Out-of-Facility Multimonth Dispensing of Antiretroviral Treatment: A Pooled Analysis Using Individual Patient Data From Cluster-Randomized Trials in Southern Africa

Background: Out-of-facility multi-month dispensing (MMD) is a differentiated service delivery model which provides antiretroviral treatment (ART) at intervals of up to 6 monthly in the community. Limited randomized evidence investigating out-of-facility MMD is available. We evaluated participant outcomes and compared out-offacility MMD models using data from cluster-randomized trials in Southern Africa. Setting: Eight districts in Zimbabwe and Lesotho. Methods: Individual-level participant data from 2 cluster-randomized trials that included stable adults receiving ART at 60 facilities were pooled. Both trials had 3 arms: ART collected 3-monthly at healthcare facilities (3MF, control); ART provided three-monthly in community ART groups (CAGs) (3MC); and ART provided 6-monthly in either CAGs or on an individual provider-patient basis (6MC). Participant retention, viral suppression and incidence of unscheduled facility visits were compared. Results: Ten thousand one hundred thirty-six participants were included, 3817 (37.7%), 2893 (28.5%) and 3426 (33.8%) in arms 3MF, 3MC and 6MC, respectively. After 12 months, retention was non-inferior for 3MC (95.7%) vs. 3MF (95.0%) {adjusted risk difference (aRD) = -0.3 [95% confidence interval (CI): 20.8 to 1.4]}; and 6MC (95.1%) vs. 3MF [aRD = 20.2 (95% CI: 21.4 to 1.0)]. Retention was greater amongst intervention arm participants in CAGs versus 6MC participants not in CAGs, aRD = 1.5% (95% CI: 0.2% to 2.9%). Viral suppression was excellent (>= 98%) and unscheduled facility visits were not increased in the intervention arms. Conclusions: Three and 6-monthly out-of-facility MMD was noninferior versus facility-based care for stable ART patients. Out-offacility 6-monthly MMD should incorporate small group peer support whenever possible.