Introducing New Technologies: Protecting Subjects of ... - CiteSeerX

World J. Surg. 27, 930–935, 2003 DOI: 10.1007/s00268-003-7096-3

WORLD Journal of

SURGERY © 2003 by the Socie´te´ Internationale de Chirurgie

Introducing New Technologies: Protecting Subjects of Surgical Innovation and Research Martin F. McKneally, M.D., Ph.D.,1 Abdallah S. Daar, D.Phil., F.R.C.S.2 1

Department of Surgery and Joint Centre for Bioethics, University of Toronto, 88 College Street, Toronto, Ontario, Canada M5G 1L4 Program in Applied Ethics and Biotechnology, Joint Centre for Bioethics, University of Toronto, 88 College Street, Toronto, Ontario, Canada M5G 1L4 2

Published Online: June 13, 2003

Abstract. The system for protecting human research subjects is under increasing pressure. Under the currently dominant Regulatory Ethics Paradigm, clinical research protocols must be reviewed and approved by an institutional review board (IRB) or equivalent. Although the IRB was introduced into health care in part to protect patients and investigators from the inherent conflict between the best clinical interest of the individual patient and the interest of science and society in answering a clinical question, its rigorous standards and rigid framework discourage surgeons from seeking potentially valuable early IRB consultation. Most of the important advances in the history of medicine, such as anesthesia, appendectomy, antibiotics, intensive care, and immunization, were introduced through an informal, unregulated innovation process that has been enormously productive but can lead to ratification of ineffective or harmful treatment by credulous physicians and patients. We propose a surgical innovation ethics paradigm that is a more nimble, flexible source of institutional and public oversight and approval of innovations that are in the gray zone prior to their conversion to formal protocols that then require IRB approval. We also discuss the management of personal and institutional conflicts of interest.

The system for protecting human research subjects has come under increasing, possibly unbearable, pressure over the past few years. Millions of Americans participate in an estimated 70,000 clinical trials conducted by industry- and government-sponsored researchers each year [1, 2]. Whereas the governance of clinical trials and the protection of human research subjects in such formal trials has been carefully defined in black and white rules and regulations, the review and oversight process has been overwhelmed in many institutions. The oversight committees cannot keep up with the volume of research. An additional area of concern that has received relatively little formal attention is that large gray zone where surgeons often find themselves—the zone of innovation—where it is unclear whether what they are doing is an evolutionary variation on a standard procedure, a unique departure from accepted standards, or the first stage of what should become recognized as a formal surgical research project. The confusion of definitions in this zone is a

Correspondence to: Martin F. McKneally, M.D., Ph.D., e-mail: martin. [email protected]

challenge with potentially serious consequences for all stakeholders. The confusion is experienced at all levels. As journal editors increasingly require institutional review board (IRB) approval for publication [3], institutions develop policies that any work that is to be published must be reviewed by the IRB. Regulators overreach, requiring that any study that might result in generalizable knowledge must be reviewed [4]. The overall effect is to overload the system. In this contribution we attempt to provide some definitions and discuss approaches that may help balance our patients’ interests with the larger societal goal: the discovery of better ways to help future patients. Formal Research under the Current Regulatory Ethics Paradigm The remarkable progress of scientific medicine derives from multiple sources, including laboratory experiments, clinical observations, and analysis of the effects of disease and its treatment on individuals and groups of patients. In a sense, all of medicine begins experimentally, as each treatment is applied to see how well it works in individual patients. Consensus develops about commonly accepted clinical practices and treatments based on informal agreement within the medical community about their apparent benefit. Because informal agreement is vulnerable to influence by many factors, consensus may develop to support erroneous or unhelpful practices, such as radical mastectomy for early-stage breast cancer or routine tonsillectomy to reduce throat infections and improve general health. During the mid-twentieth century, experimentation on prisoners, minors, and other vulnerable groups [5] led to a codification of international guidelines for the conduct of human research. Under the currently dominant regulatory ethics paradigm (REP), “innovative treatments are regarded as questionable until they are framed in a research protocol with formal mechanisms of informed consent. The protocol must be reviewed and approved by an IRB or equivalent” [6].The REP has provided a rigorous framework for the justification and testing of experimental treatments and the confirmation or refutation of their value (Table 1). Important documents in the evolution of the REP include the development of the Nuremberg code following the trial of the Nazi doctors in 1946

McKneally and Daar: Protecting Surgical Subjects from New Technologies

Table 1. Elements of the regulatory ethics paradigm. 1. The participant must be informed of the experimental nature of the treatment. 2. Much more complete discussion of potential risks is required than is customary for clinical informed consent to ordinary care. 3. Participants must be told they are free to choose standard care rather than the experimental treatment. 4. A multidisciplinary institutional review board (IRB) that includes representatives of the public must review and approve the protocol. 5. The outcomes of the experiment, including all adverse events, must be reported to the IRB.

