207 ORIGINAL ARTICLE Acta Medica Indonesiana - The Indonesian Journal of Internal Medicine HIV Patients Drop Out in Indonesia: Associated Factors and Potential Productivity Loss Adiatma YM. Siregar1, Pipit Pitriyan1, Rudi Wisaksana2 1 Department of Economics, Center for Economics and Development Studies, Faculty of Economics and Business, Padjadjaran University, Bandung, Indonesia. 2 Department of Internal Medicine, Hasan Sadikin Hospital - Faculty of Medicine, Padjadjaran University, Bandung, Indonesia. Corresponding Author: Dr. Adiatma YM. Siregar. Center for Economics and Development Studies, Faculty of Economics and Business, Padjadjaran University. Jl. Cimandiri No.8, Citarum, Bandung 40115, Indonesia. email: adiatma.siregar@unpad. ac.id. ABSTRAK Tujuan: meneliti tentang faktor-faktor yang berhubungan dengan probabilitas drop out yang lebih tinggi bagi pasien HIV, dan potensi produktifitas yang hilang akibat drop out tersebut. Metode: Kami menganalisis data dari 658 pasien HIV dari sebuah database di sebuah rumah sakit rujukan utama di kota Bandung, Jawa Barat, Indonesia dari tahun 2007-2013. Pertama, kami menggunakan metode analisis regresi probit dan mengikutsertakan, antara lain, variabel berikut: status pasien (aktif atau drop out), CD4 cell count, TB dan infeksi oportunistik (IO), status bekerja, jenis kelamin, pengalaman sebagai penasun, dan dukungan dari keluarga dan peers. Kedua, kami menggunakan data tingkat drop out dari database kami dan tingkat penurunan CD 4 cell count dari studi lain untuk mengestimasi produktifitas yang hilang akibat drop out. Hasil: CD4 cell count yang rendah diasosiaikan dengan probabilitas drop out yang lebih tinggi. Dukungan dari peers, hidup bersama keluarga, dan menderita TB diasosiasikan dengan probabilitas drop out yang lebih rendah. Produktifitas yang hilang pada tingkat nasional akibat drop out (dank arena tingkat CD 4 cell count yang menurun) dapat mencapai US$365 juta (menggunakan upah rata-rata). Kesimpulan: pertama, karena tingkat CD 4 cell count yang rendah diasosiasikan dengan probabilitas drop out yang lebih tinggi,kami merekomendasikan (untuk mengoptimasikan) pemberian ARV dini pada tingkat CD 4 cell count yang lebih tinggi, melibatkan pemberian layanan HIV di tingkat komunitas. Kedua, dukungan keluarga dan peer harus diperkuat untuk mendukung kesuksesan dari perawatan HIV. Ketiga, drop out dari layanan ART akan menyebabkan hilangnya produktifitas yang cukup besar. Kata kunci: HIV, sosial-ekonomi, terapi antiretroviral, Indonesia, drop out. ABSTRACT Aim: this study reported various factors associated with a higher probability of HIV patients drop out, and potential productivity loss due to HIV patients drop out. Methods: we analyzed data of 658 HIV patients from a database in a main referral hospital in Bandung city, West Java, Indonesia from 2007 to 2013. First, we utilized probit regression analysis and included, among others, the following variables: patients’ status (active or drop out), CD4 cell count, TB and opportunistic infection (OI), work status, sex, history of injecting drugs, and support from family and peers. Second, we used the drop out data from our database and CD 4 cell count decline rate from another study to estimate the productivity loss due to HIV patients drop out. Results: lower CD4 cell count was associated with a higher probability of drop out. Support from family/peers, living with family, and diagnosed with TB were associated with lower probability of drop out. The productivity loss at national Adiatma YM. Siregar Acta Med Indones-Indones J Intern Med 208 level due to treatment drop out (consequently, due to CD4 cell count decline) can reach US$365 million (using average wage). Conclusion: first, as lower CD 4 cell count was associated with higher probability of drop out, we recommend (to optimize) early ARV initiation at a higher CD 4 cell count, involving scaling up HIV service at the community level. Second, family/peer support should be further emphasized to further ensure treatment success. Third, dropping out from ART will result in a relatively large productivity loss. Keywords: HIV, socio-economic, antiretroviral therapy, Indonesia, drop out. INTRODUCTION Indonesia is experiencing a fast increase in HIV prevalence (estimated <0.1% in 2001 to 0.4% in 2012)1 and a change in the HIV epidemic- whereas it was previously concentrated within the intravenous drug users (IDUs) and their couples, and now it moves to general population. The percentage of HIV transmission through injecting drug use decreased from 53% in 2001- 2005 