gotiations for new assurances implementing the regulations issued May 30, 1974, began in mid-1975 and took more than two years to complete.4

Although the process of providing an assurance of compliance with the regulations might ideally provide an occasion for careful consideration by institutions of how they will meet their responsibilities toward human subjects, some aspects of the manner in which new assurances are being negotiated decrease the likelihood that grantee institutions will take advantage of the opportunity to review their responsibilities.

A sample assurance (dated July 3, 1981, rev. August 11, 1981) has been distributed by HHS to each institution having a general assurance on file with HHS under the old regulations. OPRR had described plans to offer institutions a variety of model formats for re-negotiating their general assurances; instead, one sample was provided. In order to affirm its intent to comply with 45 CFR 46 an institution following this 22 page document would have to recite each individual section and subsection of the regulations almost in their entirety." This recitation has the effect of a litany,

4 Letter from Charles R. McCarthy to Morris B. Abram (May 7, 1980) at 2.

5 See, e.g., Statement of Charles R. McCarthy at the PRIM&R Conference on "The New Federal Regulations: What They Do and Do Not Regulate," Boston (March 26-27, 1981) at 38.

"To the extent that the "sample assurance" does go beyond the regulations, questions of a different sort are raised. The sample is not formulated as a series of issues of administration and structure that OPRR believes ought to be addressed. Rather, it sets forth a structure which-by its very conjunction with many requirements of the regulations-HHS seems to expect of an institution. In effect, by including a number of items not covered by the regulations (e.g., the whole concept of an "appeals IRB"; 7 of the 15 elements of "institutional policy" in Sec. I.C. of the "sample" assurances; etc.), the Department appears to adopt policies, principles or rules that have not been subject to the usual process of scrutiny and comment that is necessary in promulgating federal regulations under the Administrative Procedure Act. The "sample" assurance notes at several points that facets which have been made up out of whole cloth (e.g., the creation of an "Office of Research Administration" to exercise administrative responsibilities, including some which, under 45 CFR 46, appear to rest with the IRB) "will not be appropriate for some institutions and are not required by the HHS regulations." Yet this disclaimer is not provided about many of the amendations to the regulations and, in any event is dissipated by the overall impression created that the model assurance represents the Department's expections. For example, the cover letter informs each institution that "In preparing this sample assurance we have attempted to include all of the elements necessary for compliance with the new regulations."

rather than focusing an institution's thinking on the specific administrative and structural arrangements through which it will carry out the commitment to abide by 45 CFR 46. Not only is such reiteration unnecessary-a one sentence promise to comply with all regulatory requirements would serve the same function-but it may discourage thoughtful self-scrutiny and actual compliance.

OPRR has explained its choice of format for negotiating assurances by stating that when such a detailed assurance is distributed within an institution "the relevant parts of the regulations would [thereby] be in the hands of the individuals with responsibility for complying with them." Yet this laudable objective is defeated by the chosen format. The sample assurance makes oblique cross references to the regulations; this would make it necessary for anyone expecting to undertand the assurance to have a copy of the regulations in hand as well.7

The Commission realizes that while another approach might have been preferable, the single sample assurance is already in use. Moreover, the Department has told the Commission that it shares the Commission's concern that the negotiation of new assurances by all research institutions be

"For example, there are many statements of the sort that the institution will act "in accordance with 45 CFR 46.116" or "will comply with the policies set forth in 45 CFR 46 Subpart C." The baldest example of the need for an accompanying set of regulations occurs on the very first page (Sec. I.A.2) of OPRR's model assurance: "Only provisions II.A.16; II.B.1.a,b,d,e,f; and III. of this assurance are applicable to the activities listed above if the only involvement of human subjects will be in one or more of the categories exempted or waived under 45 CFR 46.101 (b)(15) or 46.101(e)."

regarded as an important opportunity for self-examination and appropriate ingenuity on the part of research institutions as they develop administrative mechanisms responsive to their particular needs. The effort devoted to this can yield great dividends in institutional sensitivity to, and readiness to provide effective protection for, the subjects of research.

The Steps Taken by OPRR to Improve Implementation. A more serious limitation of the assurance procedure as it presently exists is its emphasis on providing only a prospective picture of the process of research review that an institution will follow. Recognizing that this provides very little information about the actual performance of the IRB, OPRR has taken several steps. The first, which has been in existence for some time, is to respond to any serious problems that come to the office's attention. Mechanisms exist to bring to OPRR's attention some problems in IRB operations, though they are far from complete or systematic. In any case, most of the "problems" uncovered have not proven to be serious; most are disposed of through a phone call to the institution or through an exchange of correspond

ence.

To increase its awareness of the adequacy of IRB functioning, OPRR has instructed each "study section" (initial review groups which advise the Institutes on the scientific merit of applications for grants and contracts) to evaluate the investigators' descriptions of six factors bearing on the protection of human subjects (see Appendix B). The "Summary Statement" for each research application, prepared by the executive secretary of the study section, provides a means for the section members to express any concerns about the description of risks, the adequacy of protection against risks, and the balance between risks and benefits. If problems relating to the protection of human subjects are identified by a study section, they are called to the attention of the Institute's advisory council or board when the project is under consideration for funding. OPRR plays a coordinating role in resolving any such problems before HHS funds are permitted to be expended.

