Testimony
Before the House Committee on Government Reform,
Subcommittee on Criminal Justice, Drug Policy and Human Resources

Protecting Human Subjects: Status of Recommendations

Statement of
George Grob
Deputy Inspector General for
Evaluation and Inspections

May 3, 2000

Office of Inspector General
U.S. Department of Health and Human Services

Good afternoon. I am George Grob, Deputy Inspector General for Evaluation and Inspections, in the Office of Inspector General (OIG), U.S. Department of Health and Human Services. I am pleased to testify at today's hearing on the Department's responses to our June 1998 report on the Institutional Review Board system for protecting human subjects of medical research.

Mr. Chairman, the Department has taken a number of promising steps, but few of our recommended reforms have been enacted.

Our June 1998 Report on Institutional Review Boards

Background. In June 1998, we released a series of reports on the Federal system for human-subject protections centers on institutional review boards (IRBs). This broad-based review was initiated after a 1995 OIG study, requested by the Food and Drug Administration (FDA), raised significant concerns about the adequacy of subject protections and the IRB system. The report focused on medical device research and found, for example, that in one instance that an investigator had implanted 264 investigational devices when he only had approval for 75 implants. We found another investigator had not gotten informed consent from the subjects and yet other instances in which changes in the informed consent documents that IRBs had requested were not incorporated. Another investigator moved from the city of practice without informing the subjects of his clinical trial who to report to after he was gone.

Findings. In the broad 1998 report, we warned that the effectiveness of IRBs was in jeopardy. We found that the clinical research environment had changed in the 20 years since the system was first established and the changes have had significant implications for IRBs. IRBs were becoming overwhelmed by their workloads, lacking necessary resources to keep up and becoming pressured to do more in a shorter time frame and with limited information on many trials. They provided little training to investigators and Board members regarding principles and practices of human subject protection. Of particular significance, they conducted very little oversight of clinical trials once the trials had started. In addition, Federal oversight of protections was limited, leaving Department with little sense of how well IRBs were actually doing their job.

These findings led us to present numerous recommendations to the National Institutes of Health (NIH), its Office for Protection from Research Risks (OPRR), and the Food and Drug Administration (FDA). Brief summaries of both our findings and recommendations from the June 1998 report are attached to this testimony.

A Follow-up Report

Based on the continued interest in human-subject protections and a request from this subcommittee, we recently issued a follow-up report providing an accounting of how fully our 1998 recommendations have been enacted. We drew on information obtained over the past two years from Department officials and pertinent documents, data reported to us by the NIH and FDA, and interviews with Department officials who also provided comments on a draft of this report, many of which are reflected in the final. Let me state that this report is not a further examination of the adequacy of the Federal oversight of human subject protections or of the protections themselves. Instead, we used as a starting point our prior recommendations and provide an accounting of how fully they have been carried out. Attached at the end of this testimony is a table compiling each recommendation and a brief description of what, if any, action has taken place.

The Department has taken action and initiated several promising steps.

Increased Enforcement. Since June 1998, both OPRR and FDA have significantly stepped up their on-site presence at research institutions. Between April 1997 and May 1998, OPRR had conducted an on-site investigation at only 1 institution. Between June 1998 and March 2000, it conducted on-site investigations at 10 institutions. Since June 1998, OPRR also conducted off-site investigations (document reviews) at more than 140 additional institutions. It found performance problems at a number of institutions it investigated and has required 7 of them to suspend some or all of their federally funded research. Where OPRR found weaknesses in the institutions' systems of protections, it cited institutions for substantive, broad-based deficiencies--ones that have direct consequences on the rights and safety of research subjects.

FDA's number of routine on-site investigations of IRBs increased from 213 in Fiscal Year 1997, to 253 in FY 1998, and 336 in FY 1999. The Center for Devices and Radiological Health, for example, issued eight warning letters to IRBs during FY 1999, compared to zero in FY 1998. The Center for Drug Evaluation and Research took administrative action against eight IRBs during FY 1998 and 1999.

