© 2004 American Thoracic Society
Ethics, Science, and Oversight of Critical Care ResearchThe Office for Human Research ProtectionsDepartment of Anesthesia and Critical Care, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts Correspondence and requests for reprints should be addressed to Greg Koski, M.D., Ph.D., Department of Anesthesia and Critical Care, Massachusetts General Hospital, 55 Fruit St., Boston, MA 02114. E-mail: gkoski{at}partners.org Last October, at a meeting in Amsterdam that was hosted by the European Society for Critical Care Medicine, a panel of experts from the United States and Europe explored the many challenges that investigators encounter in the intensive care unit and the impact of the rapidly changing regulatory environment posed by the recently adopted European Directive on Clinical Trials (1). Although all human subject research is challenging, that performed in intensive care units is particularly so because the patients used as subjects of investigation, like the newly born and the extremely old or mentally impaired, live on the edge of human existence. In the intensive care unit, research is often conducted at the uncertain boundary between life and death. Too often, everyday care of critically ill patients is not well supported by empirical evidence of safety and effectiveness; although such data are desperately needed, they are very difficult to obtain. Not surprisingly, many experts on the Amsterdam panel noted that regulations that are intended to protect human subjects often make it even more difficult to obtain such data, and some expressed fears that provisions of the European directive will make it impossible to do so. The final session of this meeting focused on existing regulations and proposed legislation both in Europe and in the United States that govern, or will govern, the conduct and oversight of human research on both sides of the Atlantic. My assignment was to discuss the Office for Human Research Protections (OHRP) of the United States Department of Health and Human Services, its mission, organization, and its responsibilities, looking not only at what it is but also at what it does. This subject is worth exploring because there seems to be some significant misconceptions. In a recent commentary, the editor of the New England Journal of Medicine, Dr. Jeffrey Drazen, himself a respected clinical investigator, stated, "The role of the OHRP is to ensure that the process of [emphasis added] research is performed in accordance with statute and regulation. The office oversees the function and performance of individual IRBs (institutional review boards) and the process of informed consent" (2). Drazen further asserted that it is the responsibility of OHRP to investigate "when there is reason to believe that the procedures [emphasis added] with respect to the protection of human subjects from risks associated with research have not been appropriately followed" (2). Drazen's focus on procedural matters is ironic in light of widespread criticism of the U.S. system for being concerned too much with procedure and regulatory compliance at the expense of meaningful efforts to protect human research participants. The latter point was underscored in the well known reports from the Office of Inspector General of the Department of Health and Human Services on the system for review and oversight of human research in the United States (3). These reports were in fact the underpinnings of the Department's initiatives to reform the system, initiatives that began with the creation of OHRP in June 2000. Dr. Drazen's commentary was made in the context of a controversial and sometimes contentious investigation of studies performed by a network of investigators on critically ill patients with acute lung failure at major academic research centers across the United States. OHRP played a very prominent role in this saga, but I would like to make it unmistakably clear that I was not a participant. As a condition for my accepting the directorship of OHRP, I voluntarily recused myself from involvement in any matter related to a Harvard-affiliated institution for the entire duration of my official responsibilities in Washington. The investigation in question began while I was director after the office received a formal complaint from three respected scientists from the National Institutes of Health, and it was not concluded until July 2003 (several months after my return to Harvard at the end of November 2002). Although the Massachusetts General Hospital, from which I took a leave of absence during my time as director of OHRP, was not a participating site in the research trials in question, it was the data-coordinating center for at least one of the collaborative studies. Accordingly, I was not involved in any of the activities or investigations conducted independently by the Division of Compliance Oversight of OHRP, which in this case reported to a senior official in the Office for Public Health and Science in the Office of the Secretary of the Department of Health and Human Services. Rather than debate again the controversy over issues concerning the "ARDSNet Trials," my intention here is to focus on the activities of OHRP and some of the lessons learned, or at least taught, by the controversy surrounding them. THE CREATION OF OHRP OHRP is a new office that was established within the Office of the Secretary of the Department of Health and Human Services in June 2000 by then secretary Donna Shalala as part of a comprehensive initiative to strengthen protections for human research participants (4). Unlike its predecessor, the Office for Protection from Research Risks, the new office was charged solely with protection of human subjects in research, rather than having responsibilities for both human and animal subjects. Like the U.S. regulations for protection of human subjects, OHRP was created in response to a perceived crisis. That crisis was not, as many believe, simply the result of the tragic death of Jesse Gelsinger in a gene transfer experiment (5, 6). Rather, it reflected a growing loss of confidence in the U.S. system for human research and protection of human subjects that had evolved over more than a decade. The Congressional hearings that followed in the wake of Gelsinger's death, and