Abstract
US President's Emergency Plan for HIV/AIDS has been credited with saving 25 million lives in sub-Sahara Africa and, as such, constitutes a preeminent US foreign policy achievement of the twenty-first century. However, the implementation of effective HIV/AIDS pharmacological interventions remains a challenge in rural Kenyan communities. Of particular importance are patient retention and care engagement and their interaction with age disparities that are sensitive to different socioeconomic contexts, as well as time-in-treatment. For the first time, we perform an intermediation and triple interaction intent-to-treat secondary analysis on a social network-based randomized controlled trial. We hypothesize that the temporal interactions of critical demographic features with a treatment/control indicator variable may significantly explain patient retention and that these results are intermediated by social network phenomena. We find that not only does extended time-in-treatment significantly improve primary outcomes, but the threefold interaction along with age and treatment itself is sufficiently flexible to fit the data remarkably well without unnecessary elaboration, an effect that is mediated via internalized stigma. This strongly suggests that patient retention varies by age group. Rather than deploying one-size-fits-all solutions, foreign and public policymakers should invest in research that considers how interventions might be optimized for different ages.Trial registration Clinical Trial Number. NCT02474992 (note: the main trial report was published here https://doi.org/10.1371/journal.pone.0255945 .) Date of submission: June 6, 2015.