Attributable burden of hospital-onset Clostridium difficile infection: a propensity score matching study

Abstract

Objective: To determine the attributable in-hospital mortality, length of stay (LOS), and cost of hospital-onset Clostridium difficile infection (HO-CDI).

Design: Propensity score matching.

Setting: Six Pennsylvania hospitals (2 academic centers, 1 community teaching facility, and 3 community nonteaching facilities) contributing data to a clinical research database.

Patients: Adult inpatients between 2007 and 2008.

Methods: We defined HO-CDI in adult inpatients as a positive C. difficile toxin assay result from a specimen collected more than 48 hours after admission and more than 8 weeks following any previous positive result. We developed an HO-CDI propensity model and matched cases with noncases by propensity score at a 1∶3 ratio. We further restricted matching within the same hospital, within the same principal disease group, and within a similar length of lead time from admission to onset of HO-CDI.

Results: Among 77,257 discharges, 282 HO-CDI cases were identified. The propensity score-matched rate was 90%. Compared with matched noncases, HO-CDI patients had higher mortality (11.8% vs. 7.3%; P < .05), longer LOS (median [interquartile range (IQR)], 12 [9-21] vs. 11 [8-17] days; P < .01), and higher cost (median [IQR], $20,804 [$11,059-$38,429] vs. $16,634 [$9,413-$30,319]; P < .01). The attributable effect of HO-CDI was 4.5% (95% confidence interval [CI], 0.2%-8.7%; P < .05) for mortality, 2.3 days (95% CI, 0.9-3.8; P < .01) for LOS, and $6,117 (95% CI, $1,659-$10,574; P < .01) for cost.

Conclusions: Patients with HO-CDI incur additional attributable mortality, LOS, and cost burden compared with patients with similar primary clinical condition, exposure risk, lead time of hospitalization, and baseline characteristics.