The Impact of a Computerized Clinical Decision Support Tool on Inappropriate Clostridium difficile Testing

Inappropriate diagnostic testing remains a persistent challenge in hospital settings. Prior research has shown that C. difficile testing is frequently ordered in patients receiving laxatives, which can interfere with test accuracy. No prior work had resolved how to systematically reduce such inappropriate testing. This gap motivated the development of interventions targeting test ordering behavior. Established knowledge includes the impact of laxatives on C. difficile diagnostics. This paper's contribution lies in evaluating a specific clinical decision support tool's effect on test appropriateness. The study context is a large health system with electronic health records. No prior work had resolved the effectiveness of integrated order sets for this purpose. This paper addresses that uncertainty through a structured retrospective analysis.

The aim of this study was to assess whether a computerized clinical decision support intervention could reduce inappropriate C. difficile testing. Inappropriate testing occurs when patients receiving laxatives are tested for C. difficile. The motivation stems from the known interference of laxatives with test accuracy. The researchers propose that integrating a decision support tool into electronic health records could improve test ordering. The intervention required providers to use a specific order set. The study focused on adult inpatients over a two-year period. The primary goal was to measure changes in inappropriate testing rates. The secondary goal was to evaluate changes in laxative discontinuation and testing decisions.

The study used a retrospective cohort design across three tertiary-care hospitals. The intervention involved a commercial electronic health record system. The order set flagged patients who had received laxatives in the prior 36 hours. A message prompted providers to reassess laxative use before testing. Providers could choose to continue or discontinue laxatives. Testing could be ordered or avoided based on the prompt. The primary endpoint was inappropriate testing rates. Secondary endpoints included laxative discontinuation and complication rates. Data were analyzed using statistical comparisons between baseline and intervention periods.

The intervention decreased inappropriate C. difficile testing from 29.6% to 27.3% (P=.02). More patients discontinued laxatives and avoided testing (5.8% vs 46.4%; P<.01). A larger proportion discontinued laxatives and underwent testing (5.4% vs 35.2%; P<.01). Complication rates increased nonsignificantly (5.0% vs 8.9%; P=.11). These findings suggest the order set influenced provider behavior. The largest effect was in reducing inappropriate testing. The intervention also increased laxative discontinuation decisions. No significant decrease in complications was observed. The results support the effectiveness of integrated decision support tools.

The authors stated that the order set successfully reduced inappropriate C. difficile testing. The intervention improved timely discontinuation of laxatives before testing. The largest change was in decreasing inappropriate test orders. The increase in laxative discontinuation was statistically significant. The nonsignificant rise in complications was not attributed to the intervention. The authors propose that decision support tools can improve diagnostic appropriateness. The findings suggest that electronic health records can guide provider decisions. The study supports the use of integrated order sets for diagnostic testing.