In value-based models of health care delivery, providers are financially rewarded or penalized on the basis of the quality of care they provide to patients, typically measured by patient outcomes. The specifics of value-based care models can take several forms, but in many cases—such as in accountable care organizations—they involve risk sharing or shared savings arrangements between multiple payer and provider organizations. To achieve the goals of value-based care, all involved organizations must be able to effectively share and integrate data from multiple sources.
Using qualitative case study interviews, this report aims to inform efforts to improve data integration to support value-based care. We conducted interviews with individuals across 21 organizations selected for the prevalence of value-based care or evidence of advanced data integration. These stakeholder interviews focused on the current state of data integration, the uses of data integration in value-based care, and the barriers to and facilitators of data integration. Our approach was guided by a framework that focuses on the technical, organizational, and environmental factors affecting data integration. Our main findings are as follows:
- The organizations selected had inconsistent definitions of data integration and varying levels of integration. This suggests that context is critical for interpretation, and generalizing about data integration across organization types, settings, and uses of integrated data is challenging. Interviewees’ understanding of data integration depended upon their roles within an organization and how they used or supported the use of data.
- Interviewees identified four primary uses for data integration in value-based care: point of care, care coordination, quality measurement and reporting, and population health.
- Higher levels of integration are not necessarily required or even desirable across all use cases. Interviewees across organizations indicated that viewing data from outside sources at the point of care in a consistent location within the electronic health record (EHR) is preferable to integrating (e.g., parsing) the outside data into the local incidence of the patient record.
- Interviewees identified the following as barriers to data integration in value-based care: problems with EHR developers, technical issues associated with Continuity of Care Documents and standards, the high costs of data integration relative to its benefits (i.e., value), legal concerns, and the relationships between payer and provider organizations. Interviewees also indicated the roles of the patient and workforce development within organizations are often overlooked in existing data integration frameworks.
- Facilitators of data integration included supportive public policies such as those encompassed in the 21st Century Cures Act (Cures Act), the expansion and increased support of value-based programs, the trend toward increased data sharing between provider organizations, interoperability between large vendor systems, and data sharing with state agencies.
The findings from this report suggest various interventions could further promote data integration. First, the Office of the National Coordinator for Health Information Technology could do more to hold EHR developers accountable for facilitating data integration. Second, payment reform could be designed to incentivize data sharing and the use of shared data. Third, incentive payments could be used to support infrastructure development for data sharing and the integration of other high-value data, including population health data, social determinants of health data, and imaging studies in standardized datasets. Finally, the implementation of new policies directed by the Cures Act—such as the United States Core Data for Interoperability, applications based on Fast Healthcare Interoperability Resources, information blocking rules, and the EHR Reporting Program—could facilitate data integration while simultaneously improving data liquidity, lowering costs, and enhancing health information technology.