Sentinel Effect. OPRR's oversight activities, in particular, have drawn the attention of the research community to issues of human-subject protections. The reviews at prominent medical centers, including the temporary suspension of federally funded research at 7 centers, have had a ripple effect beyond the individual institutions visited by OPRR. Many major medical journals and newspapers have given prominent attention to OPRR's enforcement actions. The adequacy of IRB oversight has been a topic at a number of national conferences and association meetings.

New Organization. In July 1999, the Secretary announced the relocation of OPRR from NIH to the Office of the Secretary and plans to establish a new advisory committee on protection from research risks to provide scientific and ethical guidance to the office. These actions, although not yet complete, have been widely publicized and are being taken to strengthen the stature and effectiveness of OPRR in its oversight role and are indicative of the Secretary's commitment to strengthening subject protections.

Other Initiatives. The Department has enacted two of our recommendations--as of October 1998, FDA now informs sponsors and IRBs associated with an investigator when FDA finds evidence of misconduct on the part of clinical investigator; and in June 1999, NIH issued a policy stating that all data and safety monitoring boards associated with NIH trials are expected to forward summary reports of adverse events to IRBs. Both actions are significant. For example, the summary information is key to an IRB's ability to ensure the continued safety of subjects. Both agencies have also increased their education outreach and resources. They have ongoing deliberations on additional proposed changes.

But overall, few of our recommended reforms have been enacted.

Despite these positive steps, there are still need for additional reforms. What follows is a brief description of any the Department actions taken in response to our six broad recommendations. A table attached to this testimony provides a quick snapshot on the status of the Department responses.

Flexibility and Accountability

We called for the Department to lessen specific procedural requirements that add questionable of value and to require that IRBs undergo regular performance-focused evaluations. Beyond one concrete action discussed below, there have been few enacted reforms along the lines we recommended. Little has been done to grant IRBs more flexibility or to establish a Federal basis for assessing the effectiveness of IRBs.

In November 1998, FDA and NIH/OPRR jointly issued new regulation expanding the categories of research that may be reviewed by IRBs through an expedited review procedure. There are also active deliberations by an NIH Advisory Group geared towards reducing regulatory burden and streamlining processes for grantee institutions. Among its recommendations, the group proposed a change, about to be implemented, to the requirement that IRBs review all protocols before funding decisions are made, an improvement we suggested.

The most notable development around IRB performance evaluations is in the private sector. A private group devoted to the ethical conduct of research, is working to develop performance standards and apply them as part of an accreditation process. Both NIH/OPRR and FDA are participating in these discussions.

Oversight of Ongoing Research

We made several recommendations to strengthen continuing protections of clinical research subjects after the research has begun. For example, we called for more explicit policies on when and how data safety and monitoring boards could be used in certain research trials and a suggested a requirement that they share summary information with IRBs. We also called for FDA to inform IRBs when it takes action against an investigator under the IRB's purview and to establish a requirement that sponsors and investigators notify IRBs of any prior IRB review of research. Finally, we called for increased IRB awareness of on-site research practices.

As stated earlier, FDA now informs sponsors and IRBs about investigator misconduct and NIH issued a policy requiring its data and safety monitoring boards (DSMBs) to forward summary information to IRBs. But, neither NIH nor FDA has issued requirements for sponsors to notify IRBs of prior reviews, nor have they issued guidance to IRBs to increase their attention to on-site research practices. FDA has not set forth regulations regarding the appropriate use for or composition of DSMBs, which can help provide and assemble valuable information for IRBs.

FDA requires all sponsors of gene transfer research, and sponsors overseen by the Center for Devices and Radiological Health, to routinely submit monitoring plans. But both of these policies do not address a large majority of clinical trials. We urge FDA and NIH take similar initiatives directed at other research areas.

