How to Leverage Population Health Strategies and Rich Data for Value-Based Care
By Julia Zhang, Senior Director of Software Engineering
The first stage of meaningful use takes patient records out of the filing cabinet so they can be accessed and shared with eligible professionals, while the second stage works to enrich patient data, relying on information from numerous healthcare partners. At the physician level, rich data can help facilitate care team collaboration and improve patient diagnoses (e.g., by connecting data points within a patient’s history that might not, at first glance, even seem related). For hospitals, clinics and transitional care facilities, rich data can also help to facilitate the move to bundled care and allow different organizations to connect, measure patient outcomes and compare results, enabling stage three: improved health. Of course, for rich data to support these results, care teams should consider a few other factors first.
Thinking More Inclusively About Data
Leveraging rich data with population health strategies can help healthcare organizations achieve value-based care goals. In fact, most chief information officers (CIOs) see population health strategies as a crucial part of their 2017 strategic healthcare information technology (IT) development plans, according to a recent focus group focus group conducted by Transcend Insights.
However, if care teams expect to take full advantage of advanced analytics and emerging prescriptive platforms, they will have to think more inclusively about data. This means that their systems should be able to aggregate data from a wider number of sources and apply analytics to pinpoint key areas of focus.
Addressing Data Integrity and Connectivity Challenges
At many healthcare organizations, technical issues—many of which are related to data integrity and connectivity—halt progress. Even within the same clinic or hospital, important information is often siloed, preventing care teams from exchanging patient data or communicating effectively on care plans. This problem exists at a time when new regulations have established performance measures as key to rating the effectiveness of physicians and organizations.
Other issues preventing the collection of rich data are that users may enter information into the electronic health record (EHR) differently or that the unstructured data may be hiding somewhere in the “notes” section of a medical record. Worse than that, the data may reside in pharmacy or medical insurance information, leaving the EHR user unaware that the data exists and unable to access it.
Clearly, the lack of one common set of standards for clinical terms that span different types of healthcare databases is a problem. Although standards have been developed and continue to improve, they remain inconsistently used across all systems. Therefore, the terms to describe a medical condition or treatment within a prescription database will often differ from those in clinical or insurance claims databases. The good news is that the Systematized Nomenclature of Medicine, the Logical Observation Identifiers Names and Codes Logical Observation Identifiers Names and Codes and RxNORM already laid some of the groundwork to respond to this issue. However, as the ubiquity of EHRs in healthcare continues to grow, the need for building upon these established efforts grows as well.
In order to more effectively use the information that already exists in EHRs for population health strategies, data will need to be complete, clean, standardized and normalized. Without the establishment of this clean data, important trends or events may go unnoticed, risking poor patient health outcomes and higher costs.
Rich Data Support at the Public Level
To support these rich data efforts at the public level, the Office of the National Coordinator for Health Information Technology (ONC) established a roadmap for interoperability roadmap for interoperability, which included the $60 million Strategic Health IT Advanced Research Projects (SHARP) program that promoted interoperability and the availability of rich data. Starting in 2011, via SHARP, the University of Illinois (Urbana), the University of Texas (Houston), Harvard University and the Mayo Clinic examined issues of data quality and integrity, interoperability and access. Specifically, the Mayo Clinic’s work focused on data normalization, natural language processing and phenotyping of patient traits. By the time this particular SHARP research came to a close in 2014, it had improved the overall understanding of data issues with EHRs.
This research did not stop altogether three years ago. The ONC continued to support the development of standards-based IT platforms with its Health Information Technology Standards Panel (HITSP). In addition, the Fast Healthcare Interoperability Resources (FHIR) initiative Fast Healthcare Interoperability Resources (FHIR) initiative and SMART currently work to develop a platform to create healthcare applications that improve care at the point of service.
Keeping up with the Demands of Value-Based Care
Population health strategies and the move to value-based care demand rich data to be extracted from and available to as many partners in the healthcare chain as possible to solve the most important problems within key patient groups. Private and public collaboration on standards and open systems promises to make it easier to standardize terminology and to help unlock the value of clean data for population health strategies. There will always be technical issues and evolving formats; however, it is essential to ensure that the best practices and standards are being used to normalize and present the data in its cleanest form.
Interested in learning more? Read our white paper “Clean Data: The Critical Component for Population Health Success” for a deeper dive.