Antiretroviral therapy (ART) has led to tremendous improvement in quality of life of people living with HIV (PLHIV)1–5 and is substantially more cost-effective today than it was a decade ago.6–9 However, lifelong medication adherence, frequent clinic visits for clinical monitoring, missed employment days, and out-of-pocket expenses pose major opportunity costs to poor PLHIV10–13 and are a major reason for loss to follow-up (LTFU) from ART programs in resource-limited settings.14–16 Despite their impact on patient retention, indirect costs are difficult to measure and are therefore often excluded in economic evaluations of ART leading to potential overestimation of cost-effectiveness ratios.6–8,11 Most of these indirect costs are influenced by the model of service delivery including the quality, fee for service, location of clinics, and comprehensiveness of the services.15–22 Alternative models of ART delivery that address the aforementioned barriers may improve patient retention and the cost-effectiveness of ART even further in the near future. However, their costs must be analyzed to determine which combination of interventions yields the greatest improvements in health at affordable costs.2–24
A model of care that integrates socioeconomic support (defined as financial and nonfinancial benefits offered to poor patients based on assessment of needs) as part of ART delivery has been shown to reduce mortality and LTFU in the Reach Out Mbuya HIV/AIDS initiative (ROM), a community-based ART program in Uganda.2,24,25 The resources required to expand this model of care are likely to be high given the growing need for socioeconomic support as part of HIV care for the poor.2,22,24,26 Furthermore, decisions to scale up socioeconomic support have been made in the absence of cost-effectiveness data, which seriously limits attempts to secure funds or to set priorities. An economic evaluation is therefore necessary to determine whether providing socioeconomic support to patients with HIV is worth scaling up compared with providing care and treatment. These questions have direct relevance to the expansion of socioeconomic support within ROM and similar ART programs funded by the United States Government through The President's Emergency Plan for AIDS Relief (PEPFAR) and will facilitate resource allocation and priority setting.
Using secondary cost data and retention outcomes from a cohort of patients receiving care from ROM,24 a cost-effectiveness analysis of socioeconomic support was conducted. We evaluated the food and education support programs, which constituted 70% and 26%, respectively, of socioeconomic support provided between 2001 and 2010.2,24 We compared the incremental benefits and costs of providing education support alone, food support alone, and both education and food support (dual support) when added onto ART. The study was approved by the institutional review boards of Makerere University School of Public Health and the National Council of Science and Technology, Uganda.
This study was conducted at ROM, an indigenous Ugandan Nongovernmental organization that has been extensively described before.2,24,25,27 In brief, ROM is one of the comprehensive ART programs in Kampala, Uganda. By March 2004, it was providing ART using the Ugandan Ministry of Health national guidelines; and by December 2010, it had cumulatively cared for nearly 10,000 patients. Most of the funding for care, treatment, and support is provided by PEPFAR but some ART and laboratory support is also provided by the Ministry of Health through The Global Fund to Fight Tuberculosis, AIDS, and Malaria. ROM provides a full range of HIV services including socioeconomic support, which is provided to the poor after a needs assessment. A “needs-most-gets-first” principle is applied in prioritizing socioeconomic support with individuals who score the highest vulnerability score index receiving support first.24,25,27 The primary beneficiary of socioeconomic support is an HIV-positive individual enrolled in ROM for care and treatment, who after the needs assessment receives socioeconomic support as an intervention to improve adherence and retention in care.24 The secondary beneficiaries are HIV-negative individuals in the same household who may also benefit from socioeconomic support.
The Socioeconomic Support Interventions
In 2002, ROM initiated an orphans and other vulnerable children (OVC) program majorly comprising school fees and psychosocial support. By 2007, the OVC program had expanded to include all the 7 core OVC program areas, ie, health care, education (school fees and scholastic materials), food (nutrition), legal, and child protection, shelter, economic empowerment, and psychosocial support.24 The underlying imperative for the program is household economic support to women receiving care and treatment from ROM. Women comprise more than 80% of ROM's patients with a significant proportion of them being either widowed or divorced. The impact of the social and economic effects of HIV on adherence in such vulnerable groups is significant.7,8,15,16,20,28–31 The supported patients are selected using scored economic indicators used to assign a vulnerability index score to their household.32–34 Initial support for the OVC program came from individuals, but funding markedly expanded through PEPFAR, which now funds more than 95% of the program.
