Background: There were nearly 33,000 admissions to Department of Veterans Affairs hospitals for alcohol withdrawal syndrome (AWS) in fiscal year 2017. Symptom-triggered management is the standard of care and, when employed effectively, the number of medication doses during admission is a good proxy for clinical severity of withdrawal. Several evidence-based algorithms for outpatient management of low risk withdrawal have been described in the literature. Chart review of 424 AWS admissions over 18 months showed 65% received three or fewer doses of symptom triggered medication, indicating low risk and good candidacy for outpatient management.

Purpose: The purpose of this project was to increase the number of patients treated for AWS as outpatients and decrease hospital admissions without increasing readmissions or clinical deterioration requiring transfer to a higher level of care.

Description: Our innovation consists of four core operational changes:1. Standardized risk stratification in the Emergency Department(ED) to identify low risk patients for outpatient treatment
2. Benzodiazepine sparing symptom triggered medication regimen
3. Daily clinical dashboard surveillance and risk stratification for continued hospital stay
4. Telephone follow up for patients discharged from the ED or hospital
We chose a risk stratification algorithm previously described by Stephens et. Al. with the only modification being a change in the outpatient fixed dose regimens to five days of Gabapentin or three days of Lorazepam. Heart rate > 100 AND systolic blood pressure >165 were added to the standard Clinical Institute Withdrawal Assessment (CIWA) score as thresholds for medication dosing, dramatically decreasing the number of doses given. An operational dashboard was developed to identify current inpatients on CIWA which a hospitalist surveilled daily and reminded attendings to risk stratify and discharge low risk patients. Placement of an electronic-consult triggered three days of hospitalist telephone follow up for all homegoing ED patients and selected homegoing floor patients.
The project was implemented mid-February 2018 and 8 months of follow up data are now available:
1. Reduction in admissions from 24 to 14 per month (50%)
2. Reduction in hospital wide all cause readmissions from 4/month to less than 1
3. Reduction in length of stay from 3.5 to 2.0 days (40%)
4. Reduction in the number of CIWA notes by nursing staff from 810/month to 430(47%)
5. Reduction in intensive care unit transfers from the floor for severe AWS

Conclusions: A combination of clinical best practice, electronic medical record innovation and use of an operational dashboard decreased hospital admissions, readmissions, length of stay and nursing workload for AWS while improving patient safety. Although this was accomplished in a Veterans Affairs hospital, implementation would be feasible at any hospital with an electronic medical record and a corporate data warehouse.