to 34% in 2011, while the heterosexual transmission rose from 37% to 71% within the same period.2 Indonesia’s response to the epidemic have started since the first domestic case was detected which included the continuing support for care and treatment programs.2,3 In accordance, the need for antiretroviral therapy (ART) was expected to reach around 87,000 patients in 2014.4 Although the availability of ART in Indonesia is increasing2, drop out rate is still a significant issue.5–7 In a more specific context, this phenomenon is apparent in a clinic within a main referral hospital in West Java. Although this clinic has been successful in increasing the quality of HIV-care and reducing mortality rate, it is still facing challenges in reducing a high dropout rate, which is only slightly lower than the national estimate.6,8 This is crucial as the success of ART will increase quality of life (QoL), productivity, and physical and emotional health.9–11 These factors in the end will influence the survival of patients.12 Given the importance of HIV patients drop out, there is an urgency to answer the following questions: 1) What are the factors influencing the HIV patients drop out rate in Indonesia?, and 2) What is the potential economic impact? There has been a number of studies trying to pinpoint the reasons underlining treatment drop out or non-adherence. Some local studies suspected socio-economic factors as possible causes13,14, while others suspected psycho-social and biomedical factors5,7,15, or simply patients’ error (e.g. forget to take medication).16 These factors have also been explored by international studies.17–21 The results were varied from all of these studies, and required further confirmation to apply in a specific context. Understanding the factors influencing dropout rate will provide valuable input in providing better service and policy input.6 The research on economic impact of drop out, on the other hand, still limited. This is a very important point for Indonesia since HIV patients drop out is still a problem. T h i s s t u d y a n a l y z e d t h e p a t i e n t s ’ socioeconomic, demographic, and clinical characteristics, and explored whether these characteristics were associated with the higher/ lower probability of HIV patients drop out. This study also estimated the economic impact of drop out in terms of productivity loss. This study was unique for four reasons. First, to our knowledge, studies exploring the factors causing HIV patients dropouts in Indonesia is limited. Second, our study utilizes data from a large number of patients’ medical record as a part of a larger cohort database owned by the clinic, providing a lot of valuable information otherwise unattainable. Third, the clinic in our study is utilized by patients from all over the province (and in some cases beyond the province), which gives it a pivotal role in providing evidence for policy making. Fourth, this study estimates the productivity loss as the impact of drop out to provide evidence for policy makers. METHODS Study Setting and Study Population The study was conducted in Bandung, at an HIV/AIDS clinic in the largest public referral Vol 48 • Number 3 • July 2016 HIV patients drop out in Indonesia 209 and teaching hospital in West Java province (43 million inhabitants). The clinic is visited by high risk group of patients and the general population, and deliver HIV-related services such as voluntary counseling and testing, ART, and sexually transmitted infections services. The clinic operates at full capacity because it is among the few clinics that deliver ART in Bandung. The clinic generates its own revenues through government, hospital, and private funding. The ART-related services are free, except for hospitalization and the registration fee. Data Collection and Analysis We utilized the database of a clinic within a main referral hospital in Bandung city, West Java, Indonesia (2007-2013). Out of the data of 8,184 HIV patients, we only included the data of 658 patients due to incomplete data set. We extracted data related to four categories: patient’s current status, socioeconomic, demographic, and clinical characteristics. The detailed variables list from each category is presented in Table 1. We used Microsoft Excel 2013 and STATA 13 for data cleaning and analysis. We utilized probit regression analysis to determine the influence of each factor. We only used the patients’ clinic ID number throughout the analysis, and none of patients’ personal identification is used. We also combined our dropout rate data and