Recurrent Clostridium difficile infection among Medicare patients in nursing homes: A population-based cohort study

Abstract

We explored the epidemiology and outcomes of Clostridium difficile infection (CDI) recurrence among Medicare patients in a nursing home (NH) whose CDI originated in acute care hospitals. We conducted a retrospective, population-based matched cohort combining Medicare claims with Minimum Data Set 3.0, including all hospitalized patients age > /=65 years transferred to an NH after hospitalization with CDI 1/2011-11/2012. Incident CDI was defined as ICD-9-CM code 008.45 with no others in prior 60 days. CDI recurrence was defined as (within 60 days of last day of CDI treatment): oral metronidazole, oral vancomycin, or fidaxomicin for > /=3 days in part D file; or an ICD-9-CM code for CDI (008.45) during a rehospitalization. Cox proportional hazards and linear models, adjusted for age, gender, race, and comorbidities, examined mortality within 60 days and excess hospital days and costs, in patients with recurrent CDI compared to those without. Among 14,472 survivors of index CDI hospitalization discharged to an NH, 4775 suffered a recurrence. Demographics and clinical characteristics at baseline were similar, as was the risk of death (24.2% with vs 24.4% without). Median number of hospitalizations was 2 (IQR 1-3) among those with and 0 (IQR 0-1) among those without recurrence. Adjusted excess hospital days per patient were 20.3 (95% CI 19.1-21.4) and Medicare reimbursements $12,043 (95% CI $11,469-$12,617) in the group with a recurrence.Although recurrent CDI did not increase the risk of death, it was associated with a far higher risk of rehospitalization, excess hospital days, and costs to Medicare.