A Multicenter Cohort Study of Falls Among Patients Admitted to the ICU*

Abstract

Objectives: To determine the incidence of falls, risk factors, and adverse outcomes, among patients admitted to the ICU. Design: Retrospective cohort study. Setting: Seventeen ICUs in Alberta, Canada. Patients: Seventy-three thousand four hundred ninety-five consecutive adult patient admissions between January 1, 2014, and December 31, 2019. Measurements and Main Results: A mixed-effects negative binomial regression model was used to examine risk factors associated with falls. Linear and logistic regression models were used to evaluate adverse outcomes. Six hundred forty patients experienced 710 falls over 398,223 patient days (incidence rate of 1.78 falls per 1,000 patient days [95% CI, 1.65–1.91]). The daily incidence of falls increased during the ICU stay (e.g., day 1 vs day 7; 0.51 vs 2.43 falls per 1,000 patient days) and varied significantly between ICUs (range, 0.37–4.64 falls per 1,000 patient days). Male sex (incidence rate ratio [IRR], 1.37; 95% CI, 1.15–1.63), previous invasive mechanical ventilation (IRR, 1.82; 95% CI, 1.40–2.38), previous sedative and analgesic medication infusions (IRR, 1.60; 95% CI, 1.15–2.24), delirium (IRR, 3.85; 95% CI, 3.23–4.58), and patient mobilization (IRR, 1.26; 95% CI, 1.21–1.30) were risk factors for falling. Falls were associated with longer ICU (ratio of means [RM], 3.10; 95% CI, 2.86–3.36) and hospital (RM, 2.21; 95% CI, 2.01–2.42) stays, but lower odds of death in the ICU (odds ratio [OR], 0.09; 95% CI, 0.05–0.17) and hospital (OR, 0.21; 95% CI, 0.14–0.30). Conclusions: We observed that among ICU patients, falls occur frequently, vary substantially between ICUs, and are associated with modifiable risk factors, longer ICU and hospital stays, and lower risk of death. Our study suggests that fall prevention strategies should be considered for critically ill patients admitted to ICU.