) 58.6 (10.9) 0.08 0.12 0.3 CD56+NT 3.3 (1.3) 4.7 (4.8) 4.7 (3.5) 0.8 0.4 0.7 CD3+CD56–T 60.8 (10.7) 45.5 (15.8) 53.9 (12.2) 0.04 0.04 0.2 Total NK 9.5 (4.5) 3.3 (1.9) 9.3 (6.3) 0.001 0.5 0.01 NK(dim) 94.1 (5.5) 87.8 (9.6) 89.4 (9.4) 0.05 0.1 0.3 NK(dim) CD16+ 92.2 (6.2) 86.0 (8.9) 90.7 (6.03) 0.13 0.5 0.2 NK(bright) 5.9 (5.5) 12.2 (7.1) 10.6 (9.4) 0.04 0.1 0.4 NK(bright) CD16+ 38.9 (13.3) 41.1 (9.5) 47.9 (18.1) 0.6 0.14 0.5 aPositive for both anti-HCV and HCV-RNA. bPositive for anti-HCV and negative for HCV-RNA. SD = standard deviation; NT = natural T cells; NK = natural killer cells. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 972 this change [25]. IL-15 is thought to regulate the function of NK cells and to promote their survival ex vivo [26], and abnormalities in the production of this cytokine contribute to a reduction in NK cell number and a decrease in the CD56(dim) population in HCV infection. Th e f r e q u e n c y o f t h e CD56(dim)16+ subset was low in our chronic hepatitis C patients compared to the resolved or healthy control groups, which is in agreement with Dessouki et al. [8], suggesting the expansion of NK cells is to produce interferon-gamma, while dim cells de- crease to minimize cytotoxicity and limit liver damage [27]. In agreement with Corado et al. [28], there was no statistically sig- nificant difference in the percentage of total T cells between the different studied groups. However, the per- centage of CD3+CD56– cells was significantly lower in chronic hepatitis C patients and spontaneous resolvers than in healthy controls. This might be due to a decrease in other T cell populations including CD4+ and/or CD8+ T cells [29]. Collectively, the results of our study confirm a potential role of NK cells and the different subsets in chronic HCV infection. More focused research on this cell popu- lation with its different stimulatory, co- stimulatory and inhibitory receptors might answer questions concerning the pathogenesis of chronicity in HCV infection. This knowledge could help in the design of more effective thera- peutic approaches for HCV infection and contribute to the improvement of antiviral therapy and the eradication of the virus. References 1. Hepatitis viruses. In: Ryan KJ, Ray CG. Sherris medical microbi- ology: an introduction to infectious diseases, 4th ed. New York, Mcgraw–Hill, 2004:551. 2. Golden-Mason L, Rosen HR. Natural killer cells: primary target for hepatitis C virus immune evasion strategies? Liver Trans- plantation, 2006, 12:363–372. 3. Alter MJ. Epidemiology of hepatitis C virus infection. World Journal of Gastroenterology, 2007, 13:2436–2441. 4. Bowen DG, Walker CM. Adaptive immune responses in acute and chronic hepatitis C virus infection. Nature, 2005, 436:946–952. 5. Pawlotsky JM. Pathophysiology of hepatitis C virus infec- tion and related liver disease. Trends in Microbiology, 2004, 12:96–102. 6. Hiroishi K, Ito T, Imawari M. Immune responses in hepatitis C virus infection and mechanisms of hepatitis C virus per- sistence. Journal of Gastroenterology and Hepatology, 2008, 23:1473–1482. 7. Lanier LL. NK cell recognition. Annual Review of Immunology, 2005, 23:225–274. 8. Dessouki O et al. Chronic hepatitis C viral infection reduces NK cell frequency and suppresses cytokine secretion: Rever- sion by anti-viral treatment. Biochemical and Biophysical Re- search Communications, 2010, 393:331–337. 9. Caligiuri MA. Human natural killer cells. Blood, 2008, 112:461– 469. 10. Cooper MA, Fehniger TA, Turner SC. Human natural killer cells: A unique innate immunoregulatory role for the CD56 (bright) subset. Blood, 2001, 97:3146–3151. 11. Campbell JJ, Qin S, Unutmaz D. Unique subpopulations of CD56+ NK and NK-T peripheral blood lymphocytes identified by chemokine receptor expression repertoire. Journal of Im- munology (Baltimore, MD: 1950), 2001, 166:6477–6482. 12. Doherty DG, O’Farrelly C. Innate and adaptive lymphoid cells in the human liver. Immunological Reviews, 2000, 174:5–20. 13. McMahon CW, Raulet DH. Expression and function of NK cell receptors in CD8+ T cells. Current Opinion in Immunology, 2001, 13:465–470. 14. Deignan T et al. Decrease in hepatic CD56+ T cells and Va24+ natural killer T cells in chronic hepatitis C viral infection. Jour- nal of Hepatology, 2002, 37:101–108. 15. Lucas M et al. Frequency and phenotype of circulating double- positive natural killer T cells during hepatitis C virus infection. Journal of Virology, 2003, 77:2251–2257. 16. Morishima C et al. Decreased NK cell frequency in chronic hepatitis C does not affect ex vivo cytolytic killing. Hepatology (Baltimore, Md.), 2006, 43:573–580. 17. Thimme R, Lohmann V, Weber F. A target on the move: innate and adaptive immune escape strategies of hepatitis C virus. Antiviral Research, 2006, 69:129–141. 18. Golden-Mason L, Rosen HR. Natural killer cells: primary target for hepatitis C virus immune evasion strategies. Liver Transplan- tation, 2006, 12:363–372. 19. Khakoo SI et al. HLA and NK cell inhibitory receptor genes in resolving hepatitis C virus infection. Science, 2004, 305:872– 874. 20. Jinushi M et al. Negative regulation of NK cell activities by inhibitory receptor CD94/NKG2A leads to altered NK cell- induced modulation of dendritic cell functions in chronic hepatitis C virus infection. Journal of Immunology (Baltimore, MD: 1950), 2004, 15:6072–6081. 21. Kronenberg M. Toward an understanding of NKT cell biology: progress and paradoxes. Annual Review of Immunology, 2005, 23:877–900. 22. Golden-Mason L et al. Hepatic interleuklin 15 (IL-15) expres- sion: implications for local NK/NKT cell homeostasis and development. Clinical and Experimental Immunology, 2004, 138:94–101. 23. Golden-Mason L et al. Altered natural killer cell subset distri- butions in resolved and persistent hepatitis C virus infection following single source exposure. Gut, 2008, 57:1121–1128. 24. Lin AW et al. CD56+dim and CD56+bright cell activation and apoptosis in hepatitis C virus infection. Clinical and Experimen- tal Immunology, 2004, 137:408–416. 25. Meier UC et al. Shared alterations in NK cell frequency, phe- notype, and function in chronic human immune-deficiency virus and hepatitis C virus infections. Journal of Virology, 2005, 79:12365–12374. 26. Cooper MA, Fehniger TA, Caligiuri MA. The biology of hue- man natural killer-cell subsets. Trends in Immunology, 2001, 22:633–640. 27. Lee S et al. Increased proportion of the CD56(bright) NK cell subset in patients chronically infected with hepatitis C virus طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 973 (HCV) receiving interferon-alpha and ribavirin therapy. Journal of Medical Virology, 2010, 82:568–574. 28. Corado J et al. Impairment of natural killer (NK) cytotoxic activ- ity in hepatitis C virus (HCV) infection. Clinical and Experimental Immunology, 1997, 109:451–457. 29. Chisari FV. Unscrambling hepatitis C virus–host interactions. Nature, 2005, 436:930–932 Prevention and control of viral hepatitis infection: framework for global action Around 500 000 000 people are chronically infected with hepatitis B virus (HBV) or hepatitis C virus (HCV). Approximately 1 000 000 people die each year (about 2.7% of all deaths) from causes related to viral hepatitis, most commonly liver disease, including liver cancer. An estimated 57% of cases of liver cirrhosis and 78% of cases of primary liver cancer result from HBV or HCV infection. Despite affecting such a huge number of people worldwide, these diseases remain largely unknown, undiagnosed and untreated. WHO’s Prevention and control of viral hepatitis infection: framework for global action describes the 4 areas of work to prevent, treat and save the lives of people infected with hepatitis through targeted regional and country-specific strategies. The 4 strategic axes are: i) Raising awareness, promoting partnerships, and mobilizing resources; ii) Evidence- based policy and data for action; iii) Prevention of transmission; and iv) Screening, care and treatment. Further information about this document and the Global Hepatitis Programmes is available at: http://www.who.int/ topics/hepatitis/en/ EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 974 Lead exposure among the general population of Duhok governorate, Kurdistan region, Iraq A.H. Al-Dosky,1 D.J. Al-Timimi 1 and S.A. Al-Dabbag 2 ABSTRACT This study was carried out to establish baseline data on lead exposure in a sample of the general population from Duhok governorate, Iraq in 2011. A cross-sectional study was made of blood lead levels of 820 individuals and of dust and of air lead levels in urban, suburban and rural residential areas. Mean blood lead (BPb) value in the overall sample was 7.3 (SD 2.8) µg/dL, range 1.6–17.0 µg/dL; significant differences were found with respect to age, area of residence, altitude and distance from a gasoline generator. The prevalence of toxic levels of lead (BPb > 10 µg/dL) was 22.8% overall and 2.4% in children. People living in suburban areas (39.7%) and < 50 m from a gasoline generator (93.1%) were exposed to high environmental lead levels. A linear relationship was found between BPb and air lead levels (r = 0.8). While lead exposure does not constitute a great health problem for adults in our community, 2.4% of the children had toxic lead levels and measures to reduce exposure to environmental heavy metals should be considered. 1Department of Clinical Biochemistry; 2Department of Family and Community Medicine, College of Medicine, University of Duhok, Duhok, Iraq. (Correspondence to S.A. Al-Dabbagh: samimaldabbagh@yahoo.com). Received: 29/08/11; accepted: 02/11/11 قارعلا ،ناتسدرك ميلقإ ،كوهد ةظفامح في ناكسلا ةماع ىدل صاصرلل ضرعتلا غابدلا ميمص ،يميمتلا ءايض ،يقسدلا ليع قارعلا في كوهد ةظفامح في ناكسلا ةماع نم ةنيع ىدل صاصرلل ضرعتلل ساسلأا طخ تايطعم عضول ةساردلا هذه نوثحابلا ىرجأ :ةـصلالخا قطانلما في ءاولها في صاصرلا تايوتسمو رابغلا سايقلو ؛ًاصخش 820 مد في صاصرلا تايوتسم سايقل ةضرعتسم ةسارد يهو .2011 ماع في فارحنلاا( ترل سييد/مارغوركم 7.3 ةنيعلا لممج في مدلا في صاصرلل ةيطسولا ةميقلا نأ ينثحابلل حضتاو .ضراولحاو يحاوضلاو ةيفيرلا ةينكسلا دعبلاو رحبلا حطس نع عافترلااو ةماقلإا ناكمو رمعلاب لصتي ظوحلم توافت كانه ناكو ،ترل سييد/مارغوركم 17.0-1.6 لاجلماو ،)2.8 يرايعلما )ترل سييد/مارغوركم 10 نم رثكأ مدلا في صاصرلا ىوتسم( صاصرلل ةماّسلا تايوتسلما راشتنا تلادعم تغلب دقو .ينلوزاغلا جاتنإ عقاوم نع ديلوت عقاوم نم ًاترم 50 نع لقت ةفاسم لىعو )%39.7( يحاوضلا في نوشيعي نيذلا سانلا نأ حضتاو .لافطلأا ىدل %2.4و لمجلما في %22.8 ءاولها في هتايوتسمو مدلا في صاصرلا ىوتسم ينب ةيطخ ةقلاع كانه ناك ماك ،صاصرلا نم ةعفترم ةيئيب تايوتسلم نوضرعم )%93.1( ينلوزاغلا نم ةيمس تايوتسم نم نوناعي لافطلأا نم %2.4 نإف ،انعمتمج في ينغلابلا ىدل ةماه ةيحص ةلكشم لّثمي لا صاصرلل ضرعتلا ناك اذإو .)r = 0.8( .