[7], the Declaration of Helsinki by the World Medical Association in 1964 [8], and the Belmont Report in 1978 [9] in the United States. The widespread adoption of the IRB approval process was a major accomplishment of scientific medicine over the last four decades. Shortcomings of the REP Formal research based on rigorous analytic techniques leads to more reliable, evidence-based practice [10], but application of the scientific method to the analysis of patient outcomes can become ethically problematic. Adherence to a protocol required for the uniformity and reliability of aggregated study data can compromise the best clinical interest of individual patients. Although the IRB was introduced into health care in part to protect patients and investigators from the inherent conflict between the best clinical interest of the individual patient and the interest of science and society when answering a clinical question, its rigorous standards and rigid framework discourage surgeons from seeking potentially valuable early IRB consultation [11]. In the United States a recent report from the Institute of Medicine has found that the existing regulatory framework—the IRB system and the Common Rule— has not and probably now cannot react adequately to the constantly evolving research environment. The report found that the IRB system is in important respects a failure [12, 13]. The rigidity of the IRB may be rooted in its origin as an antidote to abusive, exploitative research practices [14]. Surgeons, accustomed to immediate adaptation to unexpected findings in their daily practice, seek a more nimble, flexible source of institutional and public oversight and approval. The problems associated with the review of complex treatment protocols and volume overload of the IRB’s agenda, resulting in superficial or slow reviews and sketchy oversight, have been well described [15, 16]. In addition to review and approval of protocols, the IRB is required to conduct continuing review of the research [17]. Between the IRB and the reporting of adverse events to regulatory authorities, there is now emerging a category of professionals described as research monitors. These individuals ensure that IRB requirements are being implemented as the research proceeds on a day-to-day basis. All these regulations, though necessary, add to the complexity of clinical research and the bureaucracy and expense associated with it. We believe that, particularly in the gray zone of innovation, it is important to explore nimble mechanisms that, at least initially, bypass the formal IRB process. Gray Zone of Innovation and Informal Research Beside, around, and under but rarely through the regulatory ethics paradigm, there has been a proliferation of new, nonvalidated

931

treatments. In the “borderland . . . outside the pale of the REP” [6] these innovations have incrementally but dramatically altered approaches to treatment and introduced entirely new interventions such as open-heart surgery, coronary bypass surgery, minimally invasive surgical techniques, organ transplants, and other major developments in contemporary surgery. Most of the important advances in the history of medicine, such as anesthesia, appendectomy, antibiotics, intensive care, and immunization, were similarly introduced through an informal, unregulated research process that has been enormously productive but can lead to ratification of ineffective or harmful treatment by credulous physicians and patients. Subsequent application of the framework of the REP has slowed or halted the enthusiastic adoption of some harmful or unhelpful treatments (e.g., gastric freezing to treat peptic ulcers) when they are proven to be no better than placebos. Innovation Ethics Paradigm for Surgical Research For the purpose of this discussion we define a surgical innovation as an intervention that is not yet viewed by the institution, community, or profession as meeting the accepted standards of safety, reliability, and familiarity (on the part of the patients and the health care team) with effects, side effects, and complications. To protect the interests of patients, staff, and institutions, innovations should be introduced through feasibility studies that ensure patient safety, rapid dissemination of lessons learned, timely assessment of side effects and complications, and management of benefits and risks. Other important dimensions that should be assessed during feasibility testing include cost and the impact on institutional programs and personnel. Feasibility Studies Feasibility testing of innovative surgical interventions is a developmental process. It includes identifying the patients who might benefit most from the innovation, standardizing the procedures used, and resolving technical problems [6]. Feasibility testing of the innovation is an experiment in clinical care conducted without a formal, fixed protocol. Based on our own experience and that of other, similar centers, we believe that the consent of the institution and the community should be obtained from the surgeon-in-chief, who serves as a representative of both and is accountable to both through the board of trustees in most public and university hospitals [18]. An application should be submitted to the surgeon-inchief by the innovator. Two informed members of the service should endorse the initiation of the feasibility study by signing the application. The consent form should be reviewed and approved by the IRB. Patients should be clearly informed that they are participating in an experiment to improve health care, not simply receiving the latest and best form of standard care. Patients are not the same; they cover a spectrum of attitudes toward risk, its management, and the opportunity to serve others. Some patients are risk-averse, some are risk-avid, some are heroes. Philip Blaiberg, Christian Barnard’s second heart transplant patient said, “Dr. Barnard, I want to go through with it now more than ever, not only for my sake, but for you and your team” [19]. Patients who participate in experimental therapy form the heroic vanguard in the ranks of patients. They should be protected from an excess of zeal—their own and that of

