Introduction

Unhealthy alcohol use increases cancer incidence and mortality and is common among hospitalized patients: between 5-30% of hospitalized patients screen positive for unhealthy alcohol use, and approximately 10-15% of patients have alcohol use disorder (AUD). Brief counseling and medications have been shown to reduce alcohol consumption, yet over 90% of patients nationally are not effectively screened and treated for unhealthy alcohol use.

Methods

We implemented a comprehensive clinical protocol for unhealthy alcohol use at a safety-net hospital in Austin, Texas. Our approach is based on the SBIRT (Screening, Brief Intervention, and Referral to Treatment) framework, which has been shown to reduce readmissions and increase abstinence rates. Nurses screen patients at intake with the AUDIT-PC, a shortened form of the validated AUDIT instrument for unhealthy alcohol use. Patients who score positive (≥5) on initial screening are evaluated by a dedicated substance use navigator (SUN) who performs a full AUDIT, delivers a brief intervention using motivational interviewing principles, and, if appropriate, provides education about medications. If the patient is interested and eligible, the primary care team starts medications (e.g. naltrexone). The SUN collaborates with our outpatient partners to arrange follow-up care, specialty referrals, and recovery resources. The SUN also contacts patients after discharge to provide further support and navigation as needed. Four weeks after discharge, the SUN attempts to contact the patient to conduct a modified AUDIT via phone to assess for changes in drinking behaviors. Three attempts are made before the record is closed.

Results

Over the first 10 months of the program (September 2022-June 2023), 10,994 patients were admitted to the hospital, and 6,703 patients (61%) were screened with the AUDIT-PC. Of those screened, 412 patients (6.1%) screened positive, and 339 (82%) were seen and had a full AUDIT performed. Of these 339, 61% had a score of 12 or greater, signifying increased risk of severe AUD.

Of the 412 patients screening positive, 79% were male. The mean age was 50 years, with a range of 20 to 90 years. About 50% of patients identified as white, 30% identified as Latinx/Hispanic, 10% identified as black or African-American, and <1% identified as American Indian, Asian, or native Hawaiian. Approximately 25% of patients had temporary housing or were unhoused. 

Of the 412 screen-positive patients, the SUNs delivered behavioral interventions to 325 (79%); 179 of them (55%) were eligible and interested in medication for alcohol cessation, and 124 patients (69%) were started on one of these medications.

Prior to discharge, the SUNs provided ongoing resource navigation services for the 325 patients they evaluated. Of these patients, 167 (51%) were discharged with a scheduled outpatient primary care appointment.  

At four weeks post-discharge, 56 patients (17%) have been successfully contacted via phone. We primarily focused on calling high-risk patients with AUDIT scores of 12 or greater. Forty-nine of these patients (88%) fit that category, but by follow-up, only 12 patients had AUDIT scores of 12 or greater. 59% of contacted patients were able to attend their follow-up appointment; however, 32% had had a repeat emergency room visit or a readmission. Of those contacted, 33 (59%) had reduced their reported alcohol use, and 19 (32%) had stopped drinking entirely. Currently, low follow-up rates are due to lack of patient response despite multiple calls, frequent re-admissions, and a significant unhoused population with no stable form of communication.

Conclusion

Screening, brief counseling, and appropriate medication treatment initiation for unhealthy alcohol use is feasible to implement in an inpatient setting with a multidisciplinary team, and is associated with changing patient behavior to decrease alcohol use.