ةئيبلا في ةليقثلا نداعملل ضرعتلا ضفخ فدهتست يتلا تاءارجلإاب مماتهلاا ءلايإ يغبنيو ،صاصرلا Exposition au plomb dans la population générale du gouvernorat de Duhok, dans la région du Kurdistan (Iraq) RÉSUMÉ La présente étude a été menée pour établir une base de données sur l'exposition au plomb d'un échantillon de la population générale du Gouvernorat de Duhok (Iraq) en 2011. Une étude transversale a été réalisée sur la concentration sanguine de plomb chez 820 personnes ainsi que sur les taux de poussière et de plomb dans l'air des zones résidentielles urbaines, suburbaines et rurales. La concentration moyenne de plomb dans le sang de l'échantillon global était de 7,3 µg/dl (E.T. 2,8 ; extrêmes 1,6–17,0 µg/dl) ; des différences significatives ont été retrouvées en fonction de l'âge, du lieu d'habitation, de l'altitude et de la distance par rapport à un groupe électrogène à essence. La prévalence de niveaux toxiques de plomb (concentration moyenne de plomb dans le sang > 10 µg/dl) était de 22,8 % dans l'ensemble de l'échantillon et de 2,4 % chez les enfants. Les personnes vivant dans des zones suburbaines (39,7 %) et à moins de 50 m d'un groupe électrogène à essence (93,1 %) étaient exposées à des niveaux élevés de plomb dans l'environnement. Une relation linéaire a été retrouvée entre la concentration moyenne de plomb dans le sang et les niveaux de plomb dans l'air (r = 0,8). Alors que l'exposition au plomb ne constitue pas un problème de santé majeur pour les adultes de notre communauté, 2,4 % des enfants présentaient des concentrations de plomb toxiques. Des mesures de réduction de cette exposition aux métaux lourds dans l'environnement doivent être envisagées. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 975 Introduction Exposure to low doses of lead is par- ticularly harmful to children, whose neurological development may be per- sistently and irreversibly affected [1–3]. Exposure to lead causes a number of diseases in later age groups, including mild mental retardation, as shown by loss of IQ points, as well as increased blood pressure, anaemia and gastro- intestinal effects. Several other disease outcomes have been associated with exposure to lead [4,5]. The only effica- cious method of avoiding these toxic effects is to control the population’s ex- posure to lead by eradicating its sources, which are not always obvious. Hence it is essential to know the level of exposure of the population so that the necessary control measures can be established. In Iraq, most studies on lead expo- sure have been conducted in exposure risk groups [6]. However, considering that there are still many gasoline power generators and vehicles using leaded petrol in the country, it is possible that high lead levels exist in the general population living in areas of heavy traf- fic and high emission of lead into the atmosphere. Therefore, this study was carried out to establish baseline data on blood and environmental lead levels in a sample of the general population from the Duhok governorate area. Methods Setting and sample A cross-sectional study design was con- ducted on 820 individuals who were not occupationally exposed to lead. They were apparently healthy individuals of both sexes with an age range from 1 month to 64 years. All had no history of chronic illness or of personal or family history of lead exposure. The sample size was calculated according to the World Health Or- ganization formula to be around 767 [7]. Individuals living in 3 distinct living areas (urban, suburban and ru- ral) were included in this study. The urban areas were in Duhok city, the main urban area in the centre of Duhok governorate. The city is located in a valley between 2 mountains and areas were chosen at different altitudes in the valley: low (540– < 560 m), moder- ate (560– < 630 m), and high (> 630 m). The suburban areas chosen were districts 60 km west of Duhok, 100 km east of Duhok and 70 km north Duhok. The rural areas chosen were 5 villages situated in the districts of Duhok governorate where there were no municipality and no industrial sources of lead pollution. Among the individuals selected, 420 were urban living in different parts of Duhok city, 300 were living in suburban areas and 100 were living in rural areas. Urban samples were collected from Azadi main secondary care hospital and the primary health centre in Duhok, sub- urban samples were collected from the local hospital while rural samples were collected directly from in the village main hall. Data collection A pre-tested questionnaire was de- signed to obtain information on age, sex, residence, cigarette smoking and current occupation. Samples Since blood lead estimation is eas- ily affected by contamination suitable precautions were taken to avoid any contamination and a trace-element-free technique was used during the handling and analysis of blood samples. Venous blood samples (5 mL) were withdrawn from each individual under study and transferred to EDTA tubes for the esti- mation of blood lead level. For determination of environmental dust lead levels roadside dust samples were collected from different residential areas (urban, suburban and rural) using methods specified by the manufacturer of the test kit. For determination of environment air lead levels roadside air samples were collected from the same urban and ru- ral areas investigated. A low flow-rate sampler with maximum flow rate of 6 m2/24 hours was used. This sampler was equipped with a pump, a counter, a 3.7 cm disk holder and fitted with glass fibre filter (Whitman GF/A). The col- lection time was 72 hours in the urban areas and 144 hours in the rural areas. Verbal consent for participation and approval to supply blood samples for testing was obtained from the parents of the children and from the adults, af- ter the nature of the study had been explained to them. The study protocol was approved by the local scientific and ethical committees. Analysis Blood lead was analysed by flame atomic absorption spectrophotometer (Perkin Elmer) using a standardized procedure published by the company. Radiation source absorbance was measured at a wavelength of 283 nm, using a band pass of 0.7 nm. Briefly EDTA–blood was precipitated with 20% trichloroacetic acid (TCA) so- lution. The supernatant solution was aspirated directly to the spectropho- tometer and the samples were run in batches using standard lead solution of 10, 25 and 50 µg/dL to correct the sensitivity of the instrument. A reagent blank was prepared by mixing an equal volume of 20% TCA with deionized water. The blank gave a reading of 1 µg/ dL and this value was subtracted from the reading of the samples examined. A pooled normal blood was run with every batch and gave a reading of 12 µg/ dL. Serial replication of aliquots from a pooled blood sample and an internal control standard were used to check the precision and accuracy of the analytical method. The coefficient of variation for lead in pooled blood samples was 3.5% (n = 30). Values for the internal control EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 976 standards (Radian Corporation) were calculated between every 10 samples in the rack of samples with each new batch. Samples which did not differ by more than 2% of the standard value were considered acceptable. All samples were analysed in triplicate and results referred to the standard curve previ- ously prepared. The dust lead levels in each area were estimate using a standard lead test kit (Abbott Laboratories). This test utilizes a patented leach method for testing and gives semi-quantitative results which indicate the approximate lead release in samples. After the comparison of the resultant colour produced, the concen- tration of lead is shown into parts per million (ppm). For the air samples, lead from the Whitman glass fibre filter was extracted with 5% nitric acid. Similar to the blood samples, the lead concentration was estimated by direct aspiration into the flame of the atomic absorption spectrophotometer. The air lead con- tent was calculated from the American Society for Testing Materials formula [8]. Assessment of lead exposure The highest values for toxic levels of BPb lead according to the recommenda- tions of the Centers for Disease Control and Prevention [9] were selected as the cutoffs for this study: > 10 µg/dL in children and > 25 µg/dL in adults. Dust lead levels > 5 ppm was considered as the cutoff value for low level of lead exposure whereas a level of < 25 ppm indicated high level of lead exposure. Statistical analysis Data were translated into a computer- ized database structure. Statistical analy- ses were computer assisted using SPSS, version 13. The correlations between blood lead level (BPb) and air lead level was estimated by linear regression analysis. Results Blood lead level The mean BPb value in the total sam- ple was 7.3 µg/dL [standard deviation (SD) 2.8 µg/dL], range 1.6 to 17.0 µg/ dL. The mean BPb concentration in adults (ages 17–64 years) was 8.6 (SD 3.3) µg/dL and in children (ages > 1–16 years) was 5.3 (SE 1.9) µg/dL, a differ- ence which was statistically significant (P < 0.01). Mean BPb levels of males [8.6 (SD 3.5) µg/dL] were significantly higher than those of females [4.7 (SD 1.4) µg/dL] (P < 0.01). The mean BPb concentration in those living in rural areas [4.2 (SD 1.2) µg/dL] was signifi- cantly lower compared with the urban and suburban groups, who had mean levels of 6.9 (SD 3.0) µg/dL and 9.1 (SD 3.2) µg/dL respectively (P < 0.001 for all) (Table 1). Comparing the mean BPb level of urban males (n = 320) and females (n = 100) showed higher values for males Table 1 Mean blood lead levels in the study population according to age, sex and residence Variable Total Blood lead concentration (µg/dL) P-valuea No. Mean (SD) Range Age of children (years) < 0.001b < 1 22 3.3 (0.7) 1.6–4.4 < 5 46 4.3 (0.5) 1.8–5.0 5–9 66 5.5 (1.4) 2.1–7.8 10–16 73 6.2 (2.4) 2.5–10.1 All 207 5.3 (1.9) 1.6–10.1 Age of adults (years) < 0.001 17–9 349 7.7 (2.9) 3.3–13.0 30–39 168 9.2 (3.2) 3.8–14.9 40+ 96 11.1 (2.9) 5.8–17.0 All 613 8.6 (3.3) 3.3–17.0 Sex < 0.001 Male 541 8.6 (3.5) 2.7–17.0 Female 279 4.7 (1.4) 1.6–9.4 Residence area < 0.001 Urban 420 6.9 (3.0) 1.6–15.0 Suburban 300 9.1 (3.2) 4.0–17.0 Rural 100 4.2 (1.2) 2.1–7.6 Total 820 7.3 (2.8) 1.6–17.0 aAnalysis of variance; bAge group < 1 year versus 10–16 years. SD = standard deviation. