Christophe Faisy1, Cindy Davagnar2, Dominique Ladiray3, Juliette Djadi-Prat4, Maxime Esvan4, Emilie Lenain4, Pierre Durieux5, Jean-François Leforestier4, Carine Marlet2, Marta Seijo2, Anne Guillou2. 1. Service de Réanimation Médicale, Hôpital Européen Georges Pompidou, Assistance Publique - Hôpitaux de Paris, Université Paris Descartes, Sorbonne Paris Cité, Paris, France. Electronic address: christophe.faisy@egp.aphp.fr. 2. Service de Réanimation Médicale, Hôpital Européen Georges Pompidou, Assistance Publique - Hôpitaux de Paris, Université Paris Descartes, Sorbonne Paris Cité, Paris, France. 3. Institut National de la Statistique et des Etudes Economiques, Malakoff, France. 4. Santé Publique et Informatique Médicale, Université Paris Descartes, Sorbonne Paris Cité, Faculté de Médecine, INSERM U729, and French Cochrane Network and Centre, Paris, France. 5. Centre d'Investigations Epidémiologiques 4, INSERM, Hôpital Européen Georges Pompidou, Paris, France.
Abstract
OBJECTIVE: Our purpose was to identify potential organizational factors that contributed to life-threatening adverse events in adult intensive care unit. METHODS: A prospective, observational, dynamic cohort study was carried out from January 2006 to December 2013 in a 20-bed adult medical intensive care unit. All patients admitted to the intensive care unit and who experienced one or more selected life-threatening adverse events (mainly unexpected cardiac arrest, unplanned extubation, reintubation after planned extubation, and readmission within 48h of intensive care unit discharge) were included in the analysis. Negative binomial regression was used to model how human resources, work organization, and intensive care activity influenced the monthly rate of selected severe adverse events. Data were collected from local and national databases. RESULTS: Overall, 638 severe adverse events involving 498 patients were recorded. Adverse events increased seasonally in May, November and December (p<.001 vs other months). The proportion of inexperienced nurses and doctors' working hours could not explain these seasonal peaks of adverse events. Multivariate analysis identified bed-to-nurse ratio and the arrival of inexperienced residents or senior registrars as being independently associated with the rate of adverse events (incidence risk ratio=1.36 (95% confidence interval, 1.05-1.75), and 1.07 (95% confidence interval, 1.01-1.13), respectively; p=.01 in both cases). According to this model, a one-unit increase in the day-night shifts carried out by each nurse per month tended to reduce the rate of adverse events (incidence risk ratio=0.60 (95% confidence interval, 0.36-1.01), p=.05). Severity at intensive care unit admission did not influence the rate of adverse events (incidence risk ratio=1.02 (95% confidence interval, 1.00-1.04), p=.12). CONCLUSIONS: Results identify nurse workload and the arrival of inexperienced residents or senior registrars as risk factors for the occurrence of life-threatening adverse events in the adult medical intensive care unit. Limiting fluctuations in bed-to-nurse ratio and providing inexperienced medical staff members with sufficient supervision may decrease severe adverse events in critically ill patients.
OBJECTIVE: Our purpose was to identify potential organizational factors that contributed to life-threatening adverse events in adult intensive care unit. METHODS: A prospective, observational, dynamic cohort study was carried out from January 2006 to December 2013 in a 20-bed adult medical intensive care unit. All patients admitted to the intensive care unit and who experienced one or more selected life-threatening adverse events (mainly unexpected cardiac arrest, unplanned extubation, reintubation after planned extubation, and readmission within 48h of intensive care unit discharge) were included in the analysis. Negative binomial regression was used to model how human resources, work organization, and intensive care activity influenced the monthly rate of selected severe adverse events. Data were collected from local and national databases. RESULTS: Overall, 638 severe adverse events involving 498 patients were recorded. Adverse events increased seasonally in May, November and December (p<.001 vs other months). The proportion of inexperienced nurses and doctors' working hours could not explain these seasonal peaks of adverse events. Multivariate analysis identified bed-to-nurse ratio and the arrival of inexperienced residents or senior registrars as being independently associated with the rate of adverse events (incidence risk ratio=1.36 (95% confidence interval, 1.05-1.75), and 1.07 (95% confidence interval, 1.01-1.13), respectively; p=.01 in both cases). According to this model, a one-unit increase in the day-night shifts carried out by each nurse per month tended to reduce the rate of adverse events (incidence risk ratio=0.60 (95% confidence interval, 0.36-1.01), p=.05). Severity at intensive care unit admission did not influence the rate of adverse events (incidence risk ratio=1.02 (95% confidence interval, 1.00-1.04), p=.12). CONCLUSIONS: Results identify nurse workload and the arrival of inexperienced residents or senior registrars as risk factors for the occurrence of life-threatening adverse events in the adult medical intensive care unit. Limiting fluctuations in bed-to-nurse ratio and providing inexperienced medical staff members with sufficient supervision may decrease severe adverse events in critically illpatients.
Authors: Kirsten Brubakk; Martin Veel Svendsen; Dag Hofoss; Tonya Moen Hansen; Paul Barach; Ole Tjomsland Journal: BMJ Open Date: 2019-12-15 Impact factor: 2.692