More than 85% of the OVC funding is for education and psychosocial support, whereas 15% is distributed between food (10%) and other core program areas (5%). The focus of this evaluation will thus be education and psychosocial support referred to together as “education support” henceforth. Children of PLHIV who receive food or both food and education will be evaluated under the alternative interventions.
Children found eligible for education support are enrolled into collaborating preselected schools with consent of their parents/guardians. School fees and scholastic materials (uniforms, shoes bags, books, and pencils) are provided for each school term. Additionally, each child is visited at school by social workers who assess school attendance and performance and address any social issues that may affect school performance and completion rates. Furthermore, each child is visited at home by community health workers based on their health and prevailing socioeconomic situation, with the HIV-infected children being visited more often.27 Psychosocial support activities included individual counseling, peer-led support clubs, and play therapy.
The food support program was initiated in 2002 in collaboration with the World Food Program and targets food insecure households as a strategy to improve adherence to medication. Cumulatively, more than 2000 primary beneficiaries and more than 10,000 secondary beneficiaries have been served. The validated household Food Insecurity Access Scale is used to screen for food insecurity among eligible households.32,33 Food beneficiaries received monthly food rations according to household size. A household of 1–4 individuals receives 1.85 L of cooking oil, 12.4 kg of posho, 4.16 kg of beans, and 8.33 kg of corn soya blend (CSB); a household of 5–7 individuals receives 3.7 L of cooking oil, 25 kg of posho, 8.33 kg of beans, and 16.66 kg of CSB; and a household of 8+ individuals receives 5.5 L of cooking oil, 37.5 kg of posho, 25 kg of beans, and 50 kg of CSB.
Food is distributed from the community clinics and from within the communities for villages that are more than 5 km from the clinics.2,24,26 Patients continuously received food as long as they were in care until 2008 when a phase-out plan was implemented during which households were reassessed every 6 months. Those found to have recovered from food insecurity are replaced to allow more vulnerable beneficiaries to be supported.
Both education and food support is given to households that are found very vulnerable based on assessment of needs. Typically, these households have unemployed HIV-positive adults who are too ill to work and take care of their children.
An empirical cost-effectiveness analysis was performed and is based on a cohort analysis that has been described previously.2,24 In brief, patients from this cohort were accrued from 3 community-based clinics managed by ROM. During the study period of May 2001 to May 2010, socioeconomic support was availed to all eligible patients in care.24 For the economic evaluation, patients who were no longer active in care by May 2004 were excluded as large-scale ART first became available at ROM in May 2004.
We compared the costs and outcomes between the 3 different types of socioeconomic support (education support only, food support only, or dual support) given to patients with HIV receiving services from ROM. The principal measure of effect is the averted number of cases LTFU. Mortality was not considered as an endpoint because the intervention groups were not comparable clinically, with the group receiving food support having lower CD4 counts and being more likely to be on ART than the education support group.24 The study adopts a health care provider perspective.
Health Outcomes Data
The health outcome was the number of patients LTFU categorized by the type of socioeconomic support they received during the study period. Patients were classified as “LTFU” if they were enrolled with ROM but had not had contact with the facility for 90 days or more after their scheduled follow-up date and were not known to have died or transferred out to another facility.2,24,29,30 Patients were categorized as “active” if they were still continuing care at ROM and had had at least 1 contact with the facility during the 90 days before the end of the study follow-up period. Health outcomes were counted over the complete study follow-up period (May 2004–May 2010).
Detailed cost data of socioeconomic services received by the patients during the study period were extracted from the computerized financial information systems supplemented by procurement records using standard microcosting methods and included the following:
- Cost per patient accrued over the total follow-up time was obtained for all patients for the alternative forms of socioeconomic support and included supplies and recurrent services (assessment cost, postdistribution costs, school visits, and psychosocial support). Additionally, cost per secondary beneficiary was calculated (cost per child per household), which provided information that was supplementary to our analytical focus, ie, their incremental cost per LTFU case averted.
