Characteristics of Family Physicians Practicing Collaboratively With Behavioral Health Professionals

INTRODUCTION

The complexity of patients presenting in primary care continues to grow with an estimated 1 in 3 adults in the United States presenting with multiple chronic conditions.¹ The parallel rise in mental health conditions and substance use disorders contributes to the poor control and growing rates of physical health conditions.² Behavioral health care access often remains limited. The COVID-19 pandemic has worsened this situation with many patients delaying care for multiple chronic conditions while deaths from suicide, alcohol, or other drugs increase.³

Family physicians (FPs) play an essential role in managing not only physical but also mental health conditions. Studies suggest that between 30% and 80% of primary care visits include a mental health complaint with this proportion increasing in the past few years.⁴

The National Academies of Sciences, Engineering, and Medicine has called for greater behavioral health integration into primary care as a way to increase behavioral health services access and manage multimorbidity.⁵ Integrated behavioral health is blended care in one setting for medical conditions and related behavioral health factors as a part of whole-person health through the collaboration of medical and behavioral health professionals.⁶ Integrated behavioral health has been shown to improve mental health and overall outcomes, patient care experience, and clinician satisfaction, while reducing health care utilization and costs.⁷

As a first step to understanding how to overcome the multiple barriers challenging the implementation of integrated behavioral health, more information is needed on the prevalence of behavioral health integration in primary care and the characteristics of FPs who have adapted integrated behavioral health. Prior work has focused on colocation using claims data rather than integration or focused on limited settings.⁸

In our study, we examined the prevalence of FPs working collaboratively with behavioral health professionals (BHPs) and described the associated demographic, practice, and geographic characteristics.

METHODS

Using a cross-sectional design, we analyzed 2017-2021 data collected from the American Board of Family Medicine’s Continuing Certification Examination Registration Questionnaire, which is a required component of registration for board-certified FPs to continue certification. The questionnaire asked FPs if they worked collaboratively with each of psychiatrists, psychiatric nurse practitioners, licensed social workers, psychologists, and/or other BHPs. Questionnaire details can be found here.⁹ Family physicians who were not providing continuity care, practiced outside the United States, or had missing geographic data were excluded. We performed χ² tests to determine associations between working collaboratively with any BHP and demographic, practice, and geographic variables (see Supplemental Appendix for these results). We then used logistic regression to calculate odds ratios and CIs controlling for differences noted by year. Analyses were completed using SAS Version 9.4 (SAS Institute Inc). State level variations were then mapped in R (R Foundation for Statistical Computing). The American Academy of Family Physicians Institutional Review Board approved this study.

RESULTS

The response rate was 100%. Of 34,949 respondents, we excluded 2,440 because they did not provide direct patient care, 6,957 for not providing outpatient continuity care, and 330 who either practiced outside the United States or had incomplete geographic data. The remaining sample included 25,222 respondents. Of these, 38.8% reported working collaboratively with any BHP; 13.4% working with psychiatrists, 5.6% with psychiatric nurse practitioners, 26.9% with licensed social workers, and 27.2% with psychologists and/or other BHPs. The mean age was 50.32 years with 46.1% identifying as female and 70.3% as White. The proportion of FPs working collaboratively with any BHP increased from 34.8% in 2017 to 43.0% in 2021 (P <0.001).

Physician characteristics significantly associated with increased odds of BHP collaboration were identifying as female (OR = 1.09) and working as core/salaried faculty (OR = 2.32 when compared with not being faculty). The strongest practice variable was working in a federal practice site and the strongest negative association was working in an independently owned practice. For geographic associations, working in the South was associated with decreased odds of BHP collaboration. Working in a county with more psychiatrists was also associated with higher likelihood of BHP collaboration. Other non-significant associations can be found in Table 1.

Prevalence of BHP collaboration by state ranged from 17.6% in Mississippi to 78.0% in Vermont. The 5 states with the lowest BHP collaboration were all located in the South while the 5 states with highest collaboration were distributed across the Midwest, Northeast, and West (Figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

State-level variation of family physicians who work collaboratively with behavioral health professionals, 2017-2021.

DISCUSSION

Although the overall proportion of FPs practicing collaboratively with BHPs increased over the past few years, significant differences exist for certain settings, notably for those working in independently owned practices and in the South. One possible reason for lower rates in the South is the overall lower number of psychiatrists, psychologists, and psychiatric nurse practitioners practicing in the South.¹⁰ Work is needed on the barriers that independent practices and practices in the South face with integrated behavioral health implementation and on how to support overcoming these barriers.

Potential novel methods to provide increased access to behavioral health include the use of behavioral telehealth services that could overcome issues with geographic variations in workforce and promote the use of shared BHP resources between practices.¹¹ Models from settings with higher rates of integration, like the federally run Veterans Affairs System and the Indian Health Service could be adapted to other settings. Because family physicians who are core/salaried faculty are more likely to work with behavioral health professionals, family medicine residencies, which are mandated by the Accreditation Council for Graduate Medical Education to integrate behavioral health into their programs, could serve as integrative behavioral health models.¹² A variety of innovative models in both the federal health care settings^13-16 and residency settings^17-20 have been previously described in the literature.

Several limitations exist. First, since the questionnaire is cross-sectional in nature, no causality can be established between the variables and working with BHPs. Second, physicians could have responded with different conceptions of what it means to be “working collaboratively with you at their principal practice site” and may have different conceptions of “integrated care.” Third, “working collaboratively” with a BHP may not align directly with “integrated care.”

Integrating behavioral health into primary care can improve behavioral health and overall health outcomes, decrease costs and utilization, and improve both patient and clinician satisfaction. Despite increases over the past few years in prevalence of integrated behavioral health in primary care, significant disparities remain. Understanding why these disparities exist and exploring how to best support behavioral health integration in these settings could help improve behavioral health access for patients in these settings and improve overall patient outcomes.