the CD4 decline rate from another study within the same clinic to estimate the productivity loss due to drop out from HIV treatment. We estimated the length of time it took for the CD4 cell count of patients who drop out to reach approximately 50 cells/mm3 (assuming that they do not seek any treatment after they drop out). We assumed that patients can no longer be productive afterwards due to the onset of various opportunistic infections as patients with CD4 cell count lower than 100 cells/mm3 was considered to experience immunological failure.23 The productivity loss was then measured by the loss of income from the moment patients reach the CD4 cell count of 50 cells/mm3 until the end of productive age using Microsoft Excel 2013. We used average and minimum wage from a publication by ILO24, and we assumed the productive age of 15-64 years as stated by Bureau of Statistics of Indonesia Table 1. List of variables Variable List Definition Type Patient Status (dependent variable) 0 = active, 1= drop out Dummy Socioeconomic characteristics - Age Absolute Continuous - Sex 0 = female, 1= male Dummy - Employment status 0 = unemployed, students, housewife, 1 = employed Dummy - Education 0 = primary school/no education, 1 = junior high, 2 = senior high, 3 = university/college Ordered dummy - Marital status 0 = not married/divorced/widowed, 1 = married Dummy - Have children 0 = no children, 1 = have children Dummy - Social insurance membership 0 = not on social insurance, 1 = on social insurance Dummy - Live with family 0 = not live with family, 1 = live with family Dummy - Support from family/peers 0 = not satisfied, 1 = satisfied Dummy Demographic characteristics - Place of stay 0 = live in Bandung city, 1 = live outside Bandung city Dummy Clinical characteristics - CD4 cell count 0 = >349 cells/mm3, 1 = 150-349 cells/mm3, 2 = 50-149 cells/ mm3, 3 = 0-49 cells/mm3 Ordered dummy - OI 0 = not infected by OI, 1 = infected by OI Dummy - IDU history 0 = no history, 1 = with history Dummy - TB history 0 = no history, 1 = with history Dummy - Length of treatment 0 = less than 1 year, 1 = more than 1 year Dummy Adiatma YM. Siregar Acta Med Indones-Indones J Intern Med 210 (BPS). Lastly, we used the discount rate of 3%25, and the exchange rate of Rp13,230/1US$.26 We then combined the estimation of productivity loss per person to the whole number of (estimated) drop out patients at the clinic and at national level. Table 2 summarizes the indicators used in the productivity loss analysis. The estimation of patients undergoing ART and drop out rate at national level were taken from Spiritia.8 RESULTS Table 3 summarizes patient characteristics based on factors presented in Table 1. Most patients are at their productive age, and mostly are male, employed, and receive secondary or tertiary education. More than a half of patients are married and have children. Almost all of patients are living with their family and satisfied with peer support. Most patients visit the clinic for the first time with higher than 50 cells/mm3 CD4 cell count, and around 60% of the patients have IDU history. Table 2. Productivity loss calculation indicators Items Values Average CD 4 cell count, drop out patients (95% C.I) 123 (91–155) Number of drop out HIV patients - Male/female 55 - Female 29 - National male** 9,510 - National female** 2,506 Monthly wage24 - Minimum US$ 97.38 - Average US$ 144.91 Productive age 15 – 64 years (50 years) Time length for CD 4 cell count to reach approximately 50 cells/mm3 after dropout* - IDU 3.42 years - Non IDU 3.67 years IDU*** - IDU male*** Dropout rate - Clinic rate 12.77% - National rate8 17.90% Unemployment rate - Clinic rate (male)*** 36.36% - National rate8 6.10% 27 Female labor participation rate - Clinic rate*** 31.03% - National rate8 51% 27 Exchange rate26 Rp 13,230/US$ Discount rate25 3% * Calculated based on results from Meijerink et al.22 ** Estimated based on Spiritia8 *** from drop out patients Table 3. Patient characteristics (n=658) Variable list Values Patients who drop out 84 (12.77%) Socioeconomic characteristics - Mean of age (95% C.I) 30 (29.7 – 30.6) - Male 443 (67.33%) - Employed 425 (64.59%) - Education (secondary and tertiary) 635 (96.50%) - Marital status (married) 393 (59.73%) - Have children 402 (61.09%) - Have social insurance membership 87 (13.22%) - Live with family 585 (88.81%) - Satisfied with support from family/peers 625 (94.98%) Demographic characteristics - Place of stay (live outside Bandung city) 236 (35.87%) Clinical characteristics - CD4 cell count (<50 cells/mm3) 213 (32.37%) - OI 