Medication Adherence and Incident Preventable Hospitalizations for Hypertension.

Abstract

Potentially preventable hospitalizations (PPHs) for hypertension (HTN) is one indicator of possible failed ambulatory care. Rates of PPHs for HTN have remained fairly level since the late 1980s, which may reflect a lack of understanding of the drivers of these hospitalizations. Anti-HTN medication non-adherence has been studied as a potential risk factor for other cardiovascular disease outcomes but not for PPHs for HTN. A cohort analysis was conducted during 2005-2012 of people with HTN enrolled in commercial and employee health plans with claims in the MarketScan database. PPH for HTN was defined according to specifications published by the Agency for Healthcare Research and Quality. The proportion of days covered (PDC) algorithm was used to assess adherence to antihypertensives. Crude- and multivariate-adjusted incident PPHs for HTN rates were calculated, as well as third-party payments for selected PPH for HTN-related expenses. During 9,344,528 person-years of follow-up (mean=3 years), 6,008 incident PPHs for HTN were identified among 3,099,291 people. The crude rate for good adherence (PDC ≥80%) was 23.2 per 100,000 person-years compared with 102.6 per 100,000 person-years for poor adherence (PDC <40%). Over the 8-year study, PPH for HTN-associated payments equaled $41 million. Payments for those with poor adherence were four times higher than for those with good adherence. Poor anti-HTN medication adherence is strongly associated with PPHs for HTN. Improving the percentage of people who achieve good medication adherence is one possible approach to reducing the burden of PPHs for HTN in the U.S.