Conflicts of Interest Among Clinical Investigators

The following article appeared in the November 1993 edition of Health Letter. The topic remains as relevant today as it was over 13 years ago because there has been no progress made on this issue.

Imagine the following: You are involved in a lawsuit against a manufacturer whose product you say caused you injury. During the trial, you get the impression that the judge is favoring your opponent – ruling to his or her advantage on motions, objections, evidence and the like, and you end up losing. Later, you learn that the judge had a significant financial interest — stocks and stock options — in the company you sued. Would you suspect that you had been had by a biased judge? You bet! And would you be justified in appealing on grounds of conflict of interest? Again, yes; judges in this situation have a duty to disclose any interest and to remove themselves from the case.

So, what about doctors who act as clinical investigators, testing drugs for pharmaceutical companies? Are they also required by law or custom to disclose financial interest in the companies sponsoring their work? No way! There is no law or regulation preventing a clinical investigator from holding stock, stock options, patents, licenses or any other financial interest in the company that makes the drug or product he or she is testing. And the Food and Drug Administration (FDA), which oversees drug trials, does not even require that such arrangements be disclosed.

More than 34 years ago, Dr. Sidney Wolfe testified before a Senate hearing on clinical investigators’ possible financial conflicts of interest.

What brought the issue to the Senate’s attention was the clinical investigation of a soft contact lens. The doctors doing the trials were paid not in money (as is customary) but in stock and stock options. The investigators wrote optimistic reports about the lens’ potential and the manufacturer’s stock rose in response. Dr. Wolfe stated then:

How, under these circumstances, can objective scientific inquiry flourish? ... Why are practitioners now (compensated only by the patient, rather than by the patient and the company, as before) reporting what appears to be a larger number and greater variety of adverse reactions than did their doubly-compensated counterparts during the pre-market investigational stage? ... [A]s long as the manufacturers, medical schools and the FDA continue to condone such practices ... the American public will be the victims.

More than two decades later – on September 9, 1993 – the FDA convened its Science Board to address this long-standing question of financial disclosure by investigators in clinical trials. In attendance were speakers representing the academic and scientific communities, consumer organizations including Public Citizen’s Health Research Group, manufacturers of drugs and medical devices, and the federal government. Public Citizen’s responses to questions raised by the Board were as follows:

1. Do financial interests of a clinical investigator in the product under study or the product sponsor have the potential to bias the outcome of studies undertaken by the sponsor? 
   Yes. Investigators should not gain financially from companies whose products they are evaluating. For the FDA not to have a policy in this area is to invite investigator bias. An investigator may be unconsciously affected by an economic incentive, causing him or her to downplay negative data and exaggerate favorable data. A financial interest may affect the way research is carried out, analyzed or reported.
   
   According to an article in the August 1993 New England Journal of Medicine, the basic purposes of conflict of interest rules are to maintain the objectivity of professional judgment and to maintain public confidence in professional judgment. In many areas of life, restrictions on conflicts of interest are the norm. For example, as previously mentioned, judges are expected to excuse themselves from cases in which they have an interest, not only to eliminate bias, but to eliminate the appearance of bias or partiality.
   
2. Is the potential for bias in this area sufficient to warrant requiring disclosure of these interests to the FDA in some or all cases?
   Yes. Disclosure is warranted, but disclosure alone is not enough – it only announces that the risk of bias is present. Clinical researchers should disclose all financial arrangements as well as any ancillary ties to companies whose products they are investigating, such as educational activities supported by the companies, participation in other research projects funded by the companies, and consulting arrangements. Disclosure should be to the FDA, to the medical center where research is conducted, to organizations that are funding the research, and to journals that publish the results.
   
   Furthermore, financial disclosures ought to be made to the FDA early in the review process. We believe that there is enough risk to subjects and data integrity during the early stages that financial disclosure ought to be made at that time.
   
   Investigators ought to have to disclose information about themselves and family members, such as parents, children, and spouses.
   
3. Do certain types of financial interest pose a greater potential for bias (e.g., equity interests or other forms of compensation where the value of interest/compensation may be influenced by the outcome of the study)?
   Yes. Researchers who might benefit financially by distorting their work have a conflict of interest even if they do not actually distort [it]. The circumstances determine whether there is a conflict of interest, not the outcome. Associations with business which could affect the outcome of a study include direct employment and consultancy, stock ownership, and patent-licensing agreements.
   
4. Apart from outcome-dependent interests, are there interests that should be of special concern to the FDA (e.g., large retainer or consulting fees)?
   Yes. Of these, can a financial threshold be identified below which the FDA may reasonably assume that such financial interests are unlikely to have influenced the outcome of the study? It would be very difficult to set a safe threshold.
   
5. If the FDA were to require disclosure of specified financial interests held by investigators, what steps should the FDA take to minimize the potential for bias resulting from such studies?
   Disclosure alone is not enough. Researchers with outcome-dependent interests should be banned from taking part in studies. Any other payment by the company, if allowed at all, should be commensurate with actual efforts expended on behalf of the company, and should be disclosed to the FDA. The FDA is already operating under an enormous statutory burden and should not be responsible for reducing the potential for bias resulting from financial interests.
   
6. Are there forms of compensation or financial interests that create a significant enough risk of biasing a study to cause the FDA not to rely on the study?
   Yes. When the sponsor and the investigator are the same person, or are closely related (business partners), as is the case with many start-up device and drug companies.
   
7. What could be the effects of such a determination on the development of drugs, biotech products and medical devices?
   It may delay the approval of some drugs and devices, because the companies will be forced to repeat the studies.
   
8. Are there methods of minimizing bias, e.g., blinding, independent assessments of study endpoints, participation of multiple investigators (most of whom have no financial interest) that are or could in some cases be adequate to protect against the potential for biases created by an investigator’s financial interests? Should utilization of such methods be required (or shown not to be needed) whenever an investigator holds and interest in the product or the sponsor?
   No. The FDA should not get involved in determining whether or not various methods employed by companies are adequate to sufficiently eliminate bias, or even whether such methods are needed. Rather, there should be a ban on such arrangements, and a disclosure requirement. It is safer and more responsible to decide in advance to remove factors that tend to distract researchers from concentrating on medical and scholarly goals.
   
9. If the FDA were to require disclosure of specified financial interests, should the FDA disclose these financial interests to the public? If so, what is the appropriate forum to release this information publicly?Public disclosure of this information in some useful form is critical. The FDA cannot always do its job alone. It is imperative that the public be able to play some policing function in this area. Our organization, and others, often examine data submitted to the FDA and bring problems to the public’s and agency’s attention. Public disclosure is a positive good in itself, and it has a way of preventing substantive abuses in the first place.