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 977 [7.6 (SD 3.7) µg/dL] than females [4.4 (SD 0.6) µg/dL)] (P < 0.001). How- ever, the difference was smaller and non-significant comparing rural males (n = 54) and rural females (n = 46) [4.5 (SD 1.8) µg/dL versus 3.9 (SD 0.5) µg/ dL] (P > 0.05). Current guidelines for lead poison- ing consider cut-offs for toxic levels when BPb levels are > 10 µg/dL for a child and > 25 µg/dL for an adult. We found 5/207 children (2.4%) had BPb level > 10 µg/dL, whereas 182/613 adults (29.7%) had BPb level 10–25 µg/dL. Therefore out of the total tested, 187 (22.8%) individuals had toxic BPb levels > 10.0 µg/dL (Table 2). BPb levels according to distance in which participants living from a gasoline generator and altitude are shown in Ta- ble 3. The mean BPb levels of individu- als living in areas 100–149, 50–99 and > 50 m distance from gasoline generators were 3.3, 6.8 and 9.6 µg/dL respectively (P < 0.001 for all). BPb levels were also significantly higher in individuals of living at low altitudes compared with moderate or high altitude levels (P < 0.001). Dust lead level Table 4 presents the distribution of the study sample according to median dust lead values found in the areas where they lived. The highest proportion of people living in areas with lead level in the range 25–50 ppm was found in those living in suburban areas (39.7%), at low altitudes (62.8%) and < 50 m from a gasoline generator (93.1%). Air lead level Table 5 shows the mean lead levels in environmental air. Mean lead levels were significantly higher in urban ar- eas than suburban and rural areas (P < 0.001). The mean blood lead levels of participants were positively correlated with mean lead levels in air (r = 0.8, P < 0.01). Discussion The BPb level is commonly used as an indicator of lead exposure in humans [10]. Among the individuals in the pre- sent study, mean BPb concentration in adults was 8.6 µg/dL and in children was 5.3 µg/dL. The results suggest that, overall, individuals in our study had BPb levels within the acceptable limit reported by others (7.3 µg/dL) [11]. Previous studies in other parts of Iraq found higher BPb levels in the population [12,13]. For example, a study conducted on 629 elementary- school children in Baghdad found a mean BPb level of 13.9 µg/dL [14]; this value was higher than ours, probably due to the origin of their samples, which were taken mainly from the centre of the city. Furthermore, several studies on children in the Eastern Mediter- ranean region (EMR) indicated high levels of lead exposure. For example, an investigation of environmental lead concentrations in Cairo indicated that the BPb levels of children ranged from 14.4 µg/dL for those < 1 year old to 10.8 µg/dL for those aged 5–6 years old, and that approximately 64% of children had BPb levels > 10 µg/dL. Similar findings have been reported from Pakistan and Saudi Arabia, as well as other countries in the EMR [15]. However, our results, in comparison, reflected a lower level of exposure, probably due to the geo- graphical nature of the residential areas, the low number of vehicles, no indus- trial emissions and low occupational exposure [16]. With respect to BPb levels, our re- sults overall showed higher values in urban individuals living near gasoline generators or at low altitudes. Other studies concur with this observation of higher values in urban areas with greater atmospheric contamination by Table 2 Distribution of blood lead levels in the study population according to age Age (years) Total Blood lead concentration (µg/dL) < 5.0 5–10 > 10–25 No. No. % No. % No. % Children < 1 22 22 100.0 0 0.0 0 0.0 < 5 46 36 78.3 10 21.7 0 0.0 5–9 66 36 54.5 30 45.5 0 0.0 10–16 73 20 27.4 48 65.8 5 6.8 All 207 114 55.1 88 42.5 5 2.4 Adults 17–29 349 89 25.5 172 49.3 88 25.2 30–39 168 17 10.1 93 55.4 58 34.5 40+ 96 0 0.0 60 62.5 36 37.5 All 613 106 17.3 325 53.0 182 29.7 Total 820 220 26.8 413 50.4 187 22.8 EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 978 the metal [17], and in crowded areas in which exposure to lead is higher owing to the greater number of gasoline gener- ators and vehicles. The results obtained for prevalence of intoxication were also similar, since a greater prevalence was observed in individuals residing in downtown areas and those living near gasoline generators. Regarding age and sex, we found a tendency towards increased BPb levels with higher age, probably due to higher exposure to lead from the contami- nated atmosphere owing to increased Table 3 Blood lead levels according to distance from gasoline generators and individuals' living altitude levels (n = 420) Variable Total Blood lead concentration (µg/dL) P-valuea No. Mean (SD) Range Distance from gasoline generator (m) < 50 158 9.8 (3.5) 4.0–15.0 < 0.00150–99 140 6.8 (1.5) 2.7–12.7 100–149 122 3.3 (1.4) 1.6–11.8 Altitude Low (540–> 560 m) 163 9.1 (4.4) 3.7–15.0 < 0.001Moderate (560–> 630 m) 189 6.3 (2.2) 2.3–11.8 High over (≥ 630 m) 68 3.3 (1.9) 1.6–7.6 aAnalysis of variance. SD = standard deviation. Table 4 Distribution of study population by median dust lead concentration in their area of residence Variable Total Dust lead concentration (ppm) 1–5 > 5–10 > 10–25 > 25–50 Mediana No. No. % No. % No. % No. % Residence area Urban 420 9 2.1 156 37.1 153 36.4 102 24.0 10–25 Suburban 300 0 0.0 27 9.0 154 51.3 119 39.7 10–25 Rural 100 18 18.0 55 55.0 27 27.0 0 0.0 5–10 Distance from gasoline generator (m) < 50 158 0 0.0 0 0.0 11 7.0 147 93.0 25–50 50–99 140 0 0.0 0 0.0 47 33.6 93 66.4 25–50 100–149 122 0 0.0 6 4.9 113 92.6 3 2.5 10–25 Altitude Low 317 0 0.0 37 11.7 81 25.6 199 62.8 25–50 Moderate 369 3 0.8 82 22.2 109 29.5 175 47.5 10–25 High 134 7 5.2 34 25.4 90 67.2 3 2.2 10–25 aP < 0.01, Kruskal–Wallis: urban and suburban versus rural; distance from gasoline generators < 50 m and 50–99 m versus 100–149 m; low and moderate altitude versus high altitude. outdoor activity. A statistically signifi- cant difference was found in the mean lead levels of males and females. Other authors have also demonstrated lower lead levels in females, below the toxic limit [18]. In fact, none of the females in our series had BPb levels within the elevated range (> 10 µg/dL). How- ever, large numbers of males had BPb values > 10 µg/dL. This could be also attributed to increased outdoor activity and to cigarettes smoking (none of the females in this study reported being a cigarette smoker). Blood lead level determination is the method currently recommended Table 5 Mean lead levels in air of the urban, suburban and rural residential areas Area Total Air lead concentration (µg/m3) No. Mean (SD) Range Urban 34 3.80 (0.51) 1.30–7.02 Suburban 2 2.29 (0.24) 2.04–2.52 Rural 4 0.44 (0.06) 0.37–0.55 SD = standard deviation. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 979 References 1. American Academy of Pediatrics. Committee on Environmen- tal Hazards: Committee on Accident And Poison Prevention. Statement on childhood lead poisoning. Pediatrics, 1987, 79:457–465. 2. Bellinger D. Lead neurotoxicity and socioeconomic sta- tus: conceptual and analytic issues. Neurotoxicology, 2008, 29(5):828–832. 3. Jones RL et al. Trends in blood lead levels and blood lead test- ing among US children aged 1 to 5 years, 1988–2004. Pediatrics, 2009, 123:e376–e385. 4. Schwartz J. Low-level lead exposure and children’s IQ: a meta- analysis and search for a threshold. Environmental Research, 1994, 65:42–55. 5. Koller K et al. Recent developments in low-level lead exposure and intellectual impairment in children. Environmental Health Perspectives, 2004, 112:987–994. 6. Al-Timimi DJ. The adverse health effects of direct and indirect exposure to lead among battery factory workers. Journal of Community Medicine, Baghdad, 1990, 3(20):103–110. 7. Introduction to basic epidemiology and principles of statistics for tropical disease control. Geneva, World Health Organization, 2002. 8. Toxic materials in the atmosphere: sampling and analysis. A symposium sponsored by ASTM Committee D.22 on sampling and analysis of atmosphere. Boulder, Colorado, 2–5 August, 1981 (ASTM Special Technical Publication No. 786). Philadelphia, American Society for Testing and Materials, 1982. 9. Preventing lead poisoning in young children. Atlanta, Georgia, Centers for Disease Control and Prevention, 2005. 10. Screening young children for lead poisoning: guidance for state and local public health officials. Atlanta, Georgia, Centers for Disease Control, 1997. 11. Needleman HL. Lead poisoning. Annual Review of Medicine, 2004, 55:209–222. 12. Jamil H et al. Lead absorption in battery factory workers. Jour- nal of the Faculty of Medicine, Baghdad, 1987, 29:211–222. 13. Al-Timimi DJ et al. Lead exposure among the general popula- tion. Iraqi Medical Journal, 1988, 37:111–116. 14. Al-Timimi DJ, Jamil H, Abu-timman AK. Effect of traffic on lead absorption among children. Journal of the Faculty of Medicine, Baghdad, 1988, 30:95–103. 15. A review of literature on healthy environments for children in the Eastern Mediterranean Region. Status of childhood lead exposure. Cairo, World Health Organization Regional Office for the East- ern Mediterranean, 2003. 16. Al-Dosky AH, Al-Timimi DJ, Al-Dabbagh SA. Occupational exposure to lead in Duhok city, Kurdistan region, Iraq. Duhok Medical Journal, 2011, 5(2):76–85. 17. Sole E, Ballabriga A, Dominguez C. Lead exposure in the gen- eral population of the Metropolitan area of Barcelona: Blood levels and related factors. Science of the Total Environment, 1988, 224:19–27. 18. Binns H, Campbell C, Brown M. Interpreting and managing blood lead levels of less than 10 microg/dL in children and reduced exposure to lead: recommendations of the Cent- ers for Disease Control and Prevention Advisory Committee on Childhood Lead Poisoning Prevention. Pediatrics, 2007, 120:1285–1298. 19. National report on human exposure to environmental chemicals. Atlanta, Georgia, Centers for Disease Control and Prevention, 2001. 20. World health report 2001. Mental health: new understanding, new hope. Geneva, World Health Organization, 2001. for lead exposure assessment [19]. The use of dust and air lead assessment has been previously recommended as ad- ditional markers for lead source and exposure [20]. We found a tendency towards increased BPb values with in- creasing level of exposure indicated by dust and air levels. In our sample, 57.8% of individuals were living in areas with dust lead levels 25–50 ppm and most of them in urban and suburban areas of high air lead levels. We conclude from our results that lead exposure does not constitute a great health problem for the adult population of our community, since average blood lead levels were below toxic levels and the prevalence of intoxication was similar to that reported in other studies. Neverthe- less, 2.4% of the children had toxic levels of lead and measures to reduce exposure to environmental heavy metals should be considered, especially in urban areas where children are particularly at risk. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 980 Consumption of iodized salt among households of Basra city, south Iraq S.M. Ebrahim 1 and N.K. Muhammed 2 ABSTRACT Progress towards universal salt iodization in Iraq is uncertain. A cross-sectional study with cluster sampling was carried out to investigate the sources and packaging of salt in 900 households in Basra and assess the knowledge and practice of household members regarding iodized salt and iodine deficiency disorders. Adequately iodized salt was found in only 68.3% of households. Salt from local sources was significantly less likely to be iodized than salt from imported sources (15.3% versus 86.9% had ≥ 15 ppm). Salt in plastic packages was mostly adequately iodized (95.0%) compared with loose packages (70.3%). Most respondents had heard about iodized salt (92.6%), yet only 27.1% knew about the health benefits. Significantly more household members with university education had iodized salt in the household than those who were illiterate or only could read/write (82.0% versus 57.7%). Better monitoring of local salt supplies and more education to improve the population’s awareness of iodization are needed. 