Education and Training

We called for the Department to require institutions receiving Public Health Service Act funds establish an education program for investigators and that IRBs be required to educate their members. No Federal regulations have been enacted requiring institutions to establish education programs for clinical investigators. Similarly, no Federal requirements have been enacted calling for education for IRB members.

NIH intramural researchers are required to complete a web-based tutorial in order to conduct human-subjects research on campus. We urge NIH to consider expanding this policy to all extramural researchers, who conduct the majority of research funded by NIH. Both NIH and FDA have increased their educational outreach through numerous training presentations and seminars for IRBs and professional groups. NIH has constructed a website containing bioethics resources and has launched new training grants in subject protection issues and bioethics. FDA is in the process of updating its Information Sheets, an important source of IRB guidance. NIH/OPRR hired a full-time educational staff person in January 1999, and is in the process of further expanding this staff.

Conflicts of Interest

We called for a number of steps to mitigate the potential influence of conflicts on IRB reviews such as a requirement increasing the number of nonscientific and noninstitutional members on IRBs and a prohibition on equity owners of independent IRBs from participating in the IRB review process. We did not identify any significant action in the Department to mitigate potential IRB conflicts of interest. We continue to regard this as a significant area warranting attention. In the increasingly commercialized research environment, potential for conflicts within research institutions loom larger than ever and it is ever more important that IRB reviews be sufficiently independent, both in reality and in appearance.

Workload Pressures

We called for OPRR to hold institutions accountable for the resource commitments they made in their assurances and for FDA to modify its site visit protocol to more readily identify situations in which limited resources may jeopardize subject protections. OPRR's enforcement efforts have brought attention to IRB resource shortages at individual institutions. However, no further action has been taken to develop indicators of adequate resource levels or to enable greater investments to support IRB functions.

One approach that NIH reports is under consideration, and is worthy of attention, would be to allow institutions to allocate an additional increment of grant funds to provide necessary resources for IRBs. A well-supported IRB should be considered a necessary cost of doing business.

Reengineering Federal Oversight

We called for NIH to revamp its oversight and assurance process and for FDA to revamp its on-site inspection process. We also called for the Department to require IRBs to register before being allowed to review research under the Department jurisdiction.

OPRR's and FDA's response in increasing their enforcement efforts is significant, as we have already indicated. By increasing their presence in research institutions, they have fostered compliance with Federal regulations there-- and most likely at others as well. But neither body has yet enacted any significant revamping of their oversight processes as we have called for. This is unfortunate because stepped up enforcement without a more efficient, performance-oriented enforcement process will still leave us with an oversight system that falls well short of its potential.

FDA has substantially increased its IRB inspections, but it has not engineered any significant changes in its approach to these inspections. The inspections remain narrow and focused on procedural compliance, not results. NIH/OPRR reports that it has been developing plans to streamline the assurance process as we called for in our June 1998 report. But no actual change has taken place to date.

It is important to note that FDA has formed a working group to establish an IRB registration system. The working group has agreed upon specifications for this registry that reflect many of our suggestions. However, nothing has been implemented yet. We encourage the group to follow through with this effort.

The Common Rule: A Significant Barrier to Departmental Progress

The HHS core regulations concerning IRBs and human-subject protections are the basis of a common Federal policy on human-subject protections. The Federal policy, known as the Common Rule, is adhered to by the Department and 16 other Federal agencies.

Because any changes to the Rule call for the concurrence of all 17 Federal agencies, we must acknowledge that the reality of gaining concurrence among 17 Federal agencies inhibits a timely and effective the Department response to a number of our recommendations.

Several of our recommendations can be carried out through administrative changes, for example requiring education through contract and grant language or altering the assurance or inspection processes. However, other recommended changes, are subsumed in the Common Rule. A requirement, for instance, for more extensive representation on IRBs of nonscientific and noninstitutional members or stronger requirements on IRBs having sufficient independence, could be difficult to carry out without the agreement of the other 16 agencies.

