Attributable inpatient costs of recurrent Clostridium difficile infections

Erik R. Dubberke, Eric Schaefer, Kimberly A. Reske, Marya Zilberberg, Christopher S. Hollenbeak, Margaret A. Olsen

Research output: Contribution to journalArticle

31 Scopus citations

Abstract

Objective: To determine the attributable inpatient costs of recurrent Clostridium difficile infections (CDIs).

Design: Retrospective cohort study.

Setting: Academic, urban, tertiary care hospital.

Patients: A total of 3,958 patients aged 18 years or more who developed an initial CDI episode from 2003 through 2009.

Methods: Data were collected electronically from hospital administrative databases and were supplemented with chart review. Patients with an index CDI episode during the study period were followed up for 180 days from the end of their index hospitalization or the end of their index CDI antibiotic treatment (whichever occurred later). Total hospital costs during the outcome period for patients with recurrent versus a single episode of CDI were analyzed using zero-inflated lognormal models.

Results: There were 421 persons with recurrent CDI (recurrence rate, 10.6%). Recurrent CDI case patients were significantly more likely than persons without recurrence to have any hospital costs during the outcome period (P <.001). The estimated attributable cost of recurrent CDI was $11,631 (95% confidence interval, $8,937–$14,588).

Conclusions: The attributable costs of recurrent CDI are considerable. Patients with recurrent CDI are significantly more likely to have inpatient hospital costs than patients who do not develop recurrences. Better strategies to predict and prevent CDI recurrences are needed.

Original languageEnglish
Pages (from-to)1400-1407
Number of pages8
JournalInfection Control and Hospital Epidemiology
Volume35
Issue number11
DOIs
StatePublished - Nov 1 2014

Fingerprint Dive into the research topics of 'Attributable inpatient costs of recurrent Clostridium difficile infections'. Together they form a unique fingerprint.

  • Cite this