1Department of Basic Medical Science, College of Nursing, University of Basra, Basra, Iraq (Correspondence to S.M. Ebrahim: samira_m_ ebrahim@yahoo.com). 2Department of Public Health, Directorate General of Basra Health Services, Basra, Iraq. Received: 06/05/11; accepted: 23/08/11 قارعلا بونج ،ةصربلا ةنيدم في تلائاعلا ينب نَدْوَيُلما حللما كلاهتسا دممح مساق دانه ،ميهاربإ دممح ةيرمس ةيدوقنع تانيع لىع ةضرعتسم ةسارد ناتثحابلا ترجأ دقو .قارعلا في ةلماشلا حللما ةَنَدْوَي في زرحلما مدقتلا نع ةدكؤم تامولعم دجوت لا :ةـصلالخا تابارطضا نعو نَدْوَيُلما حللما نع تلائاعلا دارفأ ىدل تاسرمالماو فراعلما مييقتلو ،ةصربلا ةنيدم في ةسرأ 900 ىدل هتئبعتو حللما رداصم ءاصقتسلا نوكي نأ بلغي ةيلمح رداصم نم دمتسلما حللما نأ ماله َّنيبتو .تلائاعلا نم %68.3 ىدل ةيفاك تايمكب نَدْوَيُلما حللما ناتثحابلا تدجوو .دويلا زوع ديزت وأ لداعت ةيمك لىع دروتسلما حللما نم 86.9 لباقم ليحلما حللما نم %15.3 نمضت ذإ( ابه دَتْعُي ةجردلو ،دروتسلما حللما نم ةَنَدْوَيلا ثيح نم لقأ ةنراقم )%95.0( ةيفاك ةجردب ًانَدْوَيُم نايحلأا بلاغ في ناك ةيكيتسلابلا تاوبعلا في حللما نأ ناتثحابلا تدجو ماك .)دويلا نم نويللماب ءزج 15 نع اوناك طقف مهنم 27.1 نأ لاإ ،)%92.6( نَدْوَيُلما حللما نع اوعمس دق ةساردلل يبيجتسلما مظعم نأ يثحابلل حضتاو .)%70.3( ةقلغلما يرغ تاوبعلاب نيذلا وأ ييملأا ىدل امم رثكأ تلائاعلا لخاد نَدْوَيُم حلم ميهدل يعماج ميلعت لىع يلصالحا تلائاعلا دارفأ نأ يثحابلل حضتا ماك .هعفانم نوفرعي ةيلحلما تادادملإل لضفأ دصرل ةسام ةجالحا نأ ناتثحابلا تجتنتساو .)%57.7 لباقم %82.0( هب دَتْعُي قرفلاو ،ةباتكلاو ةءارقلا داكلاب نوعيطتسي .حللما ةَنَدْوَيب سانلا يعو يسحتل فيقثتلا نم ديزمللو حللما نم Consommation de sel iodé des ménages de la ville de Bassora (sud de l'Iraq) RÉSUMÉ Les progrès tendant vers une iodation universelle du sel en Iraq sont incertains. Une étude transversale d'un échantillon en grappes a été menée pour connaître les sources d'approvisionnement et le conditionnement du sel dans 900 ménages de la ville de Bassora, mais aussi pour évaluer les habitudes de consommation de sel iodé des membres des ménages et leurs connaissances des troubles dus à une carence en iode. La présence de sel adéquatement iodé a été observée dans seulement 68,3 % des ménages. Le sel de sources locales était significativement moins iodé que le sel importé (15,3 % contre 86,9 % contenait plus de 15 ppm). Le sel conditionné en emballage plastique avait le plus souvent une teneur en iode adéquate (95,0 %) par rapport au sel vendu en vrac (70,3 %). La plupart des répondants avaient entendu parler du sel iodé (92,6 %) ; pourtant, seulement 27,1 % en connaissaient les bienfaits pour la santé. Les membres des ménages qui avaient fait des études universitaires étaient plus nombreux à consommer du sel iodé que ceux qui étaient analphabètes ou savaient seulement lire ou écrire (82,0 % contre 57,7 %). Un meilleur suivi de l'approvisionnement local en sel et une meilleure éducation sont nécessaires pour améliorer la sensibilisation de la population à l'iodation. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 981 Introduction The greatest cause of preventable brain damage in childhood is iodine defi- ciency, caused by a low dietary supply of iodine [1]. This is the primary motiva- tion behind the current worldwide drive to eliminate iodine deficiency through the strategy of universal salt iodization [2–4]. The World Health Organization (WHO)/United Nations Children's Fund (UNICEF)/International Coun- cil for the Control of Iodine Deficiency Disorders (ICCIDD) have approved a cut-off point of 20–40 parts per million (ppm) for iodine in salt. The goal is to have 90% of households using ad- equately iodized salt (≥ 15 ppm iodine) and to use salt iodine testing as an indi- cator for monitoring progress towards universal salt iodization [5–7]. In 1993 the Iraqi government be- gan a plan of action towards iodization of all salt in the country and to ensure that only iodized salt was distributed in food rations [8]. Yet inadequacies in the programme were reported in 2000 [9] and when the status of io- dine deficiency disorders (IDD) in the Eastern Mediterranean countries was monitored in the beginning of 2001, Iraq was classified as having severe IDD along with Afghanistan and Pakistan [8]. Following the invasion of Iraq by the Allied Forces in 2003 and the cur- rent political difficulties, the status of iodine nutrition in the Iraqi population is still uncertain. Salt is no longer distrib- uted in food rations and people tend to meet all their salt requirements from the local markets, with most people choosing low-price salt brands which may be partially iodized or even totally uniodized. This was confirmed by the World Food Programme’s food secu- rity survey in late 2007, which revealed that only 23.4% of Iraqi households had adequately iodized salt and for Basra governorate in the south of Iraq the rate was even lower (20.7%) [10]. The present study was carried out to investigate the sources and packaging of salt in households in Basra and assess the knowledge and practice of household members in Basra regarding iodized salt and IDD in order to understand reasons behind the low consumption of iodized salt. Methods Setting and sample The study was a cross-sectional, multi- stage study conducted in Basra city, which is located in the centre of Basra governorate in the south of Iraq and is the biggest of the 7 administrative districts. The study was conducted dur- ing March and April 2011. The survey used probability proportionate to size cluster sampling, a methodology widely used for health indicators. The sample size was 900 households derived from 30 clusters, with 30 households from each cluster. In the areas selected, 2 field teams coordinated closely with health visitors at local primary health care centres to identify and invite household members for the assessments. Prior to the fieldwork, each team was trained in interviewing techniques, standardized coding and completion of forms and spot-testing of household salt. Data collection In this survey, interviewers tested samples of salt found in households for iodine levels using spot test kits (MBI Chemicals) obtained through UNICEF. Adequately iodized salt has been defined as containing ≥ 15 ppm of iodine, and according to manufacturer’s specifications these kits were accurate for visual detection of iodine concentra- tions at 0, 7, 15 and 30 ppm. At the time of testing, information about the brand name (local or imported) and the type of packaging (plastic non-transparent or transparent), and whether it was la- belled as iodized or not, was noted. The person in the household prin- cipally responsible for food cooking was identified and given a series of knowledge and practice questions. Re- spondents were asked about their level of education and if they knew about the effects of iodine deficiency; had heard about iodized salt; and were aware of the health benefits of iodized salt. Re- spondents were also asked about their salt storage practices. Data analysis All data were expressed as numbers and percentages. SPSS, version 16 was used for the statistical analysis. The chi- squared test was used to examine the association between the groups and a P-value < 0.05 was considered to be significant Results Adequately iodized salt (≥ 15 ppm) was found in 614 (68.3%) of the 900 tested salt samples, 75 (8.3%) were partially io- dized and 211 (23.4%) were uniodized. Imported brands of salt were consumed by 665 households (73.9%), out of which 578 (86.9%) were adequately iodized. The remaining 235 households (26.1%) consumed locally produced salt from factories in Basra and nearby governorates, out of which only 36 sam- ple (15.3%) were adequately iodized. This difference was statistically signifi- cant (P < 0.01) (Table 1). Original salt packages were available in 802 households (89.1%) and all were labelled as iodized. Table 1 showed that of 161 salt samples packed in non-trans- parent plastic packages and labelled as iodized, 95.0% were adequately iodized, and this was significantly higher than in the 641 salts packed in transparent packages (70.3%) (P < 0.01). Significantly more of the house- hold respondents with higher educa- tion level used adequately iodized salt than those with lower education levels (82.0% of those with university educa- tion versus 57.7% of those where were illiterate or only read/write) (P < 0.01) (Table 1). Although the great majority EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 982 of household respondents had heard about iodized salt (92.6%), Table 2 shows that only 27.1% were aware of the health benefits of iodized salt. Even among those who consumed iodized salt only 33.6% were aware of the ben- efits. Most of the households (804, 89.3%) had good storage practices, keeping their salt covered, in a place away from humidity. Discussion Rapid test kits were used for the as- sessment of iodine levels in the house- hold salt samples because these give immediate results and may also play a valuable educational role in that they provide a visible indication to the public that salt actually is iodized [5]. Their sensitivity for determining a specific salt iodine concentration (e.g. 15 ppm) has been less firmly established [11]. More than two-thirds (68.3%) of households in Basra city were consuming iodized salt, which was a higher than the rate in 2007. Yet the current rate was still considerably below the national goal of 90% coverage. Our figure did not differ much from the results of household surveys in Egypt (68.3%) and Oman (78%) but was much higher than what was reported for Yemen [12]. The higher rate of consumption of adequately iodized salt in the current study may be because the Ministry of Health in the last few years required all imported salts to be checked at import for the certificate of conformity and this may have been more effective in promoting adequate iodized salt sup- plies than the food inspections of the domestic salt enterprises [13]. Imported salts were consumed more than locally produced salt brands due to the fact that salt factories in Iraq supply only about 10% of the salt needed for the country and not all them produce iodized salt [14]. All salt factories in Iraq are private and some of them have no of- ficial license and are still producing uni- odized salt. However, local monitoring Table 1 levels of iodized salt consumption in sampled household by the source of salt production, type of packaging and education of the household respondents (n = 900) Variable No. Iodine content (ppm) P-value 0 < 15 ≥ 15 No. % No. % No. % Source of salt tested < 0.01 Imported 665 40 6.0 47 7.1 578 86.9 Local 235 171 72.8 28 11.9 36 15.3 Total 900 211 23.4 75 8.3 614 68.3 Type of salt package < 0.01 Transparent 641 128 20.0 62 9.7 451 70.3 Plastic 161 3 1.9 5 3.1 153 95.0 Total 802 131 16.3 67 8.4 604 75.3 Respondent’s educational level < 0.01 Illiterate/read and write 345 107 31.0 39 11.3 199 57.7 Primary 214 46 21.5 15 7.0 153 71.5 Secondary 208 40 19.2 15 7.2 153 73.6 Higher 133 18 13.5 6 4.5 109 82.0 Total 900 211 23.4 75 8.3 614 68.3 Table 2 Knowledge of household respondents about iodized salt (n = 900) Variable No. % Knew effects of iodine deficiency Yes, at least 1 188 20.9 Yes, 2 or more 54 6.0 No 658 73.1 Heard about iodized salt Yes 833 92.6 No 67 7.4 Aware of health benefits of iodized salt Yes 244 27.1 No 656 72.9 Consumed iodized salt (n = 614) And aware of its benefits 206 33.6 But not aware of its benefits 408 66.4 طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 983 has also shown that there are a number of fake products on the market, with uniodized salt wrongly labelled as io- dized in Basra city. In this study not all the imported salt which was labelled as iodized was found to be iodized when reaching the consumers and this may be related to the following: variability in the amount of iodine added during the io- dization process; uneven distribution of iodine in the iodized salt within batches and individual bags due to insufficient mixing of salt after the salt iodization process and/or variation in particle size of salt crystals in a batch; loss of iodine due to salt impurities, packaging and environmental conditions during stor- age and distribution; and the availability of uniodized salt (which could be past its expiry date) from unconventional marketing sources [5]. In this study salt iodine levels dif- fered significantly between the types of salt packaging materials, similar to a household survey in Kazakhstan [11] and a national micronutrient survey in Oman in 2004 [12]. A study conducted in Canada showed the effect of high hu- midity coupled with porous packaging that resulted in 30%–80% loss of iodine within a period of 6 months. However, longer storage—beyond 6 months— aggravated losses. Therefore, it is rec- ommended that the time required for distribution, sale and consumption of iodized salt be minimized as far as pos- sible, to ensure effective use of the added iodine [15]. This result agrees with the results of a study in China which sug- gested that waterproof sealed sacks protect the salt from temperature and humidity [16]. In this study women were targeted in particular because of their role in controlling family food purchasing and consumption in the majority of households. A higher education level of household respondents was associated with higher rate of consumption of io- dized salt. That could be because better education may influence good practice through better awareness about iodized salt [17,18] or because those with bet- ter education have a higher incomes, which enables them to afford more expensive, imported brands of salt. The rate of knowledge about the benefits of iodine in our study was lower than in a household survey of food caterers in Ethiopia but the proportion who had heard about iodized salt was higher in our study [17]. Respondents’ aware- ness about the benefits of consuming iodized salt was lower than for female respondents in a household survey in Oman [12] and much lower than in Kazakhstan [11]. The rate of consump- tion of iodized salt among those who aware of the benefits of iodized salt was not much different from a study done in Turkey on the effect of mass media in promoting iodized salt consumption among married women [19]. Most of the households stored their salt with a cover and away from humid- ity. These good practices may not be due to care of iodized salt but to prevent salt contamination and to keep it from moisture. More public information is needed about storage practices at the retailer and household level to prevent loss of iodine from salt. Conclusions and recommendations Only about two-thirds of salt samples consumed by Basra city households were adequately iodized and we therefore recommend developing and implementing an on-going monitoring system for quality assurance and quality control for importation, production, dis- tribution and marketing of iodized salt. Salt from local sources was significantly less likely to be iodized and therefore efforts should be focused on increasing and ensuring the production of only io- dized salt by factories inside Iraq. Only 27.1% of household members knew the benefits of iodized salt and those with better education were significantly more likely to have iodized salt in the household, suggesting that more efforts are needed to raise awareness in the community about the importance of IDD and its effective prevention with iodized salt. Salt producers could be required to place information about the health benefits of iodized salt on salt packages. Iraq should be committed to assessment and reassessment of the programme for the elimination of IDD. References 1. Delange F. The disorders induced by iodine deficiency. Thy- roid, 1994, 4:107–128. 2. The miracle of iodated salt: Ethiopia’s commitment of universal salt iodations. Addis Ababa, Ministry of Health/United Nations Children’s Fund, 1995 (US1 Circular No.9). 3. Iodine status worldwide: WHO global database on iodine defi- ciency. Geneva, World Health Organization, 2004. 4. State of the world’s children 2006. New York, United Nations Children’s Fund, 2006. 5. Elimination of iodine deficiency disorders. A manual for health workers. Geneva, World Health Organization, 2008. 6. Recommended iodine levels in salt and guidelines for monitor- ing their adequacy and effectiveness. Geneva, World Health Organization/ United Nations Children’s Fund/ International Council for the Control of Iodine Deficiency Disorders, 1996 (WHO/NUT/96.13). 7. Guiding for a national program for the control of iodine defi- ciency disorders in the Eastern Mediterranean. Alexandria, World Health Organization Regional Office for the Eastern Mediter- ranean, 1988 (EMRO Technical Publication No. 12). 8. Azizi F. IDD in the Middle East. IDD Newsletter, 2001, 17(3) (http://www.iccidd.org/media/IDD%20Newslet- ter/1991-2006/aug2001.html, accessed 12 August 2012). 9. Multiple indicator cluster survey for the year 2000 (detailed re- port). Baghdad, United Nations Children’s Fund, 2001 (http:// www.childinfo.org/files/iraq1.pdf, accessed 11 July 2012). EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 984 10. Comprehensive food security and vulnerability analysis in Iraq. Baghdad, United Nations World Food Programme, 2008 (http://www.uniraq.org/documents/Iraq_CFSVA_Flyer_ EN.pdf, accessed 11 July 2012). 11. An assessment of the household use and adequacy of iodized salt in the republic of Kazakhstan. Geneva, United Nations Chil- dren’s Fund, 2005. 12. Izzeldin SH. Latest status of iodine nutrition. World Health Organization Regional Office for the Eastern Mediterranean, Permanent Advisory Committee in Nutrition, 2010. 13. Haar F et al. Renewed stimulus for USI in Kyrgyzstan. IDD News- letter, 2010, 35(1):10–12. 14. Azizi F. Current status of iodine nutrition in Iraq. IDD Newsletter, 2010, 36(2):1–3. 15. Diosady LL et al. Stability of iodine in iodized salt used for cor- rection of iodine-deficiency disorders, II. Food and Nutrition Bulletin, 1998, 19:240–250. 16. Yuqin Y et al. Long-term stability of iodized salt: the multiple province study in China. IDD Newsletter, 2002, 18(1) (http:// www.iccidd.org/media/IDD%20Newsletter/1991-2006/ feb2002.htm, accessed 12 August 2012). 17. Takele L, Belachew T, Bekele T. Iodine concentration in salt at household and retail shop levels in Shebe town, south west Ethiopia. East African Medical Journal, 2003, 80:532–539. 18. Agarwal S et al. Consumption of iodized salt among slum households of north-east Delhi, India. Indian Journal of Com- munity Medicine, 2009, 34(4):368–369. 19. Gamze C et al. The role of local mass media in promoting the consumption of iodized table salt. Health Education Research, 2001, 16:603–607. Regional strategy on nutrition 2010–2019 and plan of action The Regional strategy on nutrition 2010–2019 and plan of action proposes approaches to address the major health and nutrition problems in the Region. It aims to support countries in establishing and implementing action in nutrition in accordance with their national situation and resources. The overall goal of the strategy is to improve the nutritional status of people throughout the life-cycle by encouraging countries to reposition nutrition as central to their development agenda. It provides a framework to assist countries to decide which nutrition actions are appropriate for a particular context and according to the most prevalent health problems. The strategic approaches target undernutrition, micronutrient deficiencies, prevention and control of obesity and noncommunicable diseases. The strategy emphasizes support for the most vulnerable groups and alleviation of poverty and hunger, the root causes of malnutrition. This document is available online at: http://applications.emro.who.int/dsaf/dsa1230.pdf طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 985 Oral and pharyngeal cancers in Yemen: a retrospective study E.S. Halboub,1,2 M.Abdulhuq 1 and A. Al-Mandili 3 ABSTRACT Hospital-based studies have revealed very high relative frequencies of oral and pharyngeal cancers in Yemen. This study estimated the relative frequencies of oral and pharyngeal cancers among Yemeni cancer patients registered in 2007 and 2008 and determined patients’ demographic and tumour characteristics. Of the registered 7515 cases, 302 (4.0%) were oral cancer and 239 (3.2%) pharyngeal cancer. Oral cancer was significantly more frequent among females while pharyngeal cancer was significantly more frequent among males. Oral cancer patients were significantly older than pharyngeal cancer patients. The tongue was the most affected oral site (53.6%) while the nasopharynx comprised 89.5% of pharyngeal cancers. The most frequent morphological type was squamous cell carcinoma (93.2%). High proportions of oral cancer (71.5%) and pharyngeal cancer (77.4%) patients were diagnosed at advanced stages. Compared with other countries in the region, oral cancer and nasopharyngeal cancer represent substantial national health burdens in Yemen. 1Department of Oral Medicine; 3Department of Oral Pathology, Faculty of Dentistry, University of Damascus, Damascus, Syrian Arab Republic (Correspondence to E.S. Halboub: helboub@yahoo.com). 