The intention of having a common Federal policy on human subject protections is an important one. One set of standards fosters a level of consistency in protections of human subjects in many different areas of research and makes complying with Federal regulations easier for researchers and sponsors. However, with the clinical research environment changing rapidly, we believe it is essential for the Federal policy and regulatory actions to keep pace. Therefore, legislative change may be necessary to achieve a timely implementation of many of our recommendations.

The Need for Action on a Broad Front

Our June 1998 inquiry and continuing work in this area convinces us that IRBs alone cannot do the job; other parties in the clinical research process, including sponsors and investigators, must take responsibility for subject protections. That is why, for example, our recommendations include actions that call for investigator education. Investigators, while they are in a position to do the most harm to patients, are also in a position to do the most good. More explicit Federal guidelines on recruiting subjects and the use and composition of data safety and monitoring boards can also help. The National Bioethics Advisory Commission is also likely to offer further guidance in the near future on the kind of changes that need to be made in the Federal regulatory system.

New Opportunity for Leadership

The Department has a significant new opportunity to exert Federal leadership in protecting human subjects with the new office in the Office of the Secretary and a new advisory committee on subject protection issues. We urge that the new office give significant attention to our prior recommendations. We also urge that it continue the important enforcement efforts undertaken by OPRR over the past 2 years. The NIH/OPRR efforts have served as a reminder to research institutions, sponsors, individual investigators, and IRBs that the reviews must still be substantive in order to ensure adequate protections for human subjects. In the important quest to reduce regulatory burdens, it is important not to lose sight of this underlying protection function.

Both FDA and NIH will retain significant statutory and operational responsibilities for protecting human subjects despite the impending establishment of a Departmental office for subject protections. FDA still has the most visible on-site presence and thus is in the best position to identify shortcomings and opportunities for improvement. NIH retains significant roles in ensuring human-subject protections as a major sponsor of clinical research and as a conduit to the research community.

Conclusion

We continue to support the recommendations we made in our earlier report and call for a greater sense of urgency in carrying them out. They offer actions that would strengthen human-subject protections without impeding vital clinical research. They reflect a respect for the largely collegial manner in which IRBs operate. Yet, they also recognize that verification and accountability must also be important features of a system intended to protect human subjects.

Our recommendations are not a complete blueprint for action. In the months ahead, we will be conducting further inquiry that more closely examines how the Department oversight can enhance human-subject protections.

Thank you for the opportunity to testify on this most important topic. At this time, I would be happy to answer any questions which you or the other members of the Subcommittee may have.

FINDINGS

Institutional Review Boards: A Time for Reform
(OEI-01-97-00193), June 1998

The Effectiveness of IRBs Is in Jeopardy.

They Face Major Changes in the Research Environment. The current framework of IRB practices was shaped in the 1970s in an environment where research typically was carried out by a single investigator working under government funding with a small cohort of human subjects in a university teaching hospital. In recent years, that environment has been changing dramatically as a result of the expansion of managed care, the increased commercialization of research, the proliferation of multi-site trials, new types of research, the increased number of research proposals, and the rise of patient consumerism. Each of these developments has presented major disruptions and challenges for IRBs. "Never before," concluded one recent review, "has such a pressure-cooker atmosphere prevailed within the IRB system."

They Review Too Much, Too Quickly, with Too Little Expertise. This is especially apparent in many of the larger institutions. Expanded workloads, resource constraints, and extensive Federal mandates contribute to a rushed atmosphere where sufficient deliberation often is not possible. At the same time, the IRBs frequently are hard-pressed to gain access to the scientific expertise they need to reach informed judgments about the research taking place under their jurisdiction.

They Conduct Minimal Continuing Review of Approved Research. In the environment described above, continuing review often loses out. Even where there is the will, there often is not the time to go beyond the perfunctory obligations. A lack of feedback from other entities that oversee multi-site trials contributes to the problem. The result is that IRBs have all too little information about how the informed consent process really works and about how well the interests of subjects are being protected during the course of research.

