payments to research subjects

Monash Bioethics Review Vol. 23 No.4 70 Ethics Committee Supplement

ETHICS COMMITTEE REFLECTIONPayments to research subjectsMARTIN Wn.KINSONSchool of Population HealthUniversity of AucklandNew Zealand

ABSTRACTThere is strong opposition in bioethics to paying research subjects.This paper, building on earlier work, gives arguments on behalf ofthe permissibility of payment. It develops an analogy betweenpayment to research subjects. And payment and regulation in thelabour market. Few seriously oppose payment in the labour market,and the reasons to allow payment carry over to payment to researchsubjects. The paper then considers and rejects an allegeddisanalogy, that research is special in that it involves subjects'bodies. At greater length, it assesses the argument from the giftrelationship: that the value of giving is good reason not to extendmarket norms into research. This argument has some force inpointing out the unattractiveness of some market motivations, butthose who offer it usually overlook the coordination advantages ofthe market. Still, the gift relationshp argument against payment hassomething going for it, which is more than can be said for virtuallyall the other major anti-payment arguments.

There is widespread opposition in bioethics to the payment of researchsubjects. Many of the arguments are based on the assumption thatpayments would attract the poor and claim that payments would beunethical for various reasons: that the desperation of the poor wouldmake the offer of money coercive, and hence their consent invalid; thatpayment would make the subjects act against their own interests; thatpoor subjects would be exploited; and that research done primarily onthe poor is inequitable. Other arguments against payments do not relyon the assumption that paid subjects would be poor. Instead, theyclaim that commercialisation has drawbacks: that it would transformthe relation of subjects to research and researchers into a commercialone, rather than a gift relationship; and that payments would drive upthe price of research and crowd out research that is not backed up by alarge budget.

A few years ago, Andrew Moore and I published a pair of papersthat defended in detail the permissibility of payment against all theseobjections, most of which we found to be hopelessly misguided.' One ofthe arguments we used pointed to an analogy between being a subjectin research and working. It is permissible to pay people for work, sowhy not for being in research? This paper develops the analogy further


to make the point that much of the anti-payment argumentation isdecidedly peculiar.

Some initial points, also from the earlier papers: the subjects Iam talking about are competent ones. The research I am talking abouthas passed all the relevant ethics committee hurdles, especially thoseon likely harmfulness. So we are concerned with research that ethicscommittees would allow to be done on subjects who are unpaid, andthe question is whether it should be permissible to pay them.

Consider now the labour market. Like being a research subject,being a worker can be dangerous, safe, uncomfortable, comfortable,boring, interesting, a useful contribution, a pointless exercise, and soon. Like research subjects, workers can be poor, or not, desperate forwork, or not, impulsive or short sighted about risk, or not, and so on .Of course, there are always disanalogies, and I get to a couple ofpurported ones at the end. But the similarities imply that anti-paymentworries that are based on a concern for the poor largely apply to thelabour market too.

Now for the oddity. While most accept that the labour marketshould be regulated, virtually no one calls for an end to payment forwork.P Apart from the fact that it would be unfair, it would obviouslymake people much worse off. Freedom of contract is a way of lettingpeople make deals that improve their lives, and iterated countlesstimes, it makes people a lot better off than they would otherwise be .Occupational freedom is also an aspect of autonomy, of people beingable to choose for themselves what to do .

This is not to say that freedom of contract is sacred. There areusually and quite properly regulations, especially in the labour market.Employers have to take steps to provide safe workplaces. The statepasses minimum wage and maximum hours legislation. Going back along way, the state requires payment in money and prevents paymentin kind. Trades unions are permitted and given legal immunity fromdamages if they call strikes. The state prevents indentured labour bymaking specific performance of labour contracts unenforceable.

But labour regulations are obviously limited. They do not insistthat there be only pleasant and well-paid jobs. They do not even insistthat there be only safe jobs, since they do not ban firefighting, militarywork, police work, or being a steeplejack, a construction worker, aprofessional rugby player, or an astronaut. These jobs carry the risk ofsudden catastrophe and of long-run harmful effects on health.

The reasons are obvious. What some find unpleasant orexcessively risky, others do not, so regulation would be too blunt, givendifferent preferences. More fundamentally, regulation alone cannotprovide all good things and, indeed, overdoing it harms those it aims tohelp, in particular by causing unemployment. (This is why evenenthusiasts for the minimum wage do not favour a $100 per hour rateat 2005 prices.)

