Implementation of a Process for Initiating Naltrexone in Patients Hospitalized for Alcohol Detoxification or Withdrawal.

Background: Naltrexone trials have demonstrated improved outcomes for patients with alcohol use disorders. Hospital initiation of naltrexone has had limited study.

Objective: To describe the implementation and impact of a process for counseling hospitalized patients with alcohol withdrawal about naltrexone.

Design: A pre-post study analysis.

Setting: A tertiary academic center.

Patients: Patients hospitalized for alcohol withdrawal.

Intervention: (1) Provider education about the efficacy and contraindications of naltrexone and (2) algorithms for evaluating patients for naltrexone.

Measurements: The percentages of patients counseled about and prescribed naltrexone before discharge and the percentages of pre- and postintervention patients with 30-day emergency department (ED) revisits and rehospitalizations.

Results: We identified 128 patient encounters before and 114 after implementation. The percentage of patients counseled about naltrexone rose from 1.6% preimplementation to 63.2% postimplementation (P<.001); the percentage of patients prescribed naltrexone rose from 1.6% to 28.1% (P<.001). Comparing preintervention versus postintervention groups, there were no unadjusted differences in 30-day ED revisits (25.8% vs 19.3%; P=.23) or rehospitalizations (10.2% vs 11.4%; P=.75). When adjusted for demographics and comorbidities, postintervention patients had lower odds of 30-day ED revisits (odds ratio [OR]=0.47; 95% confidence interval [CI], 0.24-0.94) but no significant difference in rehospitalizations (OR=0.76; 95% CI, 0.30-1.92). In subgroup analysis, postintervention patients counseled versus those not counseled about naltrexone were less likely to have 30-day ED revisits (9.7% vs 35.7%; P=.001) and rehospitalizations (2.8% vs 26.2%; P<.001).

Conclusions: The implementation of a process for counseling patients hospitalized for alcohol withdrawal about using naltrexone for the maintenance of sobriety was associated with lower 30-day ED revisits but no statistically significant difference in rehospitalizations.