Comparing objective measures of adherence to HIV antiretroviral therapy: Electronic medication monitors and unannounced pill counts

Abstract

We compared electronic medication monitoring and electronic monitoring adjusted by interview with an unannounced pill-count adherence measure, in a cross-sectional analysis “n = 42”. We used the following measures “a” electronic monitored doses “EMD: # medication monitor openings/# doses prescribed”, “b” adjusted electronically monitored doses “AEMD: EMD adjusted by the number of extra openings and extra doses removed”, “c” percentage of days dose taken “PDDT: % days the medication monitor was opened at least once, and “d” unannounced pill count “PC”. The results showed that AEMD, EMD, and PDDT were significantly associated with PC adherence. AEMD explained 83% of the variation in PC adherence, EMD 49%, and PDDT 58%. EMD adherence was significantly lower than PC adherence “p = .02”. Electronic medication monitoring measures were closely related to PC adherence. AEMD was more closely associated than EMD or PDDT. EMD may underestimate adherence due to the removal of multiple doses from the bottle at one time.