Gender-Dependent Differences in Outcome After the Treatment of Infection in Hospitalized Patients

CONTEXT: While it is established that management strategies and outcomes differ by gender for many diseases, its effect on infection has not been adequately studied. OBJECTIVE: To investigate the role of gender among hospitalized patients treated for infection. DESIGN: Observational cohort study conducted during a 26-month period from December 1996 through January 1999. SETTING: University-affiliated hospital. PARTICIPANTS: A total of 892 patients in the surgical units of the hospital with 1470 consecutive infectious episodes (782 in men and 688 in women). MAIN OUTCOME MEASURES: Mortality during hospitalization by gender for infection episodes overall and for specific infectious sites, including lung, peritoneum, bloodstream, catheter, urine, surgical site, and skin/soft tissue. RESULTS: Among all infections, there was no significant difference in mortality based on gender (men, 11.1% vs women, 14.2%; P = .07). After logistic regression analysis, factors independently associated with mortality included higher APACHE (Acute Physiology and Chronic Health Evaluation) II score, older age, malignancy, blood transfusion, and diagnosis of infection more than 7 days after admission, but not gender (female odds ratio OR for death, 1.32; 95% confidence interval CI, 0.90-1.94; P = .16). Mortality was higher in women for lung (men, 18% vs women, 34%; P = .002) and soft tissue (men, 2% vs women, 10%; P < or = .05) infection; for other infectious sites, mortality did not differ by gender. Factors associated with mortality due to pneumonia by logistic regression included higher APACHE II score, malignancy, diabetes mellitus, diagnosis of infection more than 7 days after admission, older age, transplantation, and female gender (OR for death, 2.25; 95% CI, 1.17-4.32; P = .02). CONCLUSIONS: Although gender may not be predictive of mortality among all infections, women appear to be at increased risk for death from hospital-acquired pneumonia, even after controlling for other comorbidities.