- Patient-independent program costs were collected for each month as far as possible and divided into 2 categories: personnel costs and administrative costs.
Economic cost was computed as the true resource cost incurred rather than just the financial cost. Salary cost was recorded for all personnel contributing directly to implementing either education or food support services. In addition, cash compensation paid to workers for transport and overtime hours was computed. However, costs of administrative and other staff shared between the 2 socioeconomic support programs were excluded.
Costs were adjusted for inflation to financial year 2010 prices using the Bank of Uganda Consumer Price Index35,36 and were converted from Uganda shillings to USD (“$” henceforth) using the average exchange rate of 1$ to 2369 Uganda shillings for the year 2010.35,36
We developed a computer-based, numerical cost-effectiveness model (decision tree) implemented in Microsoft Excel. Calculated costs and health outcomes were used to estimate the incremental cost-effectiveness between the 3 alternative interventions when added onto the existing ART program. The total costs for each intervention consisted of the service costs for all the patients receiving that type of support for the complete duration of the follow-up period and the patient-independent fixed program costs. The total costs and health outcomes were scaled by the number of patients within each of the programs. Cost-effectiveness is portrayed as the net program cost divided by the net intended effect of one optional intervention compared with another. Thus, the 3 types of socioeconomic support were compared with each other using costs per LTFU case averted as a measure of technical efficiency.
The patient outcomes are as shown in Figure 1. Of 2383 patients who received either education or food support, 12 had not been retained in care by the beginning of May 2004 and were excluded from the study. Thus, 2371 patients were evaluated. After 7 years of follow-up, 762 (33%) were LTFU with 42% of them being from the food support program. The average household size of the food beneficiaries was 6.6, therefore a total of 10,240 individuals received food either as primary or secondary beneficiaries. The education support program, on the other hand, targeted individual children, and an average of 1.07 children was supported per household, totaling 570 supported children belonging to 545 patients who were evaluated.
Table 1 shows the discounted annual costs for the 3 intervention groups for the period 2004–2010. The average cost per patient through the study period or education support was $237, $538 for food support, and $776 for the dual support program. The average total annual costs were $88,643 for education support, $538,005 for food support, and $103,045 for the dual support program.
Table 2 shows the cost profile for the period 2004–2010. Overall, supplies accounted for the highest proportion of the program costs across all 3 interventions with an average of 72%, 96%, and 86% for the education support, food, and dual programs, respectively. Personnel costs accounted for less than 10% of the program costs across all years and were independent of the number of patients enrolled in each program.
Incremental Costs and Cost-Effectiveness
LTFU in the education support arm was 12% at a unit cost of $831 over the study period. Whereas, the food program lost 42% of the beneficiaries to follow-up at a unit cost of $2065. The food and dual support programs are both dominated by the education support program because they are more costly and have more patients LTFU (Table 3).
The cost-effectiveness analysis showed that in the presence of providing ART, education support is a less costly and more effective intervention compared with food or the dual support programs. We could not assess the counterfactual, that is, the cost-effectiveness of education compared with no socioeconomic support at all, because socioeconomic support is the current standard of care at ROM.2,24 In the absence of a “no intervention” or control option, our study focuses on technical efficiency (ie, comparing the program options with each other) and not on allocative efficiency (ie, comparing these program options with any other health care intervention). This implies that we did not need to convert the averted LTFU cases into (quality-adjusted or disability-adjusted) life years to judge which of the 3 options is most efficient. A consequence is that we can compare these costs and effects with other programs averting HIV LTFU cases but not with programs impacting other outcomes. To enable the latter, we would need to convert each averted LTFU case in a more generic health outcome (such as DALYs).
