177. Use of Electronic Best Practice Alert (BPA) to Reduce Inappropriate Testing for Clostridium difficile infection (CDI) at a Tertiary Care Center

Abstract

Background Clostridium difficile assays are unable to differentiate between active infection and asymptomatic carriage. Failure to account for noninfectious causes of diarrhea in hospitalized patients may contribute to overdiagnosis of CDI, leading to unnecessary treatment and increased cost of care. This study assessed if a best practice alert (BPA) could improve ordering practices.MethodsA BPA was instituted in a tertiary academic medical center in the electronic ordering system on July 17, 2017 alerting providers ordering C. difficile testing if their patient had received laxatives, oral contrast, or initiation of tube feeds within the preceding 48 hours. Reevaluation of diarrhea was recommended 48 hours after discontinuation of laxatives or initiation of new tube feeds in stable patients. BPA override was available for the following scenarios: high clinical suspicion, concern for severe disease and diagnosis cannot be delayed, ileus, worsened diarrhea on chronic tube feeds, worsened diarrhea on chronic laxatives. The rate of test positivity was compared before and after BPA implementation by χ2 test.ResultsBetween July 17, 2017 and December 31, 2017, the BPA triggered in 1,284 unique clusters (all BPAs firing within 48 hours of another BPA were considered a single cluster). Chart review showed cancellation of CDI testing in 416 (32.4%) cases and delayed tests for at least 24 hours in 163 (12.7%) cases. Of 868 tests where the BPA was initially or ultimately overridden after 24 hours, 186 (21%) were positive. The most common reasons for BPA override were: 512 (39.9%) high clinical suspicion and 177 (13.8%) worsened diarrhea on tube feeds. The rate of inpatient CDI testing declined from 191.6/10,000 patient-days in the January–June 2017 period to 165.9/10,000 patient-days in July–December 2017 (P < 0.001). The proportion of positive tests increased from 8.4% to 10.6% (P = 0.006) over the same time periods. Two cases of delayed diagnosis were identified; no treatment complications were noted on chart review.ConclusionAn electronic BPA may be an effective tool to identify alternative etiologies of diarrhea in hospitalized patients and reduce inappropriate/enrich appropriate CDI testing. No adverse outcomes were noted in patients with delayed CDI diagnosis.Disclosures All authors: No reported disclosures.