Access to high-quality mental health specialty care for primary care patients has historically been problematic. Behavioral health care integration into the patient-centered medical homes has been shown to be effective, but dissemination and implementation of these team-based models of care remain challenging to healthcare systems. As such, lessons can be learned from VA's national implementation of Primary Care–Mental Health Integration (PC-MHI). Here we devise and validate a new metric for VA clinic engagement in PC-MHI (number of PC-MHI service users/number of primary care patients) during each year.
This is a retrospective longitudinal cohort study of 112,737 primary care patients in 29 Southern California VA clinics from October 1, 2008 to September 30, 2013. First, we examined a subset of 66,638 patients with mental health diagnoses to understand access to mental health care and other healthcare services. Our multilevel regression models used clinic PC-MHI engagement to predict relative rates for the full-range of patient healthcare utilization and total VA costs, adjusting for year and clinic fixed effects, other clinic interventions, and patient characteristics. Then, we constructed quality metrics for 12,663 patients who were newly diagnosed to have depression. Our fully-adjusted regression models used clinic PC-MHI engagement to predict probabilities of follow-up within 84 and 180 days and receipt of minimally appropriate treatment for these patients.
Greater clinic PC-MHI engagement was associated with significantly more mental health visits and less non-primary care based mental health specialty (MHS) visits, consistent with a substitution of PC-MHI visits for MHS visits. It was associated with less general MHS visits, rather than more specialized MHS visits. This reduction appeared targeted at patients with mild-to-moderate mental illnesses (i.e., depression), rather than with serious mental illness (i.e., schizophrenia, bipolar disorder). Despite shifting mental health care for Veterans with less complicated mental illnesses from specialty to primary care, our findings demonstrated no difference in depression care quality. We did not find adverse impacts on ED visits, hospitalizations, total patient costs, or mortality. Therefore, PC-MHI may improve mental healthcare value for primary care patients, as it may have improved realized accessibility to mental health care without necessarily increasing costs.