They Face Conflicts That Threaten Their Independence. Clinical research provides revenue and prestige to the institutions to which many IRBs belong. The institutions expect IRBs to support these interests at the same time that they protect human subjects. The resulting tension can lessen the IRBs' focus on their basic mission. The minimal "outside" representation that typically exists on IRBs deprives them of an important counterbalance to the institutional interests. For independent IRBs, the dependence on revenue from industry sponsors exerts similar possibilities for conflict.

They Provide Little Training for Investigators and Board Members. The IRB system depends heavily on research investigators' commitment to uphold human-subject protections. But as that system now operates, it offers little educational outreach to investigators to help them become informed and sensitized about these protections. Similarly, it provides minimal orientation and continuing education for IRB members--a deficiency that is especially detrimental to nonscientific and noninstitutional members.

Neither IRBs Nor The Department Devote Much Attention to Evaluating IRB Effectiveness. IRBs rarely conduct inquiries to determine how well they are accomplishing their mission; their judgments of effectiveness rely mainly on the number of protection lapses or complaints that are brought to their attention. The Department agencies conducting oversight seldom go any further. The Office for Protection from Research Risks, in the National Institutes of Health, focuses almost entirely on up-front assurances. The Food and Drug Administration relies on compliance-focused inspections.

RECOMMENDATIONS

Institutional Review Boards: A Time for Reform
(OEI-01-97-00193), June 1998

1. Recast Federal Requirements So That They Grant IRBs Greater Flexibility and Hold Them More Accountable for Results.

• Eliminate or lessen specific procedural requirements that are of questionable value. Our aim was to provide overburdened IRBs with greater discretion that would enable them to develop more innovative and strategic approaches to their reviews. There are requirements, for example, that limit what IRBs can accomplish in conducting protocol reviews outside of convened board meetings. In addition, we highlighted requirements that call for IRBs to conduct full, annual reviews of approved protocols and that call for complete reviews of Federal funding applications prior to funding decisions.
  
• Require that IRBs undergo regular performance-focused evaluations that are carried out in accordance with Federal guidelines. In our review, we were struck by how little attention Federal oversight bodies and IRBs themselves gave to evaluating how successful IRBs were in protecting human subjects. It is time, we concluded, for the Federal government to mandate self-evaluations or, better yet, evaluations conducted by independent, outside parties. We also urged that the results of such evaluations be made public.

2. Strengthen Continuing Protections for Human Subjects Participating in Research.

• Require Data Safety Monitoring Boards (DSMBs) for certain high-risk and multi-site trials. DSMBs are independent assessment bodies that provide medical, scientific and other expertise that typically is not available on IRBs, thereby serving an invaluable function in protecting human subjects. We recommended that NIH and FDA take the lead in seeing that DSMBs become more firmly established as oversight mechanisms and become more clearly accountable.
  
• Require DSMBs to provide summary information to IRBs. We urged that DSMBs provide their summary assessments of adverse event reports to IRBs. IRBs are swamped with individual adverse event reports from multi-site trials, but these reports lack the essential context to confer meaning about the relative safety of the trial. DSMBs can provide this context and thereby enhance the IRB's capacity to assess ongoing safety.
  
• Alert IRBs to corrective actions taken against investigators under the board's purview. Although FDA provides information on its website about corrective actions that result from investigator inspections, the Agency does not routinely inform the respective IRBs of such actions. We recommended that FDA inform individual IRBs when it takes corrective action against an investigator who is conducting research reviewed by the IRB.
  
• Require sponsors and investigators to notify IRBs of any prior IRB review of a research plan. Sometimes sponsors shop around for an IRB that will give their protocol a favorable review. We pointed out that such action can undermine the IRB review process and, accordingly, urged that this requirement be enacted.

• Call for increased IRB awareness of on-site research practices involving human subjects. IRBs are rarely aware of what actually takes place between investigator and subject. We called for IRBs to move beyond their focus on the informed consent document and periodically check for themselves how the actual consent process is working. For particularly sensitive or risky projects, we suggested they might call for the participation of counselors, ombudsmen, or other third parties that could be available to make certain that the consent process functions in the interest of human subjects.