932

Table 2. Elements of the Surgical Innovation Ethics Paradigm. 1. The participant must be informed of the experimental nature of the treatment. 2. Much more complete discussion of potential risks is required than is customary for clinical informed consent for ordinary care. 3. Participants must be told they are free to choose standard care rather than the innovation. 4. The surgeon-in-chief and two members of the relevant service must endorse the initiation of the study. 5. An innovation task force of institutional stakeholders should be informed of outcomes, adverse events, and the impact on institutional resources and programs. 6. Only the consent process and form should be reviewed by the IRB. 7. When the initial conditions have been defined, such as identification of appropriate patients and stabilization of techniques, a formal research proposal may be presented to the IRB to validate the procedure.

their innovative caregivers. Their heroism should be more widely recognized and celebrated. Innovation Task Force A Task Force on Innovation, representing some of the institution’s stakeholders, should serve in an advisory role to maximize the dissemination of information about the innovation, help the innovators overcome barriers to progress, and provide a balanced institutional perspective. Clear, timely reports of effects, side effects, and complications should be provided to the chief of service and the surgeon-in-chief. The Task Force on Innovation should be informed by the surgeon-in-chief of the progress of the feasibility study. When the initial conditions have been satisfactorily defined, the study should be considered for conversion to a formal research protocol and submitted to the IRB for approval and oversight (Table 2). Power of “And” The Task Force does not replace the IRB but plays a complementary role, supporting and encouraging innovation while providing communication links to the IRB and the institution, which protect the interests of all stakeholders. Similarly, the Innovation Ethics Paradigm does not supplant the Regulatory Ethics Paradigm but extends the scope of responsibility and support to include those who are responsible for establishing, during the first phase of innovation, the initial conditions needed for formal research. Ensuring Patient Safety during Surgical Innovation Managing the Learning Curve: Within the Institution Audit, Morbidity, and Mortality Conferences. Trust, transparency, continuous learning, and improvement can be fostered through regular surgical audits and frequent, even daily, morbidity and mortality miniconferences, where goals include learning from mistakes, recognizing problems, and raising awareness of risks and limitations [20] of systems and individuals. Ideally, intended interventions should be discussed in a nonthreatening atmosphere, their rationale described, criticisms and suggestions internalized, and outcome measures discussed before the procedures are undertaken. Outcomes should be reviewed at regular intervals afterward.

World J. Surg. Vol. 27, No. 8, August 2003

An independent, internal departmental system for collecting all instances of morbidity and mortality, drawing regular comparisons between individuals, teams, and procedures, provides early warning of system and individual failures. Similar approaches can be implemented on a large scale for specific innovations short of formal controlled clinical trials. Bull et al. [21] described how a new procedure, the arterial switch operation, can be evaluated and adopted despite higher early mortality rates than the operation that was being replaced, even in the sensitive area of pediatrics where learning curve mortality is difficult to accept. Predictive mathematic modeling and extrapolation from sound clinical reasoning showed that the new operation, in the long run, has a lower overall mortality rate. The approach allowed the international specialist community, the hospital team, and the individual surgeons to learn and eventually minimize the risk of the innovative operation despite the higher early mortality rate. Tracker Studies. A practical way that has been proposed to deal with innovative changes is through the use of “tracker studies” for evaluating interventions when there is a likelihood of frequent, multiple developments and variants. Here, standard randomized, controlled trials are inappropriate, but their rigor could be approached by carefully planned and coordinated flexible protocols without prefixed sample size or trial duration but with sophisticated interim analyses aimed at maximizing collections of data and ensuring standardized outcome measures across centers [22]. Real-time Learning. In a remarkable study published in the Harvard Business Review, Edmondson et al. [23] described the effectiveness of “real-time learning,” analyzing and drawing lessons from the innovative process while it is under way. They studied 16 teams that were learning minimally invasive cardiac surgery. Stable teams, willingness to learn, and robust back-and-forth communication during the procedures proved far more effective than debriefing, after-action reports and project audits. Managing the Learning Curve: Beyond the Institution Surgeons can strengthen the process of innovation and help foster patient safety through electronic communication of outcomes, lessons, and experience in online discussion groups. Visiting colleagues’ hospitals or operating rooms to review procedures and share experience dramatically accelerates group members’ learning. This has been effectively practiced by the Northern New England Cardiovascular Disease Study Group [24]. The process, fostering peer stimulation, encouragement, and pressure to improve performance, is more effective than the more competitive model of public report cards. Protecting Subjects by Managing Conflicts of Interest Innovation may lead to financial and nonfinancial conflicts of interest [25, 26]. One of the most important changes in research during the past decade has been the increasing participation of private interests in biomedical research [27]. In the United States one of the key catalysts of this change was the passage in 1980 of the BayhDole act, which transferred intellectual property rights to researchers funded by federal research money [28]. The reduction in university funding from traditional sources such as governments and