166 (25.23%) - IDU history 394 (59.88%) - TB history 163 (24.77%) - Length of treatment (>1 year) 583 (88.60%) Factors Associated with Drop Out Figure 1 presents the marginal effect of the probit regression result. From the factors associated with higher probability of drop out (and significant at 90%, 95%, or 99% C.I), we found that patients with under than 50 cells/mm3 CD4 cell count had 8.7% higher probability of drop out compared with patients with higher than 349 cells/mm3 CD4 cell count. This percentage decreased as we move up the CD 4 cell count groups. From the factors associated with lower probability of dropping out (and significant at Vol 48 • Number 3 • July 2016 HIV patients drop out in Indonesia 211 90%, 95%, or 99% C.I), we found that patients who lived with family have 10.2% lower probability of drop out compared with patients who did not. Patients who were satisfied with support from family/peers, had 8.8% lower probability of dropping out, compared to those who were not satisfied. Patients who lived outside Bandung city, and patients who were diagnosed with TB were associated with lower probability of drop out compared to their counterparts. Those who have been undergoing treatment for more than 1 year had a lower probability of dropping out compared to those who were undergoing the treatment for less than one year. Impact of Drop Out on Productivity Loss Table 4 presents the productivity loss at the clinic level and at national level. Using clinic dropout rate, the estimated productivity loss over remaining productive lifetime for patients visiting clinic can reach US$18 million. Projecting the clinic dropout rate into the number of HIV patients on ARV at national level, the estimated productivity loss over remaining productive lifetime can reach US$191 million. Using the national dropout rate, these numbers can rise to US$365 million. Figure 1. The econometric analysis result using probit regression, presented by the marginal effect results for each variables Table 4. Estimated productivity loss over remaining productive lifetime (US$) Items Amount Per person (based on clinic estimate) - High estimate* 34,226 - Low estimate** 23,001 Clinic estimation - High estimate* 18,607,499 - Low estimate** 12,504,956 National estimation Using clinic indicators - High estimate* 191,054,269 - Low estimate** 128,395,829 Using national indicators (except for % of IDU) - High estimate* 365,041,113 - Low estimate** 245,321,691 *using average wage, ** using minimum wage DISCUSSION This study has two aims. First it analyzes the factors associated with HIV patients dropout in the clinic. Second, it estimates the impact of HIV patients dropout on productivity loss over remaining productive lifetime at the clinic and Adiatma YM. Siregar Acta Med Indones-Indones J Intern Med 212 national level. Based on these objectives, five observations can be made. First, as higher CD 4 cell count are associated with lower probability of drop out, in line with a study by Blutinger et al.21 in India, optimizing early ARV initiation at higher CD 4 cell count is warranted. The benefit of potential costs saving by providing early ART has been discussed in other studies28–31, and our study adds that it may also reduce the probability of dropout. One of the potential scale up sites are the community clinics, since patients are taking HIV test at this type of clinic at a much higher CD4 cell count compared to those visiting hospitals.32 In addition, accessing HIV service at the community level will reduce patients’ costs and increase adherence of ART.5,28,33,34 However, the potential budget impact of scaling up and providing early ART35,36, and the required community staff capacity37 should be carefully assessed. Given the current prevalence rate of Indonesian HIV epidemic (0.5% for adults age 15-49)38, it seems that scaling up ART at the community should be conducted at areas with high prevalence setting 39,40 to have a large impact. Second, our results show that living with family and peer support is significantly associated with lower probability of HIV patients dropout. Another study from Indonesia and some studies from other countries also have similar finding.15,17,19,20,41,42 Thus, it is clear that family/ peer support should be further enhanced to ensure treatment success in the long run. Finding the correct scheme to enhance family/peer support should be done carefully as not all schemes may work (e.g. use of pager).43 On the other hand, interventions such as text messaging, self-forming group of patients, modified