The Kanyakla study: Randomized controlled trial of a microclinic social network intervention for promoting engagement and retention in HIV care in rural western Kenya.

Background: Existing social relationships are a potential source of "social capital" that can enhance support for sustained retention in HIV care. A previous pilot study of a social network-based 'microclinic' intervention, including group health education and facilitated HIV status disclosure, reduced disengagement from HIV care. We conducted a pragmatic randomized trial to evaluate microclinic effectiveness.

Methods: In nine rural health facilities in western Kenya, we randomized HIV-positive adults with a recent missed clinic visit to either participation in a microclinic or usual care (NCT02474992). We collected visit data at all clinics where participants accessed care and evaluated intervention effect on disengagement from care (≥90-day absence from care after a missed visit) and the proportion of time patients were adherent to clinic visits ('time-in-care'). We also evaluated changes in social support, HIV status disclosure, and HIV-associated stigma.

Results: Of 350 eligible patients, 304 (87%) enrolled, with 154 randomized to intervention and 150 to control. Over one year of follow-up, disengagement from care was similar in intervention and control (18% vs 17%, hazard ratio 1.03, 95% CI 0.61-1.75), as was time-in-care (risk difference -2.8%, 95% CI -10.0% to +4.5%). The intervention improved social support for attending clinic appointments (+0.4 units on 5-point scale, 95% CI 0.08-0.63), HIV status disclosure to close social supports (+0.3 persons, 95% CI 0.2-0.5), and reduced stigma (-0.3 units on 5-point scale, 95% CI -0.40 to -0.17).

Conclusions: The data from our pragmatic randomized trial in rural western Kenya are compatible with the null hypothesis of no difference in HIV care engagement between those who participated in a microclinic intervention and those who did not, despite improvements in proposed intervention mechanisms of action. However, some benefit or harm cannot be ruled out because the confidence intervals were wide. Results differ from a prior quasi-experimental pilot study, highlighting important implementation considerations when evaluating complex social interventions for HIV care.

Trial Registration: Clinical trial number: NCT02474992.

Competing Interests: G Ouma, B Mattah, P Mohamed, J Obanda, and A Odhiambo are employed by the Ekialo Kiona Centre, a Community Based Organization in Kenya and the local implementing organization for the microclinic/kanyakla intervention. M Hickey, B Pedersen, N DesLauriers, K Fiorella, and C Salmen are unpaid members of the Board of Directors of Organic Health Response (organichealthresponse.org), a US-based 501c3 nonprofit organization that supports the Ekialo Kiona Centre. D Zoughbie is employed by, and E Ding provides consultation for Microclinic International (microclinics.org), a US-based 501c3 nonprofit that provided technical support for intervention design and implementation. Microclinic International also received an unrestricted grant from a commercial source to support this work: Google Inc (administered via the Tides Foundation). These disclosures do not alter our adherence to PLOS ONE policies on sharing data and materials.