Integrating Multisector Data Benefits Individuals with Behavioral Health Disorders

Guest blog post by George Runger of Arizona State University

It is already well understood that a major barrier to improving health, well-being and equity in the United States is our fragmented and uncoordinated system of health care delivery, public health and social service systems. One area where system failures can be easily highlighted is in the health care coordination of individuals with mental illness and/or substance abuse disorders, which affects nearly a quarter of the US population. A research team at Arizona State University is tackling the issue with a comprehensive multisector approach focused on aligning the delivery and financing systems for medical care for this vulnerable population. Grant support is provided by the Robert Wood Johnson Foundation.

The study team includes individuals with expertise in health care financing, public health, behavioral health, informatics and other areas to ensure all sides of the issue are represented. The study is following an iterative process of Community Based Participatory Research and Participatory Design whereby the stakeholders are fully involved in all aspects of the research process from identifying the problem to planning and evaluating the interventions.

In order to begin addressing this important issue, the study team assessed the available data and defined their broad population. This was the first step in their multisector approach as they recognized that individual patients access services not only across the health system, but also through public health services and the community spectrum. They were able to demonstrate that individuals with behavioral health disorders are more likely to have interactions with the criminal justice system. These points in time add to the fragmented care provided to this population, so it was key to determine what happens to such individuals when they leave the normal flow of daily life.

The study team solicited the services of the ASU Center for Health Information & Research (CHiR), which maintains various repositories of health information. CHiR’s experinece in managing health information could be leveraged for criminal justice and homeless information. This would create a win-win for the study team and CHiR as the success of this effort could open the door for other studies seeking to integrate health and non-health data. Due to certain laws and restrictions, CHiR was not able to access every data source sought for the study, but several data sets were made available.

The data sources included state Medicaid data, emergency department and inpatient stays reported to the state, booking and release information from the county jail system, adult probation information. CHiR serves as the data collection repository and provides the data integration and analytics for the study.

Integrating health and non-health data is easier said than done. CHiR created analysis data sets from the health data it has worked with for years. The initial working data set included adults in health plans that integrate behavioral health and acute care services. From there, CHiR used probabilistic matching techniques to match the health to the criminal justice data, which consists of multiple iterations of matching on key data elements.

The good news is some gaps in care were better explained with information from criminal justice records. Rather than assuming the reasons why individuals are inconsistent with their health care for ongoing conditions, the data provide the opportunity to coordinate care with the non-health system. Models have been under ongoing development to determine better strategies and care coordination. Possibly the spectrum of behavioral health disorders have varied needs for care. There are still a number of questions.

This study expects the findings to promote innovation and transformational action at national, state and local levels. Ultimately, they seek to achieve interoperability at the individual patient level, overtime and across multiple health and non-health related information systems and to create an integrated, coordinated, cost-effective and effective delivery system for behavioral health disorders. This can only lead to improved health outcomes, accessibility, quality and efficiency, which is good for everyone dealing with a behavioral health disorder.