the discomfort that they caused within the Clinton administration—an administration in which the president had personally apologized to the surviving victims of past tragedies in human research and vowed to take action to prevent future tragedies—were a potent stimulus for long overdue reform. Gelsinger's death was only the straw that broke the camel's back. It came after 20 years of largely ignored reports and recommendations calling for reform of an inefficient and questionably effective system that was showing clear signs and symptoms of serious disease—a system itself in need of intensive care. The creation of OHRP followed recommendations made in 1998 by a special advisory panel convened by Dr. Harold Varmus, then the director of the National Institutes of Health (7). The panel had expressed concern that the Office for Protection from Research Risks could not effectively exercise its responsibilities on behalf of the Department, the U.S. government, or the international research community from its position within the administrative hierarchy of the National Institutes of Health, reporting as it did to the Deputy Director for Extramural Research. The panel believed that the mission of the office could be fulfilled only if it were placed at a higher level within the government, within the Office of the Secretary of the Department of Health and Human Services. Ultimately, the new OHRP was placed at the secretarial level within the Office for Public Health and Science, reporting to the Assistant Secretary for Health. This structure ensured that it would have the visibility, resources, authority, and autonomy needed to do its job. The Department's broad leadership responsibility and enforcement authority were delegated by the Secretary to the new office and its director, except for promulgating new regulations—an authority that cannot be delegated under Federal law. Specifically, OHRP was charged with fulfilling all of the responsibilities set forth in the Public Health Service Act, the authorizing statute, regarding the protection of human subjects. These responsibilities are established in law, published in the Federal Register, and are available on the OHRP website (http://ohrp.osophs.dhhs.gov). They include the following:
Clearly, these statutory responsibilities and authorities greatly exceed Dr. Drazen's assertion that the role of OHRP is solely to ensure that the process of research is performed in compliance with statutes and regulations. Indeed, the creation of OHRP was intended to do more than set in motion a process of administrative reform: OHRP was charged with a much broader leadership role, one that drew on the ethical principles set forth in the Belmont Report (8) to promote responsible conduct of human research while ensuring that existing policies and procedures work effectively to protect human research subjects. The primary role of OHRP and the institutional review boards it oversees must be proactive, not reactive—to prevent harm to research participants (9, 10). Nevertheless, it also has an obligation to conduct appropriate inquiries when concerns arise. Some, including Drazen, have challenged OHRP's authority to investigate and intervene when concerns for the safety, well being, or rights of research participants are raised. Drazen stated that "if there is concern about the safety of human subjects that derives from the conduct of the trial, rather than from the design of the research, the OHRP should investigate the complaint" (2). He goes on to suggest that concerns about safety of human subjects that arise from the trial design are an issue for the National Institutes of Health (2). According to this view, OHRP would have no role or authority to intervene when a poorly designed and unsafe study results in direct harm or injury to research subjects as long as the research was conducted in compliance with administrative and procedural requirements set forth in the federal regulations for protection of human subjects. This view is too narrow. Even if one accepts Drazen's assertion that OHRP's responsibilities are to ensure that all federally conducted or supported research—and, for that matter, all human research conducted under Federal assurances of compliance—is conducted in accordance with federal regulations for protection of human subjects, the regulations themselves establish a much broader set of responsibilities than Drazen suggested. Indeed, the regulations explicitly require that all research involving human subjects must be designed according to sound scientific principles, and the risks to subjects must be minimized. Under the requirements of 45 Code of Federal Regulations 46.111, institutional review boards are specifically charged with determining that
These are not merely procedural requirements, nor are they responsibilities that are left to the institutional review boards alone. These requirements are intrinsic and essential to the design and execution of sound human research studies, and the responsibility for ensuring that they are fulfilled falls primarily to those who propose and conduct the research, the investigators and the sponsors. Institutional review boards are charged with reviewing and overseeing the research to reaffirm independently that these conditions are satisfied and that subjects will be properly protected. All research conducted under a federal assurance requires ongoing institutional review board review and approval, and the regulations give institutional review boards broad authority to require changes in the proposed research to secure such approval. In actual practice, these requirements are not always met. Fully one-third of OHRP's determination letters issued in the past 3 years specifically cite failures of institutional review boards to fulfill these essential responsibilities in the course of their review, approval, and oversight of human research studies. Another third of those determinations of noncompliance are related to deficiencies in informed consent. In a recent analysis of OHRP determinations of noncompliance, deficiencies in these two critical areas were discovered at more than half of the institutions investigated (11). This strikingly high percentage no doubt reflects some selection bias resulting from the fact that OHRP