clinics and physicians in private practice to participate in clinical research. Moreover, primarily because of federal clinical testing requirements for new products, clinical trials, especially industry-funded ones, now focus on multi-site studies involving thousands of human subjects. Without doubt, this increase in research conducted at multiple sites has allowed a greater understanding of the benefits and risks of a drug or device before marketing. Now more than ever, IRBs must have sufficient staff, expertise, and technical support to meet the demands of the changing research environment.

II.

RECOMMENDATIONS FOR IMPROVING THE CURRENT SYSTEM

WIRB believes that the present IRB regulations can be effective in providing human subject protection, but because of the dramatic change in the research environment, it is appropriate to pause and reevaluate the process at this time. As the OIG draft report noted, there is no indication that the current regulatory system has resulted in harm to research subjects. However, we believe that even the perception of problems in the system will undermine public confidence and could adversely affect the advances in medical research. Therefore, the recommendations made by the OIG would be helpful in maintaining our country's leadership in the protection of the rights of research subjects.

A. IRB Registration

WIRB supports the concept of IRB registration. At present, NIH/OPRR and FDA lack basic information on the existence, location, and make-up of IRBs, and they must rely on IRB information that is either provided by sponsors or investigators in their applications to FDA or provided by institutions in assurance documents with NIH/OPRR. WIRB believes that requiring such basic information would help to improve protection of human subjects by allowing the agencies to manage their oversight efforts more effectively and to fully communicate with IRBs.

B. Information Sharing

WIRB also strongly supports the OIG's recommendation concerning "information sharing." There have been situations where certain sponsors or investigators were displeased with an IRB's review and switched to a new IRB without informing the new IRB about the previous review. While the majority of sponsors and investigators have demonstrated integrity during the IRB review process, WIRB believes that it is important that these parties be obligated to disclose any prior disapprovals of the research. Further, so that an IRB can effectively monitor the progress of research studies, copies of all FDA and OPRR inspection reports concerning clinical studies which the IRB has approved should be provided to the reviewing IRB. Finally, because clinical investigations are

already subject to government and sponsor auditing and monitoring, we believe that extending the required "information sharing" between the parties would be helpful to the IRB's oversight of research without adding more oversight to an already overworked clinical investigator.

C. Ethical Considerations: Structure of Board

Concerns have been raised about the ability of IRBs to act independently of the sponsors or the institution to which they are affiliated. The issue of independence applies equally to institutional IRBs as well as independent IRBs. We strongly support formal studies that analyze actual conflicts of interests in all IRBs. Without knowing whether conflicts of interest bias IRB decisions, it would be difficult to implement meaningful regulatory change.

It appears that the best way to control conflicts of interest is through organizational structures that eliminate the financial interest of the board members in their decision making. WIRB already practices this through separation of its administrative and review functions.

Moreover, WIRB strongly supports the OIG's recommendation that IRBS should include more non-scientific and non-institutional members. WIRB highly values the important role that these members play in protecting human subjects and understands the OIG's view that present regulations requiring merely one non-scientific member and one non-institutional member may not be adequate.

WIRB also strongly supports the OIG's recommendations for increased board member training. IRB members need initial and continuing education to understand and stay current with the complex scientific, regulatory, and ethical issues in today's research environment. WIRB also supports an increase in investigator training, both inside and outside of institutions. There are few educational resources available for investigators. and this contributes to the problem of investigator non-compliance with regulations. D. Common Policy Shared by FDA and NIH/OPRR

We strongly support the concept of a "shared" policy between FDA and NIH/OPRR in oversight of IRBs. As discussed previously, regulatory mechanisms employed by FDA and NIH/OPRR vary. FDA oversight of IRBS is included in the process of evaluating the safety and efficacy of drugs, devices, and biologics. Its approach is more compliance-based, focusing on inspection of IRB research sites. In contrast, NIH/OPRR oversight of IRBS focuses on assurances.

Each of these systems incorporates valid oversight tools. However, we agree with the OIG that similar oversight policies and close collaboration would strengthen the

-4

protections to human subjects. Moreover, a common policy would result in a more efficient regulatory scheme that would be advantageous to both the federal government and to IRBs. We believe that a shared policy between FDA and NIH/OPRR will improve the ability of IRBs to comply with federal policies, regardless of whether the research protocol is regulated by NIH/OPRR or by FDA. We also support the OIG's

recommendation that NIH/OPRR and FDA involve other departments in the Department of Health and Human Services, as well as non-federal parties such as IRBs, sponsors, and clinical investigators, in development of a shared policy.