There are obvious parallels with research. There are good reasonsfor a system of ethical regulation to protect subjects from risks ofharms. There is a parallel to the rule against enforcement of specific


performance in the right of subjects to withdraw from trials at anypoint. There are informed consent regulations to try to ensure thatsubjects know what they are getting into, something especiallyimportant in the absence of trades unions to negotiate on their behalf.But there is also a role for freedom of contract as a way of improvingpeople's lives, including-particularly-the lives of the badly off andthere are severe limitations on what can be achieved by regulation.

So consider the point sometimes made that payments to the poorwould be wrong because the poor are already victims of injustice. It isindeed an unjust world we live in, but the prohibition of payment tosubjects, like the prohibition of payment to poor workers, would notmake it less unjust. Indeed, in stripping badly off subjects of one of thefew money-making assets they have, prohibition probably makes theworld a less just place.

So much for the analogy between work and research. What nowof the disanalogies? I have seen it said that, unlike dangerous work,the risks of research are not predictable, although I have not seen anyreasons either to believe this pair of claims, nor that it would matter ifthey were true . It is also arguable that the risks of research are notgreat, on the assumption that there is the usual risk-minimizationrequirement of ethical review. Another purported disanalogy claimsthat, unlike with ordinary work, research involves people's bodies, andthere is something special about people's bodies. It is not usuallyexplained why this alleged specialness justifies banning payment.Offhand, one might think that if my body is special, it should be up tome what I do with it. But anyway, the specialness of research in thisrespect is an illusion. Most work involves people's bodies, if onlybecause people's work contracts require them to be somewhere specificwhen they work. In this respect, work is like research where people'sminds are of interest to researchers, rather than their bodily reactions.But, like some research, some work directly involves people's bodies,such as being a model. Should payments to models be banned? Itmight be said that being a model is not risky, but then the point is nolonger one about bodies. It is about risk, in which case the earliercomparison with fire fighting, being a steeplejack etc. , applies.

A different argument develops the theme of the gift relationship toattack the extension of the market.3 When subjects voluntarilyparticipate in research without being paid, they are giving; they actaltruistically in contributing to the common good of research. Tointroduce payment would undermine this gift relationship, and wouldbe an unwelcome extension of market norms into one of the few areasof social life that has remained free of them. Note that this argumentdoes not rely on the claim that payment would disproportionatelyattract the poor. It is an argument against payment to anyone, even therich.

There are many problems with the argument. Many people areeager to participate in research in order to get a last-hope treatment orbetter medical care. While their motives are understandable, this ishardly the gift relationship in action. There is already the payment of


subjects in non-harmful research, such as economic experiments.Perhaps there is not in fact a gift relationship to be undermined, orperhaps the absence of such a relationship in some cases need notundermine it in others. And banning payment in the cases where italready occurs might not recreate a gift relationship.

Let us put these problems aside and ask why there should be agift relationship in research. Should there be a gift relationship in everyarea where there is currently a market? Should jobs not be paid, but bedone for free? Or is research special? To be sure, research has itspublic good aspects, and one might say that altruism is moreappropriate when done for the public good than it would be in ordinarywork, where private employers benefit. But of course, research has itsprivate good aspects and ordinary work has its public good aspects.And to the extent that research is a public good, why should subjectsbe the only givers? One does not see anti-inducement academicscriticising their own salaries for undermining a gift relationship.

One can appreciate better the pros and cons of the giftrelationship by bearing in mind what markets do . There are two majorfunctions that markets perform better than any other mechanism likelyto be devised to run an economy. They provide incentives, based ongreed and fear . And they provide a coordinating mechanism, usingprice, for the vast mass of welfare-improving transactions. In the lightof these two roles, one should be ambivalent about a gift relationship inresearch. There is indeed something horrible about a system based ongreed and fear." It is largely this that justifies the feeling, which I share,that it is unattractive to extend the market into areas of social lifewhere we feel they do not belong. So if there is a gift relation, there is agood reason not to ruin it, and if permitting payments would ruin it ,there would be a reason not to permit them. But note the ifs : perhapspermission would not undermine; perhaps the reason not to permitwould not justify an enforceable ban; perhaps the reason isoutweighed. And, crucially, perhaps there is no significant giftrelationship but only a shortage of subjects, in which case banningpayments would not be justified on gift relationship grounds.

