Toward a learning health system

This past year, I took a leave of absence from Allscripts to serve as the Office of the National Coordinator for Health IT (ONC) coordinator for Query Health, an Open Government Initiative that is establishing standards, policies and services for distributed population queries of clinical records.  It comes at a unique moment in time – at the confluence of broad deployment of Electronic Health Records, the compelling need for standards for secondary use of that healthcare information, and a Stage 3 Meaningful Use strategy that focuses on a “learning health system.”  That is, a system in which the vast array of health data can be  aggregated, analyzed, and leveraged using real-time algorithms and functions.

I’m thrilled to be back and sharing what I learned about what we can do to implement a learning health system that benefits patients on a national scale.

Our work began in August 2011 in Washington D.C., with a “Summer Concert Series” environmental scan of the best work on distributed queries happening around the country.   I collaborated with some of the top folks in the industry from the more than 100 member organizations. It was energizing to be engaged with colleagues so deeply committed and passionate about improving health care.

My job was to lead the overall initiative representing ONC.  Clinical, operations and technical workgroups, each with around 40 members, delivered the functional and operational requirements, the technical approach, the proposed standards and reference implementations.    We actively engaged with the National Coordinator, the HIT Standards Committee, the HIT Policy Committee and the Privacy and Security Tiger team to ensure that Query Health aligned with broad national priorities and strategies.

Understanding Population Health

Distributed population queries can be applied to a variety of secondary uses.  Distributed population queries enable an understanding of population measures of health, performance, disease and quality, while respecting patient privacy, to improve patient and population health and reduce costs.

Distributed population queries are a central component of ONC’s strategy for a learning health system.  These queries “send questions to the data” and return aggregate population measures that keep patient-level information protected at the source.

We use distributed population queries today for a variety of purposes.  For example, public health tracks diseases, including flu-like illness, and evaluates optimization of scarce resources.  The FDA evaluates signals related to drug safety once drugs are released to the market.  Researchers compare the relative effectiveness of drugs and treatments.

Putting It into Practice

There are five Query Health pilots kicking off this Summer and Fall. 

1. The New York City and State public health departments are sending questions to both provider practices and RHIOs related to diabetes and hypertension.
2. The Food and Drug Administration is sending questions to a clinical data source at Beth Israel Deaconess Medical Center to evaluate which post-market drug surveillance questions can be supported by clinical data. 
3. The Massachusetts Department of Public Health is sending diabetes-related questions to community health centers and provider practices. 
4. The Centers for Disease Control is applying Query Health standards to its BioSense 2 cloud-based distributed data repository for situation awareness and disease syndromes. 
5. Allscripts is testing the applicability of Query Health to dynamically query for clinical quality measures. 

Query Health standards are being prepared for standards ballot by HL7 and ONC’s Office of Science and Technology.  The standard for Queries is based on an improved, more parsimonious version of the Health Quality Measure Format or HQMF.  The standard for Results is the Quality Reporting Document Architecture or QRDA (Categories 2 & 3).  The target data is aligned with the S&I Framework Clinical Element Data Dictionary, the National Quality Forum’s Quality Data Model and the HL7 Consolidated CDA. 

You can find more information about the project at QueryHealth.org. 

What’s your take on how the Query Health initiative can improve how we use health IT for the benefit of patient and patient populations? Do you have new ideas we haven’t yet considered? Share your thoughts below.