Intensive care unit occupancy and premature discharge rates: A cohort study assessing the reporting of quality indicators

Purpose: ICU occupancy fluctuates. High levelsmay disadvantage patients. Currently, occupancy is benchmarked annuallywhichmay inaccurately reflect strained units. Outcomes potentially sensitive to occupancy include premature (early) ICU discharge and non-clinical transfer (NCT). This study assesses the association between daily occupancy and these outcomes, and evaluates benchmarking care across Scotland using daily occupancy. Materials and methods: Population: all Scottish ICU patients, 2006-2014. Exposure: bed occupancy per unit-day; Outcomes: proportion of early discharges and NCTs. Design: Retrospective cohort study. Outcome rates were calculated above various occupancy thresholds. Polynomial regression visualised associations, and inflection points between occupancy and outcomes. Spearman's rho correlations between occupancy measures and outcomes were reported. Results: 65,472 discharges occurred over 57,812 unit-days. 1954(3.0%) discharges were early; 429 (0.7%) were NCTs. Early discharge rates above 70%, 80% and 90% occupancy were 3.9%, 5.0% and 7.5% respectively. Occupancies at which outcome rates greatly increased were near 80% for early discharge, and 90% for NCT. Mean annual occupancy was not correlated with outcomes; annual proportion of days >= 90% occupancy correlated most strongly (early discharge rho = 0.46,p < .001; NCT rho = 0.31, p < .001). Conclusions: We demonstrate a clear association between daily ICU occupancy and early discharge/NCT. Daily occupancy may better benchmark care quality than mean annual occupancy. (C) 2019 Elsevier Inc. All rights reserved.