Paying Subjects to Take Part in Research: A New Perspective on Coercion and Undue Influence

Clinical Researcher—March 2019 (Volume 33, Issue 3)

PEER REVIEWED

David Borasky, MPH, CIP; Jeffrey A. Cooper, MD, MMM;

Kelly FitzGerald, PhD

Under U.S. Department of Health and Human Services (HHS) and Food and Drug Administration (FDA) regulations and the International Council on Harmonization guidelines for Good Clinical Practice, for an institutional review board (IRB) to approve research with human subjects, it must determine that investigators will obtain informed consent from each prospective subject or the subject’s legally authorized representative, under circumstances that minimize the possibility of coercion or undue influence.{1–4}

Payment for participation in research represents a mechanism to induce subjects to take part in research when they otherwise might not take part. Therefore, payment is part of the consent process and any payment provided to subjects must take place under circumstances that minimize the possibility of coercion or undue influence. However, how should IRB members determine that this is, in fact, the case in the studies they review?

Defining Undue Influence

To determine whether any payments to participants minimize the possibility of coercion or undue influence, IRBs should apply the definitions of “coercion” and “undue influence.” The Belmont Report defines “coercion” as an overt threat of harm that is intentionally presented by one person to another to obtain compliance, where compliance in this case refers to agreeing to take part in research.{5} The Belmont Report further defines “undue influence” as an offer of excessive, unwarranted, inappropriate, or improper reward or other overture to obtain compliance. In this context, “compliance” refers to agreeing to take part in research or to continue participation in research. “Influence” means to impact, determine, guide, shape, alter, change, or transform and “undue influence” is influence that is excessive, unwarranted, inappropriate, or improper.{6–8} Under the regulations and guidance, undue influence is ethically unacceptable, whereas influence that is not undue is allowable.

As parts of the HHS, the Office for Human Research Protections (OHRP) and FDA have released guidances regarding subject incentives that state, “Paying research subjects in exchange for their participation in research is a common and, in general, acceptable practice.”{9,10} The FDA guidance goes on to state that payments made to offset or reimburse out-of-pocket expenses do not raise issues of coercion or undue influence. This guidance also informs IRBs to look carefully at payments to ensure that they are neither coercive nor unduly influential.

Proposing a Different Approach

Most IRBs are cautious about payments and reject those that are out of the norm as coercive or unduly influential. However, two recent papers cast doubt on this approach, and suggest that IRBs need to take a different approach to evaluating whether the circumstances of payments to subjects minimize the possibility of coercion or undue influence.{11,12}

The first issue noted by the authors of these papers is that studies of IRB members indicate that they commonly reject the use of monetary incentives because they categorize such incentives as coercive.{7,8} However, these authors note that IRBs should abandon the idea that incentive payments can be coercive.{11,12}

As noted in the Belmont Report, coercion involves the intentional threat of harm.{5} Although a threat to withdraw a payment to which a subject is entitled can represent an intentional threat of harm, subjects taking part in research are not entitled to incentive payments. Therefore, the offer of an incentive payment is a benefit and cannot represent a harm, no matter how large the payment.

Offering an incentive payment for participation cannot meet the definition of “coercion.” Therefore, IRBs should stop using coercion as a basis for requiring investigators to reduce the amount of payment to subjects. Instead, IRBs should focus on whether the influence presented by incentive payments represent influence that is acceptable or influence that is undue.

Considering Payment Impact

While coercion is relatively easy to recognize because subjects are threatened, distinguishing undue influence from mere influence is more difficult. Undue influence implies that individuals will agree to take part in research without a rational consideration of the information provided in the informed consent process, such as the risks and procedures involved in the research.

IRBs cannot consider an incentive payment to be unduly influential solely because an individual would not take part in the research but for the payment; this is precisely the reason investigators use recruitment incentives. Moreover, this logic would compel IRBs to determine that many acceptable incentives are unduly influential, such as advertising or a physician telling a patient that they might want to talk to the investigator to learn about taking part in a clinical trial.

If an IRB has approved a study, the IRB has determined that the risks are acceptable in the absence of payment. Adding payment to a research study cannot affect the acceptability of the research risks. In both cases, reasonable individuals can use good judgement and decide to take part in research with incentive payments, where in the absence of those payments, they would have declined.

Undue influence must lead to poor judgement. The question is: How does an IRB determine whether a payment causes prospective subjects to make a decision that is against their interests?

Recognizing Context Plays a Part

A problem with payments is that an offer of payments designed to incentivize participation in research will not affect all individuals the same way. Subjects will have different perceptions of the financial reward based on the burden of the research in terms of costs of transportation, length of study visits, and degree of risks and discomforts. The offer may result in poor judgement by some individuals, but may not affect the judgement of others.