directly observed therapy (mDOT), are able to support ART adherence.44–47 Indonesia should start to either adopt (or modify if necessary) these already proven interventions or come with a breakthrough approach to improve the current family/peer support for HIV patients. Third, as patients with TB history is associated with lower probability of drop out, it is a plausible idea to integrate TB and HIV treatment. Studies have shown that this integration have a positive impact on patients and community48–50, and models from low- and middle-income countries are available.51 However, the infrastructures and the capacity of health care workers, need to be enhanced for this integration to work.51,52 Fortunately, this kind of integrated TB/HIV models are also increasing in Indonesia2, leading to more efficient service deliveries. Therefore, consistent monitoring and evaluation of this scheme is required to produce an even more efficient and better quality HIV/ TB integrated services. Fourth, our study shows that patients with longer than a year length of treatment are associated with lower probability of drop out. This has an important implication because if we are able to reduce drop out rate and keep patients within the treatment, it will lead to an even lower probability of drop out rate, creating a continuous effect. Fifth, the costs of productivity loss over remaining productive lifetime due to dropout can be high. In our case, it can reach US$365 million: 0.042% of our gross domestic product (GDP) in 201353 or 1.36% of total health expenditure in the same period.54 We need to compare this productivity loss to the costs of treatment if patients are staying in the program. Based on the study by Siregar et al.28, the cost of treating a patient with ARV is approximately US$1,200 per patient per year for patients with CD4 cell count lower than 50 cells/mm3 (this cost decreases as the CD4 cell count increases). Multiplying this amount with the number of drop out patients at national level (using clinic dropout rate and the discount rate of 3%) for the rest of their life years (subtracting life expectancy55 with the age of drop out in the clinic) results in US$301 million, lower than the productivity loss. Thus, it seems that keeping patients within the treatment program is less costly compared to losing them. In addition, the ART provides other benefits aside of treating HIV patients. It also acts as a prevention measure for HIV epidemic56,57 and will most likely prevent spending more costs. Furthermore, bear in mind that our productivity loss calculation is based only on productivity approach using average and minimum wage. As such, in showing the potential economic impact of HIV patients dropout, we have forgone Vol 48 • Number 3 • July 2016 HIV patients drop out in Indonesia 213 other important values such as those stemming from loss of/reduced quality of life, costs to family members, hospitalization costs, OI treatment costs, possible line 2 ART costs, and indirect costs. Therefore, we believe the actual economic impact of dropout should be much higher. However, we need to take into account the currently low health budget of Indonesia if we want to provide ART to all HIV patients.58 In addition to our main observations, an important finding, although it is not statistically significant, should be carefully noted. Our analysis shows that higher education is related with higher probability of drop out. This is the opposite of the results by Schilkowsky et al.20, Waldrop-Valverde et al.59, and Unge et al.42. We suggest to explore this finding further as it has an important implication. If indeed a higher education is associated with higher probability of dropout, then a specific intervention should be designed to target patients with higher education as more than 90% of the patients in our study belong to this group. On the other note, the history of drug use, surprisingly, does not seem to have a significant influence on the drop out rate. This is supported by another study in the clinic13, although international studies show otherwise.59,60 This warrants further investigations since IDUs still hold a crucial role in the Indonesian HIV epidemic. Study Limitation There are several limitations in our study. First, the analysis on factors associated with HIV patients drop out is specific to the clinic that we studied, although similar results are found in some other local and international studies. Other clinics in Indonesia, however, may have different patterns. Therefore, applying the same study in more clinics in