Socioeconomic status may influence the diagnosis, prevention, and management of HIV as a consequence of differing access to medical services, a differing standard of service once accessed, late referral to treatment or specialist care, differing compliance with interventions, and differing outcomes of interventions. Education support is associated with LTFU through opportunity cost to pay for transport for clinic appointments or motivation to continue care in the interest of a child's education.37–40 Among PLHIV, food insecurity inhibits ART initiation, retention in care, and ART adherence37–40 and leads to worse virologic and immunologic outcomes,40–44 increased hospitalizations, opportunistic infections,40–43 and mortality.45–49 In turn, HIV/AIDS worsens food insecurity by eroding economic productivity,50–53 reducing social support due to HIV stigma,52 and increasing medical expenses.7,13,19–21,26,51,52 This cycle is particularly devastating in Uganda, where AIDS is the leading cause of mortality15,16,45–49 and where food insecurity affects nearly three-quarters of PLHIV.40–42,51,52
Our study is the first economic evaluation of food and education support added onto an ART program within the Ugandan context and may have implications for scaling up these services. The high costs of the food program in particular have direct implications for its scale-up and sustainability. Overall, the unit cost of education support remained stable through the study period with an average unit cost of $237. In addition to being higher than the unit cost for education support across all the years, there were also wide inter-year variations in unit cost per household of food reflecting the higher influence of inflation on the cost of food inputs, which accounts for 96% of the program costs. The unit cost for the dual program was $882, making it a much more expensive program to fund.
ROM will have to seek higher efficiencies in the food program by identifying savings across the range of costs without compromising the quality. It is unlikely that ROM can improve efficiency through improved personnel productivity, which only accounted for on average 1.8% of the food program costs and 9.9% of the education program costs. The low personnel costs are explained by the task-shifting model and the use of community health workers, whose health care roles are integrated at the household level to include assessments, monitoring of OVCs, and food monitoring leading to gain in efficiencies. Hiring from within the community boosts the community capacity while reinforcing a deep sense of community ownership.10,26,27,54–57 The relatively low psychosocial support costs reflected in the dual program could be explained by the fact that these households are the most vulnerable and their children are engaged in household chores with no time to attend psychosocial support activities.
With improvement in quality of life of PLHIV, many will return to employment or become more productive.2 Major gains in cost-effectiveness in the food program are likely to derive from shorter durations of support or a reduction in the food rations to households. In addition, there is an emerging consensus on the need to scale down on food support and, where feasible, combine food support with livelihood strengthening, which is increasingly regarded as essential for improving food security and resilience over the long-term.39,44,58 ROM needs to embrace this paradigm shift.
Coverage is a crucial contextual factor for the cost-effectiveness of interventions. We compared the interventions in light of the current supply of services. Although education support was found to be cost-effective, it has a much lower coverage per patient than the food support program (9% vs 21%). This is a neglected opportunity that ROM should address.
This analysis is not without limitations. We used LTFU as a measure of health effect. Several studies1,3,4,16,20 have reported high mortality among LTFU. However, a tracing study3 found only 11% mortality among our LTFU patients, which is attributed to the presence of an organized community system of care that facilitates early diagnosis, recognition, and management of side effects and prompt referrals. Additionally, same-day tracing of LTFU patients ensures that those who have died are documented within few days of a missed clinic appointment.
We chose a narrow definition of costs, focused exclusively on the direct variable costs of providing services, and excluded costs associated with the use of existing infrastructure. ROM is predominantly a health facility and socioeconomic support that comprises less than 15% of the HIV care and treatment budget. In addition, the interventions being compared would incur the same costs on infrastructure further justifying the exclusion of these costs. We did not include the indirect costs of transport and time because we considered them negligible. Clinics are located within the communities and more than 90% of the patients are unemployed.2,24,25
Overall, only 16% of ROMs patients received food support, which in addition to retention, benefits the household through short-term child survival, an aspect that may improve the cost-effectiveness of food support. However, the analysis did not consider the food program's phase-out strategy, which was implemented in 2008. This would ambitiously allow twice the number of households to be supported over the same duration hence reducing the unit cost per household to $269, which is still more expensive than the education program. In contrast, education support has a long-term impact on child survival. We did not include long-term gains in household productivity associated with education support, and therefore it is likely that this aspect would have also shown greater cost-effectiveness of the intervention. Additionally, OVC community-based programming helps to reduce stigma and discrimination and creates an enabling environment for people infected and affected by HIV to access services. By addressing socioemotional effects of the epidemic, OVC programs reduce the likelihood of children and adolescents moving from being affected by the epidemic to being infected. These disregarded benefits make the OVC program even more attractive.