Congressional efforts in the 1970s established an excellent system of oversight. The United States continues to be an international leader in both oversight of human subjects participating in clinical trials and biomedical research and development. However, WIRB supports a review that seeks to address changes in the research environment and to provide consistency in oversight of clinical trials, whether funded by the federal government or the private sector. The central issue is to ensure human subject protection through the use of an efficient and consistent oversight policy.

Independent IRBs have responded to the legitimate needs of the medical research community by providing an independently based assurance of human subject protection. As independent IRBs, we play a critical role in protecting rights of research subjects as well as facilitating the development of new medical therapies.

We thank you for the opportunity to present our views and we look forward to continuing to work with Congress, HHS and FDA to protect the rights of human subjects who choose to participate in clinical research.

Mr. SNOWBARGER. Thank you, Dr. Bowen.

Next is Dr. Timothy Walsh from the Institutional Review Board for the New York Psychiatric Institute. Dr. Walsh, welcome to the committee.

Dr. WALSH. Mr. Chairman, and distinguished members of the committee, I'm here with Dr. John Oldham, who is chairman of our Institute. We very much appreciate the opportunity to testify about the critical issue of protection of human subjects in research.

In my remarks, I'd like to highlight a few important issues about the study that have already been raised in the remarks this morning. The research study that we've been asked to discuss was focused on violence and anti-social behavior among youth. The recent deadly shootings at schools across our country and the rising tide of youth suicide are tragic reminders that we must do more to understand and to prevent violence among young people. The purpose of the study conducted at Psychiatric Institute was to learn more about the origins of such problems. The broad goal of the project was to identify psychological, environmental, and biological factors which increase a child's risk of developing anti-social behavior. The hope was that, if successful, this research would lead to targeted interventions to help prevent these difficulties from developing in youngsters at risk.

In its review of this proposal, our IRB carefully considered both the scientific merits of the study and the potential risks and benefits to the subjects, and concluded that approval of the study was consistent with Federal and State regulations, as well as with accepted ethical principles. The study was carried out and to the best of our knowledge, it was concluded without any harm to any participant.

The fenfluramine study was one part of a larger project which involved 126 boys with an older brother who had been adjudicated a juvenile delinquent. Because the research involved numerous meetings with families over several years, the investigators sought potential participants living in proximity to our Institute, which is at the northern end of Manhattan adjacent to the Bronx, and so they obtained information on eligible families from the family courts of those two boroughs. The overwhelming majority of individuals in this system, and, therefore, the overwhelming majority of participants in these studies were from minority ethnic groups. The investigators recognized that such work should be more broadly based and this study was planned as the first phase of a larger research program. Consistent with this plan, the investigators later submitted to the IRB and to NIH a proposal to conduct similar research in a larger and more geographically and ethnically diverse sample.

One major concern which has been raised about this study is its use of fenfluramine. There is strong evidence that the brain chemical, serotonin, plays an important role in the regulation of violent behavior. And by giving participants a single oral dose of fenfluramine, a pill, on one occasion, and then measuring changes in hormone levels in the blood, investigators could obtain an indirect measure of brain serotonin function. The procedure is very analogous to the widely used glucose-tolerance test.

At the time the study was proposed, fenfluramine had been on the market as Pondimin for approximately 20 years for obesity, and was commonly used in research to assess brain serotonin function. The IRB thoroughly reviewed the potential risks, and concluded they were minor at most. Fenfluramine studies were carried out in 36 youngsters and to the best of our knowledge, none experienced significant problems.

Concern about the use of fenfluramine subsequently developed because of the association between its use as Redux and heart valve abnormalities. These concerns first emerged in 1997, well after the conclusion of the Psychiatric Institute study in 1995. The concerns about valve damage are about obese individuals who took fenfluramine for months, almost always in combination with another medication, phentermine, the Fen-Phen combination. FenPhen was not used in the studies at Psychiatric Institute, only a small single dose with a pill of fenfluramine. There is no indication that this use of fenfluramine is associated with any risk of cardiac damage.

In closing, we would emphasize our belief that research on the development and the prevention of violent behavior among young people is critical for our country. Studies in this sensitive area, as in all research, must be carried out with the strictest attention to safeguards for all the participants. We welcome the opportunity to provide information to the committee, and we'll be happy to answer your questions.

[The prepared statement of Dr. Walsh follows:]