Lost to follow-up in MDR-TB patients in Armenia in 2013–2018

432 VOLUME 5 | ISSUE 4 | DECEMBER 2019 | 369–613PUBLIC HEALTH PANORAMA ORIGINAL RESEARCH Lost to follow-up in MDR-TB patients in Armenia in 2013–2018 Anush Khachatryan1, Karapet Davtyan2, Ruzanna Grigoryan3, Lusine Kocharyan1, Andrei Dadu2 1 National Tuberculosis Control Center of the Ministry of Health, Yerevan, Armenia 2 WHO Regional Office for Europe, Copenhagen, Denmark 3 TB Research and Prevention Center NGO, Yerevan Armenia Corresponding author: Anush Khachatryan (email: arm_anush@mail.ru) ABSTRACT Introduction: Tuberculosis (TB) and particularly multidrug resistance (MDR)- TB is one of the major threats to public health in Armenia. Only 42% of MDR- TB patients were successfully treated in 2015 – well below the WHO target of 75% – with one third of patients lost to follow-up (LTFU). In order to reduce the number of LTFU MDR-TB patients and ultimately improve efforts to control MDR-TB in the country, the factors leading to treatment interruptions require further investigation. Methods: This retrospective cohort study included all MDR-TB patients registered in Armenia in 2013–2018. The rate of, and factors contributing to, a LTFU outcome were determined, with the impact of each factor evaluated via survival analysis. Results: The study included 652 MDR-TB cases, of which, 234 (35.9%) were successfully treated, 161 (24.7%) were LTFU, 112 (17.0%) were still on treatment, in 82 (12.7%) treatment failed and 62 (9.5%) died. The mean duration of treatment among LTFU patients was 7.7 (SD ±6.4) months, which was about 14 months less than the mean duration of treatment among successfully treated patients at 21.9 (SD ±5.1) months. The risk factors for LTFU included migrant work (OR=1.7, CI=1.1–2.5, P <0.01) as well as having a history of previous treatment (OR=3.6, CI=1.6–8.0, P <0.01). Conclusion: The study identified a small number of risk factors associated with LTFU among MDR-TB patients in Armenia, which could be targeted by interventions to improve the treatment success of these patients and the TB situation in the country overall. Keywords: TUBERCULOSIS, MULTIDRUG-RESISTANT TUBERCULOSIS, LOST TO FOLLOW-UP, SUCCESSFUL TREATMENT, NATIONAL TB PROGRAMME, ARMENIA BACKGROUND Tuberculosis (TB) is one of the most important public health problems globally. According to the World Health Organization (WHO) it was one of the top 10 causes of death in the world and the leading cause of death from a single infectious agent in 2017 (1, 2). More than 80% of active TB cases in the world are concentrated in low and middle-income countries (1, 3). Armenia faced a re-emergence of TB after it had declared its independence from the Soviet Union in 1991 and experienced a drastic socioeconomic crisis (4, 5). The burden of the disease had been persistently high in the country for decades. Even in 2014 the TB incidence rate (including HIV/TB cases) was 45 per 100 000 people in Armenia with the estimated TB mortality rate equal to 5.1 per 100  000 (6, 7). However, the TB epidemiological situation has been improving in the country over the past few years, with the TB incidence rate at 31 per 100 000 population and the TB mortality rate at 1.3 per 100 000 population in 2018 (8, 9), but despite this, TB remains the major threat to public health in Armenia. The successful treatment of drug-sensitive TB cases in Armenia in 2013–2017 was relatively high at 79–81%, compared to the WHO target of 90%. However, the situation for those with multidrug-resistant (MDR)-TB has been more alarming, with only 42–49% successfully treated between 2013–2016, well below the WHO target of 75% (1, 2, 9, 10). A significant barrier in reaching this target in Armenia is the incident of lost to follow-up (LTFU) patients, who