Abstract 343: Electronically-measured, but Not Self-reported, Medication Adherence is Associated with Healthcare Utilization in Patients with Uncontrolled Hypertension

Introduction: Non-adherence to antihypertensive medications can lead to poorly controlled hypertension and increased cardiovascular risk. Prior studies utilizing pharmacy refill data have shown an association between medication non-adherence and increased healthcare utilization. However, data are lacking as to whether electronically measured, day-to-day adherence is associated with healthcare utilization.Methods: Between 2011 and 2014, we enrolled a convenience sample of 161 patients with treated but uncontrolled hypertension from a primary care clinic. Adherence to up to 4 blood pressure (BP) medications was electronically measured during the interval between two clinic visits using a 4-compartment electronic pillbox (MedSignals) that records the date and time each compartment is opened. Adherence to each medication was calculated as the percent of days the correct number of doses of a particular medication was taken as prescribed; regimen adherence was the mean adherence to monitored medications. Adherence was also measured after completion of the electronic monitoring period using the 8-item Morisky Medication Adherence Scale. Healthcare utilization (cardiac procedures, including EKGs, stress tests and echocardiograms; ED visits; all-cause hospitalizations; scheduled primary care visits) was assessed via review of the electronic medical record in the 1 year after electronic monitoring. Logistic regression and negative binomial regression analyses were used to model the association between electronic adherence and healthcare utilization adjusting for covariates (age, gender, ethnicity, race, Charlson Comorbidity Index, Medicaid status, BP, PTSD, depression).Results: The mean age was 65 years, 75% were female, 80% Hispanic, 86% had Medicaid, and 52% were non-adherent by electronic monitoring. Non-adherent patients were more likely to have ≥1 cardiac procedures (27% vs. 10%, p=0.004), ED visits (35% vs. 13%, p=0.001), and hospitalizations (27% vs. 12%, p=0.01) compared to adherent patients, but did not have more primary care visits (p=0.55). In adjusted analyses, non-adherent patients had more ED visits (adjusted rate ratio [aRR] 2.4 [95% CI 1.3-4.6], p 0.40 for all measures).Discussion: We found associations between electronically-measured, but not self-reported, medication adherence and healthcare utilization. These findings suggest that electronic adherence monitoring may be a useful tool for identifying high utilizers of healthcare resources and targeting such patients for cost-reduction strategies.