The impact of at-risk drinking on the outcomes of nontrauma patients is not well characterized. The aim of this study was to determine whether at-risk drinking is independently associated with the survival of nontrauma patients in an ICU and within 1 year following ICU discharge.
Observational cohort study.
A 21-bed mixed ICU in a university hospital.
A total of 662 patients who experienced an ICU stay of 3 days or more and for whom alcohol consumption could be assessed.
ICU-related variables were collected prospectively, and a 1-year follow-up was determined retrospectively. Analyses were adjusted based on prognostic determinants of short- and long-term outcomes, as previously described in ICU patients and alcohol abusers. Two hundred and eight patients (33%) were identified as at-risk drinkers according to the National Institute on Alcohol Abuse and Alcoholism criteria. Additionally, 111 patients (17%) died in the ICU, and 97 (15%) died after ICU discharge. From the ICU admission until the end of the 1-year follow-up period, the at-risk drinkers exhibited poorer survival than the non–at-risk drinkers (p = 0.0004, as determined by the log-rank test). More specifically, 50 at-risk drinkers (24%) versus 61 non–at-risk drinkers (13%) died in the ICU (p = 0.0009 for the comparison). After adjustment, at-risk drinking remained independently associated with mortality in the ICU (adjusted odds ratio of 1.83; 95% CI of 1.16–2.89; p = 0.01) and with mortality within the year following ICU discharge (adjusted hazard ratio of 1.70; 95% CI of 1.15–2.52; p = 0.008). The causes of death in the at-risk and non–at-risk drinkers were similar.
In this population of critically ill nontrauma patients, at-risk drinking was independently associated with death in the ICU and within the year following ICU discharge.
1CHU Rennes, Maladies Infectieuses et Réanimation Médicale, Rennes, France.
2Inserm-CIC-0203, Faculté de Médecine, Université Rennes I, IFRI 40, Rennes, France.
3Biosit, Faculté de Médecine, Université Rennes 1, Rennes, France.
* See also p. 988.
The authors have disclosed that they do not have any potential conflicts of interest.
For information regarding this article, E-mail: firstname.lastname@example.org