Socioeconomic differentials in health are widely recognized with individuals of lower socioeconomic status having a higher risk for LTFU compared with those of higher socioeconomic status. For our patient cohort enrolled in the ART program, we found that investing in education support is less costly and more effective (fewer LTFU) than investing in food support. The findings strongly suggest the need for a paradigm shift by ROM to channel resources more predictably, efficiently, and effectively to support its response capacity to the education support program. This paradigm shift would better meet the direct, short-term retention needs for patients in care while also laying the foundation for long-term sustainability of households through poverty alleviation.
1. Egger M, May M, Chene G, et al.. Prognosis of HIV-1-infected patients starting highly active antiretroviral therapy: a collaborative analysis of prospective studies. Lancet. 2002;360:119–129.
2. Alamo ST, Colebunders R, Ouma J, et al.. Return to normal life after AIDS as a reason for lost to follow up in a community-based antiretroviral treatment programme, 2012 reason for lost to follow up in a community-based antiretroviral treatment programme. J Acquir Immune Defic Syndr. 2012;60:e36–e45.
3. Hogg RS, Yip B, Kully C, et al.. Improved survival among HIV-infected patients after initiation of triple-drug antiretroviral regimens. CMAJ. 1999;160:659–665.
4. Mocroft A, Vella S, Benfield TL, et al.. Changing patterns of mortality across Europe in patients infected with HIV-1. EuroSIDA Study Group. Lancet. 1998;352:1725–1730.
5. Sterne JA, Hernan MA, Ledergerber B, et al.. Long-term effectiveness of potent antiretroviral therapy in preventing AIDS and death: a prospective cohort study. Lancet. 2005;366:378–384.
6. Badri M, Maartens G, Mandalia S, et al.. Cost-effectiveness
of highly active antiretroviral therapy in South Africa. PLoS Med. 2006;3:e4.
7. Goudge J, Govender V. A Review of Experience Concerning Household Ability to Cope with the Resource Demands of Ill-health and Health Care Utilization. Regional Network for Equity in Health in Southern Africa (Equinet) Policy Series No. 3. Harare, Zimbabwe: Equinet Policy Series. Available at: http://www.equinetafrica.org
. Accessed August 19, 2014.
8. Hogan DR, Baltussen R, Hayashi C, et al.. Cost effectiveness analysis of strategies to combat HIV/AIDS in developing countries. BMJ. 2005;331:1431–1437.
9. Akileswaran C, Lurie MN, Flanigan TP, et al.. Lessons learned from use of highly active antiretroviral therapy in Africa. Clin Infect Dis. 2005;41:376–385.
10. Weidle PJ, Wamai N, Solberg P, et al.. Adherence to antiretroviral therapy in a home-based AIDS care programme in rural Uganda. Lancet. 2006;368:1587–1594.
11. Colebunders R, Kamya M, Semitala F, et al.. Free antiretrovirals must not be restricted only to treatment-naive patients: experience in Uganda suggests that restricting access is not the way forward. PLoS Med. 2005;2:e276.
12. Which resources pay for treatment? A model for estimating the informal economy of health. Soc Sci Med. 1996;42:671–679.
13. Hardon AP, Akurut D, Comoro C, et al.. Hunger, waiting time and transport costs: time to confront challenges to ART adherence in Africa. AIDS Care. 2007;19:658–665.
14. Colebunders R, Ronald A, Katabira E, et al.. Rolling out antiretrovirals in Africa: there are still challenges ahead. Clin Infect Dis. 2005;41:386–389.
15. Merten S, Kenter E, McKenzie O, et al.. Patient-reported barriers and drivers of adherence to antiretrovirals in sub-Saharan Africa: a meta-ethnography. Trop Med Int Health. 15(suppl 1);2010:16–33.
16. Unge C, Sodergard 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:e13613.
17. 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–S133.
18. Stout BD, Leon MP, Niccolai LM. Nonadherence to antiretroviral therapy in HIV-positive patients in Costa Rica. AIDS Patient Care STDs. 2004;18:297–304.