McKneally and Daar: Protecting Surgical Subjects from New Technologies

nongovernmental organizations has meant that universities must now attract more private funding to function in a highly competitive environment. We must find ways to deal with the potential conflicts of interest (COI) for both institutions [29–33] and individual investigators. Personal COI for clinicians performing research is a complex issue and a constant source of ethical tension that all clinicianinvestigators must honestly confront [34–36]. Disclosure and Recusal for Institutional COI Emanuel and Steiner argued that even a pilot trial should not be conducted in an institution that has a major financial interest in the outcome [30]. The minimum requirement is full disclosure followed by recusal if it appears that the COI is unmanageable. The most stringent step, recusal, may not be necessary if the financial interest is modest. Disclosure means that: (1) all patients must be informed of the COI, and this information must be stated in the consent form; (2) the financial interest must be disclosed to the IRB; (3) the interests of all collaborators, co-investigators, and institutions and persons who have supported the research must be included; (4) all publications, including oral presentations and abstracts, must disclose the institution’s financial interest; and (5) the safeguards that have been put into place to deal with the COI must be disclosed. Management of Institutional COI through External Monitoring Committees Regarding the institutional COI, the best safeguard would be to establish an ad hoc external committee to monitor the conflict. This may be needed in certain circumstances, for example, when the IRB determines that a particular study is compelling but that there is a conflict of interest involving a major financial interest or a compelling study was approved by the IRB and the conflict arose or was discovered later. External monitoring committees can be cumbersome, and there may be resistance to this approach, especially as it is not yet common practice. If an institution establishes such a committee, it should ensure that its members are knowledgeable enough to review the research. At the outset, the external committee should review the research design; later it should review the data generated and be able to require modifications or stop the trial. Its members should not have any direct or indirect financial interest in the outcome of the trial. They should not be paid beyond reasonable compensation for expenses incurred. The members might include a biomedical researcher, a lawyer, and a bioethicist [30]. There is a risk that as more accountability is required of institutions regulatory agencies may come to insist that data from trials conducted at institutions with a financial COI be inadmissible evidence for drug and device approval [30]. This would be “strong medicine” but would ultimately set a high ethical and scientific standard that would benefit the quality of research and therefore benefit institutions and patients. New Approach: From Regulatory Compliance to Prevention of Harm Greg Koski, the former director of the U.S. Office of Human Research Protection (OHRP), was publicly quoted as saying that the