different settings in Indonesia (e.g. region, culture, religion, and politics) may yield more varied result and may enrich the discussion. Second, we assume that the drop out patients will not join other ART program at other clinics and that they will not seek treatment afterwards. Of course this may not be true to all patients, as some may visit other clinics to obtain ART once their disease has become more severe and they feel the need for treatment. A more detailed analysis following patients who drop out may provide further insight on the potential economic impact of HIV patients drop out. Third, there are much incomplete data set per person in our database, cutting our observation from 8,184 to 658 patients. This, will have implication on our results, and having better dataset may have improved our findings. Regardless, given the current number of observation, we have managed to show the significant factors that influence the probability of HIV patients dropout, adding information for the clinic managers and/or policy makers in designing appropriate policy. Fourth, our productivity loss calculation is based only on productivity approach using average and minimum wage. We believe actual economic impact of drop out should be much higher as we forgone important values such as loss of/reduced quality of life, costs to family members, and hospitalization costs. However, the purpose of our productivity loss analysis is to show the potential economic impact due to the drop out at its most basic state. Our study has shown this and clarifies that this amount can only be higher if it is further calculated. CONCLUSION Controlling drop out rate is crucial as the success of ART will increase QoL, productivity, and physical and emotional health which influences the survival of patients. This study has shown that higher CD 4 cell count, satisfied with family/peer support, have been in treatment longer than a year, and have TB history are associated with lower probability of HIV patients dropout. We have also presented the severity of dropout in terms of its consequences on productivity loss. The estimated productivity loss over productive lifetime due to dropout can reach US$312 million for IDU and US$306 million for non-IDU at national level. We recommend to optimize early ART, family/peer support, and TB/HIV service integration to reduce drop out rate and reduce its economic impact. Adiatma YM. Siregar Acta Med Indones-Indones J Intern Med 214 ACKNOWLEDGMENTS We would like to thank all of Teratai clinic staff (Hasan Sadikin Hospital) for their contribution in providing the available data for analysis. This research was financially supported by Universitas Padjadjaran. Authors declare no conflict of interest. REFERENCES 1 UNAIDS. Global report: UNAIDS report on the Global AIDS epidemic 2013. 2013. 2 National AIDS Commission (NAC). Republic of Indonesia Country report on the Follow up to the Declaration of Commitment on HIV/AIDS (UNGASS), Reporting Period 2010-2011. 2012. 3 Coordinating Minister for People’s Welfare/Chair of the National AIDS Commission Government of Indonesia. Indonesian National AIDS Strategy. 1994. 4 Ministry of Health of Indonesia. Mathematic model of HIV epidemic in Indonesia. 2008. http://spiritia.or.id/ Doc/model0814.pdf. 5 Haroen H, Handayani M, Puspitasari TS, Pinxten L, Van Den Bome B, Halfens R. Lost to follow up from ARV treatment: reasons and true outcome of patients in Bandung, Indonesia. IMPACt results. 2011. 6 Wisaksana R, Alisjahbana B, van Crevel R, Kesumah N, Sudjana P, Sumantri R. Challenges in delivering HIV-care in Indonesia: experience from a referral hospital. Acta Med Indones-Indones J Intern Med. 2009;41 (Suppl 1):45–51. 7 Azmir M. Factors affecting adherence to anti retroviral therapy among HIV patients attending HIV clinic at Hasan Sadikin Hospital. Oral Present XVI Congr Indones Soc Trop Med Bali. 2010. 8 Yayasan Spiritia. ARV treatment in Indonesia. 2014. http://www.spiritia.or.id/Stats/statart.php?lang=id (accessed June 5, 2015). 9 Mannheimer SB, Matts J, Telzak E, et al. Quality of life in HIV-infected individuals receiving antiretroviral therapy is related to adherence. 2005 DOI:10.1080/09 540120412331305098. 10 Parsons TD, Braaten AJ, Hall CD, Robertson KR. Better quality of life with neuropsychological improvement on HAART. Health Qual Life Outcomes 2006;4:11. 11 Beard J, Feeley F, Rosen S. Economic and quality of life outcomes of antiretroviral therapy for HIV/AIDS in developing countries: a systematic literature review. AIDS Care. 2009;21:1343–56. 