Self-reported adherence and reasons for nonadherence among patients with low proportion of days covered for antihypertension medications

BACKGROUND: Incorporation of pharmacy fill data into the electronic health record has enabled calculations of medication adherence, as measured by proportion of days covered (PDC), to be displayed to clinicians. Although PDC values help identify patients who may be nonadherent to their medications, it does not provide information on the reasons for medication-taking behaviors.OBJECTIVE: To characterize self-reported adherence status to antihypertensive medications among patients with low refill medication adherence. Our secondary objective was to identify the most common reasons for nonadherence and examine the patient sociodemographic characteristics associated with these barriers. METHODS: Participants were adult patients seen in primary care clinics of a large, urban health system and on antihyperten-sive therapy with a PDC of less than 80% based on 6-month linked electronic health record-pharmacy fill data. We administered a validated medication adherence screener and a survey assessing reasons for antihyper-tensive medication nonadherence. We used descriptive statistics to characterize these data and logistic and Poisson regression models to assess the relationship between sociodemographic characteristics and adherence barriers.RESULTS: The survey was completed by 242 patients (57% female; 61.2% White; 79.8% not Latino/a or Hispanic). Of these patients, 45% reported missing doses of their medications in the last 7 days. In addition, 48% endorsed having at least 1 barrier to adherence and 38.4% endorsed 2 or more barriers. The most common barriers were being busy and having difficulty remembering to take medications. Compared with White participants, Black participants (incident rate ratio = 2.49; 95% CI = 1.93-3.22) and participants of other races (incident rate ratio = 2.16; 95% CI = 1.62-2.89) experienced a greater number of barriers.CONCLUSIONS: Nearly half of patients with low PDC reported nonad-herence in the prior week, suggesting PDC can be used as a screening tool. Augmenting PDC with brief self-report tools can provide insights into the reasons for nonadherence.