EHR Opioid Decision Support Had Low Adoption and Prescribing Fidelity

Researchers at one academic health system evaluated an electronic health record (EHR) tool intended to tailor postoperative opioid prescriptions to patients’ inpatient opioid consumption. The mixed-methods pilot included an initial randomized comparison of procedure-based and oral morphine equivalent (OME)-based recommendations, interviews with 20 clinicians, and a redesigned tool piloted across three general surgery teams. The report dates the second pilot to February–July 2026; phase-specific measures were not directly comparable.

In the second phase, 143 of 215 encounters (66.5%) resulted in an opioid prescription. The decision-support order set was used in 48 of those 143 encounters (33.6%), but only 11 prescriptions (7.7% of all opioid prescriptions) met the defined fidelity threshold of within ±7.5 OME of expected daily requirements. Among tool users, 11 of 48 prescriptions (22.9%) matched; 25 of 37 nonmatching prescriptions (68%) exceeded expected requirements. Of 72 opioid-free discharges, 61 (84.7%) met the study’s appropriateness criteria.

Clinician interviews identified alert fatigue, prescribing habits, rigid taper instructions, fragmented information, and service-level variation as barriers. The small, single-center pilot suggests that measuring tool use alone may overstate implementation success. Its findings do not establish that the tool improves patient outcomes or support a change in prescribing practice.