Specification of implementation interventions to address the cascade of HIV care and treatment in resource-limited settings: a systematic review

Matthew D. Hickey, Thomas A. Odeny, Maya Petersen, Torsten B. Neilands, Nancy Padian, Nathan Ford, Zachary Matthay, David Hoos, Meg Doherty, Chris Beryer, Stefan Baral, Elvin H. Geng

The global response to HIV has started over 18 million persons on life-saving antiretroviral therapy (ART)-the vast majority in low- and middle-income countries (LMIC)-yet substantial gaps remain: up to 40% of persons living with HIV (PLHIV) know their status, while another 30% of those who enter care are inadequately retained after starting treatment. Identifying strategies to enhance use of treatment is urgently needed, but the conceptualization and specification of implementation interventions is not always complete. We sought to assess the completeness of intervention reporting in research to advance uptake of treatment for HIV globally. We carried out a systematic review to identify interventions targeting the adult HIV care cascade in LMIC dating from 1990 to 2017. We identified components of each intervention as "intervention types" to decompose interventions into common components. We grouped "intervention types" into a smaller number of more general "implementation approaches" to aid summarization. We assessed the reporting of six intervention characteristics adapted from the implementation science literature: the actor, action, action dose, action temporality, action target, and behavioral target in each study. In 157 unique studies, we identified 34 intervention "types," which were empirically grouped into six generally understandable "approaches." Overall, 42% of interventions defined the actor, 64% reported the action, 41% specified the intervention "dose," 43% reported action temporality, 61% defined the action target, and 69% reported a target behavior. Average completeness of reporting varied across approaches from a low of 50% to a high of 72%. Dimensions that involved conceptualization of the practices themselves (e.g., actor, dose, temporality) were in general less well specified than consequences (e.g., action target and behavioral target). The conceptualization and Reporting of implementation interventions to advance treatment for HIV in LMIC is not always complete. Dissemination of standards for reporting intervention characteristics can potentially promote transparency, reproducibility, and scientific accumulation in the area of implementation science to address HIV in low- and middle-income countries.