An inpatient treatment and discharge planning protocol for alcohol dependence: efficacy in reducing 30-day readmissions and emergency department visits

Abstract

Background: Alcohol dependence results in multiple hospital readmissions, but no discharge planning protocol has been studied to improve outcomes. The inpatient setting is a frequently missed opportunity to discuss treatment of alcohol dependence and initiate medication-assisted treatment, which is effective yet rarely utilized.

Aim: Our aim was to implement and evaluate a discharge planning protocol for patients admitted with alcohol dependence.

Setting: The study took place at the San Francisco General Hospital (SFGH), a university-affiliated, large urban county hospital.

Participants: Learner participants included Internal Medicine residents at the University of California, San Francisco (UCSF) who staff the teaching service at SFGH. Patient participants included inpatients with alcohol dependence admitted to the Internal Medicine teaching service.

Program description: We developed and implemented a discharge planning protocol for patients admitted with alcohol dependence that included eligibility assessment and initiation of medication-assisted treatment.

Program evaluation: Rates of medication-assisted treatment increased from 0% to 64% (p value < 0.001). All-cause 30-day readmission rates to SFGH decreased from 23.4% to 8.2% (p value = 0.042). All-cause emergency department visits to SFGH within 30 days of discharge decreased from 18.8% to 6.1% (p value = 0.056).

Discussion: Through implementation of a discharge planning protocol by Internal Medicine residents for patients admitted with alcohol dependence, there was a statistically significant increase in medication-assisted treatment and a statistically significant decrease in both 30-day readmission rates and emergency department visits.

Figures

Figure 1

Comparison of pre-intervention group (June 2011) (a), and post-intervention group (March 2012) (b). This figure breaks down the characteristics of all patients with alcohol dependence admitted during the two comparison months. Regarding eligibility criteria for treatment with naltrexone, over 50 % of patients in both groups were eligible for treatment. The primary process measure was the percent of those patients eligible for naltrexone who were prescribed naltrexone. This increased from 0 % to 64 % (p < 0.001) after the intervention. For the primary outcome measures, the rate of readmissions within 30 days decreased from 23.4 % to 8.2 % (p = 0.042) and the rate of patients with emergency department (ED) visits within 30 days decreased from 18.8 % to 6.1 % (p = 0.056).