investigates only complaints that are deemed to have merit. However, to quote a statement by the recently elected governor of California after allegations of inappropriate behavior surfaced during the campaign, "Where there's smoke, there's fire." These findings indicate that there is ample room for improvement. SEPARATION OF SCIENCE AND ETHICS IN HUMAN SUBJECTS RESEARCH AND THE RESPONSIBILITY OF OHRP The notion that scientific and ethical issues in human subjects research can be segregated into discrete domains is untenable. Indeed, one of the most serious flaws in the current system for human subjects research is a direct but unintended consequence of the implementation of the institutional review board review process, namely that investigators have been allowed to delegate their personal responsibility for the ethical considerations of their work to the institutional review boards (9). This approach essentially means, from the perspective of investigators under the current system, that if an institutional review board grants approval to do a study, then the study must be ethical. In all cases, comprehensive institutional review board review begins with the science. Expert consultants or scientific review groups often facilitate this review with expertise in the specific area or investigation (12). If a study design is not sound or if it will not yield valid and meaningful data, it is unethical from the outset. Of course, a study design may be procedurally and statistically sound, but it may not be ethical. Often, an optimal study design in human research requires some compromise to ensure the well-being of the participants. The views expressed by Drazen that institutional review boards and OHRP ought not to be involved in the domain of science and study design are not surprising. They reflect attitudes that have been all too common within the human research community in the past and, unfortunately, remain so in some quarters. Although the vast majority of all human research studies conducted today are probably sound both scientifically and ethically, history has demonstrated all too clearly that some scientists, institutions, sponsors, and institutional review boards, even those with good intentions, can and do engage in activities and practices that may threaten both the integrity of science and the well-being of research participants. Even when proposed research is extensively vetted within the scientific community and approved by an institutional review board, questions may arise regarding the adequacy of the design of studies on both scientific and ethical grounds. Indeed, responsible scientists have an obligation to question their own work and the work of others. This is particularly true when the safety of research participants and the ethics or scientific integrity of research are questionable. As Albert Einstein once said, "The world is a dangerous place, not because of those who do evil, but because of those who look on and do nothing." As research becomes increasingly sensitive and complex, given the new tools and capabilities of modern science and medicine, many challenging questions arise. In many instances, neither the investigators nor the institutional review boards have the expertise that is necessary to address and resolve these complicated issues alone. When such questions are raised by knowledgeable individuals, based on valid information and substantive criticism, OHRP has both the responsibility and the authority to respond, provided that the work falls within its broad statutory jurisdiction. Perhaps even more importantly, the scientific community also shares a responsibility to respond to such expressions of concern in an open-minded and constructive manner. When science uses human beings as the subjects of investigation within an ethical framework that embodies respect for individuals, beneficence, and justice, scientists must accept their moral duty for ensuring safety and well-being of their research subjects. In most instances, even very challenging situations can be resolved reasonably, thereby allowing proposed research to proceed with appropriate safeguards for participating subjects. If, in the end, no easy resolution is possible, then the interests of the individual participants must prevail over those of science and society. No doubt there will be times when the burden of making this determination will fall to OHRP. In such cases, the decision may not be popular with individuals who would prefer to have the research go forward as proposed. In exercising its authority, OHRP must conduct its inquiry in a manner that engenders respect for its decision. Recognizing that OHRP does not itself have sufficient expertise to deal with all of the complexities of modern science, the office must (and does) call on outside experts to help. These consultants must consider very carefully the relevant issues, with broad input from all stakeholders to ensure that OHRP's decision is sound and well reasoned. RESPONSIBILITIES BEYOND COMPLIANCE To some it might appear that compliance oversight is OHRP's principal responsibility, but that is an inaccurate characterization. Compliance oversight activities account for only approximately 10% of the staff and budget resources used by the office in any given year. The Division of Compliance Oversight is only one of the operating divisions of the office. Other divisions include the Office of the Director, the Division of Policy, Planning and Special Projects, and the Division of Education and Development. Earlier this year, under the direction of the acting director, Dr. Bernard Schwetz, the Division of Assurances and Quality Improvement was reorganized, placing responsibility for assurances and compliance within the policy division, and the quality improvement activities were reassigned to the Division of Education as part of an effort to streamline the operational effectiveness of the office. Since its creation, the bulk of OHRP's resources have been devoted to education, support activities, and, particularly, new quality improvement initiatives, all of which are conducted outside of the compliance domain. From the outset, an operational "firewall" was established between the Division of Compliance Oversight and the other operating divisions as they developed