0 avis
Practices of end-of-life decisions in 66 southern French ICUs 4years after an official legal framework: A 1-day audit
Archive ouverte
Edité par CCSD ; Elsevier Masson -
International audience. Since the implementation of two French laws in 2002 and 2005 and the implementation of guidelines about End-of-Life (EoL) decisions, few studies concerning EoL practices in French intensive care units (ICUs) have been reported. This study was aimed at assessing compliance with recommendations and current legislation concerning EoL decisions. Prospective observational study based on 1-day audit conducted from January to May 2009 in 66 southern French ICUs. Six hundred and twenty-five patients were included (median age: 63 [52-76] years, median SAPS II: 46 [34-58]). The written designation of a surrogate decision-maker was reported for 87 (15%) patients. Advance directives were completed for only 4% of patients. The EoL decision-making process consisted in a multidisciplinary approach for 99 (47%) patients and was recorded in the medical chart for 63 (64%) cases. Families were informed about medical decisions in 58% of cases. This proportion was higher (87%) if a decision to forego life-sustaining therapy was made. EoL decisions consisted of withholding treatments for 72 (94%) patients and withdrawal of treatments for 5 (6%) patients. In the multivariate stepwise logistic regression, four variables were independently associated with a decision to forego life support: preexisting dependence on others (P<0.0001), advance directives (P=0.01), age (P=0.008) and the SAPS 2 score (P=0.009). The major finding of the present study is the existence of a gap between the widely approved EoL recommendations made by scientific societies and the daily practice of southern French ICUs. Even if EoL decisions are mostly shared with relatives, their written documentation in medical charts remains insufficient. Concerning EoL practices, the withdrawal of treatment remains an uncommon decision.
Consulter en ligne
Suggestions
Du même auteur
