Objective: To assess the efficacy of 2 adherence interventions, medication managers (MM) and medication alarms (ALR), among antiretroviral (ARV)-naive persons with HIV initiating ARV therapy.
Methods: A multicenter, randomized, adherence intervention clinical trial was conducted among participants coenrolled in an HIV ARV strategy study for ARV-naive individuals. Sites were assigned by cluster randomization using a 2 x 2 factorial design to administer MM, ALR, MM + ALR, or neither (control). MM participants received individualized, structured, long-term adherence support from trained MMs. ALR participants received individually programmed ALR alarms for use throughout the study.
Results: The 928 participants, followed a median of 30 months, included 22% women and 75% nonwhites; the median baseline CD4 count was 155 cells/mm. First virologic failure was 13% lower in all MM versus no-MM groups (P = 0.13) and 28% lower in MM versus no-MM subgroups randomized to 2-class ARV arms in the parent ARV study (P = 0.01). MM (vs. no-MM) participants had significantly better CD4 cells count (P = 0.01) and adherence (P < 0.001) outcomes. ALR (vs. no-ALR) participants had worse virologic outcomes.
Conclusion: This large randomized clinical trial demonstrated that interpersonal structured adherence support was associated with improved long-term medication adherence and virologic and immunologic HIV outcomes.