The PHS Act requires that DHHS issue regulations for the protection of human subjects in research and to implement a program of instruction and guidance in ethical issues related to research. The key provisions of Subpart A of these regulations are as follows:

• "Research" is defined as a systematic investigation designed to develop or contribute to generalizable knowledge; "Human Subject" is defined as a living individual about whom the investigator conducting the research obtains either data through intervention or interaction with the individual, or identifiable private information.
• Each institution engaged in the conduct of Federally-supported research involving human subjects must provide a written assurance that it will comply with the requirements of the regulations. This assurance document must be approved by the Agency Head or designee, and must contain the following information at a minimum:

• A statement of principles governing how the institution will protect human subjects in research projects conducted or sponsored by that institution.
• Designation of one or more IRBs meeting the requirements of the regulation, and for which sufficient meeting space and support staff are provided to support the required review and record-keeping functions.
• A listing of IRB members by name, degree, representative capacity, experience, and institutional affiliation. Changes in IRB membership are to be reported to OPRR.
• Written procedures for initial and continuing review; for determining when a review more frequently than yearly is required; and for assuring prompt reporting to the IRB of protocol changes by the investigator, and approval of those changes before implementation.
• Written procedures for reporting any unanticipated problems with human subjects protections, and serious or continuing non-compliance with the policy or IRB decisions.
• The IRB must review and approve all covered research and must require documentation of informed consent where appropriate.
• The IRB must conduct continuing review of protocols at intervals appropriate to the degree of risk but no less than once a year.
• The IRB may use expedited review procedures (review by the Chair or designee(s)) to approve items eligible for such review or minor changes to previously approved protocols.
• In order to approve a research project, the IRB must be satisfied that:
• Risks to the subjects are minimized.
• Risks are reasonable in relation to the anticipated benefits.
• Selection of the subjects is equitable.
• Informed consent will be sought from the individual subject or the legal representative.
• Informed consent will be appropriately documented.
• When appropriate, the research plan makes adequate provision for monitoring the data collected to ensure the safety of subjects.
• When appropriate, there are adequate provisions to protect the privacy of subjects and to maintain the confidentiality of data.
• Special protections are in place for populations that are vulnerable, e.g., children, prisoners, the mentally disabled, pregnant women, etc.

Additional protections apply to DHHS-supported programs pertaining to research, development, and related activities involving fetuses, pregnant women, and human in vitro fertilization (Subpart B); biomedical or behavioral research involving prisoners (Subpart C); and children (Subpart D).

As noted above, FDA has published its own regulations for IRBs and Informed Consent for evaluation of FDA-regulated test articles. In general, these regulations are consistent with the Common Rule but differences do exist beyond those dictated by the different missions of the agencies. Some of the significant differences are:

In order to put potential issues and solutions into the appropriate context, workgroup members identified the intended level of protections, i.e., the objectives that must be achieved if a program of human subjects protection is to be considered effective. It is the members’ view that the protections that should achieved by the federal regulations and the objectives of a credible program of human subjects protections are as follows:

• At its core, a program must embody the basic ethical principles of the Belmont Report – respect for persons, i.e., that individuals should be treated as autonomous individuals and that persons with diminished capacity are entitled to additional protection; beneficence, i.e., do no harm, and maximize benefits and minimize possible harms; and justice, i.e., fairness in who is selected to participate in, and bear the burdens of, research.
• The program must acknowledge that the central purpose of an IRB is to review research through a group process with balanced expertise to ensure adequate protection of the rights and welfare of the subjects, and to advise the scientist on the ethical conduct of research.
• The program must ensure that research participants understand that, while risks have been minimized, research is not risk-free. As such, the program must ensure that these participants receive sufficient information upon which to make an informed decision regarding their participation.
• For the program to operate efficiently and effectively, it should operate locally with sufficient flexibility to foster creative strategies to protect human subjects. To that end, Federal agencies have a responsibility to operate in the spirit of good will to assure that the regulations and policies are clear and flexible, consistent across the multiple agencies, and consistently applied and interpreted both within and across agencies.
• The program must operate on the basis of trust and on the assumption that researchers and IRB members are ethical and guided by their professional codes of conduct.
• The program of human subjects protection must have institutional support. The IRB’s primary advocacy is for the research participants, not the institution or its investigators. Secondary protections are afforded to the institution and its investigators.
• The program must have a comprehensive educational component reaching all who are involved – IRB chairs, members, administrators, investigators and research staff, institutional officials, students, and the community of research subjects.

