Utilizing a real-time discussion approach to improve the appropriateness of Clostridioides difficile testing and the potential unintended consequences of this strategy.

We report electronic medical record interventions to reduce Clostridioides difficile testing risk 'alert fatigue.' We used a behavioral approach to diagnostic stewardship and observed a decrease in the number of tests ordered of ~4.5 per month (P < .0001). Although the number of inappropriate tests decreased during the study period, delayed testing increased.

Why do clinicians order inappropriate Clostridium difficile testing? An exploratory study.

BACKGROUND: The drivers behind Clostridium difficile testing are not well understood. METHODS: C difficile testing orders were reviewed. An algorithm that sequentially considered the presence of diarrhea, leukocytosis, fever, and laxative use was created. Tests deemed potentially inappropriate were discussed with the ordering team. RESULTS: Of 128 orders reviewed, 62% (n = 79) were appropriate. Patients whose testing was deemed inappropriate had longer lengths of stay. Diarrhea and laxative use were common and similarly distributed in those appropriately or inappropriately tested. The most commonly cited reason for ordering an inappropriate test was the reporting of diarrhea to the clinician by the patient or nursing that was not documented in the electronic health record. The next most common reason was clinician perception of risk. Demographics, laxative use, fever, leukocytosis, and diarrhea were similarly distributed between patients testing positive or negative by C difficile polymerase chain reaction. DISCUSSION: The discriminating value of diarrhea or laxative use in assessing for C difficile infection is poor. Multiple streams of information add to the complexities of decision making around C difficile testing. Inconsistent definitions of diarrhea likely contribute to this complexity. Clinician-perceived risk to the patient, partially driven by length of stay, may be a large driver of testing practices. CONCLUSIONS: Without understanding the knowledge, attitudes, and values that underlie clinician behavior, interventions targeting ordering practices may not succeed.