Clostridiodes difficile Testing Protocol: An Effective Strategy to Decrease Over-Testing Patients for C. difficile Without Missing True Positive Infections

TL;DR

This study shows a protocol effectively reduces unnecessary C. difficile testing in hospitals without missing true infections.

Contribution

A new diagnostic stewardship protocol for C. difficile testing is shown to reduce over-testing while maintaining high accuracy.

Findings

The protocol canceled 391 out of 443 C. difficile tests, significantly reducing unnecessary testing.

The protocol had a 96.82% negative predictive value, indicating it rarely missed true infections.

Only 12 false negatives occurred within 30 days of canceled tests, confirming protocol effectiveness.

Abstract

Objectives: Patients presenting with symptomatic diarrhea are frequently tested for Clostridioides difficile infection (CDI). However, increasing numbers of false positive results could lead to unnecessary testing and misuse of hospital resources. Additionally, physicians are often skeptical about the diagnostic stewardship protocols established by clinical institutions because of their uncertain negative predictive value and often err on the side of ordering the test despite the protocol (error of commission rather than error of omission). The objective of this study is to evaluate the implemented screening test protocol at Baylor Scott & White All Saints Medical Center in terms of its efficacy in mitigating the overuse of hospital testing resources while appropriately triaging true cases of CDI and the negative predictive value of the diagnostic stewardship protocol.

Methods: We performed a retrospective, observational study at Baylor Scott & White All Saints Medical Center, Fort Worth, between January 1, 2021, and January 1, 2022, where C. difficile testing was analyzed after implementation of our protocol.

Results: 443 patient encounters contained C. difficile stool testing orders. Of the 443 initial orders, 391 were confirmed to be canceled. Fifty-two orders were carried out, of which 7 were positive for C. difficile and 45 were negative. Of the 391 C. difficile tests canceled, 377 were due to the study protocol during the index encounter, and 14 were canceled for undocumented reasons. Of these 377 canceled orders, there were 12 false negatives due to a positive C. difficile result on the repeat test performed within 30 days of the index order canceled based on the protocol criteria, thus leaving 365 true negatives. The negative predictive value (NPV) of the protocol was 96.82% (365/377).

Conclusions: Our protocol proves effective in avoiding false positive hospital-related C. difficile infections, and at the same time, very effective in not missing out on the true positive C. difficile infections (excellent NPV). It is an important quality improvement initiative that can be implemented at any hospital.

Figures