Does Unprecedented ICU Capacity Strain, As Experienced During the COVID-19 Pandemic, Impact Patient Outcome?

Abstract

Objectives: To determine whether patients admitted to an ICU during times of unprecedented ICU capacity strain, during the COVID-19 pandemic in the United Kingdom, experienced a higher risk of death.

Design: Multicenter, observational cohort study using routine clinical audit data.

Setting: Adult general ICUs participating the Intensive Care National Audit & Research Centre Case Mix Programme in England, Wales, and Northern Ireland.

Patients: One-hundred thirty-thousand six-hundred eighty-nine patients admitted to 210 adult general ICUs in 207 hospitals.

Interventions: Multilevel, mixed effects, logistic regression models were used to examine the relationship between levels of ICU capacity strain on the day of admission (typical low, typical, typical high, pandemic high, and pandemic extreme) and risk-adjusted hospital mortality.

Measurements and main results: In adjusted analyses, compared with patients admitted during periods of typical ICU capacity strain, we found that COVID-19 patients admitted during periods of pandemic high or pandemic extreme ICU capacity strain during the first wave had no difference in hospital mortality, whereas those admitted during the pandemic high or pandemic extreme ICU capacity strain in the second wave had a 17% (odds ratio [OR], 1.17; 95% CI, 1.05-1.30) and 15% (OR, 1.15; 95% CI, 1.00-1.31) higher odds of hospital mortality, respectively. For non-COVID-19 patients, there was little difference in trend between waves, with those admitted during periods of pandemic high and pandemic extreme ICU capacity strain having 16% (OR, 1.16; 95% CI, 1.08-1.25) and 30% (OR, 1.30; 95% CI, 1.14-1.48) higher overall odds of acute hospital mortality, respectively.

Conclusions: For patients admitted to ICU during the pandemic, unprecedented levels of ICU capacity strain were significantly associated with higher acute hospital mortality, after accounting for differences in baseline characteristics. Further study into possible differences in the provision of care and outcome for COVID-19 and non-COVID-19 patients is needed.

Figure 1.

Flow diagram of admissions during pre-pandemic reference and analysis periods.

Figure 2.

ICU capacity strain between March 1, 2019, and February 28, 2021. A, Mean bed census (the number of other patients in ICU on the index patient’s day of admission, averaged across all ICUs). B, ICU capacity strain: pandemic low, less than observed range of values for the ICU during the pre-pandemic period; typical low, more than 10% below the pre-pandemic median bed census for each ICU but within the observed range; typical, within ± 10% of the pre-pandemic median; typical high, more than 10% above the pre-pandemic median up to the pre-pandemic maximum; pandemic high, up to 50% greater than the pre-pandemic maximum; and pandemic extreme, more than 50% greater than the pre-pandemic maximum bed census for the ICU.

Figure 3.

Distributions of patient-level ICU bed census on date of admission compared with the maximums observed for the admitting ICU during the pre-pandemic reference period.