Gender differences in case mix and outcome of critically ill patients

Introduction: The potential for gender-related bias in the provision of medical treatments has gained increased interest in recent years. The aim of this retrospective, observational study was to evaluate the association between gender and clinical outcome in an Italian mixed medical-surgical ICU population. Methods: Data on 1978 patients admitted to the ICU during a 3-year period were analyzed. Demographics, diagnosis, and hospital stay details were recorded. Results: Male ICU admissions were predominant over female ones (64% vs 36%). Neither ICU survival rate (80% in group male, 79% in group female; P = 0.602) nor hospital survival rate (72% in group male, 72% in group female; P = 0.820) showed gender-related differences. A statistically significant difference was found in terms of mean (SD) age (57 [19] years in group male, 62 [18] years in group female; P < 0.001), ICU length of stay (7.0 [9.1] days in group male, 5.7 [7.7] days in group female; P < 0.001) and length of mechanical ventilation (6.3 [8.4] days in group male, 5.3 [7.5] days in group female; P = 0.001). Severity of illness, measured through the simplified acute physiology score II, was not statistically different between gender groups; nor was the incidence of infective complications. After stratifying for diagnostic subgroups a few gender differences were pointed out, but none of them affecting ICU and hospital survival rates. A Kaplan-Meier 30-day ICU survival analysis revealed no differences between the male and female groups of the study population. Conclusions: According to our results, mortality among critically ill patients was not influenced by gender. Despite a higher frequency of men admitted, women were older than men. Moreover, men were treated for a longer period of time than women. Limitations of the study were the inability to establish causal relations and to account for variables with important effects on the reported associations. Moreover, the sample size was small if compared to similar multicenter studies. © 2011 Elsevier HS Journals, Inc. All rights reserved