190. Promise and Pitfalls: Analyzing the Accuracy and Perceptions of Documenting Antibiotic Indication During Order Entry

BACKGROUND: The CDC and The Joint Commission recommend that prescribers document an indication (IND) when ordering antibiotics (ABX). However, to assist with ABX stewardship efforts, the IND needs to be accurate. Previous data with ABX order forms showed that the IND provided by prescribers is often erroneous. This study sought to assess the accuracy of IND documentation as well as prescriber perceptions of the indication requirement at our institution. METHODS: Patients who received ≥1 dose of ceftriaxone, ciprofloxacin, ertapenem, oxacillin, piperacillin/tazobactam, or vancomycin from May 1, 2017 to June 30, 2017 were identified. The electronic medical records of 30 randomly selected patients who received each study ABX were retrospectively reviewed. Patients receiving multiple ABX were evaluated only once. The primary endpoint was IND accuracy defined as agreement of IND entered during order entry and the IND documented in progress notes at the time of order entry. Secondary endpoints included correlation of entered IND and final diagnosis (DX) for empiric ABX and IND accuracy according to prescribing service and specific IND. A brief survey was emailed to prescribers to assess the burden and perceptions of requiring an IND during order entry. RESULTS: A total of 4,524 patients received ≥1 dose of a study ABX. Of the 180 patients selected for evaluation, 89.4% of INDs were accurate. 51.7% of ABX were classified as empiric, 41.1% as definitive, and 7.2% as prophylactic. ABX entered as prophylaxis were more likely to be inaccurate than empiric or definitive ABX (46%, 94%, 92%, respectively, P < 0.05). For empiric ABX, 78.5% of INDs documented at order entry matched the final DX. IND accuracy varied among services (81.8–94.7%, P = NS). Of 863 (29%) prescribers, 254 responded to the survey. Respondents felt that documenting the IND took ≤10 (33%) or 11–20 seconds (33%). They considered it a “minor nuisance” (46%) or “occasionally burdensome” (27%). Most (55%) felt it had no impact on their consideration of ABX appropriateness. CONCLUSION: With the exception of prophylaxis, the accuracy of ABX IND documented during order entry was sufficiently reliable to assist stewardship efforts. Although IND documentation is perceived as only a minor burden, prescribers attributed minimal beneficial effect on ABX prescribing. DISCLOSURES: All authors: No reported disclosures.