A large increase in lay healthcare workers has occurred in response to shortages of professional healthcare staff in sub-Saharan African antiretroviral treatment (ART) programs. However, little effectiveness data of the large-scale implementation of these programs is available. We evaluated the effect of a community-based adherence-support (CBAS) program on ART outcomes across 57 South African sites.
CBAS workers provide adherence and psychosocial support for patients and undertake home visits to address household challenges affecting adherence. An observational multicohort study of adults enrolling for ART between 2004 and 2010 was performed. Mortality, loss to follow-up (LTFU) and virological suppression were compared by intention-to-treat between patients who received and did not receive CBAS until five years of ART, using multiple imputation of missing covariate values.
66,953 patients were included, of whom 19,668 (29.4%) received CBAS and 47,285 (70.6%) did not. Complete-case covariate data were available for 54.3% patients. After five years, patient retention was 79.1% (95% CI: 77.7%-80.4%) in CBAS patients vs. 73.6% (95% CI: 72.6%-74.5%) in non CBAS patients; crude hazard ratio (HR) for attrition 0.68 (95% CI: 0.65-0.72). Mortality and LTFU were independently lower in CBAS patients, adjusted HR 0.65 (95% CI: 0.59-0.72) and aHR 0.63 (95% CI: 0.59-0.68), respectively. After six months of ART, virological suppression was 76.6% (95% CI: 75.8%-77.5%) in CBAS patients vs. 72.0% (95% CI: 71.3%-72.5%) in non-CBAS patients (P<0.0001), adjusted odds ratio (aOR) 1.22 (95% CI: 1.14-1.30). Improvement in virological suppression occurred progressively for longer durations of ART (aOR 2.66 [95% CI: 1.61-4.40] by 5 years).
Patients receiving CBAS had considerably better ART outcomes. Further scale-up these programs should be considered in low-income settings.