Researchers and physicians have different obligations to patients

Investigators are not fiduciaries to subjects

Health care patients, the general public, and sometimes even physicians and researchers at times misconstrue the relationship investigators have to research participants, an ethics expert says.

Investigators have obligations to perform research according to ethical and legal guidelines regarding the protection of human subjects, but they are not fiduciaries of research subjects, says E. Haavi Morreim, PhD, a professor in the College of Medicine at the University of Tennessee Health Science Center in Memphis, TN. Morreim was the chair of the independent patient advocacy council (IPAC), established to represent study participants' interests in the first clinical trials of the AbioCor Implantable Replacement Heart, manufactured by ABIOMED of Danvers, MA.

"A fiduciary is someone whose No. 1 obligation, whose whole purpose of the relationship, is for the fiduciary to promote the interests of the person who is beneficiary," Morreim says. "The classic example is a trustee who is overseeing assets on behalf of a beneficiary of a trust fund."

In this sense, a physician is a fiduciary of his or her patient because the physician's goal is to benefit the patient, Morreim explains. But this is not true of the relationship between investigators and research participants.

"By definition, the investigator's job is to do the research," Morreim says. "If you mistakenly think the researcher's No. 1 duty is to the research subject, then you'd have to modify the protocol anytime it would disservice the patient's interests."

Researchers' duties to subjects are as strong as those that fiduciaries have, but they aren't specifically fiduciary in character, Morreim says. Investigators have an obligation to follow the protocol and ensure scientific quality.1

However, medical care patients, the public, and even some physician investigators sometimes erroneously believe that a fiduciary type of relationship does exist between investigators and subjects.

Part of this myth is due to how physicians sometimes become investigators to their own patients. "A physician could be in conflict of interest in using patient trust to sign up patients for research," Morreim says.

This change in relationship can be appropriate when the patient has a disease for which there is no known cure or effective treatment, and the experimental treatment is all that is available, Morreim says.

"Then the physician might encourage the patient to sign up, but also to consider and look into the protocol," Morreim adds.

To prevent conflicts of interest, investigators generally should not recruit their own patients.

"Admittedly, in a community setting this becomes complicated, but I can envision some alternatives," Morreim says.

For instance, physician investigators could place information in their waiting rooms about research and provide a phone number to call if a patient is interested in participating.

"Normally, the treating physician doesn't have any business encouraging someone to sign up, but making this information available is not the same thing as recruiting," Morreim explains.

"We talk about therapeutic misconception and how patients/subjects may adopt the therapeutic misconception," Morreim says. "But frankly I think physician investigators are at least as susceptible to this."

Any physician who tells a patient that he or she thinks the patient should be in a particular clinical trial is quite possibly buying into therapeutic misconception, Morreim says.

Physicians might justify this type of interaction by thinking that they are doing what's best for the patient, but it's still a conflict of interest since the patient's participation in a clinical trial will benefit the physician investigator, Morreim adds.

There are alternative recruitment methods available. For instance, an investigator could place posters and other information in medical centers or at venues outside of his or her own clinic to describe a research trial and ask that volunteers call an information number if they're interested, Morreim suggests.

Or they can conduct advertising in the community through newspapers, television, and other media.

Having a physician's employee provide direct recruitment is not an ethically acceptable recruitment method in most cases because the employee is beholden to the doctor, Morreim says.

"You might describe the project and say, 'If you're interested I can set you up,' but that in itself carries the potential hazard of being weighted with encouragement by the physician," Morreim explains. "It's not an easy issue, and there can't be an absolute prohibition because there are times it's not feasible to have an independent entity."

Referrals made by physicians to another doctor investigator's trial are acceptable ethically so long as the referring physician is not entitled to a finder's fee, Morreim notes.

"If the physician is simply relaying information to the patient and saying, 'There is a study, and it's neither here nor there to me if you sign up, but you seem to be eligible, and if you're interested I can hook you up with people who give you information,' then it's okay," Morreim explains. "But if there's a finder's fee then that's a different situation than if someone is just paid to conduct extra tests to see if a patient is eligible for a trial."

Once participants are recruited, the physician investigator also might avoid ongoing issues with conflicts of interest and therapeutic misconception by having a neutral third party involved as a patient advocate.

In some cases, there are neutral persons on site who describe the proposed research project. This is what ABIOMED did with the AbioCor artificial heart transplant trials when they set up an independent group to work with trial participants, providing extensive education and reinforcing informed consent.

For the artificial heart transplant trials, there was a patient advocate available to trial participants. This person typically was a medical doctor who could carefully explain the risks, including what might happen if the participant had a stroke.2

The AbioCor IPAC recommended that the informed consent document be changed to reflect some of the actual experiences of participants in the trial, once the trial had enrolled some heart patients and various side effects or outcomes had occurred.2

The IPAC also identified issues regarding whether certain nonmedical costs would be reasonable to provide to participants and their families without these becoming undue inducements.2

Whether it's feasible to use patient advocates, it's important for researchers and clinical trial professionals to consider the potential for therapeutic misconception and how they can avoid it.

"In the bioethics community you find a number of people who think the investigator is a fiduciary, but they don't understand what is packed into the word fiduciary," Morreim says.

Even IRBs are not fiduciaries to research participants, although they come closer to filling those criteria than do investigators, Morreim adds.

What investigators do have are constraints on how they can treat research subjects, and these are powerful constraints, both positive and negative, Morreim says.


  1. Morreim EH. The clinical investigator as fiduciary: Discarding a misguided idea. J Law Med Ethics 2005;33:586-598.
  2. Morreim EH, Webb GE, Gordon HL, et al. Innovation in human research protection: The AbioCor Artificial Heart Trial. Am J Bioeth 2006;6:W6-W16.