3. Enact Federal Requirements That Help Ensure That Investigators and IRB Members Are Adequately Educated About and Sensitized to Human-Subject Protections.

• Require institutions to establish an education program for investigators in human-subject protections. Such a requirement exists for research involving animal subjects. We found the case for education requirements no less compelling for research involving humans. The mandatory education we called for could be provided through media such as seminars, individual instruction, videos, or on-line tutorials.
  
• Require investigators receiving funding under the Public Health Service Act for research involving human subjects to provide a written attestation indicating that they will uphold Federal policies concerning human-subject protections. We recommended such an attestation as a way of heightening investigators' awareness of their responsibilities as investigators and interest in participating in educational programs addressing human-subject protections.
  
• Require IRBs to educate their members about human-subject protections. In order for IRBs to adequately review research protocols to ensure human-subject protections, each board member must be educated in both applicable Federal regulations and ethical principles. We called for a specific mandate that IRBs and their parent institutions provide initial and continuing education.

4. Help Insulate IRBs from Conflicts That Can Compromise Their Mission in Protecting Human Subjects.

• Require more extensive representation on IRBs of nonscientific and noninstitutional members. At present just one IRB member can wear both of these hats and satisfy the requirement. We found that to be an untenable situation, one that can deprive IRBs of a valuable counterbalance to internal, institutional pressures that can threaten their independence.
  
• Reinforce to IRBs and their parent institutions the importance of IRBs maintaining sufficient independence. It is particularly important that IRBs not report to a part of an institution responsible for bringing in research funds.
  
• Prohibit equity owners from participating in the IRB review process. Such a practice does not necessarily inhibit the independence of the review process, but it establishes a situation that can undermine a perception of impartiality. We recommended that it should be disallowed.

5. Recognize the Seriousness of the Workload Pressures That Many IRBs Face and Take Actions That Aim to Moderate Them.

• Require that IRBs have access to sufficient resources to adequately carry out their duties. Our recommendation was directed not only to staff and board member resources, but also to space, computers, and other essential elements. We urged OPRR to hold institutions accountable for the resource commitments they made in their assurances and for FDA to modify its site visit protocol so that it could more readily identify situations where resource shortages jeopardize an IRB's ability to oversee research.

6. Reengineer the Federal Oversight Process

• Revamp the NIH/OPRR assurance process. NIH/OPRR's oversight process has been concentrated on reviewing up-front assurances aimed at obtaining an institution's commitment to adhere to Federal requirements. We found that assurance process to be paperwork-laden with little effect on IRB functioning. We urged that NIH/OPRR reorient the assurance process so that it rests on an institutional attestation to conform to Federal requirements, and then devote more NIH/OPRR resources to periodic performance-based reviews of institutions and their IRBs.
  
• Revamp the FDA on-site inspection process. We recognized that FDA has a much greater on-site presence than NIH/OPRR, but urged that FDA transform its site visit protocol from a narrow compliance orientation to one that is much more performance-based. Such an approach would pay particular attention to how individuals were actually being approached about participating as human subjects and to how IRBs were making risk-benefit trade-offs.
  
• Require IRBs to register with the Federal government. We found that one of the major impediments to Federal oversight of human-subjects research is that there is no complete registry of the IRBs reviewing this research. Such a registry would be invaluable for FDA and NIH/OPRR as it would allow them to target their oversight and communicate more effectively with IRBs.

Current Status of FDA and NIH/NIH/OPRR Response to Recommendations

Institutional Review Boards: A Time for Reform (OEI-01-97-00193), June 1998

E-Mail List | HIPDB | Hotline | Privacy Notice | Disclaimers

FOIA Information | Contact Us | What's New | Exclusions Database

HHS Homepage | IGnet | FirstGov | Accessibility | Adobe Acrobat

Go to the top of the page