In examining existing legislation, regulations, policies and practices, the members agreed that a credible policy of human subjects protection had to achieve these objectives. Any changes proposed to reduce regulatory burden should be consistent with, and foster, the achievement of these objectives.

Following the development of the objectives that a human subjects protection program must achieve, the members identified specific issues that were of concern. The following list includes issues related to the legislation, the regulations and the inconsistency between the agencies on some aspects of the regulations, the interpretation of the regulations, and policies and practices of the agencies. It should be noted that the purpose of the workgroup was not necessarily to come to consensus. While there was general agreement on the issues and the proposed solutions, there was not necessarily unanimity on each.

As noted above, in developing solutions, the workgroup emphasized the need to assure, at the outset, that these solutions must not compromise the objectives identified earlier as the intended levels human subjects protection. Many of these solutions are not new and have been recommended by others in various reviews. Recommendations from the members are as follows:

1. Take actions to assure that IRBs receive useful off-site adverse event data. Several recommendations were offered by members to assure that IRBs receive useful data on off-site adverse events, upon which an informed judgment and reasoned action can be taken. Members noted that the recommendations proposed below will require coordinated action among institutions and their IRBs, FDA, NIH, the NIH-supported cancer and AIDS cooperative groups, and the DSMBs.
1. Clarify the criteria for off-site reporting of adverse events. Members suggested that the definition be events that are "both serious and unexpected" and related to the research, the definition provided by the FDA in 21 CFR 312.32(c)(1)(i). As noted above, the reporting criteria may be clouded by the different definitions for reporting events to IRBs that are contained in 45 CFR 46.103(b)(5)(i) and 21 CFR 56.108(b), and in the NIH cooperative group policies, so that consistent, explicit clarification is needed.
2. FDA should require the establishment of DSMBs and NIH should ensure adherence to its policy. NIH should ensure adherence to its June 10, 1998 policy requiring the establishment of a DSMB for multi-site clinical trials that entail a potential risk to participants. In addition, members recommend that FDA provide guidance to require sponsors to establish a DSMB for each IND application and IDE and develop methods to communicate the DSMB’s deliberations to the IRBs.
3. The DSMBs should provide aggregated and interpreted data to the IRBs at regular and defined times. Individual off-site adverse event reports should not be provided to IRBs. Instead, data from the off-site adverse events should be aggregated and interpreted by the DSMB or the FDA and provided to the IRB. These reports should be provided at regular and defined times. On-site reporting of adverse events should continue to be provided to the local IRB but be presented in a format that allows the individual report to be considered in the context of other previously reported events.
4. Should the policy continue to be the submission of individual off-site reports, expedited review procedures or staff review should be permitted.
2. NIH and FDA should both reconcile conflicting regulations and policies and better coordinate their programs for human subjects protections. As noted above, while FDA and NIH have achieved significant commonality in their regulations and policies, differences do exist in several areas beyond those resulting from their different missions. Both agencies should commit to a process to reconcile these differences.

In addition, members recommend that each agency develop a mechanism to better coordinate and internally manage the various components of these complex programs within their own agency. During the workgroup meetings, NIH cooperative groups, Contract Research Organizations, and agency staff were cited as providing conflicting interpretations of regulations. Additional training by NIH and FDA was suggested as an approach although it would appear that more proactive management and coordination is also necessary to avoid the inefficiencies that result from reported disconnects. In addition, an external advisory board might be useful in this regard, and it could be useful also as a source of advice and counsel on a broader set of policy issues.