More specifically, what is unduly influential to one subject might be merely influential to another. As OHRP guidance states, “because influence is contextual, and undue influence is likely to depend on an individual situation, it is often difficult for IRBs to draw a bright line delimiting undue influence.”{9} Meanwhile, the IRB must approve the research for all individuals who meet the selection criteria, even if an incentive payment may cause poor judgement in a minority of subjects.

Since the theoretical likelihood of undue influence goes down with a decreasing incentive, many IRBs minimize the possibility of undue influence by mandating payments low enough to prevent any possibility of an incentive payment affecting someone’s judgement. This is problematic for two reasons. First, there is no threshold of payment below which no person would ever be unduly influenced. If the goal is to protect against such outliers, no payment can ever be free of undue influence. Second, there are other influences on a subject’s participation in research, and if the same paradigm is applied to these influences, one reaches an illogical conclusion.

Acknowledging Influence of Potential Benefits

Lee points out factors other than incentive payments that influence subjects, and notes that if payments can be unduly influential, so can these other factors.{11} For example, a common influence on subjects’ decisions to enroll in research would be potential benefit.

The potential benefit may be a therapeutic misconception or may be reasonably expected, based on reality. For example, early trials may have shown a high rate of complete remission in a lethal cancer that otherwise had no effective treatment, such as seen with early trials of imatinib and ipilimumab. This potential benefit, or the incorrectly perceived potential benefit, may be enough to cause an individual to make a decision contrary to his or her own best interest. Research staff who run active Phase I oncology units see this not infrequently.

Minimizing All Possible Sources of Influence

Lee observes that the standard IRB approach to reducing the likelihood that payments will unduly influence subjects is to reduce them.{11} If IRBs were to apply their standard approach to payments to other research benefits, Lee notes that IRBs would have to reduce potential benefits to minimize undue influence. For example, the IRB would require protocols to randomize a greater percentage of subjects to placebo or require lower doses of drug in the treatment group to the point where individuals would be less motivated to take part.

Factors such as the possibility of closer follow-up, less expensive medical care, or access to medical care also influence subjects to join trials. IRBs could minimize the possibility of undue influence by reducing follow-up visits or requiring subjects to pay for the study drug; clearly, however, this is absurd. The IRB would be penalizing all subjects based on the behavior of a subset of individuals, and would be interfering with science that otherwise meets the criteria for approval.

Factoring in Fairness

There is an issue of fairness at stake when denying a subject reasonable compensation because of the behavior of a minority. If an insufficient number of subjects enroll in a research study, the likelihood of gaining important knowledge is reduced.

One can argue that reducing payments to subjects is different from reducing benefits in research. However, that does not solve the problem that potential benefits may present undue influence and that, to approve research, the IRB is obligated to ensure that such undue influence is minimized. Can IRBs use a single strategy to minimize undue influence related to incentive payments and potential benefits?

Obtaining Legally Effective Consent

One solution to this issue is to go back to the definition of undue influence. Undue influence implies that individuals will agree to take part in research against their own best interest. The unduly influenced subject may have agreed to participate in the research without any consideration of the risks or the procedures that would take place.

The FDA and HHS regulations refer to “legally effective consent,” which generally means that an individual has been provided sufficient information about the research, understands that information, has considered that information, can understand the implications of a decision to take part, can make a decision, and can communicate that decision.{1,2,5} The unduly influenced subject fails these criteria, based on not considering the information and not understanding the implications of a decision to take part.

Essentially, unduly influenced subjects are in a state where they cannot provide legally effective consent. This applies not only to subjects who are unduly influenced by incentive payments, but also to subjects unduly influenced by actual or perceived potential benefits, advertisements, and recommendations of a treating physician that participation in a specific research trial would be in that patient’s best interests.

Determining Capacity to Consent

Another situation in which individuals are unable to provide legally effective informed consent arises when subjects are cognitively impaired and lack the capacity to provide informed consent. Typically, protocols require subjects to be able to personally provide informed consent, or in some cases, require a legally authorized representative to provide consent on behalf of such subjects.

Although some protocols involve the expectation that prospective subjects may or may not lack capacity to consent, almost all protocols involving adults are based on the idea that the research team knows whether a subject taking part has the capacity to provide legally effective informed consent. If they do, subjects must personally provide informed consent and permission by a legally authorized representative for the subject to take part in the research is insufficient. If they don’t, the investigator cannot enroll the subject or must obtain the permission of a legally authorized representative, depending on the specifics of the protocol.

Many IRBs have dealt with situations that require the research team to evaluate during the consent discussion whether the prospective subject has the capacity to consent. This involves the research team listening to the prospective subject’s statements and questions to assess whether he or she understands the information being provided, is considering that information, understands the implications of a decision to take part, can make a decision, and can communicate that decision.

IRBs can apply the same process to undue influence, which like capacity to consent, is another factor that interferes with the ability of an individual to provide legally effective informed consent.

