• Garrett McDougall and Osama Loubani.
• Department of Emergency Medicine, McMaster University, Hamilton, ON, Canada.
• J Crit Care. 2024 Apr 16; 82: 154813154813.

PurposeTo estimate differences in case-mix adjusted hospital mortality between adult ICU patients who are transferred during their ICU-stay and those who are not.Methods19,260 visits to 12 ICUs in Nova Scotia (NS), Canada April 2018-September 2023 were analyzed. Data were obtained from the NS Provincial ICU database. Generalized additive models (GAMs) were used to estimate differences in case-mix adjusted hospital mortality between patients who underwent transfer and those who did not.Results1040/19,260 (5%) ICU visits involved interfacility-transfer. No difference in hospital mortality was identified between transferred and non-transferred patients by GAM (OR, 0.99, 95% CI, 0.82 to 1.19; p = 0.91). No mortality difference was observed between patients undergoing a single transfer versus multiple (OR, 0.87; 95% CI, 0.45 to -1.69; p = 0.68). A GAM including the categories no transfer, one transfer, and multiple transfers identified a difference in hospital mortality for patients that underwent multiple transfers compared to non-transferred patients (OR, 0.68; 95% CI, 0.46 to 1.00, p = 0.05), but no difference was identified in a post-hoc matched cohort sensitivity analysis (OR, 0.68; 95% CI, 0.46 to 1.01, p = 0.05).ConclusionThe transfer of critically ill patients between ICUs in Nova Scotia did not impact case-mix adjusted hospital mortality.Copyright © 2024. Published by Elsevier Inc.

3 keywords...

hide…