are defined as those whose treatment is interrupted for two consecutive months or more. Recent data indicate that about a  third of MDR-TB patients in Armenia are LTFU, an increase from previous years (20% in 2014 and 28% in 2016). One particular factor thought to be behind this high number of LTFU cases is that 433 ТОМ 5 | ВЫПУСК 4 | ДЕКАБРЬ 2019 Г. | 369–613ПАНОРАМА ОБЩЕСТВЕННОГО ЗДРАВООХРАНЕНИЯ LOST TO FOLLOW-UP IN MDR-TB PATIENTS IN ARMENIA IN 2013–2018 many Armenians migrate, primarily on a temporary basis, to other countries to work. According to the latest study from the Ministry of Territorial Development and administration of Armenia, the Russian Federation is the most popular destination for Armenian migrants, with more than 95% of seasonal and 75% of long-term migrants working in the country. Annually, more than 200 000 Armenians go to Russia for seasonal employment (11). This study aims to identify and analyse the risk factors contributing to a  LTFU outcome in MDR-TB patients registered in Armenia in 2013–2018. The specific objectives were to identify: • the proportion of patients LTFU, • the proportion of migrant workers among the LTFU patients, • additional factors associated with a LTFU outcome, • the dropout rate over the course of MDR-TB treatment. The study findings may help to develop more factor and time- specific interventions to prevent treatment interruptions and to improve the MDR-TB treatment success rate. METHODS DESIGN This is a  retrospective cohort study carried out using the electronic national TB database obtained from the Armenian National TB Programme (NTP) as the source of information. GENERAL SETTINGS Armenia has a  total population of around 3 million people. The country is divided into 11 administrative regions with diverse geographical features and both urban and rural residential areas (12, 13). According to the World Bank’s classification is currently an upper middle-income country having been upgraded from lower middle in 2018. About 30% of the population live below the poverty line, despite a literacy rate of 99% (12, 14) and it is this poverty that often drives Armenians to work abroad. The implementation of the NTP (the management, financing, monitoring and evaluation of TB control services) is the responsibility of the National TB Control Centre (NTCC) under the supervision of the Ministry of Health (MOH). Following WHO recommendations, the Armenian NTP assures free TB health-care services, which includes the diagnosis and treatment of drug-sensitive and drug-resistant TB cases, in order to achieve the global TB targets in the WHO European Region under the National Strategic Plan 2016–2020 (5, 12). SETTINGS TB care in Armenia is provided through six inpatient TB departments at general hospitals and 59 specialized TB outpatient centres located in Primary Healthcare (PHC) facilities, along with 24 microscopy laboratories. STUDY POPULATION All MDR-TB patients registered by the NTP in Armenia in 2013–2018 made up the study cohort. The registration of TB cases is mandatory in Armenia and we therefore assume that all MDR-TB cases diagnosed the country within the study period are included in our study. In order to identify risk factors associated with LTFU patients, the LTFU group was compared to successfully treated patients. Those patients who were still undergoing treatment as well as those in which treatment failed or who died were excluded from this part of the analysis. DATA VARIABLES Variables included in the analysis comprised demographic and clinical characteristics including treatment outcome, treatment duration any, history of previous treatment, results of smear analysis at the beginning of the treatment, use of new drug regimen (bedaquiline and/or delamanid), comorbidities (diabetes, hepatitis C and HIV), TB type, reasons for LTFU as reported by the patient’s doctor (for example being a migrant worker or refusing to continue the treatment), gender, age and region. DATA MANAGEMENT AND ANALYSIS Statistical analysis was conducted using STATA 14 statistics software and the EasySTAT online statistical application (15, 16). The proportion and mean differences between different groups were tested by chi-squared tests and t-tests respectively, with the significance level set at 5%. The Kaplan-Meier curve and Cox (proportional hazards) regression were used for survival analysis. Sensitivity analysis was performed using an adjusted Fine-Gray model to confirm the significance levels observed using the Cox proportional hazards model. ETHICAL CONSIDERATIONS Local ethical approval was obtained from the Institutional Ethics Review Board of the Centre of Medical Genetics and Primary Health Care. Ethical exemption was also received from the WHO Research Ethics Review Committee. 434 VOLUME 5 | ISSUE 4 | DECEMBER 2019 | 369–613PUBLIC HEALTH PANORAMA LOST TO FOLLOW-UP IN MDR-TB PATIENTS IN ARMENIA IN 2013–2018 RESULTS The study cohort comprised 651 MDR-TB cases registered during the study period, of which 234 (35.9%) were successfully treated, 161 (24.7%) were LTFU, 112 (17.0%) were still on treatment, in 82 (12.7%) treatment failed and 62 (9.5%) died. The treatment outcomes of MDR-TB patients by year are presented in Table 1. There were considerable differences in the distribution of treatment outcomes depending on the region (Table 2, Fig.1). Historically, TB prevalence in general has been low in the Vayots Dzor region and likewise this study revealed only two cases of MDR-TB with a LTFU status in the region, so it was excluded from statistical comparison based on these low numbers. Among the other regions, the highest percentage of LTFU patients was observed in the Gegharquniq region (35.5%), which was the only region where the level of LTFU was significantly higher (P<0.01) than that of the capital city Yerevan. Across Armenia, the most common reason for LTFU, as reported by doctors, was leaving the country for the purpose of migrant work, with 104 cases (64.6%). This was followed by the refusal to continue treatment in 52 cases (32.3%). In addition, alcohol abuse was reported by doctors as being the primary ground for LTFU in three patients, one case of LTFU was reported to be the result of pregnancy and one case the result of drug use (Table 3). To further examine the risk factors behind a LTFU outcome we excluded all outcomes except successfully treated and LTFU. This new cohort comprised 395 patients, of which 318 (80.5%) were male and 77 (19.5%) were female. The mean age of this group was 42.2 (SD ± 13.7), 272 (68.9%) had worked abroad as a migrant worker and 147 (37.2%) were from rural areas of Armenia. The characteristics of this group can be found in Table 4. A LTFU outcome was significantly associated with the migrant work status compared to a  successful treatment outcome, with the odds of a LTFU outcome around 1.7 times higher for migrant workers (OR=1.7, CI=1.1-2.5, p<0.01) than those who did not travel to work abroad. In addition, a history of previous treatment for TB was also associated with the LTFU outcome, TABLE 1. TREATMENT OUTCOME BY YEAR (2013–2018) Outcome Total N=651 (%) 2013 N=115 (%) 2014 N=117 (%) 2015 N=103 (%) 2016 N=125 (%) 2017 N=105 (%) 2018 N=86 (%) Successful treatment 234 (35.9%) 56 (48.7%) 67 (57.3%) 46 (44.7%) 56 (44.8%) 9 (8.6%) 0 (0%) Death 62 (9.5%) 11 (9.6%) 8 (6.8%) 8 (7.8%) 16 (12.8%) 10 (9.5%) 9 (10.5%) Failure 82 (12.6%) 17 (14.8%) 19 (16.2%) 16 (15.5%) 14 (11.2%) 12 (11.4%) 4 (4.7%) Lost to follow-up 161 (24.7%) 31 (27%) 23 (19.7%) 33 (32%) 36 (28.8%) 27 (25.7%) 11 (12.8%) Still