Recognizing Undue Influence

Some IRBs may scoff at the idea that the research team can evaluate whether a subject is providing legally effective informed consent, but the fact is that IRBs often rely on research teams to make this decision. If the research team can determine whether a subject is providing legally effective informed consent based on capacity to consent, it can follow the same process to exclude subjects being unduly influenced by incentive payments or by perceived or actual prospect of benefit.

In our travels, we have met investigators from clinics that exclusively conduct Phase I oncology trials. They commonly run into patients who are looking for a cure and are not interested in learning about or considering the risks. These investigators recognize undue influence, and will not enroll subjects who ignore the information required to provide legally effective informed consent.

Researchers can follow the same process for incentive payments, and it is not hard to detect subjects whose judgement is impaired by the promise of dollar signs. There is no doubt that many researchers would benefit from training and mentoring in this regard. However, changing the behavior of research teams is the only way to address undue influence caused by perceived or potential benefit, and has the side effect of addressing undue influence caused by incentive payments.

Identifying the One Percent

This process directly addresses OHRP’s aforementioned observation that “because influence is contextual, and undue influence is likely to depend on an individual situation, it is often difficult for IRBs to draw a bright line delimiting undue influence.”{9} Most IRBs understand that capacity to consent depends on both the protocol and the individual.

A subject may have the capacity to consent to a study involving a single blood draw, but may lack the capacity to agree to take part in a complicated clinical trial. Ninety-nine percent of individuals may have the capacity to take part in a trial, while one percent do not. The same is true for undue influence—a subject may be unduly influenced by a $2,000 incentive for one trial, but not another. Ninety-nine percent of individuals might not be unduly influenced by a $2,000 incentive for one trial while one percent are. The IRB can never draw a bright line, but the research team is in the ideal situation to detect and minimize undue influence.

Conclusion

In summary, IRBs should understand that incentive payments can never be coercive. The issue with incentive payments is that they can be unduly influential. Undue influence is not unique to payments, but also occurs because of perceived or actual potential benefit.

Because undue influence depends on the subject, IRBs cannot define when a payment, a perceived potential benefit, or actual potential benefit will be unduly influential to a specific subject. Nonetheless, IRBs are required to determine that the consent process is conducted in such a way that incentive payments minimize the possibility of undue influence.

For undue influence related to perceived or actual potential benefits, the current process is for researchers to not enroll potential subjects who are being unduly influenced. Rather than restricting incentive payments to low levels, IRBs can use the same process to address undue influence caused by incentive payments as a unified and reasonable approach to minimize all forms of undue influence.

References

  1. Code of Federal Regulations. Protection of Human Subjects, 45 CFR §46 (January 19, 2017). https://www.hhs.gov/ohrp/regulations-and-policy/regulations/45-cfr-46/index.html
  2. Code of Federal Regulations. Protection of Human Subjects, 21 CFR §50 (April 24, 2001). https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?CFRPart=50
  3. Code of Federal Regulations. Institutional Review Boards, 21 CFR §56 (April 24, 2001). https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?CFRPart=56
  4. International Council for Harmonization. 2015. Guideline for good clinical practice E6(R2). ema.europa.eu/docs/en_GB/document_library/Scientific_ guideline/2009/09/WC500002874.pdf
  5. S. Department of Health, Education, and Welfare. 1978. The Belmont report: ethical principles and guidelines for the protection of human subjects of research. Bethesda, Md.: The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. https://www.hhs.gov/ohrp/regulations-and-policy/belmont-report/index.html
  6. com. https://www.dictionary.com/
  7. Largent E, Grady C, Miller FG, Wertheimer A. 2013. Misconceptions about coercion and undue influence: reflections on the views of IRB members. Bioethics 27:500–7.
  8. Klitzman R. 2013. How IRBs view and make decisions about coercion and undue influence. J Med Ethics 39:224–9.
  9. Office for Human Research Protections. Informed Consent FAQs. https://www.hhs.gov/ohrp/regulations-and-policy/guidance/faq/informed-consent/index.html
  10. S. Food and Drug Administration. Payment and Reimbursement to Research Subjects—Information Sheet. https://www.fda.gov/RegulatoryInformation/Guidances/ucm126429.htm
  11. Lee E. 2019. Our flawed approach to undue inducement in medical research. Bioethics 33(1):13–8. https://www.ncbi.nlm.nih.gov/pubmed/30014475
  12. Gelinas L, et al. 2018. A framework for ethical payment to research subjects. NEJM 378(8):766–71. https://www.nejm.org/doi/full/10.1056/NEJMsb1710591

David Borasky, MPH, CIP, (dborasky@cgirb.com) is Vice President for IRB Compliance at WIRB-Copernicus Group in Durham, N.C.

Jeffrey A. Cooper, MD, MMM, is Vice President for Process and Strategic Improvement at WIRB-Copernicus Group in Washington, D.C.

Kelly FitzGerald, PhD, is Vice President for IRB and IBC Affairs at WIRB-Copernicus Group in Puyallup, Wash.