Comparing quality of care outcomes between assisted living and nursing homes before and during the COVID‐19 pandemic

Abstract

AbstractBackgroundWhile assisted living (AL) and nursing home (NHs) residents in share vulnerabilities, AL provides fewer staffing resources and services. Research has largely neglected AL, especially during the COVID‐19 pandemic. Our study compared trends of practice‐sensitive, risk‐adjusted quality indicators between AL and NHs, and changes in these trends after the start of the pandemic.MethodsThis repeated cross‐sectional study used population‐based resident data in Alberta, Canada. Using Resident Assessment Instrument data (01/2017–12/2021), we created quarterly cohorts, using each resident's latest assessment in each quarter. We applied validated inclusion/exclusion criteria and risk‐adjustments to create nine quality indicators and their 95% confidence intervals (CIs): potentially inappropriate antipsychotic use, pain, depressive symptoms, total dependency in late‐loss activities of daily living, physical restraint use, pressure ulcers, delirium, weight loss, urinary tract infections. Run charts compared quality indicators between AL and NHs over time and segmented regressions assessed whether these trends changed after the start of the pandemic.ResultsQuarterly samples included 2015–2710 AL residents and 12,881–13,807 NH residents. Antipsychotic use (21%–26%), pain (20%–24%), and depressive symptoms (17%–25%) were most common in AL. In NHs, they were physical dependency (33%–36%), depressive symptoms (26%–32%), and antipsychotic use (17%–22%). Antipsychotic use and pain were consistently higher in AL. Depressive symptoms, physical dependency, physical restraint use, delirium, weight loss were consistently lower in AL. The most notable segmented regression findings were an increase in antipsychotic use during the pandemic in both settings (AL: change in slope = 0.6% [95% CI: 0.1%–1.0%], p = 0.0140; NHs: change in slope = 0.4% [95% CI: 0.3%–0.5%], p < 0.0001), and an increase in physical dependency in AL only (change in slope = 0.5% [95% CI: 0.1%–0.8%], p = 0.0222).ConclusionsQIs differed significantly between AL and NHs before and during the pandemic. Any changes implemented to address deficiencies in either setting need to account for these differences and require monitoring to assess their impact.