continuous management and improvement of these data are necessary to enable the transition to a continuously learning health system.

The applications of digital health data in a learning system are multiple, including care coordination; management of patient populations; associated care and business processes; outcome, quality, and value assessments; generation of clinical evidence, including clinical trials, clinical effectiveness, and genomic studies; surveillance and trend detection, including medical products safety, syndromic and actionable surveillance, and hypothesis generation; and public health program management. These differing uses vary in their requirements for data quality and characteristics, but all share common challenges related to data access, liquidity, interoperability, and the development of innovative methods for analysis. These issues formed the foundation for the presentations and discussions at the IOM public workshop on Digital Data Priorities for Continuous Learning in Health and Health Care.

WORKSHOP SCOPE AND OBJECTIVES

Workshop participants included experts from across medicine, public health, informatics, health information technology, health care services research, health care quality reporting, biomedical research, clinical research, statistics, medical product manufacturing, health care payment and financing, and patient advocacy. Content was structured to explore the data quality challenges and opportunities in a learning health system, highlighting the opportunities and priorities beyond care coordination such as population and care process management, clinical research, translational informatics, and public health support at the national and state level. The workshop also explored the potential for learning from large-scale health datasets, focusing on innovative approaches to overcoming the challenges of distributed data, data harmonization, and identity resolution.

The workshop statement of task can be found in Box 1-1, and the elements are reflected in the stated meeting objectives:

1.  Discuss the current quality status of digital health data.

2.  Explore challenges, and identify key questions related to data quality in the use of EHRs, patient registries, administrative data, and public health sources for learning—continuous and episodic—and for system operational and improvement purposes.

3.  Engage individuals and organizations leading the way in improving the reliability, availability, and usability of digital health data for real-time knowledge generation and health improvement in a continuously learning health system.

4.  Identify and characterize the current deficiencies and consider strategies, priorities, and responsibilities to address the deficiencies.

5.  Initiate the development of a strategic framework for integrated and networked stewardship of efforts to continuously increase digital data utility.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement