Abstract
Background/Aims Identifying factors related to readmission is important for successfully targeting appropriate interventions to groups at risk for readmission. The objective of this study was to investigate the association of long-term medication adherence with hospital readmission in a cohort of beneficiaries enrolled in a Medicare cost plan.
Methods The study employed a retrospective cohort design using pharmacy and healthcare utilization claims from a Medicare Cost Contract plan for January 2009 through December 2009. Inpatient hospitalization was identified based on the revenue code (100–169, and 200–219). Eligible members were continuously enrolled through the study period, and experienced at least one hospitalization in 2009 after which they were discharged to home. About 1767 members were eligible and included in the analysis. Approximately 13% had a claim for a subsequent readmission during the study period. Medication adherence in the year before the index hospitalization was measured with the medication possession ratio (MPR), defined as the supply of medications in days minus the last fill days supply divided by the total number of days between the last fill date and the first fill date for drugs for chronic medication. Likelihood of readmission within 90 days was estimated using the logistic regression. Covariates entered into the model included demographics, MPR (both continuous and categorical with 3 categories: low (MPR<0.5), medium (.5 < MPR<.8), and high (MPR>.8 adherence), and having an office visits within 30 days of discharge.
Results Members with high medication adherence were less likely than those with low adherence to have a claim indicating 90-day readmission (OR=.35, p=0.01). When considered as continuous, higher MPR was associated with decreasing risk of readmission (OR=0.24, p=0.02). Having an office visit within 30 days from the date of discharge was as associated with a decreased risk of readmission (OR=0.06, p<0.001).
Discussion The health behavior of long-term adherence to medications was associated with risk of readmission. A major study limitation is the underestimation of readmission rates due to lack of complete capture of hospital claims. An important next step will be to replicate the study with a larger population for which all claims are captured.
