[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Vital Directions from the National Academy of Medicine
October 25, 2016

Data Acquisition, Curation, and Use for a Continuously Learning Health System

Author Affiliations
  • 1Section of Cardiovascular Medicine, Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut
  • 2Genetic Alliance, Washington, DC
  • 3Johnson & Johnson, New Brunswick, New Jersey
JAMA. 2016;316(16):1669-1670. doi:10.1001/jama.2016.12537

Health-related data and research data are vital resources for clinical care, informed clinical choice, quality improvement, drug and device safety, effectiveness assessments, and scientific discovery. Such data are the reagents that can be used to produce information to support personal choices about health care, system choices about optimizing medical and public health strategies, and policy choices about current and future laws and regulations. Data provide the necessary ingredients for medical breakthroughs.

Sharing, curation, and use of data for a continuously learning health system hold great potential to better engage people in their health and health care. Initially introduced by the Institute of Medicine, a learning health system is described by the Office of the National Coordinator for Health Information Technology (ONC) as an ecosystem in which all stakeholders can contribute, share, and analyze data and where continuous learning cycles encourage the creation of new knowledge that can be used by a variety of health information systems.1,2 In part owing to an inability to fully leverage relevant data, a learning health system remains more an aspiration than an achievement.

There are formidable impediments to leveraging existing data. Data holders, including health systems, researchers, health companies, and others, currently bear most of the costs relative to the benefits of data sharing. The incentives of data holders are not always aligned with those of society, patients, and researchers. Moreover, there are technological and cultural barriers. This leads to difficulty in coalescing health-related data that reside in disparate venues and formats within the health ecosystem. The ability to access these data is not sufficient to produce benefit; advances in analytics and application are also required.

Recent Progress on Health Data Access and Use

In recent years, policy makers, organizations, and individuals have advanced efforts to promote the culture and infrastructure needed to support the secure accessibility of health and health care data.3 For example, the companies that are part of the Pharmaceutical Research and Manufacturers of America have committed to sharing their data.4 The International Committee of Medical Journal Editors released a proposal stating the belief that there is “an ethical obligation to responsibly share data generated by interventional clinical trials.”5 The promulgation of common standards, implementation of appropriate legislation and regulations, growth of public activism regarding health information, and technological advancements have sped changes in expectations and capabilities.6 Funders are increasingly making support contingent on data sharing. Agencies such as the National Institutes of Health and the Patient-Centered Outcomes Research Institute and private foundations including the Wellcome Trust and the Bill & Melinda Gates Foundation have mandated forms of data sharing as a condition of funding. On the clinical side, companies that provide 90% of the country’s electronic health records and several large health systems have signed the ONC interoperability pledge and committed to consumer access, no blocking/ensuring transparency, and standards.7

Summary Recommendations for Vital Directions

To enable data to fuel a learning health system, progress is important across several domains. Change is needed in the culture and incentive structures of the health system to move away from a status quo with little opportunity for data sharing. Barriers to digital health data sharing create inefficiencies and errors that cost lives and resources. Strategically important in this respect is encouraging individuals’ access to their data by clarifying individuals’ rights to their data. This will require the creation of the tools and infrastructure needed for patients to put their data to work for them—and may also require regulatory changes. A final vital direction is developing seamless means to synthesize data from disparate sources. Priority considerations in promoting these vital directions therefore include those that follow.

  • Foster a culture of data sharing. For data sharing to become more common, the culture of health care, public health, and medical science will need to evolve such that refusing to share is understood as counter to the best interests of individuals and society. There should be a broad appreciation for an individual’s right to access and share his or her own health data. In research, there should be expectations that good science and good scientific citizenship require that participant-level data be available for evaluation and reuse. Incentives should be aligned to reflect that culture. It will be important to create support and rewards for sharing data and penalties for not sharing data. Financial incentives should reward health systems that facilitate sharing, companies with data-sharing programs, and vendors with interoperability features. Academic promotion could consider data sharing and downstream use of the shared data. Metrics on ease of data accessibility at the hospital, health system, or office level should be publicly reported.

  • Create the operational functionality for data sharing. The first step in this respect is for government to establish the legal and regulatory tailwinds for data sharing. Data sharing should be mandated for studies that use public funds and for those published in journals. Authentication systems for patient portal access should be standardized. The ONC effort to support application programming interfaces to facilitate the exchange of health data should be continued.8 It will also be important to advance ways in which data could be shared with the individual’s explicit permission. Movement of data without an individual’s permission should be minimized to the extent possible and comply with permitted uses and appropriate disclosures.

    The platform is critical. Today’s platforms are inadequate to store and manage the increasing range of health data. Stakeholders should work together to seek common requirements for infrastructure and support the development of participant-centric data-sharing solutions. Owing to the possibility of economies of scale and without overlooking the benefits of market competition, government solutions for data-sharing infrastructure should be investigated.

  • Build the continuous data-sharing improvement capacity. Strategic federal initiatives are needed for issues like health data sharing that have substantial consequences across multiple levels of influence. Because many of the efforts recommended herein are already under way in the federal government, such an initiative could be started de novo or be combined with existing efforts—and may be best accomplished as a White House Initiative spanning the government. The initiative would also seek to support market forces to leverage governmental efforts by creating products that facilitate the use of increasingly available data. The visibility of the efforts can create opportunities to recognize achievements, promote education about rights and laws, institute standards, penalize infractions, and protect individuals.