2Department of Oral Medicine and Periodontology, Faculty of Dentistry, University of Sana’a, Sana’a, Yemen. Received: 24/05/11; accepted: 26/07/11 Cancers de la cavité buccale et du pharynx au Yémen : étude rétrospective RÉSUMÉ Des études en milieu hospitalier ont révélé des fréquences relatives très élevées du cancer oropharyngé au Yémen. La présente étude a estimé les fréquences relatives du cancer de la cavité buccale et du pharynx chez des patients yéménites atteints de cancers enregistrés en 2007 et 2008. Leurs données démographiques et les caractéristiques de leur tumeur ont été relevées. Sur 7515 cas enregistrés, 302 concernaient des cancers de la cavité buccale (4,0 %) et 239 des cancers du pha rynx (3,2 %). Les cancers de la cavité buccale étaient nettement plus fréquents chez les femmes alors que les cancers du pharynx étaient beaucoup plus nombreux chez les hommes. Les patients atteints d'un cancer de la cavité buccale étaient bien plus âgés que les patients souffrant d'un cancer du pharynx. La langue était le site le plus souvent touché de la cavité buccale (53,6 %) alors que le nasopharynx était impliqué dans 89,5 % des cancers du pharynx. Le type morphologique le plus fréquent était l'épithéliome malpighien (93,2 %). Des proportions élevées de patients atteints d'un cancer de la cavité buccale (71,5 %) ou du pharynx (77,4 %) ont reçu un diagnostic à un stade avancé de la maladie. Par rapport à d'autres pays de la Région, le cancer de la cavité buccale et du nasopharynx représente une charge sanitaire importante pour le Yémen au niveau national. ةيداعتسا ةسارد :نميلا في موعلبلاو مفلا اناطسر لييدنلما دحمأ ،قلحا دبع دوممح ،بوبلح حلاص ماصع تارتاوتلا ةساردلا هذه ردقتو .نميلا في موعلبلاو مفلا نياطسرل ًايبسن ًادج ةعفترم تارتاوت تايفشتسلما في ةزكترلما تاساردلا ترهظأ :ةـصلالخا .مرولا تماسو ضىرملل ةيفارغميدلا تماسلا ددتحو ،2008و 2007 يماع في ينلجسلما نميلا في ناطسرلا ضىرم ينب موعلبلاو مفلا نياطسرل ةيبسنلا ًارُتاوت رثكأ مفلا ناطسر ناك دقو .موعلبلا ناطسرل )%3.2( ةلاح 239و ،مفلا ناطسرل )%4.0( 302 اهنم ،ةلاح 7515 ةلجسلما تلاالحا ددع غلب دقو ناطسر ضىرم نم هب ُّدَتْعُي ردقب ًارمع بركأ مفلا ناطسر ضىرم ناكو .روكذلا ينب )ًاراركت( ًارُتاوت رثكأ موعلبلا ناطسر ناك ماك ،ءاسنلا ينب )ًاراركت( رثكأ ناكو .موعلبلا تاناطسر نم %89.5 يفنلأا موعلبلا ناطسر لكش مانيب ،)%53.6( مفلا ناطسر في ًةباصإ قطانلما رثكأ ناسللا ناكو .موعلبلا تاناطسر نمو )%71.5( مفلا تاناطسر نم ةعفترم ةيوئم ةبسن صيخشت مت دقو .)%93.2( ةكئاشلا ايلالخا ةناطسر ًاعويش ةيجولوفرولما طمانلأا موعلبلاو مفلا نياطسر نأ نوثحابلا دجو ،ميلقلإا في ىرخلأا نادلبلا ىدل ام عم جئاتنلا ةنراقمبو .ةمدقتم لحارم في ضىرلما ىدل )%77.4( موعلبلا .نميلا في ةينطولا ةحصلا لىع ًلايقث ًائبع نلاثمي يفنلأا EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 986 Introduction Oral and pharyngeal cancers were es- timated to affect more than 482 000 people globally in 2008, more than half of whom (273 000) died of the disease [1]. Unfortunately, two-thirds of these cases occurred in developing countries [2]. The prevalence of oral cancer and pharyngeal cancer shows a wide dispar- ity by sex [1,3] and geographical area [3,4]. While nasopharyngeal cancer is prevalent among the Chinese [5], oral cancer has been known to be preva- lent in South Asian countries such as India [6] and Pakistan [7]. Worldwide variations in oral cancer distribution are believed to reflect different preva- lences and patterns of risk habits such as khat chewing and waterpipe smoking [8–11]. Hospital-based studies have also re- vealed very high relative frequencies of oral cancer [12] and head and neck can- cer [13,14] in Yemen, where such risk habits are also common. As yet there has been no national population-based cancer report issued by the National Oncology Center (NOC) though many years have elapsed since its es- tablishment in 2002. The objectives of this study therefore were to determine the relative frequency of oral cancer and pharyngeal cancer among Yemeni cancer patients who were registered at NOC in 2007 and 2008 and to analyse these patients by demographic and tu- mour characteristics. Methods This descriptive, epidemiological study was carried out on retrospective data provided by the NOC, Sana’a, Yemen. For all patients registered in 2007 and 2008, data regarding all patients with head and neck cancers was provided. The data included demographic details (sex, age and address) and tumour details (primary site, histology, grade and stage). The first author (H.E.) had access to the patients’ hard files to validate the provided data. When there were discrepancies between data sources, the data in the hard files were used. The primary site (topogra- phy) and histology (morphology) of the neoplasms were reclassified and coded according to the International classification of diseases for oncology, 3rd edition (ICD-O-3) [15]. Hence only patients with malignant oral cancer or pharyngeal cancer (code C00-C14) were included in the present analysis. Tumours with other codes (i.e. not epi- thelial tumours) were excluded, even those affecting the oral cavity, pharynx or both sites. For the included cases, tumour staging was reclassified and coded according to the SEER summary staging manual 2000 [16]. The main morphology of the malignant tumours were reclassified further as: squamous cell carcinoma; adenoid cystic carci- noma; and papillary carcinoma. The less frequent morphological types were reclassified as other. The statistical description and analyses were carried out using SPSS software, version 19. The descriptive statistics were presented as mean and standard deviation (SD) and relative frequencies as appropriate. The differ- ences in age, were analysed by inde- pendent t-tests, while the associations of the cancer distribution (oral cancer versus pharyngeal cancer) with the dif- ferent grouping factors were analysed using chi-squared tests. P < 0.05 was taken as the level of significance. Results Of the 7515 patients registered with neoplasms in 2007 and 2008, 1004 (13.4%) had head and neck cancers. Of these 541 were primary malignancies of the oral and pharyngeal tissues (code C00-C14); 302 (4.0%) were oral can- cer and 239 (3.2%) pharyngeal cancer. Patients with tumours of the mesothe- lial and soft tissues (code C45-C49), lymphoid, haematopoietic and related tissues (C81-C96) and ill-defined sec- ondary and unspecified sites (C76- C80) were erroneously registered as oral cancer or pharyngeal cancer and were excluded (n = 110). Distribution by site The tongue was the most affected oral subsite (162/302, 53.6%), followed by the floor of the mouth (55/302, 18.2%). Nasopharyngeal cancer constituted the major proportion of pharyngeal cancers (214/239, 89.5%) (Table 1). Distribution by region Of the 541 cases of oropharyngeal cancer 19.2%, 11.8%, 8.7% and 7.8% were referred from Al-Hodeida, Taiz, Hajja and Al-Amana provinces respec- tively (Table 2). The distribution of oral cancer and pharyngeal cancer was statistically different across different provinces (P < 0.001). Oral cancer was more frequent in Al-Hodeida, Hajja and Al-Mahweet provinces while pharyn- geal cancer was more frequent in almost all other provinces. Distribution by age and sex The overall mean (SD) age of oro- pharyngeal cancer cases was 52.8 (SD 17.0) years. Although the difference in age between males and females within each cancer type was not significant, oral cancer patients (males, females or both) were significantly older than their pharyngeal cancer peers (P < 0.001) (Table 3). More males were affected by all oropharyngeal cancers (297, 54.9%) than were females (244, 45.1%) (Table 1). The male to female ratios for oral cancer and pharyngeal cancer were 1.0 (152:150) and 1.5 (145:94) respec- tively. There was also a difference in the distribution of oral and pharyngeal cancers by sex. Women were more sus- ceptible to oral cancer and males to pharyngeal cancer (P = 0.016) (Table 4). Neither the subsite distribution of oral cancer nor of pharyngeal cancer طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 987 was significantly associated with pa- tient’s sex (Table 1). One-quarter of oropharyngeal can- cer patients (135, 25.0%) were aged ≤ 40 years. Significantly more of this younger group had pharyngeal cancer than oral cancer, whereas among the older age group pharyngeal cancer was more common (P < 0.001) (Table 4). Distribution by grade and stage It is noteworthy that all cases were his- tologically verified. The most frequent morphological type of oropharyngeal cancer was squamous cell carcinoma (93.2%) and this was true for both oral and pharyngeal cases. The distribution by histological grade was significantly different; more of the well- and mod- erately-differentiated cancers were in oral cancer (258/302, 85.4%) com- pared with pharyngeal cancer (32/239 13.4%) (P < 0.001) (Table 4). A high proportion of both oral (216, 71.5%) and pharyngeal (285, 77.4%) cancer patients were diagnosed at ad- vanced (regional/metastasized) stages; however, the proportion of pharyngeal cancer patients at an advanced stage was significantly higher (Table 4) (P < 0.001). Discussion The cancer registry is an essential part of any rational programme of cancer control, benefiting both individuals and the society in which they live. Establish- ing the NOC was therefore a big step towards improving cancer control in Yemen. Nevertheless, providing chem- otherapy and radiotherapy to patients has been the main priority of the health service and the work of the registry cen- tre has been neglected. Hence, many patients with early stage cancer or who had chemotherapy elsewhere might not have been recorded. Addition- ally, many cancer patients in Yemen still prefer to go abroad for treatment and are consequently not recorded. Therefore it would have been unreli- able to attempt to calculate the cancer incidence in this study as it would have been an underestimate. The relative fre- quency is still a valid measure, however. The comparison of relative frequencies of a given cancer between populations is at best an indirect measure of the absolute risk difference and will be more reliable when the cancer site of interest accounts for a small proportion of all cancer cases [17] as it is the case in the present study. Another limitation of the present study was the quality of the provided data. No data on patients’ habits were available. Moreover, there was no stand- ard coding and classification of tumour topography, morphology and extent of disease at the time of diagnosis. The authors, however, made efforts to com- pare the provided data with what was re- corded in the patient’s files and recoded them according to the international standards. In addition, all the registered cases were histologically verified which makes the present data reasonably valid and reliable. The present analysis contradicts hospital-based studies which reported higher relative frequencies of oral can- cer in the central and northern regions of Yemen: 18% [12] and 8.7% [14]. However, the figure in this study slightly exceeds what was previously reported in the southern region of Yemen: 2.9% [18]. Such a regional disparity within an individual country was reported previously in Yemen [19] and Saudi Table 1 Subsite distribution of 541 cases of oral cancer and pharyngeal cancer by sex Cancer type Total (n = 541) Males (n = 297) Females (n = 244) No. % No. % No. % Oral cancer Lip 36 6.7 19 6.4 17 7.0 Tongue 162 29.9 88 29.6 74 30.3 Gum 37 6.8 15 5.1 22 9.0 Floor of mouth 55 10.2 27 9.1 28 11.5 Other & unspecified parts of the mouth 3 0.6 0 0.0 3 1.2 Parotid gland 3 0.6 0 0.0 3 1.2 Other & unspecified major salivary glands 6 1.1 3 1.0 3 1.2 Subtotal 302 55.9 152 51.2 150 61.4 Pharyngeal cancer Tonsil 