Methods: We implemented an electronic medical record (EMR) based ATO in August 2016 at a tertiary academic medical center. We leveraged native alerting functionality in the EMR to present a notification to the primary clinician after a patient had received intravenous vancomycin (V) and/or piperacillin/tazobactam (Z) for more than 48 hours. The ‘best practice alert’ (BPA) reminded the clinician to review the appropriateness of antibiotic selection and to attest to this action.
Results: After implementation of this ATO, we reviewed a random sampling of 100 patients initiated on V and/or Z therapy for whom providers received an automated ATO notification and compared antibiotic use to 100 patients who received the same therapy prior to the ATO implementation. We found an overall increase in de-escalation (defined as either changing or stopping therapy) on day 3 of V and/or Z therapy from 27% pre-ATO to 70% for post-ATO. For overall V therapy, we found a de-escalation from 27% pre-ATO to 68% for post-ATO and for overall Z therapy, we found a de-escalation from 25% pre-ATO to 72% for post-ATO.
Conclusion: An EMR generated BPA ATO had a significant impact on de-escalation of V and/or Z on day 3 of therapy.
B. Clay, None
S. Abeles, None