Background: Poor adherence to medications for cardiovascular disease (CVD) is associated with adverse health outcomes, but little is known about its association with healthcare utilization (HCU).
Objective: To examine whether adherence is associated with a long-term decrease in HCU.
Methods: This is a retrospective cohort study of 1582 patients with CVD who enrolled in Maccabi Healthcare Services in Israel, initiating CVD medication therapy in 2006. Adherence was assessed by the proportion of days covered (PDC) with medications. Patients were defined as: non-adherent (PDC <0.4), partially adherent (0.4 ≤ PDC < 0.8), and fully adherent (PDC ≥0.8). HCU was estimated for 4 years following treatment initiation. Multivariable GEE models were used to analyze predictors of HCU. Model I included total adherence during the entire follow-up period as well as the interaction between this measure and the follow-up year. Model II included previous and current year's adherence as well as previous year's HCU cost. Both models were adjusted for potential confounders including: patient's age, gender, socioeconomic status, ownership of voluntary supplementary health insurance, and comorbidities.
Results: The median age of patients was 63 (69 % males). Fifty-four percent of patients (n = 860) were defined as adherent, 24 % as partially adherent and 22 % as non-adherent. Model I: the annual HCU costs of adherent patients decreased by 10 % following treatment initiation [rate ratio (RR) = 0.90, 95 % confidence interval (CI) 0.86-0.94, P < 0.001]. This decrease stemmed predominantly from reduction in hospitalization costs. No significant changes in annual costs following treatment initiation were observed among partially adherent (RR = 1.00, 95 % CI 0.90-1.10, P = 0.935) and non-adherent (RR = 0.98, 95 % CI 0.87-1.10, P = 0.681) patients. Model II: no temporal association was found between adherence and HCU.
Conclusions: Adherence to CVD medications is relatively low. Adherence is associated with long-term decrease in healthcare expenditure. Exploring reasons for the high non-adherence and ways to improve adherence may optimize utilization of health systems' scarce resources.
Keywords: Adherence; Cardiovascular disease; Healthcare utilization.