19. Catz SL, Kelly JA, Bogart LM, et al.. Patterns, correlates, and barriers to medication adherence among persons prescribed new treatments for HIV disease. Health Psychol. 2000;19:124–133.
20. Olowookere SA, Fatiregun AA, Akinyemi JO, et al.. Prevalence and determinants of nonadherence to highly active antiretroviral therapy among people living with HIV/AIDS in Ibadan, Nigeria. J Infect Dev Ctries. 2008;2:369–372.
21. Coetzee D, Hildebrand K, Boulle A, et al.. Outcomes after two years of providing antiretroviral treatment in Khayelitsha, South Africa. AIDS. 2004;18:887–895.
22. Simoni JM, Amico KR, Smith L, et al.. Antiretroviral adherence interventions: translating research findings to the real world clinic. Curr HIV/AIDS Rep. 2010;7:44–51.
23. Maskew M, MacPhail P, Menezes C, et al.. Lost to follow-up: contributing factors and challenges in South African patients on antiretroviral therapy. S Afr Med J. 2007;97:853–857.
24. Talisuna-Alamo S, Colebunders R, Ouma J, et al.. Socio-economic support reduces non-retention in care in a comprehensive community based programme in Uganda. J Acquir Immune Defic Syndr. 2012;59:e52–e59.
25. Larry WC, Alamo S, Guma S, et al.. Two year virologic outcomes of an alternative AIDS care model: evaluation of a peer health worker and nurse-staffed community-based programme in Uganda. J Acquir Immune Defic Syndr. 2009;50:276–282.
26. Donahue J. Community-Based Economic Support for Households Affected by HIV/AIDS. Washington, DC: United States Agency for International Development; 1998.
27. Alamo S, Wabwire-Mangen F, Kenneth E, et al.. Task-shifting to community health workers: evaluation of a peer-led model in an antiretroviral programme in Uganda. AIDS Patient Care STDs. 2012;26:101–107.
28. Dalal RP, Macphail C, Mqhayi M, et al. Characteristics and outcomes of adult patients lost to follow-up at an antiretroviral treatment clinic in Johannesburg, South Africa. J Acquir Immune Defic Syndr. 2008;47:101–107.
29. Fox MP, Rosen S. Patient retention in antiretroviral therapy programs up to three years on treatment in sub-Saharan Africa, 2007-2009: systematic review. Trop Med Int Health. 2010;15(suppl 1):1–15.
30. Nachega JB, Mills EJ, Schechter M. Antiretroviral therapy adherence and retention in care in middle-income and low-income countries: current status of knowledge and research priorities. Curr Opin HIV AIDS. 2010;5:70–77.
31. Rosen S, Fox MP, Gill CJ. Patient retention in antiretroviral therapy programs in sub-Saharan Africa: a systematic review. PLoS Med. 2007;4:e298.
32. WFP/UNHCR. UN/Inter-agency Guidelines WFP/UNHCR Joint Assessment Guidelines (JAMs). 2009.
33. FAO/WFP. Joint Guidelines for Crop & Food
Security Assessment Missions (CFSAMs). IASC; 2009. Initial Rapid Assessment (IRA): Guidance Notes & Field assessment form. Available at: http://www.wfp.org
. Accessed August 20, 2014.
34. UNAIDS, UNICEF, USAID, WFP, and FAO. An Update on the National Plans of Action for Orphans and Children Made Vulnerable by HIV/AIDS and the Second Phase of the Rapid Assessment, Analysis and Action Planning Process. New York, NY: UNICEF; 2005.
35. Uganda beaureau of statistics Official Website. Available at: http://www.UBOS.org
. Accessed March 21, 2013.
37. Weiser SD, Tuller DM, Frongillo EA, et al.. Food
insecurity as a barrier to sustained antiretroviral therapy adherence in Uganda. PLoS One. 2010;5:e10340.
38. Goudge J, Ngoma B. Exploring antiretroviral treatment adherence in an urban setting in South Africa. J Public Health Policy. 2011;32(suppl 1):S52–S64.
39. Nagata JM, Magerenge RO, Young SL, et al.. Social determinants, lived experiences, and consequences of household food
insecurity among persons living with HIV/AIDS on the shore of Lake Victoria, Kenya. AIDS Care. 2011;24:1–9.