    Research into data-sharing tools is also important. Optimizing the organization and use of data depends on the ability to generate new knowledge in the fields of data science, health economics, and implementation science. The capability to do so will require investments in research germane to data sharing. To embrace data-driven health care, a culture shift is needed across academia and the research community in what is considered science, rather than infrastructure development or implementation.

    In addition, interventions that aspire to promote data sharing should be evaluated by quantitative metrics that assess progress and monitor for unintended consequences. The development of these metrics requires stakeholder input, data sources to enable the calculations, and specifications that promote a true reflection of the domain under assessment. Examples of metrics are percentage of late-stage clinical trials by funder with complete and accurate reporting at ClinicalTrials.gov within 12 months and published within 18 months of completion, and percentage of the nation’s 200 largest health systems with Blue Button capability and ability for patients to view, download, and transmit their own comprehensive data. Another measure could be the number and quality of articles produced by data sharing.

At this vital juncture in health care and research, the secure sharing of data has great potential. However, achievement of such a grand strategy for change will require unprecedented levels of collaboration among and communication between all stakeholders in the health system, and systems to evaluate effects and iterate for improvement.

Back to top
Article Information

Corresponding Author: Harlan M. Krumholz, MD, SM, Section of Cardiovascular Medicine, Department of Internal Medicine, Yale School of Medicine, One Church Street, Ste 200, New Haven, CT 06510 (harlan.krumholz@yale.edu).

Published Online: September 26, 2016. doi:10.1001/jama.2016.12537

Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Krumholz reported having research agreements with Medtronic and Johnson & Johnson (Janssen), through Yale, to develop methods of clinical trial data sharing; serving as chair of the cardiac scientific advisory board for UnitedHealth; being a founder of Hugo, a personal health information platform; and receiving research funding from the US Food and Drug Administration and Medtronic, through Yale, to develop methods for postmarket surveillance of medical devices. Dr Waldstreicher reported being an employee of and stockholder in Johnson & Johnson; a former employee of Merck and Co; and a shareholder in a common stock fund for Merck and Co. No other disclosures were reported.

Funding/Support: The National Academy of Medicine’s Vital Directions initiative is sponsored by the California Health Care Foundation, The John A. Hartford Foundation, the Robert Wood Johnson Foundation, and the National Academy of Medicine’s Harvey V. Fineberg Impact Fund.

Disclaimer: This Viewpoint on data acquisition, curation, and use for a continuously learning health system provides a summary of a discussion paper developed as part of the National Academy of Medicine’s initiative on Vital Directions for Health & Health Care (http://nam.edu/vitaldirections). Discussion papers presented in this initiative reflect the views of leading authorities on the important issues engaged, and do not represent formal consensus positions of the National Academy of Medicine or the organizations of the participating authors.

Additional Contributions: Coauthors of the National Academy of Medicine discussion paper included Philip E. Bourne, PhD (National Institutes of Health), Richard E. Kuntz, MD (Medtronic), Mark B. McClellan, MD, PhD (Duke University), and Harold L. Paz, MD (Aetna). Elizabeth Finkelman, MPP (National Academy of Medicine) served as the initiative director. Maria A. Johnson, MBA (Yale Center for Outcomes Research and Evaluation) and Pranammya Dey (Yale Center for Outcomes Research and Evaluation) provided editorial assistance; they received no compensation.

Institute of Medicine. Best Care at Lower Cost: The Path to Continuously Learning Health Care in America. Washington, DC: National Academies Press; 2012.
DeSalvo  KB. Connecting Health and Care for the Nation: A Shared Nationwide Interoperability Roadmap, Version 1.0. Washington, DC: Office of the National Coordinator for Health Information Technology; 2015.
Ross  JS, Krumholz  HM.  Ushering in a new era of open science through data sharing: the wall must come down. JAMA. 2013;309(13):1355-1356.
Pharmaceutical Research and Manufacturers of America. Principles for responsible clinical trial data sharing. http://www.phrma.org/phrmapedia/responsible-clinical-trial-data-sharing. Accessed May 16, 2016.
Taichman  DB, Backus  J, Baethge  C,  et al.  Sharing clinical trial data: a proposal from the International Committee of Medical Journal Editors. JAMA. 2016;315(5):467-468.
National Institutes of Health Office of Extramural Research. NIH sharing policies and related guidance on NIH-funded research resources. https://grants.nih.gov/policy/sharing.htm. Accessed May 16, 2016.
HealthIT.gov. Interoperability pledge. https://www.healthit.gov/commitment. Accessed August 22, 2016.
Office of the Federal Register. 2015 edition health information technology (Health IT) certification criteria, 2015 edition base electronic health record (EHR) definition, and ONC health IT certification program modifications. http://www.federalregister.gov/articles/2015/10/16/2015-25597/2015-edition-health-information-technology-health-it-certification-criteria-2015-edition-base. Accessed August 22, 2016.