933

U.S. system of protecting human subjects in medical research is in “critical” condition and that it is “like a nuclear reactor . . . actions need to be taken to prevent the system from meltdown” [1]. As a result of this unsatisfactory state of affairs, OHRP proposed a dramatic shift away from its previous approach of developing ever more detailed regulatory instruments coupled with a focus on their enforcement and on the oversight of research programs. That approach, as evidenced by a number of recently publicized oversight failures (e.g. the Gelsinger gene therapy case in Pennsylvania, which in its wake highlighted a large number of systems failures), can be judged to be inadequate. The new approach proposed by Koski takes into account the fact that researchers on the whole are not only motivated to do their best for research subjects but that their own enlightened self-interest demands that they improve their performance. Koski’s open letter says, “Many times over the past year and a half we have heard the sound bites and catch phrases—‘shared goals and shared responsibilities,’ ‘flexibility with accountability,’ ‘do it right because it’s the right thing to do’” [37]. The emphasis of the new approach is on quality improvement, with the new program based on self-evaluation and collegial constructive consultations with OHRP staff, adopting an attitude of collaboration rather than confrontation. A directed self-assessment tool has already been developed [38]. In the first instance, OHRP is calling on federally funded research centers to volunteer for this new program. No doubt there will be resistance to this approach, but it certainly is worth a trial. It seems preferable to the current situation, especially for surgeons performing innovative surgery, who on the whole appear to be unaware of the Department of Health and Human Services regulations and rarely seek IRB review. They also generally oppose outside regulation and are often unaware of what constitutes innovation and research [11]. It is time for a new approach to surgical innovation. Re´sume´. Le syste`me de protection de la recherche chez l’homme est sous une pression croissante. Sous le Regulatory Ethics Paradigm, le syste`me couramment en rigueur, le protocole de recherche´ clinique doit ˆetre vu et approuve´ par un panel d’inspection, appele´ «Institutional Review Board» (IRB). Alors qu’a ` l’origine, I’IRB a ´ete´ introduit dans le syste`me de sante´ pour prote´ger les patients et les investigateurs du conflit entre le meilleur inte´reˆt pour le patient individuel d’un cote´, et l’inte´reˆt pour la science et la socie´te´, qui veulent une re´ponse a ` une question clinique, de l’autre, les standards rigoureux et ses re`gles tre`s strictes de´couragent souvent les chirurgiens dans leur recherche et les encouragent a ` ne pas consulter I’IRB to ˆt dans leur ´etude. La plupart des progre`s importants dans l’histoire de la me´decine, comme l’anesthe´sie, l’appendicectomie, les antibiotiques, les soins intensifs et l’immunisation ont vu le jour gra ˆce a ` un proce´de´ informel, non re´glemente´, qui, de temps a ` autre, laisser entrer dans la pratique un traitement inefficace voire dangereux, meˆme s’il est parfois productif. Nous proposons un Paradigme de l’e´thique de l’innovation chirurgicale qui est plus souple et flexible, permettant l’approbation des innovations dans la zone grise avant qu’il ne devienne protocole formalise´ approve´ par I’IRB. Nous abordons ´egalement la gestion du personnel et les conflits d’inte´reˆt a ` l’origine des difficulte´s. Resumen. El sistema de proteccio ´n en investigacio ´n en humanos se encuentra ´ tica bajo presio ´n creciente. Bajo el actual Paradigma de Regulacio´n E (Regulatory Ethics Paradigm), los protocolos de investigacio ´n clı´nica deben ser revisados y aprobados por un Consejo Institucional de Revisio ´n (Institutional Review Board) o su equivalente. Si bien tal Consejo fue introducido en el campo de la atencio ´n de la salud en parte para proteger pacientes e investigadores del inherente conflicto entre el mejor intere´s clı´nico del paciente individual y el intere´s de la ciencia y de la sociedad en cuanto a esclarecer interrogantes clı´nicos, sus rigurosos esta ´ndares y su rı´gido marco de referencia desestimulan a los cirujanos en cuanto a hacer consultas precoces potencialmente valiosas al Consejo. La mayor parte de los avances en la historia de la medicina, como la anestesia, la apendicectomı´a, los

934

antibio ´ticos, el cuidado intensivo y la inmunizacio´n fueron logrados mediante procesos innovativos informales y no regulados que han sido enormemente productivos, porque pueden llevar a la ratificacio´n de tratamientos ineficaces o dan ˜inos por me´dicos y pacientes cre´dulos. Proponemos un Paradigma ´ tico de Innovacio´n Quiru E ´rgica que sea una fuente ma ´s flexible de veedurı´a institucional y pu ´blica y la aprobacio ´n de innovaciones que se hallen en la zona gris antes de convertirlos en protocolos formales que requieran la aprobacio ´n del Consejo. En este artı´culo tambie´n se discute el manejo personal e institucional de los conflictos de intere´s.

References

World J. Surg. Vol. 27, No. 8, August 2003

14. 15. 16. 17. 18. 19. 20. 21.

