Degree of Behavioral Health Integration and Patient Outcomes

Earlier this month Dr. Mullin and colleagues published an important paper demonstrating the value of integrating primary care and behavioral health services.  The paper, Degree of Behavioral Health Integration and Patient Outcomes, was published in the Journal of the American Board of Family Medicine and is available for free from the publisher.  

What the researchers studied

Our team looked at adults with multiple chronic medical and behavioral health conditions who received care in one of 44 primary care practices across the United States. All practices had at least some behavioral health care collocated in their practice. The key question was not whether integration helps, but how much integration is enough to see meaningful benefits for patients.

How integration was measured

Practices were scored using the Practice Integration Profile (PIP), a validated tool that captures how well behavioral health and primary care are woven together across workflows, teamwork, patient identification, clinical services, and patient engagement. Scores range from 0 to 100.

The main finding

Practices that reached a PIP score of 65 or higher consistently had better patient-reported outcomes than practices below that level. This suggests an important “threshold” of integration beyond which meaningful improvements in patient outcomes can be observed.

PROMIS-29 domains

Patients completed the PROMIS-29, which measures physical, emotional, and social functioning using standardized questions. In highly integrated practices, patients reported:

• Less pain, especially pain intensity
• Better physical functioning, such as walking, climbing stairs, and completing daily tasks
• Better social participation, meaning fewer limitations in work, family, and social roles
• Lower burden of anxiety, depression, fatigue, and sleep disturbance

Although not every individual symptom difference reached statistical significance after adjustment, every domain moved in a favorable direction in highly integrated practices, both at baseline and two years later.

Correlation is Certain, Causality is Possible

This study shows that higher levels of integration are associated with better patient outcomes. Although the pattern of results is consistent across outcomes and over time, and becomes stronger as integration increases, the study design does not allow for definitive conclusions about causality.

Why is this study so important?

Not every patient in a highly integrated practice actually saw a behavioral health clinician. Even so, patients still benefited, suggesting that collocated, team-based, integration changes how the entire practice functions.  This model of care has benefits beyond referring patients to a consulting specialist.