Race is Not an Independent Risk Factor for Poor Outcomes in Clostridium difficile Infection

Abstract

Purpose: To determine if race is a predictor of poor C. difficile infection (CDI) outcomes, as measured by rates of colectomy and death. Methods: Demographic and clinical data for all admissions to Rhode Island Hospital (RIH) from January 1st 2006 to December 31st 2010 were collected. All CDI cases during this period were reviewed to determine age, gender, race, mode of acquisition [hospital-acquired (HA), healthcare-associated (HCA), or community-acquired (CA)], need for colectomy, and death. Poor outcome was defined as CDI in a patient who underwent colectomy and/or died during the same hospitalization. We performed bivariate analyses of CDI cases with or without poor outcome by age, gender, mode of acquisition, relapse status, and race. Multiple logistic regression was performed to determine if race was an independent predictor of poor outcome. Results: There were 142,122 admissions to RIH. Of these, 82% were white, while 9% were black. Over the same time period, there were 2,109 cases of CDI in 1,951 unique patients. The mean (±SD) age of CDI patients was 68 (±18). Of the CDI cases, 86% occurred in whites and 8% in blacks. CDI acquisition was HA in 59%, HCA in 13%, and CA in 28%. Colectomy occurred in 42 patients (2%). There were 318 deaths (15%). Whites had a significantly higher proportion of CDI cases when compared to overall RIH admissions (86% vs. 82%, p<0.001). However, there was no significant difference in the proportions of black CDI cases and overall black admissions (8% vs. 9%, p=0.7). A subgroup analysis of CDI relapsers vs. non-relapsers revealed no significant differences in age, gender, or race. Individuals who died were significantly older than survivors (74±14 vs. 67±19, p<0.001). There were significant differences in mortality by mode of acquisition, with HA and HCA infections resulting in death in 18% and 17% of cases, respectively, while CA infection had a lower mortality rate of 8% (p<0.001). There was no significant difference in colectomy rate between blacks and whites (3% vs. 2%, p=0.7). Similarly, there was no significant difference in mortality between blacks and whites (13% vs. 15%, p=0.4). In a multiple logistic regression model, accounting for age and mode of acquisition, black race was neither protective against nor a risk factor for a poor outcome (OR 1.0, 95% CI 0.6-1.7). Conclusion: White individuals may contract CDI more frequently than would be expected from their proportion of overall hospital admissions. However, race does not appear to be an independent predictor of poor CDI outcomes. As shown in previous studies, both age and mode of acquisition are associated with poor outcomes in CDI.