Creating an interdisciplinary, modular, and policy-oriented county-level mental health index for the United States.

Background: In the United States, there is an increased focus on mental health problems and the need for effective estimates of localization of mental health problems is manifest. This study draws on behavioral, economic, social, and public health risk factors to predict county-level mental health for counties in the United States. Methods: Candidate predictors were identified from a literature review. Selected variables were condensed into relevant submodules and a mental health outcomes index using separate principal component analyses. The submodules were then regressed to predict the mental health index. County-level vulnerability scores were calculated for each predictor to assess their relative importance as suggested by the model. Findings: A systematic search of available data yielded a set of 56 predictive factors, including novel predictors. Seventeen submodules in five policy domains were derived, explaining 85.1% of the mental health index variance. Tableau was utilized to create interactive maps displaying the submodules and mental health outcomes over counties in the United States for use by policymakers and stakeholders. Interpretation: This model may be used for several purposes throughout the government and academia, the timeliest of which would be providing information to redirect mental health interventions to maximize outcomes as funders begin pursuing efficiency of spending over volume spent. By utilizing the index and submodules, policymakers can focus their mental health improvement efforts in the most impactful areas for their constituencies.