Abstract
The elderly population is particularly vulnerable to Clostridium difficile infection (CDI), but the epidemiology of CDI in long-term care facilities (LTCFs) is unknown.We performed a retrospective cohort study and used US 2011 LTCF resident data from the Minimum Data Set 3.0 linked to Medicare claims. We extracted CDI cases based on International Classification of Diseases-9 coding, and compared residents with the diagnosis of CDI to those who did not have a CDI diagnosis during their LTCF stay. We estimated CDI prevalence rates and calculated 3-month mortality rates.The study population consisted of 2,190,613 admissions (median age 82 years; interquartile range 76-88; female to male ratio 2:1; >80% whites), 45,500 of whom had a CDI diagnosis. The nationwide CDI prevalence rate was 1.85 per 100 LTCF admissions (95% confidence interval [CI] 1.83-1.87). The CDI rate was lower in the South (1.54%; 95% CI 1.51-1.57) and higher in the Northeast (2.29%; 95% CI 2.25-2.33). Older age, white race, presence of a feeding tube, unhealed pressure ulcers, end-stage renal disease, cirrhosis, bowel incontinence, prior tracheostomy, chemotherapy, and chronic obstructive pulmonary disease were independently related to "high risk" for CDI. Residents with a CDI diagnosis were more likely to be admitted to an acute care hospital (40% vs 31%, P < 0.001) and less likely to be discharged to the community (46% vs 54%, P < 0.001) than those not reported with CDI during stay. Importantly, CDI was associated with higher mortality (24.7% vs 18.1%, P = 0.001).CDI is common among the elderly residents of LTCFs and is associated with significant increase in 3-month mortality. The prevalence is higher in the Northeast and risk stratification can be used in CDI prevention policies.