Impact of automated alerts on discharge opioid overprescribing after general surgery

Elsie Rizk,Navjot Kaur,Phuong Y Duong,Ezekiel Fink,Matthew A Wanat,J Douglas Thornton,Min P Kim
DOI: https://doi.org/10.1093/ajhp/zxae185
2024-07-01
Abstract:Disclaimer: In an effort to expedite the publication of articles, AJHP is posting manuscripts online as soon as possible after acceptance. Accepted manuscripts have been peer-reviewed and copyedited, but are posted online before technical formatting and author proofing. These manuscripts are not the final version of record and will be replaced with the final article (formatted per AJHP style and proofed by the authors) at a later time. Purpose: The objectives of this study were to identify the most performed surgical procedures associated with the highest rates of discharge opioid overprescribing and to implement an electronic health record (EHR) alert to reduce discharge opioid overprescribing. Methods: This quality improvement, before-and-after study included patients undergoing one of the identified target procedures-laparoscopic cholecystectomy, unilateral open inguinal hernia repair, and laparoscopic appendectomy-at an academic medical center. The alert notified providers when the prescribed opioid quantity exceeded guideline recommendations. The preimplementation cohort included surgical encounters from January 2020 to December 2021. The EHR alert was implemented in May 2022 following provider education via email and in-person presentations. The postimplementation cohort included surgical encounters from May to August 2022. The primary outcome was the proportion of patients with a discharge opioid supply exceeding guideline recommendations (overprescribing). Results: A total of 1,478 patients were included in the preimplementation cohort, and 141 patients were included in the postimplementation cohort. The rate of discharge opioid overprescribing decreased from 48% in the preimplementation cohort to 3% in the postimplementation cohort, with an unadjusted absolute reduction of 45% (95% confidence interval, 41% to 49%; P < 0.001) and an adjusted odds ratio of 0.03 (95% confidence interval, 0.01 to 0.08; P < 0.001). Among patients who received opioids, the mean (SD) opioid supply at discharge decreased from 92 (43) oral morphine milligram equivalents (MME) (before implementation) to 57 (20) MME (after implementation) (P < 0.001). The proportion of patients who received additional opioid prescriptions within 1 to 14 days of hospital discharge did not change (P = 0.76). Conclusion: Implementation of an EHR alert along with provider education can reduce discharge opioid overprescribing following general surgery.
What problem does this paper attempt to address?