on treatment 112 (17.2%) 0 (0%) 0 (0%) 0 (0%) 3 (2.4%) 47 (44.8%) 62 (72.1%) TABLE 2. TREATMENT OUTCOMES OF MDR-TB PATIENTS BY REGION OF ARMENIA (2013–2018) Outcome Total N=651 (%) Vayots Dzor N=4 (%) Gegharquniq N=31 (%) Armavir N=90 (%) Ararat N=54 (%) Tavush N=28 (%) Shirak N=57 (%) Kotayq N=80 (%) Syuniq N=22 (%) Yerevan N=198 (%) Lori N=65 (%) Aragatsotn N=22 (%) Successful 234 (35.9) 0 (0) 7 (22.6) 29 (32.2) 20 (37) 9 (32.1) 21 (36.8) 27 (33.8) 10 (45.5) 82 (41.4) 23 (35.4) 6 (27.3) Death 62 (9.5) 2 (50.0) 1 (3.2) 5 (5.6) 1 (1.9) 3 (10.7) 8 (14.0) 11 (13.8) 1 (4.5) 18 (9.1) 10 (15.4) 2 (9.1) Failure 82 (12.6) 0 (0) 5 (16.1) 12 (13.3) 6 (11.1) 4 (14.3) 9 (15.8) 10 (12.5) 3 (13.6) 24 (12.1) 8 (12.3) 1 (4.5) Lost to follow- up 161 (24.7) 2 (50) 11 (35.5) 27 (30.0) 14 (25.9) 7 (25.0) 14 (24.6) 19 (23.8) 5 (22.7) 45 (22.7) 14 (21.5) 3 (13.6) On treatment 112 (17.2) 0 (0) 7 (22.6) 17 (18.9) 13 (24.1) 5 (17.9) 5 (8.8) 13 (16.2) 3 (13.6) 29 (14.6) 10 (15.4) 10 (45.5) 435 ТОМ 5 | ВЫПУСК 4 | ДЕКАБРЬ 2019 Г. | 369–613ПАНОРАМА ОБЩЕСТВЕННОГО ЗДРАВООХРАНЕНИЯ LOST TO FOLLOW-UP IN MDR-TB PATIENTS IN ARMENIA IN 2013–2018 FIG. 1. TREATMENT OUTCOMES OF MDR-TB PATIENTS BY REGION OF ARMENIA (2013–2018). 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Aragatsotn Ararat Armavir Gegharquniq Kotayq Lori Shirak Syuniq Tavush Vayots Dzor Yerevan Successful treatment Lost to follow up Failure Death On treatment TABLE 4. FACTORS ASSOCIATED WITH AN INCREASED ODDS OF BEING LTFU AMONG MDR-TB PATIENTS (2013–2018) Variable Level Total N=395 (mean, %) LTFU N=161 (mean, %) Successful treatment N=234 Odds ratio/ mean difference 95% CI P value Results of smear microscopy Smear negative 152 (38.5%) 59 (36.6%) 93 (39.7%) 1.1 [0.8–1.7] 0.534 Smear positive 243 (61.5%) 102 (63.4%) 141 (60.3%) 1.0 HIV status HIV-negative 348 (88.1%) 142 (88.2%) 206 (88%) 1.0 - - HIV-positive 22 (5.6%) 7 (4.3%) 15 (6.4%) 1.4 [0.6–3.3] 0.41 Missing 25 (6.3%) 12 (7.5%) 13 (5.6%) 1.3 [0.6–3.0] 0.48 New drug in regiment No 284 (71.9%) 117 (72.7%) 167 (71.4%) 1.1 [0.7-1.7] 0.78 Yes 111 (28.1%) 44 (27.3%) 67 (28.6%) 1.0 Diabetes No 375 (94.9%) 151 (93.8%) 224 (95.7%) 1.4 [0.6–3.3] 0.39 Yes 20 (5.1%) 10 (6.2%) 10 (4.3%) 1.0 Hepatitis C No 349 (88.4%) 140 (87.0%) 209 (89.3%) 1.25 [0.7–2.5] 0.47 Yes 46 (11.6%) 21 (13.0%) 25 (10.7%) 1.0 Migrant worker No 272 (68.9%) 99 (61.5%) 173 (73.9%) 1.0 Yes 123 (31.1%) 62 (38.5%) 61 (26.1%) 1.7 [1.1–2.5] <0.01 Gender Female 77 (19.5%) 24 (14.9%) 53 (22.6%) 1.7 [1.0–2.5] 0.06 Male 318 (80.5%) 137 (85.1%) 181 (77.4%) 1.0 Geographical area Rural 147 (37.2%) 70 (43.5%) 77 (32.9%) 1.6 [1.0–2.4] 0.03 Urban 248 (62.8%) 91 (56.5%) 157 (67.1%) 1.0 TB type Extrapulmonary 12 (3%) 4 (2.5%) 8 (3.4%) 0.7 [0.2–2.7] 0.77 Pulmonary 383 (97%) 157 (97.5%) 226 (96.6%) 1.0 Patient type New 217 (54.9%) 82 (50.9%) 135 (57.7%) 1.0 - - Previous failure 75 (19%) 25 (15.5%) 50 (21.4%) 0.8 [0.5–1.4] 0.49 Previous interruption 35 (8.9%) 19 (11.8%) 16 (6.8%) 2.0 [1.0–4.0] 0.06 Other previously treated 32 (8.1%) 22 (13.7%) 10 (4.3%) 3.6 [1.6–8.0] <0.01 Relapse 36 (9.1%) 13 (8.1%) 23 (9.8%) 0.9 [0.4–1.9] 0.85 Age in years (mean ± SD) 42.2 (± 13.7) 41.5 (± 13.9) 42.7 (± 13.5) –1.2 (± 1.41) [–4.0– 1.6] 0.39 Treatment duration in months (mean ± SD) 16.1 (± 9.0) 7.7 (± 6.4) 21.9 (± 5.1) –14.2 (± 0.6) [–15.4– 13.0] <0.01 LTFU – Lost to follow-up, MDR-TB – Multidrug Resistance Tuberculosis. For categorical variables Pearson’s Chi-Square test was performed if all if number of patients in comparison subgroups were >5. Otherwise Fisher’s exact test was performed. For numerical variables t-tests were used. TABLE 3. REASONS REPORTED BY DOCTORS FOR MDR- TB PATIENTS BEING LTFU (2013–2018). Reasons N 161 (100%) Leaving the country for migrant work 104 (64.6%) Refusal to continue treatment 52 (32.3%) Alcohol use 3 (1.9%) Pregnancy 1 (0.6%) Drug use 1 (0.6%) 436 VOLUME 5 | ISSUE 4 | DECEMBER 2019 | 369–613PUBLIC HEALTH PANORAMA LOST TO FOLLOW-UP IN MDR-TB PATIENTS IN ARMENIA IN 2013–2018 with the odds around 3.6 times higher than those without a  history of previous treatment (CI=1.6-8.0, p<0.01). Being from a rural area was also associated with the LTFU outcome, with the odds around 1.6 times higher (CI=1.0–2.5, p=0.03). The mean duration of treatment among LTFU patients prior to the interruption, was 7.7 (SD ± 6.4) months, which was about 14 months less than the mean duration of treatment among successfully treated patients (21.9 months, SD ± 5.1). Survival analyses in LTFU and successfully treated patient groups, confirmed that migrant work status (adjusted hazard ratio (HR)=1.6, CI=1.07–2.15, P=0.02) and a  history of successful, but not failed or relapse, TB treatment (adjusted HR=2.3, CI=1.18–3.23, P<0.01) were associated with a  LTFU outcome but also revealed that the rate of drop out from the treatment regimen (LTFU) was 40% lower in patients on newer drug treatment regimens (adjusted HR=0.6, CI=0.4–0.8, P<0.01)(Table 5). A  Kaplan-Meier survival curve (Fig. 2) indicated that over roughly a 24 month period, which is the point by which most treatment programmes are complete, there was a reasonably consistent rate of patients LTFU at about 2% a month. Sensitivity analysis using an adjusted Fine-Gray model and death as the competing factor, showed that death had no a significant effect on time to LTFU. DISCUSSION This was the first countrywide cohort study in Armenia to evaluate risk factors associated with LTFU among MDR-TB patients from all TB departments and centres in 2013–2018. According to the study findings nearly one in four MDR-TB patients were LTFU with the mean duration of an incomplete treatment at 7.7 months, which is about three times less than the complete treatment duration. These numbers are quite disturbing and point out the urgency of specific interventions to improve the number of patients completing treatment. One of the factors leading to the LTFU outcome among MDR-TB patients is being a migrant worker. The highest rate of the LTFU was observed in the Gegharquniq region of Armenia, which is consistent with its high rate of migrant workers as a result of poor socioeconomic conditions in the region (17, 18) This highlights the importance of effective collaboration between TABLE 5. HAZARD RATIOS (HR) FOR A LTFU OUTCOME FOR THE PERIOD OF 2013–2018) Variable Crude HR 95% CI P value Adjusted HR Adjusted 95% CI Adjusted P value Smear Positive 1.0 0.69–1.3 0.73 HIV-positive 0.5 0.21–0.97 0.04 0.5 0.21–1.00 0.05 Using new drug in regimen 0.6 0.4–0.8 <0.01 0.6 0.38–0.82 <0.01 Diabetes 1.2 0.7–2.3 0.52 Hepatitis C 0.8 0.52–1.31 0.42 1.0 0.6–1.62 0.95 Migrant worker 1.6 1.14–2.14 <0.01 1.5 1.07–2.15 0.02 Male 1.3 0.82–1.95 0.29 1.1 0.66–1.71 0.81 Age in years 1.0 0.98–1.01 0.44 Urban area 0.8 0.56–1.05 0.10 Pulmonary 1.4 0.51–3.71 0.53 0.8 0.27–2.09 0.59 Patient type New 1.2 0.79–1.94 0.35 1 Previous failure 2.3 1.6–3.1 <0.01 1.1 0.66–1.82 0.71 Previous interruption 1.7 0.92–3.05 0.09 1.5 0.89–2.53 0.13 Other previously treated 2.3 1.32–4.15 <0.01 2.0 1.18–3.23 <0.01 Relapse 0.9 0.45–1.72 <0.71 0.7 0.35–1.31 0.25 437 ТОМ 5 | ВЫПУСК 4 | ДЕКАБРЬ 2019 Г. | 369–613ПАНОРАМА ОБЩЕСТВЕННОГО ЗДРАВООХРАНЕНИЯ LOST TO FOLLOW-UP IN MDR-TB PATIENTS IN ARMENIA IN 2013–2018 the Armenian NTP of that of other countries, especially the Russian Federation where the majority of migrant workers head to, to make sure that successful referrals to local doctors are carried out and that patients complete their treatments in their host country. In addition, the education of MDR-TB patients, especially those with a high risk for