Time of admission to intensive care unit, strained capacity, and mortality: A retrospective cohort study.

ABSTRACT

PURPOSE: We sought to study the association between afterhours ICU admission and ICU mortality considering measures of strained ICU capacity. MATERIALS AND METHODS: Retrospective analysis of 4141 admissions to 2 ICUs in Lisbon, Portugal (06/2016-06/2018). Primary exposure was ICU admission on 2000 h-0759 h. Primary outcome was ICU mortality. Measures of strained ICU capacity were bed occupancy rate ≥ 90% and cluster of ICU admissions 2 h before or following index admission. RESULTS: There were 1581 (38.2%) afterhours ICU admissions. Median APACHE II score (19 vs. 20) was similar between patients admitted afterhours and others (P = .27). Patients admitted afterhours had higher crude ICU mortality (15.4% vs. 21.9%; P < .001), but similar adjusted ICU mortality (aOR [95%CI] = 1.15 [0.97-1.38]; P = .12). While bed occupancy rate ≥ 90% was more frequent in patients admitted afterhours (23.1% vs. 29.1%) or deceased in ICU (23.6% vs. 33.7%), cluster of ICU admissions was more frequent in patients admitted during daytime hours (75.2% vs. 58.9%) or that survived the ICU stay (70.1% vs. 63.9%; P ≤ .001 for all). These measures of strained ICU capacity were not associated with adjusted ICU mortality (P ≥ .10 for both). CONCLUSIONS: Afterhours ICU admission and measures of strained ICU capacity were associated with crude but not adjusted ICU mortality.