Drayton Trumble1, Michael A Meier2, Maryellen Doody3, Xiaoming Wang4, Sean M Bagshaw2. 1. Department of Critical Care Medicine, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada. bagshaw@ualberta.ca. 2. Department of Critical Care Medicine, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada. 3. Department of Critical Care Medicine, General Systems Intensive Care Unit, University of Alberta Hospital, Alberta Health Services, Edmonton, Alberta, Canada. 4. Research Facilitation; Research Priorities and Implementation; Research, Innovation and Analytics; Alberta Health Services, Edmonton, Alberta, Canada.
Abstract
BACKGROUND: Falls among hospitalised patients contribute to avoidable morbidity and prolonged hospital stay. We aimed to describe the incidence, circumstances and outcomes associated with patient falls occurring in intensive care units. METHODS: Retrospective cohort study of adult admissions to an academic, tertiary ICU in Edmonton, Canada between 1 January 2013 and 30 April 2016. Fall events were ascertained by interrogation of an electronic health record. Each fall was independently adjudicated by two intensivists to confirm that a fall did occur, and to determine if the fall was potentially avoidable. Agreement was assessed by intraclass correlation (ICC). A matched cohort (one fall to five non-fall) was created to compare course and outcomes. RESULTS: Of 31 events identified, 26 were adjudicated as a fall (84%), for an estimated incidence of 5.2 per 1000 ICU admissions (95% CI, 3.4-7.6). Of these, 23 were judged as avoidable (89%) (ICC, 0.69; 95% CI, 0.37-0.85). The mean age was 54.5 years (SD, 17.4 years), 20 of the falls (77%) involved men, and 10 (39%) were surgical admissions (eight [31%] for trauma). The median ICU stay before the falls was 10.6 days (interquartile range [IQR], 6.2-15.0 days), and 13 patients (50%) scored positive for delirium. At the time of the fall, three patients (12%) were invasively ventilated, and eight (31%) were classified as ward-ready. Eleven falls (42%) were witnessed and 12 (46%) occurred after hours, of which nine involved patients (75%) who had delirium. Seven falls (27%) occurred within 2 hours of a nursing shift change and 11 (42%) during a nursing break coverage. No patient sustained a major injury related to a fall but four (15%) reported minor injuries such as a laceration or abrasion. The median ICU stay after a fall was 2.2 days (IQR, 1.2-2.9 days). Compared with the matched non-fall cohort, patients who fell had a longer duration of ICU stay (median, 12.0 days [IQR, 8.3-18.6 days] v 4.7 days [IQR, 2.8-8.4 days]; P < 0.0001). CONCLUSION: Falls in the ICU are infrequent and generally perceived as avoidable. Falls often occur at night, are unwitnessed and are associated with concomitant patient delirium and nursing shift changes and cross coverage.
BACKGROUND: Falls among hospitalised patients contribute to avoidable morbidity and prolonged hospital stay. We aimed to describe the incidence, circumstances and outcomes associated with patient falls occurring in intensive care units. METHODS: Retrospective cohort study of adult admissions to an academic, tertiary ICU in Edmonton, Canada between 1 January 2013 and 30 April 2016. Fall events were ascertained by interrogation of an electronic health record. Each fall was independently adjudicated by two intensivists to confirm that a fall did occur, and to determine if the fall was potentially avoidable. Agreement was assessed by intraclass correlation (ICC). A matched cohort (one fall to five non-fall) was created to compare course and outcomes. RESULTS: Of 31 events identified, 26 were adjudicated as a fall (84%), for an estimated incidence of 5.2 per 1000 ICU admissions (95% CI, 3.4-7.6). Of these, 23 were judged as avoidable (89%) (ICC, 0.69; 95% CI, 0.37-0.85). The mean age was 54.5 years (SD, 17.4 years), 20 of the falls (77%) involved men, and 10 (39%) were surgical admissions (eight [31%] for trauma). The median ICU stay before the falls was 10.6 days (interquartile range [IQR], 6.2-15.0 days), and 13 patients (50%) scored positive for delirium. At the time of the fall, three patients (12%) were invasively ventilated, and eight (31%) were classified as ward-ready. Eleven falls (42%) were witnessed and 12 (46%) occurred after hours, of which nine involved patients (75%) who had delirium. Seven falls (27%) occurred within 2 hours of a nursing shift change and 11 (42%) during a nursing break coverage. No patient sustained a major injury related to a fall but four (15%) reported minor injuries such as a laceration or abrasion. The median ICU stay after a fall was 2.2 days (IQR, 1.2-2.9 days). Compared with the matched non-fall cohort, patients who fell had a longer duration of ICU stay (median, 12.0 days [IQR, 8.3-18.6 days] v 4.7 days [IQR, 2.8-8.4 days]; P < 0.0001). CONCLUSION: Falls in the ICU are infrequent and generally perceived as avoidable. Falls often occur at night, are unwitnessed and are associated with concomitant patientdelirium and nursing shift changes and cross coverage.
Authors: Sean M Bagshaw; Henry T Stelfox; Theodore J Iwashyna; Rinaldo Bellomo; Dan Zuege; Xioaming Wang Journal: Intensive Care Med Date: 2018-11-12 Impact factor: 17.440