OPRR reports that it is planning to systematize the information available from the "Summary Statements" into a data base which could be used to evaluate the IRB system in general and the performance of each IRB in particular. The Commission hopes to learn more about these efforts as part of its work on its next Biennial Report. The sensitivity of such a system is a matter of special concern; for example, will it be able to differentiate serious problems from clerical errors on the part of an investigator or IRB, or to separate

those instances in which concerns raised by study sections that are found to be "justified" from those which are merely "differences of opinion" between a study section and a conscientious IRB?

Study section review does not provide OPRR either with general information about IRB functioning (since each study section looks only at the "end product" of IRB action in the cases it is reviewing and not at overall IRB activities) or with particular information about the manner in which an IRB follows up on research once approved. Some firsthand information is available to OPRR, however, through various site visit mechanisms. Although OPRR itself has conducted only a few such on-site inspections, the routine institutional site visits conducted by scientific review groups in the General Clinical Research Centers Program (which is operated by the Division of Research Resources at NIH) include meetings with IRB members and review of IRB practices. Summary reports of those site visits are reviewed by OPRR; none has triggered further review of an IRB by that Office.8

In responding to the National Commission's recommendation of "compliance activities," such as IRB audits and site visits, the Department in August 1979 said that Congressional action would be unnecessary on this point, since such practices were already part of HHS procedures. As already described, systematic efforts in the direction of "compliance" (as opposed to "assurance") mechanisms are still far from complete. Indeed, representatives of the Department have explained that steps toward auditing IRB performance which were described to the Commission in May 1980 remain in the planning stages because OPRR's limited resources are largely devoted to the regular process of reviewing all NIH research proposals for compliance with the regulatory requirements and to the negotiation of new assurances under the regulations promulgated in January 1981.

Although OPRR has not yet instituted regular site visits, it reports that some site visits have been conducted. Yet, since well-defined procedures for auditing IRBs are lacking, OPRR had difficulty in giving a complete picture of the site visits it has conducted. In a letter to the Chairman of the Commission in May 1980, the Director of OPRR defined site visits to include "examination of IRB minutes and interviews with the chairmen and members of the IRB, administrative staff, and research investigators" and reported that OPRR carried out two such site visits in Fiscal Year 1979

Letter from Dr. McCarthy to Morris B. Abram (May 7, 1980) at 3; see also testimony of Dr. McCarthy, transcript of 2nd meeting of the President's Commission (May 16, 1980) at 81-82.

and three in Fiscal 1980.9 The Deputy Director of OPRR testified in November 1981, however, that his office had conducted a total of 80 site visits between 1975 and 1981, although many of these were "of a routine nature to assist institutions in complying with the regulations in circumstances of special complexity (cooperative research projects of a large scale) or to provide guidance and information on HHS policy and to discuss general problems of IRB operation."10 Clearly, the November 1981 statement reflects a very different (i.e., more expansive) notion of what constitutes a "site visit." Indeed, the OPRR officials agree that the number of such visits that could properly be termed "audits" of IRB operations was probably "very small," and that few if any of those conducted were in response to allegations of serious problems or to reports from FDA inspection teams, or from the reports of NIH study sections' concerns.11 The additional contacts with research institutions do, however, provide OPRR with "extensive general information about IRB functioning even though the information lacks the precision that might come from formal IRB audits and site visits (in the narrow sense)."12

In order to mount effective "compliance activities,' OPRR will need a schedule of, and defined procedures for conducting, either routine or "spot" audits of IRBs. Such steps would permit genuine "site visits" to be readily distinguished from visits to provide guidance or information, on the one hand, and from extraordinary investigations of alleged misconduct, on the other. In order to help HHS obtain more than sporadic glimpses of the performance of IRBs, the Commission is working with OPRR (and the FDA) to develop means of obtaining information about IRBs that are both economical and likely to promote the system's highest aspirations. (Further information on this point is contained in Chapter Four.)

Food and Drug Administration. An approach that is very different from that of NIH is followed by the Food and Drug Administration (FDA), even though it is also a compo

"Testimony of Charles MacKay, Deputy Director, OPRR, transcript of 14th meeting of the President's Commission (November 14, 1981) at 320. See also, "Response to Request for Comments and Corrections of Biennial Report" (undated, distributed by Dr. McCarthy to members of the Commission and Staff on November 14, 1981) part II at 7.

10 Testimony of Charles McCarthy, transcript of 14th meeting of the President's Commission (November 14, 1981) at 321.

11 Testimony of Dr. McCarthy and Dr. MacKay, 14th meeting of the President's Commission (November 14, 1981) at 333-4.

12 Letter to Commissioners from Dr. Charles McCarthy (December 28, 1981).