being LTFU, on the importance of completing treatment should be improved. These recommendations are also in line with the WHO End TB strategy, in which three of the targets refer to key populations, including migrants (19). The high rates of LTFU and emigration in Armenia requires systematic research and follow-up interventions in order for the country to achieve the WHO’s Global Plan to End TB targets (90% reduction in mortality and 80% reduction in TB incidence by 2030 (17, 19, 20). Patients with a history of previous TB treatment and/or those from rural areas are at higher risk of a LTFU outcome. These may be explained by the long duration and exhaustive nature of each treatment and the higher rate of emigration due to higher unemployment rates in these areas, respectively. We identified that being on a treatment regimens containing newer drugs contributed to a lower risk of a LTFU outcome. This could be explained by the better supervision and follow-up of the patients receiving the new TB drugs where there are more strict criteria to ensure treatment success. This might indicate a better follow-up interventions targeting the individual needs of a patient can improve treatment success rates. Our results, which show that the dropout rate is relatively constant over 24 months, indicate that these interventions should be equally distributed throughout the entire treatment period. The fact that all TB centres in the country were included in the study and the findings come from a  routine setting depicts a  comprehensive and complete picture of MDR-TB interruptions in Armenia. In addition, the study findings are particularly valuable since they respond to operational research priorities reflected in the NTP review report of Armenia. It should noted however, that some potential risk factors for LTFU such as employment and smoking status are not included in the analysis as these variables were missing from the database. These variables were therefore beyond the scope of this study but could serve as valuable targets for further study. In conclusion, we have identified several factors associated with a  LTFU outcome. More targeted interventions, particularly those aimed at migrant workers, could improve the overall treatment success rate of MDR-TB patients, leading to an improved TB situation in the country. Acknowledgements: We would like to thank the national TB Control Centre of Armenia for defining the research questions and providing data for this study, and the secretariat of the European TB Research Initiative (ERI-TB) at the WHO Regional Office for organizing the Structured Operational Research Training (SORT-TB) for six eastern European countries supported by the United States Agency for International Development (USAID)-WHO regional partnership project to End TB in eastern Europe (RP). The SORT-TB curriculum was an adaptation of the SORT IT course of the Special Programme for Research and Training in Tropical Diseases made for the eastern European context. Sources of funding: This study was funded by USAID. The funder had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. Conflicts of interest: None declared. Disclaimer: The authors alone are responsible for the views expressed in this publication and they do not necessarily represent the decisions or policies of the World Health Organization. REFERENCES1 1. Global Tuberculosis Report WHO 2018. Geneva: World Health Organization; 2018. (https://apps.who.int/iris/ handle/10665/274453). 1 All references were accessed 13 December 2019. FIGURE 2: KAPLAN-MAIER SURVIVAL CURVE FOR MDR-TB PATIENTS FOR THE STUDY PERIOD OF 2013-2018 0. 00 0. 25 0. 50 0. 75 1. 