Assessing ICU mortality among non-COVID-19 patients during the COVID-19 pandemic

BACKGROUND: The COVID-19 pandemic challenged healthcare infrastructure and delivery, particularly during peak periods. How these disruptions affected care and outcomes for non-COVID patients in intensive care units (ICUs) remains unclear. OBJECTIVES: To evaluate whether COVID-19 burden at the institutional and ICU level impacted ICU mortality among mechanically ventilated (MV) non-COVID patients across a health system. METHODS: Clinical data for ICU patients admitted within the University of Pittsburgh Medical Center system from March 2020 through December 2022 were included. High and low periods of COVID-19 were defined based on the average COVID-19 positivity rate across UPMC hospitals with high periods defined as months when the average positivity rate ≥10 %. A mixed-effects logistic regression evaluated the association between COVID-19 periods and ICU mortality with random intercepts for hospital and patient. Secondary models evaluated variation by ICU subtype and institution type. Models adjusted for demographic, clinical, and ICU characteristics. RESULTS: Among 19,727 MV non-COVID ICU admissions, 10,798 occurred during high COVID-19 periods and 8929 during low periods. Baseline characteristics were similar with the largest age group being 60-69 years (28.1 % low vs 28.6 % high). Patients were predominantly White males admitted to quaternary hospitals. Admission during high COVID-19 periods was associated with higher ICU mortality (OR=1.19, 95 % CI: 1.08,1.31) particularly in MICUs (OR=1.18, 95 % CI: 1.14,1.22) and quaternary hospitals (OR=1.19, 95 % CI: 1.09,1.29). CONCLUSION: High COVID-19 periods were associated with increased odds of ICU mortality among MV non-COVID-19 patients. These findings underscore the need for system-wide surge planning and incorporation of case-mix and workload measures in future evaluations.