40. Weiser SD, Frongillo EA, Ragland K, et al.. Food
insecurity is associated with incomplete HIV RNA suppression among homeless and marginally housed HIV-infected individuals in San Francisco. J Gen Intern Med. 2009;24:14–20.
41. Wang EA, McGinnis KA, Fiellin DA, et al.. Food
insecurity is associated with poor virologic response among HIV-infected patients receiving antiretroviral medications. J General Internal Medicine. 2011;26:1012–1018.
42. Weiser SD, Tsai AC, Gupta R, et al.. Food
insecurity is associated with morbidity and patterns of healthcare utilization among HIV-infected individuals in a resource-poor setting. AIDS. 2012;26:67–75.
43. Rawat R, McCoy SI, Kadiyala S. Poor diet quality is associated with low CD4 count and anemia and predicts mortality among antiretroviral therapy naive HIV-positive adults in Uganda. J Acquir Immune Defic Syndr. 2014;66:e15–e22.
44. Kadiyala S, Rawat R. Food
access and diet quality independently predict nutritional status among people living with HIV in Uganda. Public Health Nutr. 2013;16:164–170.
45. Tang AM, Forrester J, Spiegelman D, et al.. Weight loss and survival in HIV-positive patients in the era of highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2002;31:230–236.
46. Zachariah R, Fitzgerald M, Massaquoi M, et al.. Risk factors for high early mortality in patients on antiretroviral treatment in a rural district of Malawi. AIDS. 2006;20:2355–2360.
47. Paton NI, Sangeetha S, Earnest A, et al.. The impact of malnutrition on survival and the CD4 count response in HIV-infected patients starting antiretroviral therapy. HIV Med. 2006;7:323–330.
48. de Pee S, Semba RD. Role of nutrition in HIV infection: review of evidence for more effective programming in resource-limited settings. Food
Nutrition Bulletin. 2010;31:S313–S344.
49. Koethe JR, Lukusa A, Giganti MJ, et al.. Association between weight gain and clinical outcomes among malnourished adults initiating antiretroviral therapy in Lusaka, Zambia. J Acquir Immune Defic Syndr. 2010;53:507–513.
50. Larson B, Fox M, Rosen S, et al.. Early effects of antiretroviral therapy on work performance: preliminary results from a cohort study of Kenyan agricultural workers. AIDS. 2008;22:421.
51. McIntyre D, Thiede M, Dahlgren G, et al.. What are the economic consequences for households of illness and of paying for health care in low-and middle-income country contexts? Soc Sci Med. 2006;62:858–865.
52. Russell S. The economic burden of illness for households in developing countries: a review of studies focusing on malaria, tuberculosis, and human immunodeficiency virus/acquired immunodeficiency syndrome. Am J Trop Med Hyg. 2004;71(suppl 2):147–155.
53. Tsai AC, Bangsberg DR, Emenyonu N, et al.. The social context of food
insecurity among persons living with HIV/AIDS in rural Uganda. Soc Sci Med. 2011;73:1717–1724.
54. Harries AD, Zachariah R, Lawn SD, et al.. Strategies to improve patient retention on antiretroviral therapy in sub-Saharan Africa. Trop Med Int Health. 2010;15(suppl 1):70–75.
55. Rueda S, Park-Wyllie LY, Bayoumi AM, et al.. Patient support and education
for promoting adherence to highly active antiretroviral therapy for HIV/AIDS. Cochrane Database Syst Rev. 2006;3:.
56. Farmer P, Leandre F, Mukherjee JS, et al.. Community-based approaches to HIV treatment in resource-poor settings. Lancet. 2001;358:404–409.
57. Lugada E, Levin J, Abang B, et al.. Comparison of home and clinic-based HIV testing among household members of persons taking antiretroviral therapy in Uganda: results from a randomized trial. J Acquir Immune Defic Syndr. 2010;55:245–252.
58. Weiser SD, Hatcher A, Frongillo EA, et al.. Food
insecurity is associated with greater acute care utilization among HIV-infected homeless and marginally housed individuals in San Francisco. J General Internal Medicine. 2013;28:91–98.