Proposed "lottery" or deferred payment model aims to resolve undue inducement.
A question that arises in Phase I research among healthy volunteers is how much to pay the participants. The sum must be sufficient to attract and fairly reward them, but not such as to unduly induce participation. Here I put forward a suggestion for a lottery type payment model, which by both deferring and subjecting part of the reward to further contingent events attempts to mitigate undue inducement. But first, some background.
(BRIAN HAGIWARA/GETTY IMAGES)
CIOMS/WHO's international ethical guidelines for biomedical research involving human subjects1 notes the following:
"Subjects may be reimbursed for lost earnings, travel costs, and other expenses incurred in taking part in a study; they may also receive free medical services. Subjects, particularly those who receive no direct benefit from research, may also be paid or otherwise compensated for inconvenience and time spent. The payments should not be so large, however, or the medical services so extensive as to induce prospective subjects to consent to participate in the research against their better judgment ("undue inducement") (emphasis added).
Undue inducement is thus a level of inducement that persuades someone to engage in research against their better judgment.
Emanuel et al. identified four essential features of undue inducement: (1) an offered good—individuals are offered something that is valuable or desirable in order to do something; (2) excessive offer—the offered good must be so large or in excess that it is irresistible in the context; (3) poor judgment—the offer leads individuals to exercise poor judgment in an important decision; (4) risk of serious harm—the individual's poor judgment leads to sufficiently high chance that they will experience a harm that seriously contravenes his or her interests.2
Where poor judgment means people can be said to be reckless about their understanding of a trial, participating even when confused about the protocol just to get the money, or where the money leads them to misrepresent their medical history or habits in order to overcome any exclusion criteria, then both they and the trial are placed in potential jeopardy and the inducement has become undue. Instances of such events occurring have been reported.3,4,5 Undue inducement thus blinds those affected to the risks they might face, such that it obviates their properly informed consent and renders their engagement as not brave or courageous—but foolhardy.
The payment model I've created is aimed at healthy human volunteers (Phase I clinical trials), who do not anticipate a benefit from engaging in the research and are the group that are virtually always paid for their participation.
Dickert and Grady6 having discussed three plausible models of payment for research subjects—the market model, the wage-payment model, and the reimbursement model—concluded that the wage-payment model was the most appropriate method of paying participants.
Like the market model though, the wage-payment model is potentially susceptible to undue inducement, and both models can also accommodate completion bonuses. Such bonus payments are not uncommon, but can be criticized because they could both interfere with a subject's decision to withdraw at any time and also dissuade subjects from reporting side-effects or adverse reactions for fear that they be discontinued from the study. Completion bonuses can thus become aids to undue inducement.7
Wilkinson and Moore have almost argued against the existence of undue inducement: "Our question is this: If people are allowed to volunteer for no pay, why not for some pay?"8
Brody made a similar point: "If the independent review panel has already concluded that the risk-benefit ratio of the research is acceptable (otherwise, the research could not be approved), how can the large payment harm the subject?"9
And Emanuel concurred: "A research trial that otherwise fulfils the fundamental ethical requirements for human subjects research inherently cannot create the possibility of undue inducement because substantial risk is precluded."10
I disagree. To take on Emanuel, while he goes on to acknowledge three objections to his claim, his responses do not convince that undue inducement is an almost mythical notion. Firstly, he acknowledges, "there may be particular individuals who fulfil the eligibility criteria for whom participation is nevertheless excessively risky." However, he promptly brushes this aside by claiming that ethical "review of any research trial is not meant to evaluate the ethics based on special interests or idiosyncrasies of particular individuals" and suggests a role for the would-be subject's personal physician in counseling for or against enrollment in a particular trial. This may not be a practical suggestion given that not all his fellow (U.S.) citizens can afford personal physicians, least of all those for whom the burden of clinical research traditionally falls.
A second objection recognizes that Phase I studies might pose net risks to the participants, but not excessive ones:
"In such trials, high incentives might lead people to make bad judgments and enrol in trials that pose serious harms…[however, in such] trials, IRB review provides a fail-safe check ensuring that an individual who might exercise bad judgment from high incentives is still prevented from excessive risks (emphasis added)."
In this, of course, he admits that people do exercise bad judgment on the basis of high levels of payment.
The final objection—that "high inducements cloud people's ability to understand and appreciate the information provided"—he dismisses with "There are no data."
