Background Alcohol use is frequent in trauma patients and alcohol withdrawal syndrome (AWS) is associated with significant morbidity. Benzodiazepines are commonly used for AWS, but may cause neurologic and respiratory adverse events (AEs). The objective was to evaluate the effectiveness and safety of a phenobarbital-based protocol for the treatment of AWS in non-intensive care unit (ICU) trauma patients. Methods Adult non-ICU trauma patients at high risk of or experiencing AWS PRE and POST implementation of a phenobarbital-based protocol were included. Outcomes were AWS-related complications (AWS-RC), benzodiazepine use, adjunctive medication use, hospital length of stay (HLOS), and medication-related AEs. Subgroup analyses were performed on patients with traumatic brain injury (TBI), rib fractures, and at high risk of severe AWS. Results Overall, 110 patients were included (51 PRE, 59 POST). AWS-RC developed in 17 PRE patients compared to 10 POST patients (33% vs 17%; P = .05). PRE patients were more likely to receive benzodiazepines (88% vs 42%, P < .0001) and higher total dose (11 vs 4 mg lorazepam equivalent; P = .001). No difference noted in HLOS (8 vs 8 days, P = .27), adjunctive medication use (49% vs 54%, P = .60), or AEs (57% vs 39%, P = .06). There was no difference in AWS-RC in the TBI subgroup ( P = .19), less AEs in the rib fracture POST subgroup ( P = .04), and less AWS-RC in the high risk of severe AWS POST subgroup ( P = .03). Discussion A phenobarbital-based protocol in trauma patients is effective in preventing AWS-RC and decreasing benzodiazepine use without increasing AEs.