8 1.5 4 1.3 4 1.6 Nasopharynx 214 39.6 134 45.1 80 32.8 Hypopharynx 17 3.1 7 2.4 10 4.1 Subtotal 239 44.2 145 48.8 94 38.5 EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 988 Arabia [20]. In support of GLOBO- CAN estimations [21] and other re- ports [3,22–24], the relative frequency of oral cancer in Yemen was higher compared with many other developed countries and indeed other Arab coun- tries. However, it was still much lower compared with data from Myanmar [25], Pakistan [7] and India [26]. The widespread habit of betel quid chewing within Asian populations, in addition to smoking tobacco, alcohol consump- tion and poor oral hygiene, have been associated with high incidences of oral cancer [8,9]. Chewing khat, smokeless tobacco use (locally known as shamma) and waterpipe smoking (locally known as mada’a) are widely prevalent in Yemen and have been linked with oral cancer [10,11]. Further research on this is needed in Yemen. Unexpectedly, the relative fre- quency of pharyngeal cancer was higher compared with other reports from Yemen [12–14,18] and other countries [7,23–29]. Nasopharyngeal cancer, which accounted for more than 89% of the pharyngeal cancers in the present study, is relatively common among Iraqi [21] and western North African men [30] and among the Chinese [5]. It is not considered to be a tobacco-related cancer [27]. However, genetic and in- fectious etiologies have been suggested [31]. In agreement with reports from Yemen [12,18,19], Saudi Arabia [23], Kuwait [27], Myanmar [25] and many other developed and developing coun- tries [3], the tongue was the most af- fected oral subsite in men and women. Differences between regions have been reported however. In south central Asia, the buccal mucosa was the most affected site and this is consistent with being the site where betel quid is held when chewing [3,26,32]. In Iraq, the lip is reported to be the most affected oral site [33,34], while the gingiva is the pre- dominant oral site in Zimbabwe [35]. Although it is not the main site in direct contact with khat, shamma or both, the tongue is considered to be a high-risk site for oral cancer [36]. To the best of our knowledge, this was the first study which reported the distribution of oral cancer and pharyn- geal cancer among different provinces in Yemen. Relative to pharyngeal can- cer, oral cancer was significantly more common in Al-Hodeida, Hajja and Al-Mahweet provinces. Interestingly these areas are those where shamma Table 2 Distribution of 541 cases of oral cancer and pharyngeal cancer by province Province Total (n = 541) Oral cancer (n = 302) Pharyngeal cancer (n = 239) No. % No. % No. % Al-Hodaida 104 19.2 76 25.2 28 11.7 Taiz 64 11.8 30 9.9 34 14.2 Hajja 47 8.7 36 11.9 11 4.6 Ibb 32 5.9 16 5.3 16 6.7 Al-Amana 42 7.8 12 4.0 30 12.6 Sana’a 36 6.7 14 4.6 22 9.2 Dhamar 35 6.5 20 6.6 15 6.3 Amran 31 5.7 12 4.0 19 7.9 Al-Mahweet 28 5.2 20 6.6 8 3.3 Aden 19 3.5 12 4.0 7 2.9 Other provinces 103 19.0 54 17.9 49 20.5 Total 541 100.0 302 25.2 239 11.7 Chi-squared test, P < 0.001. Table 3 Distribution of 541 cases of oral cancer and pharyngeal cancer by mean age and sex Sex Oral cancer Pharyngeal cancer P-value No. of cases Mean (SD) age (years) No. of cases Mean (SD) age (years) Male 152 59.5 (15.5) 145 45.9 (14.9) < 0.001 Female 150 57.3 (12.9) 94 45.3 (17.2) < 0.001 Total 302 58.4 (14.3) 239 47.7 (17.6) < 0.001 Independent t-test. SD = standard deviation. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 989 use and waterpipe smoking are known to be widely prevalent. As no data on substance use habits were available, fu- ture research could investigate whether these habits are associated more with oral cancer than pharyngeal cancer. Both sexes in this study were found to be affected equally by oral cancer. Previous reports from Yemen [13,14,18] and other countries [3,7,23–26,34] found a higher relative frequency of oral cancer among men. However, Sawair et al. reported a higher relative frequency of oral cancer among women in Yemen [12]. Cultural norms allow more men to be exposed to risk factors for oral cancer such as smok- ing, khat and shamma compared with women and this is an area for future research. In agreement with studies in Yemen [12–14,18] and other countries [7,23–26,29], men more commonly were affected by pharyngeal cancer than were women. It is difficult at present to suggest reasons to explain the higher relative frequency of pharyngeal cancer (especially the nasopharyngeal cancer) among Yemeni men. However, face covers commonly worn by women in Yemen might protect them from infec- tion by Epstein-Barr virus [31], which is considered as an etiological factor for nasopharyngeal cancer. The mean age of oral cancer patients was comparable to those reported by previous studies in Yemen [12,19] and Iraq [33]. However it was lower than reports from Saudi Arabia [37] and most developed countries [38], but higher than reported in Pakistan [32], Myanmar [25] and Zimbabwe [35]. This disparity could be reflected partly by the onset and the intensity of practising risk habits in different regions worldwide. On average, the khat chew- ing habit starts in the early 20s and the frequency of practising it ranges from once a week to daily. Shamma use and waterpipe smoking, on the other hand, start at earlier ages and may be practised more than once a day. Patients were affected by pharyn- geal cancer at a significantly younger age compared with oral cancer. Moreo- ver, more than 37% of patients with pharyngeal cancer were aged 40 years or less. The occurrence of pharyngeal cancer in younger age populations was reported too in Saudi Arabia [31] and Kuwait [27]. This might mean that the pharyngeal tissues are more susceptible or more exposed to risk factors or both. In agreement with other studies in Pakistan [32], Myanmar [25] and Zim- babwe [35], squamous cell carcinoma Table 4 Demographic and clinical characteristics of 541 cases of oral cancer and pharyngeal cancer Variable Total (n = 541) Oral cancer (n = 302) Pharyngeal cancer (n = 239) P-valuea No. % No. % No. % Age group (years) < 0.001≤ 40 135 25.0 46 15.2 89 37.2 > 40 406 75.0 256 84.8 150 62.8 Sex Male 297 54.9 152 50.3 145 60.7 0.016 Female 244 45.1 150 49.7 94 39.3 Main morphology Squamous cell carcinoma 504 93.2 281 93.0 223 93.3 > 0.05 Adenoid cystic carcinoma 9 1.7 6 2.0 3 1.3 Papillary carcinoma 12 2.2 8 2.6 4 1.7 Other 16 3.0 7 2.3 9 3.8 Grade Well differentiated 224 41.4 211 69.9 13 5.4 < 0.001 Moderately differentiated 66 12.2 47 15.6 19 7.9 Poorly differentiated 29 5.4 17 5.6 12 5.0 Undifferentiated 114 21.1 7 2.3 107 44.8 Unspecified 108 20.0 20 6.6 88 36.8 Tumour extension (stage) Localized 34 6.3 30 9.9 4 1.7 < 0.001 Regional 334 61.7 193 63.9 141 59.0 Metastasized 67 12.4 23 7.6 44 18.4 Unspecified 106 19.6 56 18.5 50 20.9 aChi-squared test. EMHJ  •  Vol. 18  No. 9  •  2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale 990 References 1. Ferlay J et al. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. International Journal of Cancer, 2010, 127:2893–2917. 2. Warnakulasuriya S. Global epidemiology of oral and oro- pharyngeal cancer. Oral Oncology, 2009, 45:309–316. 3. de Camargo Cancela M et al. Oral cavity cancer in developed and in developing countries: population-based incidence. Head and Neck, 2010, 32:357–367. 4. GLOBOCAN 2002: Cancer incidence, mortality and prevalence worldwide. IARC CancerBase No. 5, version 2.0. Lyon, Interna- tional Agency for Research on Cancer, 2004. 5. Warnakulasuriya KA et al. Cancer of mouth, pharynx and naso- pharynx in Asian and Chinese immigrants resident in Thames regions. Oral Oncology, 1999, 35:471–475. 6. Sherin N et al. Changing trends in oral cancer. Indian Journal of Cancer, 2008, 45:93–96. 7. Bhurgri Y et al. Cancer incidence in Karachi, Pakistan: first results from Karachi Cancer Registry. International Journal of Cancer, 2000, 85:325–329. 8. Nair U, Bartsch H, Nair J. Alert for an epidemic of oral cancer due to use of the betel quid substitutes gutkha and pan masala: a review of agents and causative mechanisms. Mutagenesis, 2004, 19:251–262. 9. Rosenquist K. Risk factors in oral and oropharyngeal squamous cell carcinoma: a population-based case-control study in southern Sweden. Swedish Dental Journal. Supplement, 2005, 179:1–66. 10. Nasr AH, Khatri ML. Head and neck squamous cell carcinoma in Hajjah, Yemen. Saudi Medical Journal, 2000, 21:565–568. 11. Scheifele C, Nassar A, Reichart PA. Prevalence of oral cancer and potentially malignant lesions among shammah users in Yemen. Oral Oncology, 2007, 43:42–50. 12. Sawair FA et al. High relative frequency of oral squamous cell carcinoma in Yemen: qat and tobacco chewing as its aetiologi- cal background. International Journal of Environmental Health Research, 2007, 17:185–195. 13. Abdul-Hamid G et al. Pattern of head and neck cancer in Yemen. Gulf Journal of Oncology, 2010, 7:21–24. 14. Al-Thobhani AK, Raja’a YA, Noman TA. The pattern and distri- bution of malignant neoplasms among Yemeni patients. Saudi Medical Journal, 2001, 22:910–913. 15. International classification of diseases for oncology, 3rd ed. Ge- neva, World Health Organization, 2000. 16. Young JL et al., eds. SEER summary staging manual— 2000: codes and coding instructions. Bethesda, Maryland, National Cancer Institute, 2001 (NIH Pub. No. 01-4969). 17. Esteve J et al. Statistical methods in cancer research. Vol. IV: de- scriptive epidemiology. Lyon, International Agency for Research on Cancer, 1994. 18. Ba Saleem HO et al. Five years cancer incidence in Aden Cancer Registry, Yemen (2002–2006). Asian Pacific Journal of Cancer Prevention, 2010, 11:507–511. 19. Halboub E et al. A case series analysis of 248 oral cancers from Yemen. Oral Diseases, 2010, 16:536. 20. Brown A, Ravichandran K, Warnakulasuriya S. The unequal burden related to the risk of oral cancer in the different regions of the Kingdom of Saudi Arabia. Community Dental Health, 2006, 23:101–106. 21. GLOBECAN 2008. International Agency for Research on Can- cer [online database] (http://www-dep.iarc.fr/, accessed 11 July 2012). 22. Rawashdeh MA, Matalka I. Malignant oral tumors in Jor- danians, 1991–2001. A descriptive epidemiological study. International Journal of Oral and Maxillofacial Surgery, 2004, 33:183–188. 23. Registry SC. Cancer incidence report, Saudi Arabia 2006. Riyadh, Ministry of Health and Saudi Cancer Registry, 2006. 24. Jemal A et al. Cancer statistics, 2010. CA: a Cancer Journal for Clinicians, 2010, 60:277–300. 