Improving Outcomes Among Medical/Surgical Inpatients With Alcohol Use Disorders

In fiscal year 2014, over 57,000 Veterans with diagnosed alcohol use disorders (AUDs)
received VHA inpatient medical-surgical services. This likely underrepresents the prevalence
of AUDs among Veteran inpatients, because these conditions often go undiagnosed during
hospital stays. The high prevalence of AUDs among VHA medical-surgical inpatients is of
critical concern because AUDs and medical conditions exacerbate one another, and their
co-occurrence increases the use of costly health services. Yet, there are no evidence-based
strategies that improve outcomes in this patient population by means of increased utilization
(initiation, engagement) of AUD treatment services. The investigators have identified
Drinking Options: Motivate, Shared Decisions, Telemonitor (DO-MoST) as a potential solution
to the critical need for evidence-based strategies.

This project will evaluate DO-MoST as a new and innovative intervention to facilitate the
transition from inpatient medical-surgical care to AUD treatment in primary and specialty
care settings, thereby improving Veterans' AUD and medical outcomes. DO-MoST entails use of
motivational interviewing and a decision aid during the inpatient stay to facilitate informed
choices about post-discharge drinking options and resources for help to change drinking, if
desired, followed by telephone calls with the patient to continue to motivate and support
decisions. In addition to determining the effectiveness of DO-MoST, this project includes a
process evaluation - that is, will gather information from providers and patients on
DO-MoST's implementation at the two project sites - to inform VA's potential widespread
implementation of DO-MoST with medical-surgical patients with AUDs.

Using an effectiveness/implementation Hybrid Type 1 design, incorporating a randomized
controlled trial (RCT) and process evaluation to facilitate future implementation, this
project has three Specific Aims.

Aim 1: Adapt a decision aid being implemented with AUD patients in non-VA primary care
settings to be appropriate for Veterans with AUDs in medical-surgical inpatient settings.
With this prototype, the investigators will conduct alpha testing with patients and
providers, and adapt and pilot the decision aid to finalize it for use in the RCT.

Aim 2: Conduct DO-MoST at two VA facilities (Ann Arbor and Palo Alto) and evaluate its
effectiveness.

The primary hypotheses are: Patients in DO-MoST, compared to patients in usual care (UC),
will be more likely to (1) utilize AUD help (initiate, engage), (2) have better AUD (fewer
heavy drinking days) and medical (physical status) outcomes, and (3) have fewer and more
delayed acute care episodes (Emergency Department visits, rehospitalizations). Patients will
be assessed at baseline, and 3, 6, and 12 months post-discharge, for outcomes and non-VA
health care; VA health care will be assessed with VA databases. GLMM analyses will be
conducted to compare the UC and DO-MoST groups on course of outcomes.

Aim 3: Conduct a qualitative process evaluation to inform the wider implementation of
DO-MoST, using the RE-AIM (Reach, Effectiveness, Adoption, Implementation, Maintenance)
framework. The purpose is to provide guidance for VA facilities' broader adoption of DO-MoST
in the future, including its possible adaptation for diverse subpopulations of Veterans, such
as patients with mental health diagnoses (e.g., PTSD).

In summary, this project will develop a decision aid and comprehensively examine DO-MoST as a
novel and groundbreaking approach to providing a bridge between inpatient medical-surgical
treatment and AUD treatment. Decision Aids have been used successfully in a number of
contexts, but never with medical-AUD inpatients. The investigator's operations partners from
VHA Mental Health Services and Medical Service are committed to directly addressing the
dangerous, costly pattern of Veterans obtaining inpatient medical-surgical services but not
receiving the AUD treatment they need. The project is focused on priorities in the VA
Blueprint for Excellence, of HSR&D Service, and of the PIs' HSR&D Centers of Innovation.

Inclusion Criteria:

Medical-surgical inpatients with alcohol use disorder at the Ann Arbor or Palo Alto VA.
Specifically:

- in a current episode of inpatient medical-surgical care,

- meet DSM 5 criteria for an AUD (moderate or severe),

- no specialty addiction treatment or weekly mutual-help group attendance within 60 days
prior to the inpatient episode,

- no restricted access due to infection control requirements (e.g., TB, MRSA, C. diff),

- no significant cognitive impairment,

- ongoing access to a cell or land line telephone,

- at least one contact who will continue to know the patient's contact information, and
(8) not having participated in an interview for Aim 1 of this project.

Exclusion Criteria:

See inclusion criteria.