Integrated Behavioral Health Care in Family Medicine Residencies

Primary care has been the principal mental health system for decades, delivering up to 80% of psychiatric care in the United States.1 Family medicine has embraced this central role of providing quality behavioral health (BH) care. The Accreditation Council on Graduate Medical Education requires significant behavioral training during family medicine residency (FMR),2 and FMR graduates deliver more BH care than other primary care specialties.3

Despite this, BH care in primary care clinics has been limited in quality and reach.4,5 Integration of behavioral health professionals into primary care settings is a major organizational strategy that addresses gaps in mental health care.6 Behavioral health integration (BHI) has been found to improve patient satisfaction,7 reduce costs,8 and improve social function for patients with depression.9 In 2013, SAMSHA described six levels of BHI grouped into three broad categories: collaboration, colocation, and integration. Collaboration is characterized by physically distant primary care and behavioral providers with varying communication. Colocation describes physical proximity with diverse care structures. Full integration is intentional team care with warm handoffs, shared records, and multidisciplinary care coordination.10

FMRs are integrating BHI into their continuity clinics and curriculum, but research on BHI curricula and penetration is sparse.11 BHI training in family medicine residencies (FMRs) increases resident knowledge and confidence,12 and residents identify working with interdisciplinary behavioral providers as a strength of a BH rotation.13 Educators have proposed model BHI curricula for medical and psychiatry residents.14 After one pilot program, 97% of residents reported an intention to practice integrated care after graduation.15 Assessment of the current state of BHI in FMRs across the country will better inform curriculum. We surveyed program directors about BHI in the family medicine centers (FMCs) where residents provide continuity care, including the degree of integration, training activities, and type of educators involved in residency-based BHI curricula. We aimed to characterize the degree of BHI and its relationship to BH activities and teachers in FMRs.

FMR program directors were emailed a link to the 2017 CERA survey.16,17 A total of 261 of 478 (45%) of program director members of the Council of Academic Family Medicine contacted completed the survey. Table 1 details the demographic questions including residency type, region, community size, residency size, and program director characteristics.

Participants selected one of six multiple-choice descriptions of BHI at the residency’s FMC which were then compressed into a three-tiered model following the SAMHSA framework. Minimal coordination and basic collaboration at a distance were grouped as collaboration. Basic collaboration onsite and close collaboration onsite with some systemic integration were grouped as colocation. Close collaboration approaching an integrated practice and full collaboration in a transformed/merged practice were grouped as full integration.10 Respondents indicated which BH activities occurred in the FMC and which types of professional providers taught BH to residents. This CERA study was approved by the American Academy of Family Physicians Institutional Review Board.

Characteristics of program directors and BHI questions were summarized using one-way frequencies and means. Associations between the reported degree of BHI offered (collaboration vs colocated vs full integration) and characteristics of BHI offered at sites were assessed using chi-square tests for categorical variables and one-way ANOVA with Tukey honest significant difference (HSD) post hoc comparisons for continuous variables. If cell sizes less than five were present for categorical variable comparisons, Fisher exact tests were used. All analyses were conducted at an alpha=0.05 level using SAS v9.4 (SAS Institute, Cary, NC).

Table 1 describes the demographics of the program directors and FMRs. FMRs most often were community-based with a university affiliation, in a medium-size community, with 19 to 31 residents. FMRs from the Northeast were the smallest proportion of the sample.

Table 2 demonstrates that most FMRs reported either colocated (33.7%) or fully integrated BH care (44.1%). About 90% of FMCs had more than one type of BH faculty. Family physicians and psychiatrists most frequently taught BH skills, while licensed counselors and marriage and family therapists were least utilized. There was no significant difference in the degree of integration by program type, geographic region, community size, or program size (results not shown).

Table 3 demonstrates that the use of all individual BHI activities in FMRs was positively associated (all P<.003) with increased degree of BHI. Average number of different BH activities also increased with degree of collaboration (P<.0001). A greater proportion of both fully integrated (75.7%) and colocated (71.6%) programs used psychologists to teach BH compared to collaborative programs (43.8%) but no other differences in types of professionals were found. The mean number of BH educator types was similar for colocated and fully integrated programs (3.3 and 3.6, respectively) with both significantly higher when compared to collaborative programs (mean=2.6).

Most FMR program directors in this study report colocated or fully integrated BH care in their FMCs, and most FMRs provide residents with training from at least three different types of BH professionals. This high level of integration was seen across all program types and regions. This high penetration of BHI into FMR education was simultaneously surprising and reassuring to the authors.

The 12.3% of programs in the collaboration phase notably lack warm handoffs, same-day consultation, and shared electronic health records. Those residencies differ from more integrated residencies in the absence of psychologists, social workers, licensed counselors or marriage or medical family therapists. Inclusion of these processes and personnel would advance their BHI.

Programs were generally low in psychiatry exposure and population health. Previous work indicated that family medicine residents perceive a lack of pharmacology and family systems education during residency.13 The diversity of BH teachers indicated in this study may be better targeted to fill those gaps.

Residencies with full BHI may consider using a similar interprofessional model to provide health care in other areas, or focusing on supporting BHI-trained residents with the transition into practice.

This study was limited by the sample size and self-report design. Objective assessment of BHI within FMR clinics would add validity to the initial characterization described here.

This study demonstrates that the concept and practice of BHI in FMRs is now well established. Maximization of each of these elements of BHI can improve patient care in the FMR clinic, and better prepare family medicine residents to enter a practice that provides behavioral health care in the primary care setting.