Impact of Clostridium difficile infection on resource utilization and clinical outcomes in hospitalized patients

Abstract number: P1466

Emons M.F., Nathanson B.H., Yu H.T., Haidar T., Alemao E., Gardiner D., Spoeri R.K.

Objective: To evaluate the impact of Clostridium difficile infection (CDI) on resource utilisation and in-hospital mortality in patients with hospital-onset healthcare facility-associated (HO-HCFA) CDI.

Methods: This retrospective analysis utilised data from April 1, 2005 to December 31, 2008 within Cerner's Health Facts® database. Qualifying cases included adults with an inpatient encounter with a primary or secondary diagnosis of pseudomembranous colitis (ICD-9-CM code: 008.45) AND a positive C. difficile toxin test AND the first positive C. difficile test (index event) was ordered more than 72 hours after admission. Non-CDI controls were hospitalised for at least 72 hours with no diagnosis of pseudomembranous colitis AND no C. difficile culture or toxin assays during the study period. Primary outcomes were in-hospital mortality, hospital length-of-stay (LOS) and 90-day all-cause readmission. A propensity-score-matched sample and a non-parse regression model with the propensity score itself and all other covariates (eg, patient demographics, clinical markers of severity of illness, treatments of interest [ie, antibiotics, proton pump inhibitors, H2 antagonists, and non-steroidal anti-inflammatory drugs]) were both used to assess the effect of HO-HCFA CDI on outcomes.

Results: A 5:1 Greedy algorithm matched 1,312 cases to 1,312 controls. Patients with HO-HCFA CDI had significantly longer unadjusted hospital LOS than non-CDI patients (mean [SD]: 22.0 [19.0] versus 6.6 [5.8] days, P < 0.001). After matching, the LOS among survivors with HO-HCFA CDI was 10.8 days (95% CI, 9.7–12.0) longer than the non-CDI control group and LOS after the CDI event was equivalent to the entire LOS in the control group. No significant differences in mortality were found between matched HO-HCFA CDI and non-CDI groups (odds ratio = 0.89; 95% CI, 0.69–1.15) though the matched mortality rate of HO-HCFA CDI patients was high (8.7%). Thirty-three percent of all matched survivors required readmission within 90 days. The relative risk for readmission was significantly higher for HO-HCFA patients: 1.56 (95% CI, 1.38–1.76).

Conclusions: HO-HCFA CDI is a growing public health concern and is associated with significantly longer hospitalisations and higher risk of hospital readmission within 90 days. To reduce the higher resource utilisation among these patients, a better understanding is needed of risk factors and potential preventative measures.

Session Details

Date: 10/04/2010
Time: 00:00-00:00
Session name: Abstracts 20th European Congress of Clinical Microbiology and Infectious Diseases
Location: Vienna, Austria, 10 - 13 April 2010
Presentation type:
Back to top