Integrating Pharmacotherapy Into SBIRT for Alcohol Use Disorder in Family Medicine

Heavy alcohol use and alcohol use disorder (AUD) contribute to adverse health outcomes and overall morbidity and mortality.^1-3 While nearly 28 million individuals in the United States met criteria for AUD in the past year, less than one in 10 received pharmacotherapy.² Only one in six US adults have reported ever reviewing alcohol use with a health care professional, despite the increasing prevalence of heavy alcohol consumption and AUD.^3-5 Historically, screening and brief interventions for AUD have been underutilized in primary care.^4,6 Primary care physicians also have reported insufficient training on medications for AUD (MAUD) as well as negative attitudes about medication effectiveness.^6-8 These factors highlight the importance of expanded training on screening and medication management of AUD.

Screening, brief intervention, and referral to treatment (SBIRT), a public health preventative framework, reduces alcohol use and health care utilization.^9,10 Education on SBIRT may improve trainee attitudes regarding medication efficacy.¹⁰ However, SBIRT is underperformed due to knowledge deficits in treating substance use disorders, shortages of mentors, and provider stigma.^6,9,11 The updated Accreditation Council for Graduate Medical Education guidelines now require family medicine residency programs (FMRPs) to implement curricula on the identification and treatment of substance use disorders.¹² However, a 2017 survey conducted with FMRPs revealed that less than half of the surveyed program directors expected their residents to prescribe medications for substance use disorders.¹³

Different teaching modalities have been used in historical SBIRT curricula, such as video/web-based self-directed modules, didactic lectures, small-group sessions, case-based learning, and skills practicums with standardized patients.^14-19 Prior SBIRT curricula have some notable limitations: long training hour requirements, lack of longitudinal knowledge assessments, limited interactive teaching techniques (ITTs), and minimal focus on MAUD.^14-17 We created a concise SBIRT curriculum that emphasizes interactive teaching modalities, incorporates instruction on motivational interviewing (MI), implements a longitudinal knowledge assessment, and distinctively includes training on AUD pharmacotherapy. Our SBIRT curriculum integrates both psychosocial and pharmacotherapy interventions for AUD, which combined may be more effective than psychosocial treatments alone.^20,21 Because advanced behavioral interventions may necessitate technical training or resources not available in primary care settings, combining psychosocial and pharmacological treatment approaches to AUD may offer additional benefits in harm reduction.^20-22

All residents (family medicine, pharmacy, and family medicine-psychiatry) currently affiliated with three large, academic, urban family medicine programs were invited to participate. Two family medicine programs responded. Both programs offer basic training to family medicine interns on MI techniques and include mandatory participation in medication-assisted treatment (MAT) clinics for alcohol and opioid use disorders. The first family medicine program received the full 2.5 hour SBIRT seminar. To accommodate their site’s didactic schedule, the second family medicine program received a truncated 1 hour SBIRT seminar. Due to a low response rate (<30%) and low survey match rate from the second family medicine program, the second site was excluded from statistical analysis. We included pharmacy residents to prepare all providers in SBIRT principles given their direct patient care responsibilities in the ambulatory care program. Additional demographic information for the participating family medicine program is available in Appendix A.

A paper survey (adapted from published papers on SBIRT) was given to all attendees pre- and post-seminar, and electronically 3 months after seminar.^23,24 The survey (Appendix B) had 16 questions on both SBIRT and MAUD: five questions about confidence using a 1–7 Likert scale and 11 questions about knowledge. The principal investigator collected and de-identified the survey data by associating each study participant’s email with a unique study participant number. Three months after the intervention, the principal investigator distributed the post-postsurvey via email through Pitt Qualtrics (Qualtrics LLC). The de-identified research data were provided to the statistician for data analysis. This study had institutional review board exempt approval status (Study 24020074).

The curriculum was a three-part series: Part I focused on screening (45 minutes); Part II focused on the brief intervention/MI (35 minutes); Part III focused on referral to treatment, including information on local 12-step Alcoholics Anonymous programs, recovery communities/centers, peer support/recovery coaches, and MAUD (55 minutes). The curriculum employed a variety of ITTs to promote active learning, which can improve analytical thinking and learner engagement (Table 1).²⁵

We used descriptive statistics to examine responses at each time. We performed general linear mixed models to test adjusted mean changes in confidence scores (Table 2). The knowledge questions (Table 3) were dichotomized as “yes” (correct) or “no” (incorrect), and we used generalized regression for analysis. Each model retained the individual’s matched response across time as the unit of measure, and 95% confidence intervals were adjusted for multiple comparisons. Analyses were performed using SAS JMP Pro version 17 (SAS Institute).

Twenty-nine out of 32 participants voluntarily completed matched presurveys and postsurveys (90.62% response rate); 10 matched participants completed post-postsurveys (31.25% response rate). Approximately 85% of participants were family physician residents, 15% of participants were pharmacy residents, and 96% of participants attended the seminar in person. Confidence statistically improved from pre- to postsurvey and from pre- to post-postsurvey for all confidence questions (all P<0.01, Table 1). Knowledge statistically improved from pre- to postsurvey in eight out of 11 knowledge questions (P<0.05, Table 2) with four questions having large changes in accuracy from pre (>50%). These four knowledge questions focused on recalling key curricular elements and selecting preferred medications in applied clinical scenarios. Although the post-post knowledge results generally improved compared to the pre, the small number of responses (n = 10) for dichotomized (correct/incorrect) data yielded nonsignificant differences and wide confidence intervals, which limited interpretation.

This interactive SBIRT and MAUD curriculum resulted in statistically significant improvements in confidence and knowledge of SBIRT and MAUD. Self-directed, case-based, and team-based educational activities, based on adult learning theories, successfully promoted confidence and knowledge.^25,28,29 This curriculum has notable strengths by (a) educating family medicine/PharmD trainees in a concise curriculum, (b) implementing entirely active learning through multiple ITTs, (c) incorporating training on MAUD, and (d) providing longitudinal assessments of confidence and knowledge. Paper surveys and structured seminar time for survey completion led to high completion rates of the pre- and postsurveys. Overall, social and experiential learning theories can be leveraged to encourage adult learner engagement in addiction medicine curriculum.

The survey was not piloted, preventing analysis of survey validity or reliability. Objective skill assessment and application of SBIRT to patient encounters were not evaluated, so only self-perceived knowledge and confidence were assessed. Due to the few post-post respondents, the standard errors for the dichotomized knowledge questions were large resulting in wide confidence intervals, limiting interpretation. Additionally, few survey responses were received from Site 2, none of which could be matched, and Site 2 was omitted. Thus, evaluation of data from a single site limited study generalizability.

A practical skill assessment of SBIRT through objective OSCEs could be considered. Future curricular efforts could evaluate family medicine residents applying SBIRT principles in clinical practice. This project could be expanded to other family medicine programs through targeted recruitment efforts. Increasing program participation could enable result comparison across contexts, potentially increasing generalizability.

This research was presented at the 2025 Society of Teachers of Family Medicine Conference in Salt Lake City, Utah, and at the 2025 Family Medicine Scholarship Day at the University of Pittsburgh Medical Center in Altoona, Pennsylvania.