Primary non-adherence to newly prescribed discharge medications and 30-day adverse outcomes by mental health status: a retrospective cohort study in two tertiary care hospitals in Montréal, Québec, Canada.

ABSTRACT: OBJECTIVES: To determine whether mental health status is associated with primary non-adherence to newly prescribed medications after hospital discharge, and to examine whether primary non-adherence is associated with 30-day adverse drug events (ADEs), emergency department (ED) visits and hospital readmissions. DESIGN: This retrospective cohort study was nested within the RightRx cluster-randomised trial. SETTING: Adult medical and surgical inpatient units in two tertiary care hospitals in Montréal, Québec, Canada, with linkage to provincial administrative health databases. PARTICIPANTS: Adults aged 18 years or older discharged alive to the community between October 2014 and November 2016 were eligible if they were enrolled in the RightRx trial, had public drug coverage through the Régie de l’assurance maladie du Québec (RAMQ) and had at least one newly prescribed discharge medication. The cohort included 2672 patients, of whom 593 (22.2%) had a documented mental health condition. Prescription-level analyses included 6870 eligible newly prescribed discharge medications. PRIMARY AND SECONDARY OUTCOME MEASURES: The primary outcome was prescription-level primary non-adherence, defined as the absence of a corresponding dispensing claim within 30 days for a newly prescribed discharge medication. Secondary outcomes were 30-day ADEs, ED visits and hospital readmissions. At the patient level, primary non-adherence was examined both as at least one unfilled newly prescribed discharge medication and as the number of unfilled newly prescribed discharge medications. RESULTS: Overall, 1499 of 6870 eligible newly prescribed discharge prescriptions (21.8%) were unfilled within 30 days. In adjusted prescription-level analyses using generalised estimating equations, mental health status was not associated with primary non-adherence (adjusted OR (aOR) 1.04, 95% CI 0.88 to 1.24; p=0.636). At the patient level, having at least one unfilled newly prescribed discharge medication was not associated with 30-day ADEs (aOR 1.01, 95% CI 0.69 to 1.48), ED visits (aOR 1.03, 95% CI 0.86 to 1.24) or readmissions (aOR 1.16, 95% CI 0.91 to 1.47). However, the number of unfilled newly prescribed medications was associated with increased odds of 30-day readmission (aOR 1.09, 95% CI 1.02 to 1.16; p=0.011). CONCLUSIONS: Primary non-adherence to newly prescribed medications was common during the first 30 days after hospital discharge, but documented mental health status was not independently associated with a higher likelihood of primary non-adherence. Having at least one unfilled newly prescribed medication was not associated with 30-day ADEs, ED visits or hospital readmissions. An exploratory count-based analysis identified a modest association between the number of unfilled medications and readmission; this finding should be interpreted cautiously given the number of secondary analyses performed and requires confirmation in independent studies. TRIAL REGISTRATION: NCT01179867.