1. Kranish M. System for protecting humans in research faulted. The Boston Globe, March 25, 2002 2. Daar As, Singer PA. Human capital is key to research ethics, Sci. Dev. Net, April 25, http://www.scidev.net/archives/archivesearch.asp? searchtext=daar&Submit=Search&t=C&sm=a#, 2002 3. International Committee of Medical Journal Editors. Uniform requirements for manuscripts submitted to biomedical journals. N. Engl. J. Med. 1997;336:309–315 4. Levine R, Ethics and Regulation of Clinical Research, 2nd edition New Haven, Yale University Press, 1988;5 5. Brody G, The Ethics of Biomedical Research: An International Perspective, 31–37 6. Agich GJ. Ethics and innovation in medicine. J. Med. Ethics 2001;27: 295–296 7. Grodin M. Historical origins of the Nuremberg Code. In Annas GJ, Grodin MA, editors, The Nazi Doctors and the Nuremberg Code: Human Rights in Human Experimentation, 1992;121–144 8. World Medical Association Declaration of Helsinki. Ethical principles for medical research involving human subjects. J.A.M.A. 2000;284: 3043–3046 9. National Commission for the Protection of Human Subjects of Biological and Behavioral Research. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research. Washington, DC, US Government Printing Office, 1979 10. Evidence Based Medicine Working Group. Evidence-based medicine: a new approach to teaching the practice of medicine. J.A.M.A. 1992; 268:2420–2425 11. Reitsma AM, Moreno JD. Ethical regulations for innovative surgery: the last frontier? J. Am. Coll. Surg. 2002;194:792–801 12. Beauchamp TL. IOM report on the system for protecting human research participants. Kennedy Institute Ethics J. 2002;12:389–390 13. Federman, DD, Hanna, KE, Rodriguez, LL, editors. Committee on Assessing the System for Protecting Human Research Participants. Institute of Medicine. Responsible Research: A Systems Approach to Pro-

22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38.

tecting Research Participants, National Academy Press, Washington, DC, 2002 Beecher HK. Ethics and clinical research. N. Engl. J. Med. 1966;274: 1354–1360 Slater EE. IRB Reform. N. Engl. J. Med. 2002;346:1402–1404 Steinbrook R. Improving protection for research subjects. N. Engl. J. Med. 2002;3446:1425–1430 http://ohrp.osophs.dhhs.gov/humansubjects/guidance/45cfr46.htm McKneally MF. Bioethics for clinicians: surgery;surgical competence. Ann. R. Coll. Physicians Surg. Can. 1999;32:77 Katz J, The Silent World of Doctor and Patient, 1986, pp 132 Seiler RW. Principles of the morbidity and mortality conference. Acta Neurochir. Suppl. (Wien). 2001;78:125–126 Bull C, Yates R, Sarkar D, et al. Scientific, ethical, and logistical considerations in introducing a new operation: a retrospective cohort study from paediatric cardiac surgery. B.M.J. 2000;320:1168–1173 Lilford RJ, Braunholtz DA, Greenhalgh R, et al. Trials and fast changing technologies: the case for tracker studies. B.M.J. 2000;320:43–46 Edmondson A, Bohmer R, Pisano G. Speeding up team learning. Harvard Business Rev. 2001;Oct:125–132 Nugent W. Collaborative practice: a personal journey. Ann. Thorac. Surg. 2001;71:765 Lemmens T, Singer PA. Bioethics for clinicians. 17. Conflict of interest in research, education and patient care. CMAJ 1998;159:960–965 Levinsky NG. Nonfinancial conflicts of interest in research. N. Engl. J. Med. 2002;347:759–761 Daar AS, Bampoe J, Bernstein M. Ethical issues in molecular medicine: an outline and relevance to neurosurgery. Neurosurgery, in press http://www.cogr.edu/bayg-dole.htm (accessed January 14, 2003) Cho MK, Shohara R, Schissel A, et al. Policies on faculty conflicts of interest at US universities. J.A.M.A. 2000;284:2203–2208 Emanuel EJ, Steiner D. Institutional conflict of interest. N. Engl. J. Med. 1995;332:262–268 Friedberg M, Saffran M, Stinson TJ, et al. Evaluation of conflict of interest in economic analyses of new drugs used in oncology. J.A.M.A. 1999;282:1453–1457 Kassirer JP. Financial conflict of interest: an unresolved ethical frontier. Am. J. Law Med. 2001;27:149–162 Montaner JSG, O’Shaughnessy MV, Schecter MT. Industry-sponsored clinical research: a double-edged sword. Lancet 2001;358:1893–1895 Elks ML. Conflict of interest and the physician-researcher. J. Lab. Clin. Med. 1995;126:19–23 Levine RJ. Clinical trials and physicians as double agents. Yale J. Biol. Med. 1992;65:65–74 Lo B, Wolf LE, Berkeley A. Conflict of interest policies for investigators in clinical trials. N. Engl. J. Med. 2000;343:1616–1620 http://ohrp.osophs.dhhs.gov/humansubjects/qip/oltr.pdf http://ohrp.osophs.dhhs.gov/humansubjects/qip/qifedreg.pdf