Guideline concordance of electronic health record order sets for hospital-based treatment of alcohol withdrawal syndrome.

Background: Treatment of alcohol withdrawal syndrome (AWS) in hospitals is inconsistent. Electronic health record (EHR) order sets protocolize care.

Objective: We examined variation in AWS order sets across hospital organizations and their concordance with AWS guidelines.

Methods: We conducted a cross-sectional study of hospital organization user-created EHR order sets for AWS extracted from the December 2021 Epic® userweb community library. Hospital organizations with an acute care hospital and $\ge$ 1 AWS order set were included. We measured the proportion of guideline-concordant care practices within four categories: (1) laboratory assessment, (2) risk assessment for severe AWS and associated management changes, (3) symptom assessment and treatment of AWS, and identification and management of complications and (4) screening, diagnosis, and treatment of unhealthy alcohol use and AUD including medications for alcohol use disorder (MAUD).

Results: Ninety-five organizations with 289 order sets were included. The proportion of organizations with guideline-concordant laboratory assessments included testing of electrolytes (83%), hepatic function (75%), substance use (83%), and screening for infections (33%). Guidance for assessing risk of severe AWS (34%) and indications for care escalation (63%) used inconsistent definitions. Use of guideline-concordant medications for AWS (99%) and AWS symptom scores (91%) were nearly universal. MAUD was included by two organizations (2%). A common templated order set was used by 26% of organizations in EHR order sets.

Conclusions: We observed frequent organizational inclusion of guideline-concordant medications and symptom scores but rare and/or poorly defined guidance for evaluating risk of severe AWS, escalation of care, and MAUD.