Here is what I have come up with so far. I would welcome comments on this draft for the next couple of weeks; I'd like to submit this comment by October 15 to be sure I make the deadline.
To the Office for Human Research Protections:
Thank you for the opportunity to comment on the draft revision of the "Terms of the Federalwide Assurance for the Protection of Human Subjects." I have two comments on this draft.
1. THE DRAFT SHOULD BE REVISED TO REFLECT THE PROVISIONS OF 45 CFR 46.103(b)(1)
I am disappointed that the current draft fails to correct a longstanding discrepancy between the Common Rule and OHRP's forms. 45 CFR 46.103(b)(1) requires that each institution receiving funding from a Common Rule agency submit an assurance that includes
A statement of principles governing the institution in the discharge of its responsibilities for protecting the rights and welfare of human subjects of research conducted at or sponsored by the institution, regardless of whether the research is subject to Federal regulation. This may include an appropriate existing code, declaration, or statement of ethical principles, or a statement formulated by the institution itself.
By contrast, the draft Federalwide Assurance requires U.S. institutions to pledge that they will be guided either by the Belmont Report, the Declaration of Helsinki, or "other appropriate international ethical standards recognized by U.S. federal departments and agencies that have adopted the Common Rule." I am unaware of any documents in this third category, nor of any element of the Common Rule that requires federal approval of a statement of principles.
Thus, while the Common Rule offers institutions complete freedom in their choice of ethical principles, the current and proposed Terms of the Federalwide Assurance limit them to one or two documents. This is like guaranteeing the freedom of religion, then requiring every citizen to adhere to either the Lutheran Book of Concord or the Articles of Religion of the Methodist Church.
Instead, the first paragraph should reflect the provisions of the Common Rule. I suggest the following language:
"All of the Institution's human subjects research activities, regardless of whether the research is subject to the U.S. Federal Policy for the Protection of Human Subjects (also known as the Common Rule), will be guided by a statement of principles governing the institution in the discharge of its responsibilities for protecting the rights and welfare of human subjects of research conducted at or sponsored by the institution, regardless of whether the research is subject to Federal regulation. This may include an appropriate existing code, declaration, or statement of ethical principles, or a statement formulated by the institution itself. This requirement does not preempt provisions of this policy applicable to department- or agency-supported or regulated research and need not be applicable to any research exempted or waived under §46.101(b) or (i)."
2. THE DRAFT SHOULD BE REVISED TO REFLECT ACCURATELY THE AUTHORSHIP AND EVOLUTION OF THE TRI-COUNCIL POLICY STATEMENT
The current draft allows non-U.S. institutions to comply based on "The 1998 (with 2000, 2002, and 2005 amendments) Medical Research Council of Canada Tri-Council Policy Statement on Ethical Conduct for Research Involving Humans."
This statement has two inaccuracies. First, the Medical Research Council no longer exists; it was replaced in 2000 with the Canadian Institutes of Health Research (CIHR). Second, the TCPS is authored not only by the CIHR but also by the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Social Sciences and Humanities Research Council of Canada (SSHRC). (That is what makes it a tri-council policy.)
Moreover, the Panel on Research Ethics plans to release a second edition of the TCPS in December 2010, and may amend it further while the new Terms of the Federalwide Assurance are still in effect. Rather than limit institutions to an outdated version of the TCPS, OHRP should allow non-U.S. institutions to abide by the current version.