12 de Boer-van der Kolk IM, Sprangers MG, Prins JM, Smit C, de Wolf F, Nieuwkerk PT. Health-related quality of life and survival among HIV-infected patients receiving highly active antiretroviral therapy: a study of patients in the AIDS Therapy Evaluation in the Netherlands (ATHENA) Cohort. Clin Infect Dis. 2010;50:255–63. 13 Wisaksana R, Indrati AK, Fibriani A, et al. Response to first-line antiretroviral treatment among human immunodeficiency virus-infected patients with and without a history of injecting drug use in Indonesia. Addiction. 2010;105:1055–61. 14 Riyarto S, Hidayat B, Johns B, et al. The financial burden of HIV care, including antiretroviral therapy, on patients in three sites in Indonesia. Health Policy Plan. 2010;25:272–82. 15 Alisjahbana B, Wisaksana R, Siregar AY, et al. Barriers, challenges, and good practices in delivering antiretroviral treatment: Report of assistance on joint rapid assessment on the strategic use of ART to reduce HIV related mortality and to prevent HIV transmission in Indonesia. 16 Widjaja FF, Puspita CG, Daud F, Yudhistrie I, Tiara MR, Suwita CS. Highly active antiretroviral therapy adherence and its determinants in selected regions in Indonesia. Med J Indones. 2011;23:50–5. 17 Ammassari A, Trotta MP, Murri R, et al. Correlates and predictors of adherence to highly active antiretroviral therapy: overview of published literature. J Acquir Immune Defic Syndr. 2002;31 (Suppl 3):S123–S127. 18 Wettstein C, Mugglin C, Egger M, et al. Missed opportunities to prevent mother-to-child-transmission: systematic review and meta-analysis. AIDS. 2012;26: 2361–73. 19 Reynolds NR, Testa MA, Marc LG, et al. Factors influencing medication adherence beliefs and self- efficacy in persons naive to antiretroviral therapy: a multicenter, cross-sectional study. AIDS Behav. 2004;8:141–50. 20 Schilkowsky LB, Portela MC, Sá M de C. Factors associated with HIV/AIDS treatment dropouts in a special care unit in the City of Rio de Janeiro, RJ, Brazil. Rev Bras Epidemiol. 2011;14:187–97. 21 Blutinger EJ, Solomon S, Srikrishnan AK, et al. Dropout from care among HIV-infected patients enrolled in care at a tertiary HIV care center in Chennai, India. AIDS Care. 2014;11:6. 22 Meijerink H, Wisaksana R, Iskandar S, et al. Injecting drug use is associated with a more rapid CD4 cell decline among treatment naive HIV-positive patients in Indonesia. J Int AIDS Soc. 2014;17:1–7. 23 World Health Organization. HIV Treatment and Care: What’s New in Monitoring. Fact Sheet. 2015. DOI:10.1186/1471-2458-11-553.PMID. 24 International Labour Organization (ILO). Indonesia : Trends in Wages and Productivity. Jakarta, Indonesia, 2015. 25 Drummond MF, Sculpher MJ, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes. New York: Oxford University Press; 2005. 26 Bank Indonesia. Foreign Exchange Rates. 2015. http:// www.bi.go.id/en/moneter/informasi-kurs/referensi- Vol 48 • Number 3 • July 2016 HIV patients drop out in Indonesia 215 jisdor/Default.aspx (accessed Nov 24, 2015). 27 The World Bank. World Development Indicators. 2015. http://data.worldbank.org/products/wdi. 28 Siregar AYM, Tromp N, Komarudin D, et al. Costs of HIV/AIDS treatment in Indonesia by time of treatment and stage of disease. BMC Health Serv Res. 2015;15:440. 29 Granich RM, Gilks CF, Dye C, De Cock KM, Williams BG. Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model. Lancet. 2009;373:48–57. 30 De Cock KM, Gilks CF, Lo Y-R, Guerma T. Can antiretroviral therapy eliminate HIV transmission? Lancet. 2009;373:7–9. 31 Stover J, Bertozzi S, Gutierrez J-P, et al. The global impact of scaling up HIV/AIDS prevention programs in low- and middle-income countries. Science. 2006; 311:1474–6. 32 Siregar AYM, Komarudin D, Wisaksana R, van Crevel R, Baltussen R. Costs and outcomes of VCT delivery models in the context of scaling up services in Indonesia. Trop Med Int Heal. 2011;16:193–9. 33 Portelli MS, Tenni B, Kounnavong S, Chanthivilay P. Barriers to and facilitators of adherence to antiretroviral therapy among people living with HIV in Lao PDR: A qualitative study. Asia Pac J Public Health. 2012;53. DOI:10.1177/1010539512442082. 34 Brinkhof MWG, Pujades-Rodriguez M, Egger M. Mortality of patients lost to follow-up in antiretroviral treatment programmes in resource-limited settings: systematic review and meta-analysis. PLoS One. 2009; 4:e5790. 35 Kitajima T, Kobayashi Y, Chaipah W, Sato H, Chadbunchachai W, Thuennadee R. Costs of medical services for patients with HIV/AIDS in Khon Kaen, Thailand. AIDS. 2003;17:2375–81. 