return to TMARC

Publication Abstract Display
Type: Published Abstract
Title: Predictors of antiretroviral adherence among active methamphetamine users with HIV.
Authors: Montoya JL, Gouaux B, Rooney A, Casaletto K, Grant I, Moore DJ
Year: 2015
Publication: Society of Behavioral Medicine’s 36th Annual Meeting and Scientific Sessions
Volume: Issue: Pages:
Abstract:Objective: HIV infection and methamphetamine use disorders (HIV/MA) are highly comorbid, and MA use is associated with worse adherence to antiretroviral (ARV) therapy. The individualized texting for adherence building (iTAB) intervention aimed to improve ARV adherence among HIV/MA individuals. The present study evaluates predictors of adherence separately for days with, and days without, self-reported MA use. Participants and Methods: ARV medications of 59 HIV/MA participants were tracked for 30 days using the Medication Event Monitoring System (MEMS). Participants were randomized to iTAB (n=41) or an active comparison intervention (CTRL; n=18). Both groups received a daily text message assessing whether they had used MA in the last 24 hours; the iTAB group also received ARV medication reminder texts. Given that overall adherence did not differ by intervention arm (i.e., iTAB v. CTRL; p<.05), we collapsed the two study arms and examined predictors of adherence on days in which participants endorsed or denied MA use via text. For each participant, adherence was averaged separately for non-MA-using days and MA-using days and then a ≥90% cut-off point was used to classify each individual as adherent or non-adherent on non-MA-using (non-MA MEMS) and MA-using days (MA MEMS). Constructs of the Health Belief Model [e.g., intentions to adhere and subjective norms (belief about whether key people approve of adherence and motivation to gain their approval)], psychiatric and HIV disease characteristics were considered as covariates. Results: We identified univariate predictors of adherence within the non-MA MEMS (nadir CD4 count) and MA MEMS groups (subjective norms, nadir CD4 count, depressed mood) (p<.10). These variables and intervention group were entered into logistic regression models predicting adherence among non-MA MEMS and MA MEMS. Within the non-MA MEMS model, lower nadir CD4 count and the CTRL-arm were associated with being adherent. Among the MA MEMS group, greater subjective norms and lower nadir CD4 count were uniquely associated with being adherent. Conclusions: A stronger sense that family and friends support medication adherence and a history of more severe immunocompromise are strong predictors of ARV adherence among HIV/MA individuals in the context of active MA use. Future adherence interventions in HIV/MA may benefit from targeting modifiable health beliefs such as subjective norms.

return to publications listing