00 0 10 20 30 40 Kaplan−Meier survival estimate Time in months Pr op or tio n of p at ie nt s on tr ea tm en t analysis time 438 VOLUME 5 | ISSUE 4 | DECEMBER 2019 | 369–613PUBLIC HEALTH PANORAMA LOST TO FOLLOW-UP IN MDR-TB PATIENTS IN ARMENIA IN 2013–2018 2. Top 10 Causes of Death Worldwide. World Health Organization; 2017 [website]. (https://www.who.int/news- room/fact-sheets/detail/the-top-10-causes-of-death). 3. Gwenan M. Knight, Ulla K. Griffiths, Tom Sumner, Yoko V. Laurence, Adrian Gheorghe, Anna Vassall, Philippe Glaziou, and G. White Impact and cost-effectiveness of new tuberculosis vaccines in low- and middle-income countries. Proc Natl Acad Sci USA. 2014; PNAS first published October 6, 2014. doi:10.1073/pnas.1404386111. 4. Richardson E. Armenia: health system review. Health Syst Transit. 2013;15(4):1–99. PMID:24334656. 5. Extensive Review of TB Prevention (Dara M, Mkrtchyan Z, Ghukasyan G, editors). Care and Control Services in Armenia 21 April  – 4 May 2011. Copenhagen: WHO Regional Office for Europe; 2011 (https://apps.who.int/iris/ handle/10665/108604). 6. Hayrapetyan A. Tuberculosis epidemiology trends in Armenia. Int J  Infect Dis. 2012; June 2012. Volume 16, Supplement 1, doi:10.1016/j.ijid.2012.05.946. 7. Hayrapetyan A, Mezhlumyan N, Davtyan H, Sevoyan M. Tuberculosis trends in Armenia. EpiNorth - J Netw Commun Dis Control North East Eur; 2012. 8. Extensive review of tuberculosis prevention, control and care in Armenia  - 17–25 July 2014, Copenhagen: WHO Regional Office for Europe; 2015 (http://www.euro.who.int/ en/countries/armenia/publications/extensive-review-of- tuberculosis-prevention,-control-and-care-in-armenia-1725- july-2014-2015). 9. Tuberculosis country brief, 2016: Armenia. Copenhagen: WHO Regional Office for Europe; 2016 (http://www.euro.who. int/en/health-topics/communicable-diseases/tuberculosis/ country-work/armenia). 10. Tuberculosis Programmatic Report of National Tuberculosis Control Center of Ministry of Health of Armenia, 2017. 11. Tamara Karelidze. Russia remains preferred destination for Armenian migrants. Emerging Europe [news]; September 25 2018 (https://emerging-europe.com/news/russia-remains- preferred-destination-for-armenian-migrants/). 12. Davtyan K, Zachariah R, Davtyan H, Ramsay A, Denisiuk O, Manzi M, et al. Performance of decentralised facilities in tuberculosis case notification and treatment success in Armenia. Public Health Action. 2014 Oct 21;4 Suppl 2:13–6. doi: 10.5588/pha.14.0038 PMID:26393091. 13. Statistical Committee of the Republic of Armenia. Statistical Yearbook of Armenia 2017. (https://www.armstat.am/ en/?nid=586&year=2017). 14. The World Bank. World Bank Open Data [web-site]. (https:// data.worldbank.org/). 15. Lawrence C. Hamilton. Statistics with STATA: updated for version 9. Belmont (CA): Thomson Brooks/Cole; 2006. 16. EasySTAT: Data Science Platform for Quick and Easy Statistics [on-line tool] (https://easystat.app/). 17. Rasuly-Paleczek G, Six-Hohenbalken M, editors. Migration and its Impact on Armenia. Vienna: Department for Social and Cultural Anthropology University of Vienna; 2017 (https:// www.univie.ac.at/alumni.ksa/wp-content/uploads/ASSA- SN-2017-01_Migration-and-its-impact-on-Armenia.pdf). 18. Sinara Navoyan. The Impact of Migration on Transformation of Gender Roles in Gegharkunik Region of the RA. [Migraciai azdecutyunt genderain pokokhutyan vra HH Gebharkunik marzum]. Yerevan: Center for Gender and Leadership Studies; 2015. 19. Uplekar M, Weil D, Lonnroth K, Jaramillo E, Lienhardt C, Dias HM, et al.; for WHO’s Global TB Programme. WHO’s new end TB strategy. Lancet. 2015 May 2;385(9979):1799–801. doi: 10.1016/S0140-6736(15)60570-0. PMID:25814376. 20. Truzyan N, Crape B, Grigoryan R, Martirosyan H, Petrosyan V. Increased risk for multidrug-resistant tuberculosis in migratory workers, Armenia. Emerg Infect Dis. 2015 Mar;21(3):474–6. doi: 10.3201/eid2103.140474. PMID:25695488. n