Antiretroviral Therapy Adherence Interventions in the Era of Universal Test and Treat: A Hybrid Systematic-Narrative Literature Review of Global Evidence.

Abstract

An undetectable viral load (VL) in people living with HIV (PWH) is key to both individual and public health success. But for the millions of PWH on oral antiretroviral therapy (ART) worldwide, this requires consistent, sustained adherence. Review of interventions to support adherence published in recent literature can provide insights into promising and effective strategies. We conducted a hybrid systematic-narrative literature review to explore optimal adherence strategies in the era of universal test-and-treat. We searched PubMed, Scopus, and Web of Science according to PRISMA guidelines for peer-reviewed studies, available in English, including people ≥ 12 years old taking ART, published between 01 January 2015 and 18 January 2024. We extracted data on the included studies and the adherence interventions (strategies used -allocated to one of 14 a priori categories or 'other', measures of adherence, and intervention outcomes). Descriptive statistics were used for study information and those interventions with a positive effect were described narratively. We extracted data from 230 studies evaluating a total of 262 interventions among 97,037 PWH. Most studies enrolled participants in Africa (106, 46%) or North America (80, 35%). The majority randomized participants (215, 94%), including 30 cluster-randomized trials. Most included general HIV clinic populations, with 51 (22%) focused on youth and 23 (10%) on pregnant and post-partum women. Many (146, 64%) used VL as an outcome. Self-reported adherence was also a commonly used outcome (129, 56%), but a minority used self-reported measures alone (36, 16%). The most common intervention strategies included across the 262 interventions were eHealth/ mHealth technologies (90, 34%) and adherence-focused counseling (81, 31%). The majority of interventions had 'other' intervention features (133, 51%), typically combined with one or more of the a priori-defined strategies (107, 80%). Most studies evaluated an approach with multiple strategies packaged into a single intervention (k = 182/262, 70%).The majority of interventions had some evidence of effect on an adherence outcome (k = 159, 61%). In studies reporting VL outcomes, 52% (k = 87/166) found some evidence of effect, while 28% (k = 47/166) found significant effects. Intervention strategies demonstrating significant impact on VL included task-shifting and changing dispensing schedules (3/5, 60% in both), while nearly half the evaluations of economic strategies demonstrated significant impact on VL (10/21). A number of different adherence intervention strategies have the potential to impact viral suppression in different populations. Variability in intervention strategies and the resulting outcomes, supports calls to target interventions to PWH who are most likely to benefit, while at the same time addressing social determinants of health and reducing barriers to accessing care to make services more person-centered. Greater attention to evaluating flexible, tailored, complex interventions may offer valuable insights for moving towards the next generation of highly generalizable, sustainable adherence support.