36 Afriandi I, Aditama TY, Mustikawati D, Oktavia M, Alisjahbana B, Riono P. HIV and injecting drug use in Indonesia: epidemiology and national response. Acta Med Indones-Indones J Intern Med. 2009;41 (Suppl 1):75–8. 37 Curran J, Debas H, Arya M, Knobler S, Pray L. Scaling up treatment for the Global AIDS pandemic: Challenges and opportunities. Washington: The National Academies Press; 2005. 38 UNAIDS. HIV and AIDS estimates (2013) Indonesia. 2015. http://www.unaids.org/en/regionscountries/ countries/indonesia (accessed June 27, 2015). 39 Nelwan EJ, Crevel R Van, Alisjahbana B, Indrati AK, Dwiyana RF. Human immunodeficiency virus, hepatitis B and hepatitis C in an Indonesian prison: prevalence, risk factors and implications of HIV screening. Trop Med Int Heal. 2010;15:1491–8. 40 World Health Organization. Scaling up HIV/AIDS care: service delivery & human resources perspectives. 2004. 41 Nachega JB, Knowlton AR, Deluca A, et al. Treatment supporter to improve adherence to antiretroviral therapy in HIV-infected South African adults. A qualitative study. J Acquir Immune Defic Syndr. 2006; 43 (Suppl 1):S127–33. 42 Unge C, Södergård B, Marrone G, et al. Long-term adherence to antiretroviral treatment and program drop-out in a high-risk urban setting in sub-Saharan Africa: A prospective cohort study. PLoS One. 2010; 5. DOI:10.1371/journal.pone.0013613. 43 Simoni JM, Huh D, Frick PA, et al. Peer support and pager messaging to promote antiretroviral modifying therapy in Seattle: a randomized controlled trial. J Acquir Immune Defic Syndr. 2009;52:465–73. 44 Finitsis DJ, Pellowski J a., Johnson BT. Text message intervention designs to promote adherence to antiretroviral therapy (ART): A meta-analysis of randomized controlled trials. PLoS One. 2014;9. DOI:10.1371/journal.pone.0088166. 45 Decroo T, Telfer B, Biot M, et al. Distribution of Antiretroviral Treatment Through Self-Forming Groups of Patients in Tete Province, Mozambique. JAIDS J Acquir Immune Defic Syndr. 2011;56:e39–44. 46 Bärnighausen T, Chaiyachati K, Chimbindi N, Peoples A, Haberer J, Newell M-L. Interventions to increase antiretroviral adherence in sub-Saharan Africa: a systematic review of evaluation studies. Lancet Infect Dis. 2011;11:942–51. 47 Pearson CR, Micek MA, Simoni JM, et al. Randomized control trial of peer-delivered, modified directly observed therapy for HAART in Mozambique. JAIDS J Acquir Immune Defic Syndr. 2007;46:238–44. 48 Ikeda JM, López Tellez CA, Hudes ES, et al. Impact of integrating HIV and TB care and treatment in a regional tuberculosis hospital in rural Guatemala. AIDS Behav 2014;18. DOI:10.1007/s10461-013-0595-9. 49 Coetzee D, Hilderbrand K, Goemaere E, Matthys F, Boelaert M. Integrating tuberculosis and HIV care in the primary care setting in South Africa. Trop Med Int Health. 2004;9:A11–5. 50 Abdool Karim SS, Naidoo K, Grobler A, et al. Integration of antiretroviral therapy with tuberculosis treatment. New Engl J Med. 2011;365:1492–501. 51 Legido-Quigley H, Montgomery CM, Khan P, et al. Integrating tuberculosis and HIV services in low- and middle-income countries: a systematic review. Trop Med Int Heal. 2013;18:199–211. 52 Mahendradhata Y, Ahmad RA, Lefèvre P, Boelaert M, Van der Stuyft P. Barriers for introducing HIV testing among tuberculosis patients in Jogjakarta, Indonesia: a qualitative study. BMC Public Health. 2008;8:385. 53 World Bank. GDP (Current US$) World Development Indicators. 2015. http://data.worldbank.org/indicator/ NY.GDP.MKTP.CD (accessed June 30, 2015). 54 World Bank. Health Expenditure, Total (% of GDP) World Development Indicators. 2015. http://data. worldbank.org/indicator/SH.XPD.TOTL.ZS (accessed Adiatma YM. Siregar Acta Med Indones-Indones J Intern Med 216 June 30, 2015). 55 The World Bank. Life Expectancy at Birth, Total (years). 2015. http://data.worldbank.org/indicator/SP.DYN. LE00.IN/countries/ID-4E-XN?display=default (accessed Aug 3, 2015). 56 Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493–505. 57 Mathers BM, Degenhardt L, Ali H, et al. HIV prevention, treatment, and care services for people who inject drugs: a systematic review of global, regional, and national coverage. Lancet. 2010;375:1014–28. 58 The World Bank. Indonesia’s Health Sector Review: Overview. 2010. 59 Waldrop-Valverde D, Jones DL, Weiss S, Kumar M, Metsch L. The effects of low literacy and cognitive impairment on medication adherence in HIV-positive injecting drug users. AIDS Care. 2008;20:1202–10. 60 Braitstein P, Brinkhof MWG, Dabis F, et al. Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: comparison between low-income and high- income countries. Lancet. 2006;367:817–24.