However, data do exist. In the TGN1412 trial of 2006 at Northwick Park Hospital, London, the participants considered the inducement offered "high inducement" ("All claimed that their motivation for taking part had been the money"11 ) and seemed not to have understood the science. Their attitude appears to have been to just get on with things. If this is so, then arguably they were reckless. Subsequent testing of the participant information sheet used in that trial on members of the public has certainly suggested its "performance…would raise a question about the extent to which a participant in this Phase I trial would have been adequately informed before giving consent."12
One of the surprising outcomes of TGN1412 was the almost immediate increase in the numbers expressing an interest in joining clinical trials—the sums on offer as had been revealed were just too attractive. Given this, the level of payment involved in the TGN1412 experiment seems to have amounted to "undue inducement" for many.
Emanuel10 believes that payment levels can be disregarded by ethics committees (ECs) because if the clinical trial is safe, then the money will not make it less safe. This has a logical attractiveness to it, but common sense tells us that higher levels of payment might tempt increasing numbers of individuals to take on too many risks, or be reckless about their understanding, in order to obtain the money. While the EC may have decided the risk-benefit ratio is favorable, this assumes the protocol is adhered to. But if the money is so great as to tempt some to breach the protocol, then the benefit-risk profile alters.
For example, in the past, some individuals have enrolled in more than one trial at a time, and others have commenced a subsequent trial prematurely—and have done so solely for the money. Although ECs may have approved the research, this is likely to have included stipulations that only persons who do not meet exclusionary criteria are allowed to participate. If temptingly large sums of money are on offer, some individuals, especially those who might regard themselves as narrowly outside the eligibility criteria, may misrepresent their age, medical history or substance usage in order to inveigle themselves into a trial.
A new proposal for a deferred payment model for use in Phase I trials, outlined here, could help minimize the chances of undue inducement.
The deferred and uncertain payment model (or "moral luck" model) is for reimbursement of expenses together with a thank you element derived as a fraction of the norm for the trial based on current methods of calculation under the wage-payment model, with the addition of one further element. This additional element may encourage comparison with a lottery, but it is important to note that this is only a component of the model. If, as the evidence from the TGN1412 trial suggests, that €2000 ($3200 USD) is broadly typical of a Phase I trial of its length and complexity, then perhaps a quarter of that payment (€500 or $800) would cover the reimbursement of expenses and a reasonable thank you sum appropriate as one component of the proposed model. Certainly if €2000 was a level of payment that unduly induced participation in the TGN1412 trial, it was more than an appropriate sum to offer, at least to those who could not understand what they were consenting to.
The thank you payment is necessary because otherwise there would be no inducement to enter a trial. Classic, rationalist, economic theory would also recognize that one generally prefers a definite payment against a possible but uncertain chance of future money (a point that has also been demonstrated in a study looking at the response rates to differently incentivized surveys13 ). There thus needs to be an effective inducement, but it must not become an undue inducement. The thank you payment is at a level that acts as an inducement while not being an undue one: It covers the basics. It is, at present, a wholly theoretical sum. The actual figure should be set at a level that research in a particular location indicates can attract sufficient numbers, but no more.
Economic modelling and/or empirical research will undoubtedly be needed to arrive at the appropriate amount, but once established this sum could even become an industry-wide base for trials commenced in that year.
For the additional element, I am proposing a potential future payment, which, if paid at all, will be made some years after the participant's direct involvement in the trial has ended, and be dependent on whether the drug is ever successfully launched. This future payment could reflect perhaps the increase in the share price of the company concerned (it being considered practically too difficult to disaggregate the effect of the new drug).
For illustrative purposes, let us assume that the share price at the time of the participant's entering the Phase I trial was €15 or $24, and the share price at launch (or later) was €33 or $53; the difference of €18 or $29, multiplied by the sort of amount that the wage-payment model would have predicted at the time of the trial, should become payable. In this example the payment would be €18 x [€2000 - €500] = €27,000 or $29 x [$3200 - $800] = $69,600. As the value of shares can go down as well as up, there may in fact be no ultimate payment, and knowledge of this should quash any undue inducement—as might the fact that so few drugs go from Phase I to a successful launch, and of those that do, the average waiting time can be 10 or so years.
