Blog Post

Using an EHR for Population Health: What’s missing?

Share this post:

Anil Jain, MD is Vice President and Chief Health Informatics Officer at IBM Watson Health.

While the electronic health record (EHR) will continue to be important, it was not originally designed to serve as the hub of longitudinally organized person and population level data such as administrative (claims, benefits, etc.), clinical, sociodemographics, genomics, and patient-generated health data (PGHD). Healthcare organizations must leverage enterprise data to meet the needs of growing populations under their management.

First, we must recognize what EHRs CAN do:
• EHRs provide the mechanism for documenting encounters, procedures, vital signs, and physical exams.
• EHRs facilitate much of the workflow that clinicians follow in their everyday work.
• Practices use EHRs to provide documentation to support billing codes.
• The EHR is the basis of the medical-legal record, which is critical for discovery and defense in the event of a malpractice claim.

That said, it’s important to address what’s needed beyond the EHR, to facilitate effective population health management (PHM)

Interoperability through a Care Collaboration Platform
In the age of big data, interoperability among key sources and IT systems is critical, especially as new technology eclipses health information exchanges (HIEs). New apps based on the Fast Health Interoperability Resources (FHIR) standard are under development, but reliable and scalable solutions for safe, secure, effective data integration are still elusive.

A “care collaboration platform” could prepare organizations for PHM — today and in the future. Connecting disparate data sources, such as claims and clinical data, through a care collaboration platform would facilitate virtual collaboration across the care team. With communication tools and workflow features, the platform would ensure consistency and quick access to the latest patient information.

Overall, this platform could deliver a more comprehensive view, enabling professionals to better understand patient needs, identify and mitigate risks, and improve follow-up care.

Harnessing New Data Sources
In addition to EHR and claims data, the centralized platform would continuously ingest and synthesize data from care managers and home health nurses and other relevant observations.
Additionally, the platform would make use of structured and unstructured data. Since unstructured data, such as procedure notes, observations and discharge summaries, comprises about 80% of the information in EHRs, uncovering this data helps to provide a full patient view.

Using advanced natural language processing (NLP), the platform could “read” unstructured data. Such intelligent processing would help to paint a broader, more vivid picture of a patient’s situation.

Cognitive Computing to Empower Providers
NLP is just one example of basic building blocks of cognitive computing, a next-generation approach to big data in which systems “learn.” Cognitive computing applications can analyze enormous amounts of data from all kinds of sources and grow smarter through interacting with those applications over time. Interoperable cognitive applications can expand the function of the overall platform by utilizing standard application programming interfaces (APIs).

While managing the health of a population requires considering each person, cognitive computing can provide that personal care at scale within the context of the overall population. The system can identify data patterns across the population and use those insights to fine-tune the delivery of more optimal care. It can even help match patients and care managers whose personalities are well suited to each other by first learning and then adapting to what factors motivate behavior change.

Care plans are most effective when they are evidence-based but tailored to needs of unique needs of a patient collaborating with a provider. Providers who have access to data and other factors from a patient’s care team can improve patient engagement through this shared decision-making process.

Cognitive computing scales the transformation of big data into actionable insights, enabling professionals to create more precise care plans that can help lead to better outcomes. With PHM-ready infrastructure, healthcare organizations can take a more comprehensive view of the populations in their care.

Fill the Gaps with the Smart IT Solutions
Building the right IT infrastructure is often the first step in developing population health management programs. While EHRs can provide a basic foundation, to truly understand the whole patient and the whole population, organizations must supplement their EHRs with the right IT solutions.

For a deeper look at this topic, download our white paper series Population Health: Beyond the EHR and Population Health: Beyond the EHR Part 2.

More Value-Based Care Stories

Does your quality performance measurement give you the full picture? It may be time to expand your metrics.

Written by Watson Health | Blog Post, Value-Based Care

Atlanta-based DeKalb Medical knew there were still areas of operational and clinical performance that could be improved — they just needed help pinpointing opportunities. ...read more


Insights and Innovations: Top 10 Value-Based Care Posts of 2017

Written by Watson Health | Blog Post, Value-Based Care

As 2017 comes to an end, we’ve compiled the top 10 most popular value-based care posts for the year on Watson Health Perspectives, based on reader interest. ...read more


Are you able to assess the impact of an employee health program when data is stored in multiple, diverse systems? Whirlpool did.

Written by Watson Health | Blog Post, Value-Based Care

Comprehensive analytics enabled Whirlpool to look at the outcomes of an employee health program from a number of angles and organize results to support decisions about the future of the program. ...read more