F. James Charney
Office of Management and Budget
New Executive Office Building
Washington, DC 20503
Dear Mr. Charney:
We write to provide comments on the proposed revision of section ___.36 of OMB Circular A-110, Uniform Administrative Requirements for Grants and Agreements with Institutions of Higher Education, Hospitals, and Other Non-Profit Organizations, (published at 64 Federal Register 5684, Feb. 4, 1999)
The revision (proposed pursuant to a requirement in Public Law 105-277) would extend mandatory disclosure requirements under Freedom of Information Act (FOIA) procedures to data produced under Federal grants and used by the Federal Government in developing policy or rules.
We have carefully considered this issue with our awarding agencies and affected entities outside the Department. In addition, the National Institutes of Health (NIH) convened a working group which reviewed the proposed revision and developed comprehensive comments. Those comments (enclosed) encompassed issues identified by many of the other HHS agencies. We have included some of the NIH concerns in this letter. In addition, we have also highlighted some of the operational issues expressed by the HHS agencies and their grant recipients. Finally, in addition to the NIH comments, we are enclosing comments from the HHS Freedom of Information Officer, the Agency for Health Care Policy and Research, the Office of Minority Health, the Substance Abuse and Mental Health Services Administration, and joint comments from the Centers for Disease Control and Prevention and the Agency for Toxic Substances and Disease Registry.
This Department strongly supports the sharing of research data, but we have concluded that the requirement to share data in the fashion proposed presents several difficult problems, and is not a good long-range approach to sharing research data. We believe that FOIA is fundamentally inappropriate as a vehicle for sharing research data held by grantees. Given the complexities inherent in the problems identified, our comments recommend possible solutions to some of these problems, but they also identify areas of concern even within these solutions.
It is important to remember that while A-110 provisions apply to universities, hospitals, and other non-profit organizations, they also apply to state, local, and tribal governments when they make subawards to entities covered by A-110. More importantly, as you know, we have applied A-110 to the large HHS entitlement grant programs such as Medicaid.
We emphasize at the outset that we believe it essential to facilitate access to research data, and HHS has a long history of making research data available for reanalysis or further research. For example, HHS typically makes data from survey research available through public use data sets and research data archives. In addition, the National Institutes of Health has taken a strong stand to encourage the sharing of research tools, protecting the scientific community's access to early work in gene identification and other areas of science.
However, it is as least as important that such access occur only in the context of strong protections for research participants, protection of proprietary interests, freedom from harassment for researchers, and confidence that the process will further research, not harm it. Our recommendations for revision seek this balance, but with the caveat that it is extremely difficult to accomplish it under the FOIA approach.
There are four major areas of concern on which we offer comments:
- definitions or concepts that require clarification;
- the protection of data, in the interest of the privacy of individuals and related interests of research entities;
- the burden on both agencies and grantees; and
- some technical issues such as timing of implementation, compliance, and data retention.
Clarification of Terms
The proposed revision explicitly or implicitly uses several key concepts -- data, publication, regulation, policy, level of funding -- for which agencies and external organizations may have different definitions. The already considerable burden on agencies and grantees would be increased if grantees had to deal as well with variability across agencies. Varying definitions would also raise concerns about the lack of consistency among agencies. Therefore, we recommend definitions of these concepts to ensure certainty.
For data produced only in part with Federal funds, we strongly recommend that any new rule clarify that only those projects for which 51 percent or more of the funds were from Federal sources should be subject to the requirement. 51 percent is not an arbitrary standard as it is used in making other governmental determinations with respect to size and definitions (e.g., 51 percent is used to determine whether businesses qualify as minority business enterprises based on required minority ownership).
The proposed revision would provide the access right when data "were used by the Federal government in developing policy or rules." It is important to define with some precision the actions that constitute policy or rules. The Federal government takes many actions and makes many choices, with varying degrees of impact and with foundations in many scientific sources (grant-supported and otherwise). Uncertainty about what Federal action gives rise to the right of access could lead to much confusion and countless legal challenges. We propose that the access right be triggered only by distinct actions: formal agency regulations that go through the rulemaking process, or other actions of equivalent impact, e.g., ones that are binding, require adherence, and take on a permanent or semi-permanent status. The right should be available only when the link between the data and the action is clear and explicit.
Effect on Protection of Data and Privacy
Protecting the privacy of individual research subjects is essential. It is important for two reasons: respect for the dignity and autonomy of research subjects demands that we not harm them by disclosing information about them; and disclosure of personal information outside of research would discourage future participation in research.
The law and the proposed revision alter the fundamental relationship between Federal agencies and their grantees, with implications for altering existing policies that protect against disclosure of personal information about research subjects. We thus recommend that the revision be distinct about the primacy of privacy as a factor in making disclosure decisions.
We note that the rule applied in its most protective fashion could still be harmful to individual privacy. If the subjects' confidence in confidentiality is undermined, research will be seriously undermined.
Increased Burden on Agencies and Grantees
The new law extends the current FOIA but is different in several ways. FOIA currently applies to documents already in the
possession of the Federal government, while the new law requires that the Federal government go to the grantee and obtain the data. This administrative burden is far greater than in the past for the agencies and extends to the grantee institutions and their investigators. In addition, the grantee which is burdened by responding should be compensated directly. We recommend that the new policy be accompanied by whatever changes are necessary to compensate the grantees for bearing this burden.
Since the new possibility of data release under the circular will likely require changes in the consent forms that research subjects sign, it is essential that the law apply only to data collection activities that begin after final implementation by grant-making agencies. To apply the new disclosure policies to existing data would risk violating agencies' and grantees' promises to research subjects. In addition, agencies must have the opportunity to build in strategies for access to and protection of scientific data from the outset of a research project.
Requirements for data retention also need to be reviewed so that A-110 and Federal Records Act retention requirements are not counterproductive. While we do not want to create an incentive to dispose of data in order to avoid FOIA requests, we also do not want to put grantee personnel in the position of facing a never-ending obligation to retain and be prepared to provide data that may far exceed their tenure at an institution.
Below are some excerpts from additional comments:
- One of the most important disadvantages of the proposed revision is that it will undermine or even jeopardize AHCPR's partnerships (for data, data linkages, or shared funding) with private institutions ....
- This draft does not indicate whether the regulation would be retroactive.
- What is a reasonable time to obtain the data?
- Examples of organizations which provide data are private insurance plans, managed care companies, and health information vendors. Would these files, given to researchers only to address specific questions by the investigators, now be available to the public if the research is performed with federal funding? If so, I suspect that these groups would no longer be willing to provide their data for federally-funded research. This would seriously limit the types of populations and issues that researchers could investigate.
- The paper work that it will generate is enormous, and counter to the reduction in paper work efforts. Entire staffs of individuals will have to be recruited to make the decisions that will have to be made, with consequent great increase in costs of conducting research and reporting its results.
We recommend additionally that there be an opportunity for the scientific and other affected grantee communities to participate in open hearings on this proposal and on other strategies for the sharing of research data. We believe that such a process could lead to better access to scientific data that would provide the openness in public policymaking that we all seek, and at the same time not damage the research enterprise. We look forward to working with you in achieving solutions to some of the problems identified in this correspondence and its enclosures.