B55-01 Impact of a Standardized Phenobarbital Protocol on Hospital Length of Stay and ICU Utilization in Alcohol Withdrawal Syndrome

Background Alcohol withdrawal syndrome (AWS) remains a major cause of hospitalization and is associated with significant morbidity, mortality, and ICU utilization. Benzodiazepines have traditionally been first-line therapy; however, phenobarbital has re-emerged as an alternative due to its dual GABAergic and glutamate-inhibiting effects, predictable pharmacokinetics, and potential to reduce ICU admissions and hospital length of stay (LOS). In October 2022, Providence Portland Medical Center (PPMC) implemented a standardized phenobarbital-based protocol and order set for AWS management. This study examined if there was a significant relationship between the aforementioned policy changes on institutional uptake, resource utilization, and clinical outcomes. Methods A retrospective cohort analysis was conducted of adult patients admitted to PPMC with a primary or secondary diagnosis of AWS between January 2019 and October 2025. Patients with a diagnosis of epilepsy or seizure disorder were excluded. Outcomes before (1/1/2019-10/21/2022) and after (10/22/2022-10/28/2025) phenobarbital protocol implementation were compared using electronic health record data obtained using Structured Query Language (SQL) to interact with the Providence Clarity Database (or EPIC database). The primary outcome was hospital LOS. Secondary outcomes included phenobarbital utilization, ICU admission, and in hospital mortality. Results A total of 4,489 patients met inclusion criteria (2,546 pre-implementation and 1,943 post-implementation). Following protocolization of AWS management, institutional uptake of phenobarbital-based treatment increased nearly threefold, from 12.45% of patients in the pre-protocol cohort receiving phenobarbital to 34.57% in the post-protocol cohort (p < 0.001). The average LOS decreased from 9.51 days to 7.05 days (p < 0.001). Analysis of secondary outcomes, show decreased ICU admissions after the protocol implementation from 28.03% to 19.19% (p < 0.001), and decreased mortality from 4.08% to 2.93% (p = 0.039). Conclusion Implementation of a standardized phenobarbital-based AWS protocol at a large tertiary care hospital was associated with a marked increase in phenobarbital utilization, a significant reduction in hospital LOS, decreased ICU utilization, and decreased mortality. These findings support the safety, efficacy, and scalability of protocolized phenobarbital use for the management of AWS in hospitalized patients and highlight the potential of standardized pathways to optimize care delivery and ICU resource utilization while providing a reproducible model for other institutions. However, as this was a retrospective cohort study rather than a randomized controlled trial, causality between the protocol implementation and observed outcomes cannot be definitively established.