Alarming rates of unhealthy alcohol, non-prescription drug, and tobacco use highlight the preventable health risks of substance abuse and the urgent need to advance behavioral health systems. Screening, Brief Intervention, and Referral to Treatment (SBIRT) is an efficacious and effective strategy for the delivery of early intervention and treatment services for people with substance use disorders as well as for those who are at-risk for developing these disorders. A mixed methods phased cluster randomized approach is used to evaluate SBIRT implementation in one medical surgical unit from each of 14 hospitals to inform SBIRT implementation methods for rural, community and urban settings in Indiana. The long-term objective is to develop an SBIRT toolkit to disseminate and sustain SBIRT use to increase the screening of substance use, delivery of brief interventions, and referral to treatment services. A broader dissemination will follow based on results of this study.

Aim 1: Test if implementation of SBIRT improves processes of care for hospital inpatients currently using tobacco, alcohol or non-prescription drugs.

Aim 2: Estimate the cost of SBIRT implementation and delivery for the health system.

Design and Sample:

A phased cluster randomized mixed methods design using a wait-list usual care control on one medical surgical unit in 14 hospitals will be conducted. The intervention is an SBIRT toolkit.

Each hospital within the healthcare system is eligible to participate, and has agreed to participate (N=14). A computer generated allocation based on a stratified random sampling approach using SPSS was used to randomize hospitals into an intervention (Group 1, n=7) or wait-list usual care control (Group 2, n=7) group. Prior to randomization, hospitals were stratified by type (academic health center, community hospital, or critical access hospital). Allocation to the intervention or wait-list usual care control groups is based on cluster, and participants will be aware of their cluster allocation. One medical surgical unit within each hospital will be selected by the nurse executive to participate. All hospitals will receive the intervention. Group 1 (n=7) hospitals will participate in the implementation and intervention first, with Group 2 (n=7) following six months later.

Methods:

Smoking cessation practices will be collected from nurses (N=560) at baseline. SBIRT use will be collected from nurses on randomly selected days via survey. Observations of SBIRT use and interviews with study coordinators will also be conducted.

SBIRT process delivery for patients will be assessed through review of de-identified electronic medical records (EMRs) abstracted at three time points: baseline, six-months post and 12 months using a random sample of 61 patient EMRs from the participating hospital units at each time point (N=2,562). Data abstracted is retrospective, and consists of SBIRT process elements only.

This study protocol incorporates standardized education on using SBIRT through a train-the-trainer approach. Site coordinators will receive information about the nurse survey and secondary data abstraction procedures. Each group will receive training on the SBIRT intervention (Group 1 first, then Group 2 six months later). Training for site coordinators will include skills in how to teach SBIRT to others and systems strategies for promoting SBIRT uptake. After the site coordinator training is completed, each hospital (via the site coordinator) will train additional nurses, validate skills in SBIRT and develop a referral process specific to its needs.

1. Hospital Unit

   Inclusion Criteria:

   a. Medical surgical unit designated as study unit

   Exclusion Criteria:

   1. Medical surgical unit not designated as study unit
   2. Units that do not admit medical surgical patients

2. Nurses

   Inclusion Criteria:

   a. Nurses employed on study unit

   Exclusion Criteria:

   a. Nurses not employed on study unit

3. Patient Electronic Medical Records (EMRs)

Inclusion Criteria:

1. Age 18 or older admitted to study unit
2. Records randomly selected from each units

Exclusion Criteria:

1. Age younger than 18 admitted to study unit
2. EMRs of patients admitted to units other than the study unit
3. EMRs of patients admitted to the study unit that are not randomly selected

Hospitals: Hospitals (N=14) within one health system will participate. Group 1 (n=7) hospitals will participate in the implementation first, with Group 2 (n=7) following six months later.

Nurses: Smoking cessation practices will be collected from nurses (N=560) at baseline. SBIRT use will be collected from nurses on randomly selected days via survey.

Patients: SBIRT process data will be abstracted from electronic medical records (EMRs) at three time points: baseline, six-months post and 12 months using a random sample of 61 patient EMRs from the participating hospital units at each time point (N=2,562). Data abstracted is retrospective, and consists of SBIRT process elements only.

Only SBIRT processes will be abstracted from randomly selected EMRs and will be reported in the aggregate. There is not a plan to share individual participant data. The investigators will share the implementation toolkit publicly.