Excess mortality in women with hospital-acquired bloodstream infection

PURPOSE: We examined the outcomes of bloodstream infection in men and in wo men and whether any sex-related differences were explained by underlying di sorders, severity of disease, or clinical management. SUBJECTS AND METHODS: Using a prospectively collected database, we compared in-hospital mortality in men and women. We used multivariable logistic reg ression analysis to test whether sex-related differences could be due to po tential confounders. RESULTS: Of 4250 patients with bloodstream infections) 1750 (41%) had hospi tal-acquired infections. The overall case fatality was 31% (625 of 2032) in women and 29% (631 of 2218, P = 0.1) in men. However, 43% (325/758) of the women with hospital-acquired infections died, compared with 33% (327/992) of the men (P = 0.0001). In a multivariate analysis, female sex was associa ted with greater mortality in patients with hospital-acquired infections (o dds ratio = 1.7; 95% confidence interval: 1.1 to 2.6). The excess mortality in women was mainly seen in patients with major underlying disorders (fata lity rate of 45% [234 of 525] in women vs. 32% in men [234 of 743, P = 0.00 01). CONCLUSIONS: Mortality in women with hospital-acquired bloodstream infectio ns is substantially greater than in men. The excess mortality was concentra ted in women with severe underlying disorders, suggesting that sepsis might have accentuated differences in the outcome of underlying disorders in wom en. (C) 2001 by Excerpta Medica, Inc.