Barriers to use of electronic adherence monitoring in an HIV clinic

OBJECTIVE: To evaluate barriers to Medication Event Monitoring System (MEMS) measurement of adherence to combination antiretroviral therapy in an HIV clinic. DESIGN: Descriptive, cross-sectional study measured MEMS adherence to one antiretroviral for one month. SETTING: HIV clinic in a Veterans Affairs Medical Center. PARTICIPANTS: Sixty-four men on a stable antiretroviral treatment regimen. MAIN OUTCOME MEASURES: Decanting (removing >1 dose at a time) before and during monitoring over a 30-day observation period was used to determine the qualitative impact of MEMS on adherence. The adherence index was the proportion of prescribed doses not missed. RESULTS: Subjects were primarily white (73%) with mean CD4+ count 408 cells/mm³, log viral load 1.81 copies/mL, and duration of antiretroviral therapy 5.5 years. Twenty-seven (42%) had some decanting routine established prior to monitoring; 12 (44%) of these patients used daily decanters and 15 (56%) used weekly pillboxes. Of those who decanted prior to the study, 10 (37%) did not stop decanting during monitoring, 14 (52%) stopped decanting only the capped medication, and three (11%) stopped decanting all antiretrovirals. Other adherence strategies did not accommodate MEMS. Eight (13%) subjects said MEMS made adherence more difficult, six (9%) said MEMS was a reminder to adhere, and two (3%) mentioned both. Two subjects attributed skipped doses or time changes to the MEMS cap. The majority who refused to participate used pillboxes. CONCLUSIONS: Personal adherence strategies incompatible with MEMS are common in persons on complex treatment regimens. Although MEMS data on decanters underestimate adherence, excluding decanters erodes applicability of descriptive measures. MEMS use may have affected adherence behavior. Measures in conjunction with MEMS should include self-reported adherence and decanting assessment.