3. Change regulations and/or policies on Continuing Review to provide flexibility to IRBs. The members made various proposals to allow the IRB more flexibility to set the timing and the method of continuing review of active projects. In suggesting this flexibility, the members contend that the IRBs will be better able to accommodate to local and unique situations, better able to respond to the level of risk of the research protocol, and better able to allocate scarce resources. The intent is to increase the effectiveness of this key component of the IRB function.
1. Timing of review. The various suggestions made by members provide a menu of possibilities for providing this additional flexibility and are listed below in the descending order of the degree to which they would provide additional flexibility to the IRBs. They would, however, require changes to existing regulations and policies.
1. Allow the IRBs the authority to set the period of approval (e.g., based on initial IRB review, the grant award date, the date at which research subjects begin being entered into the protocol, etc.) as well as the review date. This would allow the IRB the maximum flexibility to assign the start date and to determine when to conduct the continuing review. It would provide maximum flexibility to the IRB to use its judgment depending on individual circumstances and the level of associated risk.
2. Eliminate mandatory annual review and allow the IRB to determine the appropriate time to undertake review based on the degree of risk that the study protocol presents to the research participants. This would allow the IRB the discretion to set the review date, allowing it to match the review date with its assessment of the level of risk.
3. Allow a 30-day grace period beyond the anniversary date for the conduct of the review. This would retain the start date and anniversary date requirements currently in effect, but would recognize scheduling or other unanticipated difficulties that may preclude a review on or before the anniversary date, and would avoid the risk of an interruption to the protocol.
2. Method of review. Members made various suggestions to increase the IRB’s flexibility to determine the method of continuing review. These proposals require a change in regulation or policy. Again, as is the case with suggestions on timing, these are intended to assure satisfactory review while allowing a more efficient allocation of IRB resources.
1. Use expedited review for studies where there have been no changes to the protocol and no additional risks have been identified, or a study in which no patients have been enrolled. This recommendation is partially accommodated by the subsequent publication of new guidelines on expedited review that occurred after the workgroup’s meeting date. This new listing allows expedited review of research where no subjects have been enrolled and no additional risks were identified. The members acknowledged the considerable assistance to IRBs that this expanded set of expedited review categories will provide.
2. Have the IRB determine the method of continuing review (full board or expedited review) at the time of initial review. This recommendation would be followed as long as no additional risks are identified and there are no significant changes to the protocol. Under the new expedited review policies cited above, expedited procedures may be used for continuing review when the research involves no greater than minimal risk and no additional risks have been identified.
3. Eliminate the requirement for annual review and replace it with a system of random audits. While this would result in fewer reviews, it could allow more intensive reviews and would detect problematic issues that may be systemic.
4. Change other policies to provide flexibility to IRBs. Members specifically cited changes to two other policies to provide additional flexibility to the IRBs, so as to allocate resources better and to respond to unique situations.

1. Allow IRBs to determine the timing of initial review. As discussed above, initial review by the IRB must occur before scientific review and, therefore, before funding decisions are made by NIH. Members contend that the workload demands are such that the IRB’s time not be wasted reviewing applications that ultimately will not be funded. Estimates vary on the workload that would be saved since some grant applications, if not funded by NIH, will be funded by other sources. The evaluation of IRBs by James Bell Associates reported that 85% of grants submitted to NIH are eventually funded by some source. However, while there was some skepticism expressed by members about the 85% figure, a savings of 15% on initial review for an overworked IRB is still substantial. Furthermore, it should be noted that grant applications initially rejected by NIH for funding may be re-written and eventually funded but this will entail still another IRB review. This proposal would give IRBs the option to conduct the review when they thought best – either at the time of submission or any time thereafter, but before the grant is awarded. This, however, will require that NIH notify the institution if a proposal has received a favorable review so that it can plan an IRB review.

Members acknowledged the argument that IRBs that choose to perform initial review of the adequacy of human subjects protections after scientific review, potentially face increased pressure since the applications have already been approved based on scientific merit and IRB approval is the last remaining step prior to receipt of funding. However, members were confident in the IRB’s ability to give first priority to the protection of human subjects, and that IRBs will only choose this option if they conclude that this more efficient process would ultimately result in better review. They also point out that non-NIH funded grants receive IRB approval after funding decisions are made with no apparent compromise to the human subjects protection program.

