Medication adherence in adolescents with behaviorally-acquired HIV: evidence for using a multimethod assessment protocol

Purpose: The present study investigated medication adherence in an understudied population, adolescents with behaviorally acquired HIV, to improve upon prior methodological limitations using concurrent collection of HIV health status markers (viral load [VL]; percentage CD4 count [CD4%]) and multimethod adherence assessment (pill count, missed doses, off-schedule dosing).

Participants: A total of 60 youth with behaviorally acquired HIV receiving routine care in a multidisciplinary specialty clinic in the Mid-Southern United States. Adherence was assessed by routine pharmacy pill count and self-reported 3-day recall of doses missed and doses taken off-schedule, collected concurrently with clinically obtained VL and CD4% indicators. Adherence measures were evaluated as predictors of VL and CD4% using logistic regression analyses.

Results: Adherence difficulties were detected by all assessment methods, with off-schedule dosing appearing the most problematic (29.4% taken off-schedule). Self-report of doses missed (p = .038) and off-schedule dosing (p = .021) significantly predicted detectable VL. For each percent increase in nonadherence by off-schedule dosing, there was a 2% increased likelihood of detectable VL. No adherence measure significantly correlated with CD4%; pharmacy pill count did not relate to either health status marker.

Conclusions: This study is the first to document multimethod medication adherence measurement in a defined sample of adolescents with behaviorally acquired HIV, using imposed concurrent collection of CD4% and VL. Adherence difficulties were detected regardless of assessment strategy, with off-schedule dosing representing the greatest nonadherence behavior. Both 3-day recall methods predicted VL. Further investigation of adherence in larger samples of youth with behaviorally acquired HIV is needed to better understand the relationship to CD4% suppression.