and implemented a new quality-improvement program. This step was taken to reassure institutions and investigators that the new office was focusing on a proactive, performance improvement rather than reactive regulatory enforcement (9). At the same time, OHRP dramatically simplified and streamlined the federal assurance process while reducing administrative burdens and paperwork wherever possible. The office developed automated Web-based systems for electronic submission of assurances and, with the Food and Drug Administration, introduced a uniform system for registration of institutional review boards. Today, in keeping with Secretary Thompson's emphasis on performance-based management, a major thrust of OHRP's current efforts is to measure performance of the institutional review board process, not just in the operational domain but also in the analytical and judgment domains, as part of an ongoing effort to improve the substantive aspects of the review process while at the same time streamlining the procedural aspects. OHRP and the Department of Health and Human Services continue to promote and facilitate development of alternative review processes, in particular, collaborative models that reduce redundancy and foster greater cost-effectiveness. Increasingly, regional, national, and other cooperative models of review are being considered and evaluated (13). Growing numbers of institutions are turning to private, independent, noninstitutional review boards to conduct review and oversight of privately supported human research activities. New mechanisms for accreditation of human research protection programs and certification of research professionals at every level, from institutional review board managers to principal investigators, have been established over the past 5 years. These events mark a growing trend toward greater professionalism in human subjects research, and greater openness and effectiveness of the process for review and oversight. As they evolve and mature, these new accreditation and certification processes, well proven tools in other professions, will help to bring greater accountability to the research process through external validation of performance excellence and responsible conduct. As the public becomes better informed about the research process, these initiatives will do much to build confidence and establish a foundation of renewed trust in the research enterprise from which all will benefit. Ultimately, the investment community may begin to look toward such documentation of performance as an indication that human research is "investment grade," a phrase coined by Leslie Platt of the consulting firm Ernst and Young, which is developing methods to establish market-driven incentives that can turn compliance from a cost center to an asset. Gradually, in the wake of recent corporate scandals in the United States, the old adage that "money prevails over ethics" is being replaced by a new sense that " good ethics is good business." There are indications that this new attitude is clearly taking hold within the research community as professional organizations such as the international Association of Clinical Research Professionals and the American Academy of Pharmaceutical Physicians vigorously promote their codes of the ethics among their members (14, 15). Both begin with protection of the well-being of research participants as the researcher's highest responsibility. There is little doubt that progress is being made, but much remains to be done (16). Concerns that ever more stringent regulatory requirements may pose unreasonable or impractical barriers to research are well justified. In the absence of empiric evidence to show that most research is conducted in a responsible manner, there is a strong tendency for policy makers to overreact. The same can be said for institutions and their review boards, still fearful of the potential consequences of failures to comply with detailed nuances of regulatory requirements that often seem to be overly prescriptive, inflexible, and too narrowly interpreted. The end result is an undesirable form of "reactive hyperprotectionism" that can have a stifling effect on research productivity without meaningfully enhancing the safety and well-being of research participants. There is immediate need for renewal of long-promised efforts by the government to work with the research community to overhaul and streamline the oversight process. Surely, as private accreditation and certification ensure that standards exceeding minimal regulatory requirements are being met, the government should be able and willing to give greater latitude to those institutions and investigators who demonstrate their commitments and competence. To the fullest extent possible, governments and regulatory authorities should express as clearly as possible the norms of conduct expected and offer practical interpretation of regulatory requirements while using education, evaluations, and quality improvement initiatives to help all parties fulfill them. Enforcement tools and sanctions are most effective when used judiciously and only as necessary and appropriate. In this regard, OHRP has established a good track record, and one hopes that it will stay this course under the leadership of my soon to be named successor. "Doing it right together." This is the mission of the Office of Human Research Protections. In human subjects research, good science, ethics, and integrity are essential and inseparable. Establishing within the research community a "culture of conscience"—in which all parties do the right thing not simply to achieve compliance with regulations but out of an awareness of responsibility and commitment to caring for those on whom scientists depend to be able to do their work—is not an impossible dream: it is a necessary and achievable goal. FOOTNOTES Conflict of Interest Statement: G.K. is a member of the boards of trustees of the Association of Clinical Research Professionals and the American Academy of Pharmaceutical Physicians. He is also the vice president for Ethics and Professional Conduct for the American Academy of Pharmaceutical Physicians. Received in original form January 7, 2004; accepted in final form February 10, 2004 REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
