Archives for Comments

Today the Biomedical Informatics Think Tank™ submitted comments on the Federal Health IT Strategic Plan: 2011-2015.  This effort involved preparing a draft and then obtaining feedback and making edits based on BITT™ member input.  The PDF of these comments can be downloaded from http://tpcaruso.com/BITT/Files/HIT Strategic Plan Comments.pdf.

—————————————————————-

May 5, 2011

Sarah Potter Office of the Assistant Secretary for Planning and Evaluation
U.S. Department of Health and Human Services
200 Independence Ave, SW, Room 404E
Washington, DC 20201

Dear Ms. Potter:

I submit this letter on behalf of the The Biomedical Informatics Think Tank™ (BITT™), a division of Projectivity, Inc. in response to a request for comments on the Federal Health IT Strategic Plan 2011-2015.  The Biomedical Informatics Think Tank brings together experts at major academic medical institutions who have been working in the fields of health information management, clinical informatics, public health informatics, clinical research informatics, bioinformatics, data mining, HIT training and high performance computing over the last thirty years.  We are focused on efforts that integrate the needs of clinicians with the needs of medical researchers to improve comparative effectiveness research (CER), public health, clinical and biomedical research, and clinical decision support system (CDSS) design.

We applaud your efforts to create health IT standards that encourage interoperability, as well as various programs that stimulate the adoption of EHRs, the growth of statewide health information exchanges, and the training of health IT experts to support this infrastructure.  Looking forward, we recommend greater emphasis on building a new health information exchange infrastructure that supports comparative effectiveness research (CER), other public health, clinical and biomedical research, and clinical decision support systems (CDSS) design.  This effort to build a stronger foundation for evidenced-based medicine (EBM) is essential to improving the health of Americans while decreasing the overall cost of health care.  We recognize that this requires giving primary attention to the ability of individuals to control the privacy and security of their health information.   The BITT has produced a white paper addressing the opportunity suggested by the President’s Council of Advisors on Science and Technology (PCAST) in their Health IT Report[i] to create a Universal Exchange Language that would support CER and other biomedical research.[ii]

In particular, we encourage strengthening of Strategy II.D.3: Ensure a mechanism for information exchange in support of research and the translation of research findings back into clinical practice.

“…information in an EHR, with the appropriate privacy protections, should be accessible by researchers, research systems, biorepositories, registries, and other types of research databases. In turn, adopted health IT interoperability standards should be used in research studies – and enhanced to cover unique research needs – in order to facilitate both the use of health care data in research and the translation of research results into clinical practice.”

Health information exchange and the secondary use of health care information are essential for research, which is critical to the success of health care reforms proposed in the Affordable Care Act.  Many of the members of the BITT have worked together to build just such a standards-based informatics infrastructure for the National Cancer Institute’s Cancer Biomedical Informatics Grid® (caBIG®), the largest single effort to integrate clinical and biomedical information for advancing biomedical research.[iii] Some of the members of BITT have also built semantic, grid and cloud infrastructures for a broader effort to increase shared use of information technology resources by biomedical research scientists.[iv] Many of the members are principle biomedical informatics experts with Clinical and Translational Science Awards (CTSAs) who are leading the charge to leverage the growing clinical and research data to allow for improvements in prevention and clinical care.

Access to clinical data has been extremely difficult primarily due to fears about privacy and security.  To solve this problem, Meaningful Use criteria for Stage 2 and 3 must include requirements that provide easy access to deidentified clinical data.  We recommend that Institutional Review Boards be required to establish criteria for such access, and that such access criteria be standardized nationally.  As part of this effort, the informed consent process should have clear standardized guidelines for increasing the secondary uses of an individual’s health information.

Once the clinical data is accessed, then making sense of this data becomes a major hurdle.  We recommend that the Federal government establish a national standard ontology that will be required in all clinical information systems in the near future, possibly as soon as the deadline for meeting Meaningful Use Stage 2 requirements.  Such a national standard is necessary for building infrastructure to support CER and other public health, clinical and biomedical research, and CDSS design.

As proposed in the PCAST Health IT Reportⁱⁱ, we recommend that the ONC create a Universal Exchange Language (UEL) Specification that defines an underlying metadata tagged graph structure based in an XML-like format that is compatible with the Semantic Web Specifications of the World Wide Web Consortium (W3C) and, like the W3C Specifications and relational databases, has a strong mathematical foundation.  The vision of a UEL supports, enhances and extends the ability of researchers and clinicians to examine large volumes of data in a fashion that does not threaten the identity of the individuals providing that information.  It creates a path to data in all relevant formats, states, and conditions.  Using a UEL, researchers will be able to see things at a macro level as never before, bringing an enormous benefit at the micro level, i.e. to patients.  Supporting Goal V, these efforts will move us toward a “learning health system”.  The UEL will power the future of health care.

A UEL in concert with Health Information Exchanges (HIEs) becomes a powerful, data resolving tool, giving a common parlance to diverse descriptions.   A UEL provides the meta-language needed by doctors, patients and researchers to reduce the barriers to confronting life and death issues that patients face, as well as halting the mounting costs of health care facing individuals, employers and the nation.

The common and readily adopted platform and data interpretation found in a UEL can be readily subsumed into many different health care delivery business models saving valuable time and money while clearing the way for further and more pertinent innovation that directly helps patients and providers make health improvements and improve health care.  A UEL can be likened to a utility service, it is not the message, it is the means by which the medical ‘message’, whatever its format , can be safely and accurately retrieved and translated for use by doctors, nurses and researchers.

Thus, we strongly recommend that methods using this UEL Specification be researched, designed, tested and then implemented for a number of use cases including securely aggregating information without the need for a universal identifier for each individual.  The Biomedical Informatics Think Tank has created one proposal for just such a technology.[v]

To repeat our key recommendations:

1. Include requirements that allow easy access to deidentified clinical data in EHRs as a part of the Stage 2 and 3 Meaningful Use requirements.
2. Require that institutional review boards establish criteria for easy access to deidentified data, and work on creating a national standard of these criteria for IRBs.
3. Establish a national standard ontology that will be required in all clinical information systems in the near future, possibly as soon as the deadline for meeting Meaningful Use Stage 2 requirements.
4. Create a Universal Exchange Language (UEL) Specification that defines an underlying metadata tagged graph structure based in an XML-like format that is compatible with the Semantic Web Specifications of the World Wide Web Consortium (W3C).
5. Research, design, test and implement first in pilots and then in national rollouts methods using this UEL Specification including methods for securely aggregating information without the need for a universal identifier for each individual.

These recommendations are critical to the ability of researchers to perform CER, other clinical and biomedical research, public health analysis, and CDSS design.  These recommendations will enhance the value of future investments made by the Patient-Centered Outcomes Research Institute (PCORI) into CER as described in the Affordable Care Act.  In particular UEL is crucial in pressing forward with the national research agenda for CER that PCORI will be tasked with managing.  As recognized in the Affordable Care Act, this research activity is essential for improving the quality of health care for Americans, while halting the increase in costs of providing health care.

Sincerely,

Thomas P. Caruso, PhD., MBA, PMP
Member, Biomedical Informatics Think Tankâ„¢ and
President, Business Ambitions