Increasing Diagnosis of Co-Occurring Mental Health and Substance Use Disorders in Community Mental Health Centers By Financial Incentive

Background and Purpose: Historically, treatment for mental health and substance use problems have not been integrated, limiting the delivery of high-quality, comprehensive care for individuals with co-occurring disorders (CODs). Mental health agencies have struggled to develop the capacity to recognize and treat substance use problems, despite federal and state mandates and incentives to expand services. Inadequate provider payments have been identified as one reason for limited co-occurring treatment capacity in mental health agencies, resulting in underdiagnosis of substance use disorders in these settings. Blended payment models, the combination of multiple payment strategies, have been identified as one policy level implementation strategy to facilitate integration of services for people complex health needs. This study assesses the impact of one blended payment strategy, combining fee-for-service substance use treatment payments and capitated payments for mental health services, in a primarily rural state, on the diagnosis of CODs in a statewide community mental health system.

Methods: The study sample consisted of services aggregated into monthly time points (n = 173889), nested within individuals (n = 19373), nested within agencies (n = 10). Electronic health records were obtained from 2017 through 2019 for all adult service users with either a mental disorder or substance use disorder within the designated community mental health system. A binary dependent variable represented whether a service user held diagnoses of CODs within a month. A logistic multilevel model was used to analyze these data. Predictor fixed effects included a variable for month, and a binary variable representing payment reform implementation. Race, gender, age, and payor were held constant as fixed effects. Level two, service user, and level three, agency, variables were modeled as random effects.

Results: Forty-six percent of the sample of service users had Medicaid as a payor. The sample was predominantly white, with only 11% of service users identified as a person of color. Fifty-one percent identified as women and 49% identified as men. The mean age was 47 (SD = 15.55). Implementation of payment reform was found to increase the likelihood of receiving co-occurring diagnoses (OR = 1.38, 95% CI 1.14-1.66, p < 0.001). People of color had a lower likelihood of receiving a co-occurring diagnosis (OR = 0.66, 95% CI 0.50 - 0.88, p < 0.001). However, we conducted a subgroup analysis of rural agencies (n = 9) and found that people of color had no greater likelihood of receiving a co-occurring diagnosis than white individuals (OR = 1.01, 95% CI, 0.71-1.42, p = 0.96).

Conclusions and Implications: This study is one of the first to assess the impacts of a blended payment model on quality of care in behavioral health. Our findings reveal that blended payment models can effectively incentivize behavioral health providers and systems to identify individuals with complex diagnoses that may go unrecognized in routine care. Furthermore, our findings highlight the need for future research that examines potential disparate impacts of payment models on diagnosis of co-occurring disorders and access to care for individuals of color.