Kipp W, Konde-Lule J, Saunders LD, Alibhai A, Houston S, Rubaale T, et al. Results of a community-based antiretroviral treatment program for HIV-1 infection in Western Uganda. Curr HIV Res. 2010 Mar;8(2):179-85
A very large proportion of those with HIV in least-developed countries (LDC), particularly in sub-Saharan Africa, live in rural areas that are poorly accessible. Finding ways to extend quality HIV/AIDS services from urban and peri-urban areas is paramount. In Uganda, for example, 80% of the population is rural, and only about 50% of all those in need of ART are receiving it. Several different models of rural-based care exist, although few have published outcomes or comparisons to standard facility-based care.(1, 2, 3)
To compare treatment outcomes and mortality between a rural community-based ART (CBART) program and a hospital-based ART program in the same district.
A Health Center (HC) in small rural sub-county in southwest Uganda, and a hospital 50 km away in the main District Hospital.
Non-randomized controlled intervention study
Treatment-naïve ART-eligible adults
HIV sero-positive adults ≥18 yrs with CD4 count of ≤200cell/µL and treatment-naïve were consecutively enrolled at both sites. Rural (intervention) patients had to be residents and willing to accept treatment support. Enrollment took place over 14 months for the 185 rural patients, and 6 months for the 200 hospital-based patients (control). Subjects in both cohorts were followed for 2 years. For the intervention, 41 community volunteers were selected and provided a 2-day training on ART, drug reactions and adherence. Each volunteer was assigned 5 patients to be visited weekly, during which they performed pill counts, assessed adverse reactions, and made referrals. Each month the volunteers delivered ARVs from the local HC. They also provided prevention information and condoms. Patients chose a family or friend to be a daily treatment supporter who completed a pill log form for twice daily pill dosing. A clinical officer familiar with the community trained and regularly supervised the volunteers. Volunteers were motivated by small non-cash incentives such as bicycles, raincoats and boots, which also allowed them to carry out their work. Patients in the control cohort received the standard of care for the facility, which consisted of monthly treatment monitoring and medication refills; adherence was not calculated systematically as part of standard care. First line treatment was stavudine, lamivudine and nevirapine (with efavirenz for those on TB treatment); cotrimoxazole was also prescribed. Viral load was measured at 6 months as a research outcome, rather than standard of care. Additional primary outcomes were mortality and change in CD4 cell count, both at 6 months. Baseline viral load and CD4 counts were measured only in the CBART cohort. The maximum observation time was 2 years for patients initially enrolled.
The two cohorts were similar in mean age (34.8-36.2 yrs), gender (41-43.5% male), and mean CD4 count (136-146 cells/µL). Control participants were more likely to have secondary education (22% vs. 12%), be single (30% vs. 14%), and be in business (52%) (all p<0.05). During the first 6 months, 24 patients died in the CBART group and 18 in the hospital group; 22 in CBART were LFU, and 16 in the hospital cohort. At 6 months, a rise in CD4 counts was similar in both groups (159 cells/µL and 145 cells/µL, p<0.001 in CBART and hospital, respectively). 124 (90%) of patients in CBART had viral suppression (<400 copies/mL, compared to 147 (80%) in hospital based cohort (p=ns). In multivariate analysis, with viral suppression as the outcome, being in the intervention arm was not significantly associated with viral suppression; only being female (OR=2.2, 95% CI: 1.0,4.7), and being widowed/divorced/separated (OR=0.22, 95% CI: 0.06,0.84) were independently predictive. Mortality was slightly higher in CBART than the control cohort, but not statistically different (27.1 vs. 20.0 deaths per 100 person-years [p100py], p=0.30). Early deaths were more frequent in the CBART cohort. Authors state that adherence in the control cohort was 98%.
The authors conclude that providing community-based ART treatment to patients in rural areas who could not otherwise travel to a health facility, is feasible and can result in comparable care to local hospital-based standard of care, at least during the first 6 months.
This was a good quality intervention study. Because data for the hospital-based cohort was obtained from routine records, adherence information was not available for this arm. How adherence was calculated in the intervention arm was not described.
This study shows that with appropriate support, using community volunteers, ART can be effectively provided with similar short-term outcomes to patients in rural Uganda. Although the study demonstrated that response to treatment and deaths were the same among patients in the community and in the hospital clinics, only treatment response at 6 months was presented. Data are still needed on the extent to which longer term outcomes can be maintained using this model. Nevertheless, other programs need to explore innovative means to deliver care to those who can otherwise not access it. Some of the success of this program is likely do to the level of community involvement provided, facilitated by the support of an indigenous organization that has been working in Uganda for more than 20 years.
- Kaleebu P, Pillay D, Walker AS, et al. Virological response to a triple nucleoside/nucleotide analogue regimen over 48 weeks in HIV-1-infected adults in Africa. AIDS 2006; 20(10): 1391-9.
- Bekker LG, Egger M, Wood R. Early antiretroviral therapy mortality in resource-poor settings: What can we do about it? Curr Opin HIV-1 AIDS 2007; 2(4): 346-51.
- Weidle P, Wamai N, Solberg P, et al. Adherence to antiretroviral therapy in a home-based AIDS care programme in rural Uganda. Lancet 2006; 368(9547): 1587-94.