Comparison of community-led distribution of HIV self-tests kits with distribution by paid distributors: a cluster randomised trial in rural Zimbabwean communities


BACKGROUND: Community-based HIV self-testing (HIVST) increases testing uptake and ART linkage. Sustainable distribution is required. We evaluated community-led HIVST in rural Zimbabwe.
METHODS: Forty 'village groups' were randomly allocated to i) community-led HIVST; ii) paid distribution (PD). In the community-led arm, communities developed and implemented HIVST models. In the PD arm, distributors were paid US$50 to distribute kits door-to-door. Distribution was for four weeks. Four months post-distribution, we conducted a population representative survey, with self-reported primary outcomes analysed using random-effects logistic regression: i) proportion of new HIV diagnoses; ii) composite outcome - linkage to confirmatory testing, pre-exposure prophylaxis (PrEP) or voluntary medical male circumcision (VMMC). We compared provider's distribution costs of community-led intervention with costs of new (<1 year's implementation) and mature (>1 year's implementation) PD programs.
We conducted a time-series analysis on monthly ART initiations in all study district clinics six months before, during and three months after HIVST distribution.
RESULTS: From January-December 2019, 27,812 and 36,699 HIVST kits were distributed in community-led and PD communities. Five community-led clusters only distributed kits door-to-door; in others kits were also available at other locations. We surveyed 11,150 participants; HIVST coverage was 21.6% and 27.5% in the community-led and PD arms respectively. There were no differences in primary outcomes: new HIV diagnosis was reported by 223 (4%) community-led arm versus 190 (3.4%) PD arm participants, AOR 1.19 (0.82-1.73); 315 (26.1%) community-led arm participants linked to confirmatory testing, PrEP or VMMC, versus 364 (23.8%) in PD arm, AOR 1.09 (0.79-1.51). Sub-group analysis showed no differences by age or sex.
We recorded 5,302 ART initiations at 133 clinics, with no difference in initiations in clinics within and outwith HIVST clusters, RR 0.94 (0.83-1.02). In post-hoc analysis ART initiations increased during HIVST distribution across all facilities, AOR 1.30 (1.24-1.37), falling to baseline levels post-distribution. Cost per HIVST kit distributed in community-led arm was US$14.52, compared with US$14.52 and US$10.63 in new and mature PD programs.
CONCLUSIONS: Community-led HIVST can perform as well as paid distribution, with similar costs in first program year. As seen with PD programs, these costs may reduce with program maturity/learning. Community-based HIVST improves ART uptake.