Association Between Medication Adherence and the Outcomes of Heart Failure

Background Previous studies of heart failure patients have demonstrated an association between cardiovascular medication adherence and hospitalizations or a composite end point of hospitalization and death. Few studies have assessed the impact of treatment adherence within large general medical populations that distinguish the health outcomes of emergency department visits, hospitalization, and death. Objective To determine the association of incremental cardiovascular medication adherence on emergency department visits, hospitalization, and death in adult heart failure patients in Indiana. Design Retrospective cohort study conducted using electronic health record data from the statewide Indiana Network for Patient Care (INPC) between 2004 and 2009. Methods Patients were at least 18 years of age with a diagnosis of heart failure and prescribed at least one cardiovascular medication for heart failure. Adherence was measured as the proportion of days covered (PDC) using pharmacy transaction data. Clinical end points included emergency department visits, hospital admissions, length of hospital stay, and mortality. Generalized linear models were used to determine the effect of a 10% increase in PDC on clinical end points adjusting for age, sex, race, Charlson comorbidity index, and medications. Results Electronic health records were available for 55,312 patients (mean age ± standard deviation [SD] 68 ± 16 years; 54% women; 65% white). Mean PDC for all heart failure medications was 63% ± 23%. For every 10% increase in PDC, emergency department visits decreased 11% (rate ratio [RR] 0.89; 95% confidence interval [CI] 0.89‐0.89), hospital admissions decreased 6% (RR 0.94; 95% CI 0.94‐0.94), total length of hospital stay decreased 1% (RR 0.99; 95% CI 0.99‐1.00), and all‐cause mortality decreased 9% (odds ratio 0.91; 95% CI 0.90‐0.92). Conclusion Incremental medication adherence was associated with reductions in emergency department visits, hospital admissions, length of hospital stay, and all‐cause mortality