Hospital-acquired Clostridioides difficile infections in Alberta: The validity of laboratory-identified event surveillance versus clinical infection surveillance

Abstract

Background: Clostridioides difficile infection (CDI) is one of the most common health care-associated infections. This study assessed the validity of a laboratory-based surveillance method as compared with a traditional, clinical surveillance method to identify hospital-acquired CDIs.

Methods: Retrospective analysis of positive C difficile laboratory records between April 2015 and March 2017 were compared with a clinical dataset of positive inpatient C difficile cases for all acute care facilities in Alberta, Canada. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated using STATA/IC 13.0.

Results: The laboratory surveillance method had a sensitivity of 96.6% (95% confidence interval [CI], 95.7%-97.3%) and a specificity of 65.7% (95% CI, 63.6%-67.8%); positive predictive value and negative predictive value were 74.3% (95% CI, 73.2%-75.5%) and 94.9% (95% CI, 93.7%-95.9%), respectively.

Discussion: To date, the breadth of research on alternate CDI surveillance systems has focused on the use of International Statistical Classification of Diseases and Related Health Problems 9th and 10th Revision coding mechanisms. Our results expand on the published literature, as a laboratory approach may provide more timely information, with a smaller amount of misclassified cases.

Conclusions: Using a laboratory surveillance method to capture hospital-acquired CDI cases is highly sensitive but not overly specific. Changes to improve the specificity of this method are provided.

Keywords: Administrative databases; Electronic surveillance; Hospital-acquired infections; Inappropriate testing; Sensitivity; Specificity.