There is no quick buck to be obtained here—indeed no further payment is guaranteed. This is still fair though because participants are made aware of this very real possibility when they enroll, and they are properly compensated by the thank you payment, which itself is couched in relation to a level of inducement that manages to attract an adequate but not excessive number of participants.
While in theory the share price could rise astronomically, this unlikely potential should not require ECs to seek any cap on the future payment, as this will only rise to the same extent as the share price. Any jackpot-sized windfall is thus remote—but even then would neither be undue (the participant could have invested in the shares anyway) nor an inducement (as it could not be known at the time of original participation). The tax liability of the participant will depend on how a jurisdiction's tax authorities view the windfall. However, because of the remoteness of the windfall the question of taxation should not affect the scheme.
The point is that the participants are not expecting any windfall. It might happen, but no one can expect it, and something like moral luck determines its payment. The sponsor company will not suffer financially because at no stage should their outlay be any more than it is at present.
Participants who withdraw, or who are withdrawn, for good reason should be eligible for the deferred payment in proportion to their involvement, in a similar way to how they are currently given pro-rata payments.
Should a small privately owned or a nonprofit concern be developing the drug, then the money that might, in this model, be used to buy shares, could instead be invested in other shares—perhaps in a pharmaceutical sector shares' tracker. If the drug of interest was for an orphan disease or for some other reason is unlikely to be profitable, then this fact need not be a hindrance to the proposal either, as it is the performance of the company (or industry) that will determine eventual payment. However, perhaps the model will be thought more useful in some situations than others.
A deferred and uncertain model of payment for research participants may offer the prospect of eliminating any undue inducement that may mar existing practices. The model, based on a drug's success (and not merely as a lottery among trialists), coupled with the price of shares is put forward as a starting point for debate and subsequent development. Such a model, when compared with the existing models, should not tempt volunteers to participate in trials where they do not fully appreciate the risks involved.
Stephen Humphreys is Lay Member for the Welwyn Clinical Pharmacology Ethics Committee, Faculty of Health & Human Sciences, The University of Hertfordshire, Hatfield Campus, Hatfield, AL10 9AB, email: email@example.com.
1. CIOMS/WHO International Ethical Guidelines for Biomedical Research Involving Human Subjects (CIOMS, Geneva, 2002).
2. E.J. Emanuel, X.E. Currie, A. Herman, "Undue Influence in Clinical Research in Developing Countries: Is it a Worry?" The Lancet, 23 (366) 336-340 (2005).
3. C.L. Tishler and S. Bartholomae, "Repeat Participation among Normal Health Research Volunteers," Perspectives in Biology and Medicine, 46 (4) 508-520 (2003).
4. C. Grady, "Vulnerability in Research: Individuals with Limited Financial and/or Social Resources," Journal of Law, Medicine & Ethics, 37, 19-27 (2009).
5. R. Dresser, "First-in-Human Trial Participants: Not a Vulnerable Population, but Vulnerable Nonetheless," Journal of Law, Medicine & Ethics, 37, 38-50 (2009).
6. N. Dickert and C. Grady, "What's the Price of a Research Subject? Approaches to Payment for Research Participation," New England Journal of Medicine, 341 (3) 198-203 (1999).
7. R.W. Grant and J. Sugarman, "Ethics in Human Subjects: Do incentives matter?" Journal of Medicine and Philosophy, 29 (6) 717-38 (2004).
8. M. Wilkinson and A. Moore, "Inducement in Research," Bioethics, 11 (5) 373-389 (1997).
9. B.A. Brody, The Ethics of Biomedical Research: An International Perspective (Oxford University Press, Oxford, 1998).
10. E.J. Emanuel "Ending Concerns About Undue Inducement," Journal of Law, Medicine & Ethics, 32, 100-105 (2004).
11. H. Biggs, Healthcare Research Ethics and Law: Regulation, Review and Responsibility (Routledge-Cavendish, London, 2010)
12. P. Knapp, D.K. Raynor, J. Silcock, B. Parkinson, "Performance-based Readability Testing of Participant Materials for a Phase I Trial: TGN1412," Journal of Medical Ethics, 35, 573-578 (2009).
13. C.M. Ulrich, M. Danis, D. Koziol et al., "Does It Pay to Pay? A Randomized Trial of Prepaid Financial Incentives and Lottery Incentives in Surveys of Nonphysician Healthcare Professionals," Nursing Research, 54 (3) 178-183 (2005).