2. Allow IRBs to utilize audio conference calls for full board reviews. Members suggested that OPRR change its policy to be consistent with FDA, and allow IRBs the option to utilize audio teleconferencing in the conduct of full board reviews. The intent is that it be an option available in limited circumstances.
5. Reengineer the assurance process. The members concluded that substantial rethinking of the purpose of the assurance process and complete redesign was in order. While it may have served an educational purpose in earlier days, it is not serving that purpose now and there is little "value added" by the size and content of the document and the length of the negotiations.
1. Simplify the document. Members generally favored a vastly simplified document, possibly one or two pages, which provides the basic information required by OPRR and which would be revised only as necessary. This new document would be the same regardless of the number of protocols performed by the institution.
2. Eliminate multiple documents. Members would eliminate the various separate inter-institutional agreements that currently require OPRR approval. Members observed that these various agreements and the inherent time lags because of the need to secure OPRR approval slow, rather than facilitate, inter-institutional alliances. Therefore, members proposed that such alliances be administered at the institutional level; if OPRR has a need to be informed of such arrangements, an informational report could be provided.

On a number of occasions, and at the meetings of these workgroups, OPRR has expressed its intent to streamline the assurance process but couple it with an enhanced education program. The members generally were very supportive of this approach.

6. Develop an expanded and comprehensive training program. Members concurred with the proposal by OPRR that there needs to be an expanded program of education.
1. Education program. Members saw the scope of this training program to be very broad. It should address the needs of all participants including IRB chairs, members, administrators, investigators and research staff, students, institutional officials and the community of research subjects. Specific suggestions were for NIH/FDA to sponsor educational conferences, to expand the current NIH-funded ethics training initiative into a broader program, to develop and implement educational strategies in association with professional organizations such as thePublic Responsibility in Medicine and Research (PRIM&R) , and for NIH/FDA to centrally design educational materials and utilize technologies to make these materials available to those at the local level, including research subjects.
2. Technical assistance program. Closely related to this educational focus was the suggestion by members for OPRR to expand its technical assistance efforts to institutions. Of particular interest to the members is a significant role in identifying and disseminating "best practices" so that IRBs may draw on the successful experiences of others. This would also include technical assistance on interpretations or applications of the regulations that facilitate new approaches to achieving regulatory compliance by removing overly restrictive and/or unnecessarily complex procedures. Additional suggestions included its becoming a central data source for national data relevant to IRBs, and sponsoring regional workshops around specific issues.

The net effect of the proposals recommended by members would significantly shift the focus and role of OPRR. Its role as a national focal point for leadership in education and technical assistance would be significantly expanded.

7. Encourage additional resources for the IRBs. As noted above, members generally observed that resources available to the IRB are not expanding relative to the increased workload. Suggestions made by the members were:
1. Provide additional Federal resources if mandating additional duties. While the mechanism for providing these resources was not defined, members observed that the Federal government should bear in mind that the role of the IRB is to ensure adequate protection of the rights and welfare of research subjects through the review of research protocols, assessment of informed consent, ongoing review and other activities. Assuming that any new requirement falls within this charter, additional resources should be provided, e.g., assuring the inclusion of women and minorities and the inclusion of children in research.
2. Amend FDA regulations to include a provision similar to the requirement in the DHHS regulations for adequate staffing and space resources. 45 CFR 46.103(b)(2) recognizes the need for sufficient staff and space to carry out IRB functions; FDA regulations carry no such requirement. Congruence between the two agencies on this issue, at a minimum, would serve to highlight its importance and necessity.
8. Pursue the development of longer-term paradigm shifts if they have the potential to reduce regulatory burden. The current assurance-based system is considered sufficiently onerous by some members that alternatives to the existing paradigm such as third-party accreditation of IRBs might be explored. While it merits long-term exploration, with a thorough evaluation of the pros and cons, it is considered a viable alternative only if it results in a net reduction in burden. Members see little advantage if it becomes an additional regulatory tool without a commensurate reduction in the current oversight activities. Some members observed that they would not proceed if the end result is similar to the outcome experienced by the Animal Care and Use community. The OPRR and the Department of Agriculture have not lessened their oversight of organizations that are accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC). Obviously, the other requirement is that such a system must provide at least an equal or better program of protections and should not be implemented unless this is assured.

Similarly other results-based techniques such as performance measures were discussed as possible alternatives. Again, it was stressed that while there is merit in pursuing alternatives to the current process-based program, their development would require a considerable long-term effort and should only be considered if it will result in a net reduction in regulatory burden.