25. Oo HN et al. Oral cancer in Myanmar: a preliminary survey based on hospital-based cancer registries. Journal of Oral Pa- thology and Medicine, 2011, 40:20–26. was the most common morphologic type diagnosed in oral and pharyngeal cancers. Moreover, most oral cancer cases were graded well to moderate dif- ferentiation whereas most pharyngeal cancer cases were graded poor to undif- ferentiated. In addition to being more accessible to diagnosis, the previously mentioned findings might explain the lower proportion of oral cancer with advanced stage (regional and distant metastasis) compared with pharyngeal cancer. Unfortunately, only 9.9% and 1.7% of oral cancer and pharyngeal can- cer cases respectively were diagnosed at the localized stage. These propor- tions might be underestimated; tumour stage was unknown in 18.5% and 20.9% of oral cancer and pharyngeal cancer cases respectively. Furthermore, many governmental and nongovernmental health facilities might not refer patients with early stage cancers to the NOC, leading to under-recording. However, patients’ lack of awareness and health knowledge along with poor resources in primary and secondary health care are possible reasons that should not be overlooked. A high relative frequency of pharyn- geal cancer (especially nasopharyngeal) among Yemenis is noted here for the first time. Overall, nasopharyngeal can- cer and oral cancer in Yemen represent national health burdens. Large-scale, matched case–control studies are needed to determine the risk factors for these cancers. The presence of a high quality national population-based can- cer registry is the cornerstone to assess the burden of each site-specific cancer and to plan and direct the available health care resources and programmes accordingly. Acknowledgements Our special thanks go to all staff mem- bers of the National Oncology Center, Sana’a, Yemen for providing the raw data and for their efforts in facilitating the task of matching these data with the patient’s hard files. طسوتلما قشرل ةيحصلا ةلجلماشرع نماثلا دلجلما عساتلا ددعلا 991 26. Sen U et al. Cancer patterns in eastern India: the first report of the Kolkata cancer registry. International Journal of Cancer, 2002, 100:86–91. 27. Morris RE et al. The epidemiology of lip, oral cavity and phar- yngeal cancers in Kuwait 1979–1988. British Journal of Oral and Maxillofacial Surgery, 2000, 38:316–319. 28. Doobaree IU et al. Head and neck cancer in South East Eng- land between 1995–1999 and 2000–2004: an estimation of incidence and distribution by site, stage and histological type. Oral Oncology, 2009, 45:809–814. 29. Attar E et al. Head and neck cancer in a developing country: a population-based perspective across 8 years. Oral Oncology, 2010, 46:591–596. 30. Salim EI et al. Cancer epidemiology and control in the Arab world – past, present and future. Asian Pacific Journal of Cancer Prevention, 2009, 10:3–16. 31. Andejani AA, Kundapur V, Malaker K. Age distribution of na- sopharyngeal cancer in Saudi Arabia. Saudi Medical Journal, 2004, 25:1579–1582. 32. Bhurgri Y. Cancer of the oral cavity—trends in Karachi South (1995–2002). Asian Pacific Journal of Cancer Prevention, 2005, 6:22–26. 33. Al-Rawi NH, Talabani NG. Squamous cell carcinoma of the oral cavity: a case series analysis of clinical presentation and histological grading of 1,425 cases from Iraq. Clinical Oral Inves- tigations, 2008, 12:15–18. 34. Al-Fouadi A, Parkin DM. Cancer in Iraq: seven years’ data from the Baghdad Tumour Registry. International Journal of Cancer, 1984, 34:207–213. 35. Chidzonga MM, Mahomva L. Squamous cell carcinoma of the oral cavity, maxillary antrum and lip in a Zimbabwean popu- lation: a descriptive epidemiological study. Oral Oncology, 2006, 42:184–189. 36. Mashberg A, Samit A. Early diagnosis of asymptomatic oral and oropharyngeal squamous cancers. CA: a Cancer Journal for Clinicians, 1995, 45:328–351. 37. Allard WF, DeVol EB, Te OB. Smokeless tobacco (shamma) and oral cancer in Saudi Arabia. Community Dentistry and Oral Epidemiology, 1999, 27:398–405. 38. Quaglia A et al. A wide difference in cancer survival between middle aged and elderly patients in Europe. International Jour- nal of Cancer, 2007, 120:2196–2201. Cancer survival in Africa, Asia, the Caribbean and Central America Population-based cancer survival rates offer an important benchmark for measuring a health care system s overall effectiveness in the fight against cancer. Cancer survival in Africa, Asia, the Caribbean and Central America presents in-depth cancer survival data from 27 population-based cancer registries in 14 low- and middle-resource countries. The differences in cancer survival reported in populations observed between and within countries studied in this volume provide valuable insights for future planning and investment by governments in primary prevention activities, early detection initiatives and tertiary care to achieve meaningful cancer control. Further information about this and other WHO publications is available at: http://www.who.int/publications/en/ Subscriptions and Distribution Enquiries regarding subscriptions and distribution of the print edition of EMHJ should be addressed to: Printing and Marketing of Publications at: email: pam@emro.who.int; tel: (+202) 2276 5000; fax: (+202) 2670 2492 or 2670 2494 Permissions Requests for permission to reproduce or translate articles, whether for sale or non-commercial distribution should be addressed to EMHJ at: emhj@emro.who.int Correspondence Editor-in-chief EMHJ 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: khayat@emro.who.int/emhj@emro.who.int 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 health services; and for the exchange of ideas, con‑ cepts, 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 Col‑ laborating Centres and individuals within and outside the Region. 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 des ser‑vices de santé ainsi qu’à l’échange d’idées, de concepts, de données épidémiologiques, de résultats de recherches et d’autres informations, 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 collabora‑ teurs de l’OMS et personnes concernés au sein et hors de la Région. 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 is abstracted/indexed in the Index Medicus and MEDLINE (Medical Literature Analysis and Retrieval Systems on Line) and the ExtraMed‑Full text on CD‑ROM, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), CAB International, Lexis Nexis, Scopus and the Index Medicus for the WHO Eastern Mediterranean Region (IMEMR). ©World Health Organization 2012 All rights reserved 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 the World Health Organization 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 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 the World Health Organization in preference to others of a similar nature that are not mentioned. All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either express or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use. The named authors alone are responsible for the views expressed in this publication. ISSN 1020‑3397 Cover designed by Diana Tawadros Internal layout designed by Emad Marji and Diana Tawadros Printed by WHO Regional Office for the Eastern Mediterranean ميدقتل برنم ىهو .ةيلماعلا ةحصلا ةمظنمب طسوتلما قشرل ىميلقلإا بتكلما نع ردصت ىتلا ةيمسرلا ةلجلما ىه ةيئابولا تايطعلماو ميهافلماو ءارلآا لدابتلو ،اله جيوترلاو ةيحصلا تامدلخا فى ةديدلجا تاردابلماو تاسايسلا لك لىإ ةهجوم ىهو .طسوتلما قشر ميلقإب اهنم قلعتي ام ةصاخو ،تامولعلما نم كلذ يرغو ثاحبلأا جئاتنو زكارلماو ،ةينعلما ةيموكلحا يرغ تماظنلما اذكو ،ةيميلعتلا دهاعلما رئاسو ةيبطلا تايلكلاو ،ةيحصلا نهلما ءاضعأ .هجراخو ميلقلإا فى ةحصلاب ينمتهلما دارفلأاو ةيلماعلا ةحصلا ةمظنم عم ةنواعتلما طسوتلما قشرل ةيحصلا ةلجلما طسوتلما قشرل ةيلماعلا ةحصلا ةمظنلم ةيميلقلإا ةنجللا ءاضعأ نادلبلا ةيملاسلإا ناريإ ةيروهجم . ايبيل . سنوت . نيرحبلا . ناتسكاب . ةدحتلما ةيبرعلا تاراملإا . ناتسناغفأ . ندرلأا صرم . نانبل . تيوكلا . رطق . ينطسلف . نماُع . قارعلا . لاموصلا . نادوسلا . تيوبيج . نادوسلا بونج . نميلا . ةيروسلا ةيبرعلا ةيروهملجا ةيدوعسلا ةيبرعلا ةكلملما . برغلما 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 . South Sudan Sudan . Syrian Arab Republic . Tunisia . United Arab Emirates . Yemen 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 . Soudan du Sud . Tunisie . Yémen Cover 18-6.indd 2 6/19/2012 8:50:23 AM Contents V olum e 18 N um ber 9 Septem ber 2012 Eastern Mediterranean Health Journal La Revue de Santé de la Méditerranée orientale Volume 18 / No. 9 September / Septembre 2012 9 ددع / شرع نماثلا دلجلما برمتبس / لوليأ Women across the generations, Morocco Globally 17.3 million people a year die of cardiovascular disease, almost 50% of whom are women. Children may also at increased risk of cardiovascular disease later in life, since in childhood they may be unable to control where and how to live a healthy heart life. Therefore, the focus of World Heart Day 2012 on 28 September is cardiovascular disease prevention among women and children. Letter from the Editor .............................................................................................................................................................. 901 Research articles In-hospital complications and 1-year outcome of acute coronary syndrome in patients with hypertension: findings from the 2nd Gulf Registry of Acute Cardiac Events ................................................................................. 902 Mapping family planning policy and programme best practices in the WHO Eastern Mediterranean Region: a step towards coordinated scale-up ..........................................................................................................911 HIV infection and related risk behaviours among female sex workers in greater Cairo, Egypt ............................ 920 Tehran dentists’ self-reported knowledge and attitudes towards HIV/AIDS and observed willingness to treat simulated HIV-positive patients .............................................................................................. 928 Knowledge, perceptions and practices towards medical ethics among physician residents of University of Alexandria hospitals, Egypt .................................................................................................................935 Study of adverse events of A/H1N1 vaccine among health care staff in selected provinces of Afghanistan, 2010 ................................................................................................................................................. 946 Estimation of malaria transmission intensity in Sennar state, central Sudan .......................................................... 951 Drug resistance pattern and outcome of treatment in recurrent episodes of tuberculosis ..................................957 Updating and validation of the socioeconomic status scale for health research in Egypt ................................... 962 Natural killer cell populations in Egyptians infected with hepatitis C virus ........................................................... 969 Lead exposure among the general population of Duhok governorate, Kurdistan region, Iraq ............................974 Consumption of iodized salt among households of Basra city, south Iraq ...........................................................980 Oral and pharyngeal cancers in Yemen: a retrospective study .............................................................................. 985 Cover 18-9.indd 1 9/6/2012 10:14:47 AM