I said there were two roles for markets: providing incentives andcoordination. The need for coordination is an inescapable one and,unlike with paying incentives, the use of a market to meet it is not aconcession to people's shabby motives. The need for coordinationwould make the abolition of the labour market a disaster even if peoplewere entirely altruistic, and the coordination reasons for markets arereasons to offer payment for research subjects." It gives a properaccounting for the value of people's time, discomfort, risk etc. Ifsubjects voluntarily and altruistically underwent considerable sufferingand loss of time and income for the sake of research that did not andwas never expected to do much good, this would be overall a bad thing.The costs would outweigh the benefits, on any view of how costs andbenefits should be weighed and shared. There would be cases wherethe mismatch between cost and benefit would be less obvious, but this


is a reason for payment, and so for the permissibility of payment: itwould make clearer the social cost of doing research.

I have concluded with the gift relationship because it is one ofthe few arguments against the permissibility of payments that has apoint, although I greatly doubt that it would justify a ban on payments.In any case, it is not the most prominent argument. The mostprominent arguments are clustered around the claim that the paid willbe the poor. In a system with ethical review of research, the argumentsbased on this claim are desperately weak.

ENDNOTES

Wilkinson M and Moore A, 'Inducement in research', Bioethics, vol. 11 , no. 5, 1997,pp. 373-89; Wilkinson M and Moore A, 'Inducements revisited', Bioethics, vol. 13,no . 2 , 1999, pp. 114-30. Overlapping and further arguments to similar conclusionscan be found in an excellent recent book by Wilkinson S (no relation), Bodies forsale, London: Routledge, 2003.

2 Some Marxists used to call for the abolition of the labour market, but that was becausethey wanted all markets abolished, and not primarily because of anti-payment-toresearch-subject-type arguments.

3 An idea most famously developed by Richard Titmuss, in his discussion of blooddonation. See The gift relationship, Harmondsworth: Pelican, 1973.

4 Cohen GA, Self-ownership, freedom, and equality, Cambridge: Cambridge UniversityPress, 1995, p . 262.

5 Wilkinson TM, Freedom, efficiency and equality, Basingstoke and New York: Palgrave,2000, esp. ch. 7. One rarely sees arguments like this in bioethics, probably becauseeconomics is not part of the base knowledge of most bioethicists. But since we aretalking about the permissibility of a market, economics is highly relevant.

Monash Bioethics Review Vol. 23 No.4

BOOK REVIEW

75 Ethics Committee Supplement

Ruth Macklin, Double standards in medical researchin developing countries, Cambridge University Press,2004, £22.99, pp. 288, ISBN 0 521 54170 O.Lawrence O. Gostin, The AIDS pandemic: Complacency,injustice, and unfulfilled expectations, Chapel Hill:University of North Carolina Press, 2004, US$35.95, pp.496, ISBN 0 8078 2830 o.

IntroductionOn the face of it, these two books seem to have little direct

overlap. Gostin's book considers the legal, ethical, political and policychallenges raised by HN/ AIDS. His focus primarily lies with Americanpolicy and history. By comparison, Macklin deals with a global issuethe question of ethical standards and exploitation in internationalmedical research. Yet, while there is little commonality in terms ofcontent, the texts complement each other well.

HIV/ AIDS is currently the greatest health threat facing the globalpoor. The burden of HN/ AIDS in developing countries highlights theenormous inequality of disease distribution around the world. Twentythree million people in Africa are now HN positive; in some countries,such as Botswana, 30% of the population is infected. Three millionpeople throughout the world die each year because of HIV/ AIDS. Let ustake a minute to compare this death toll to the devastation caused bythe Tsunami in Asia a few weeks ago. At the time of writing theTsunami was estimated to have killed between 150,000-200,000people. A staggering loss of life that has rightly ignited a sense ofgenerosity and care, not only from the world's governments but alsofrom the world's citizens. By the 6th of January the British public haddonated £100 million in aid, while public donations in Australia hadreached AUS$100m. As the world rushes to the aid of the victims andfamilies of the Tsunami, it makes one regret that the destructionwrecked by HN/ AIDS has not achieved a similar media profile.

Western governments and international agencies have beeninvolved in the fight against HN/ AIDS in the developing world over thelast decade: contributing funds, policy advice and discounted drugs.Nevertheless, one of the most direct and sustained interactions thedeveloped world has had with people living with HIV/ AIDS in thedeveloping world, is through the thousands of international clinicaltrials sponsored by entities such as pharmaceutical companies and theNational Institute of Health (NIH) in America. The ethical issues raisedby such trials are the topic of Macklin's book.


The AIDS Pandemic: Complacency, Injustice, andUnfulfilled Expectations

Gostin has been writing about AIDS since the mid-1980s and hebrings to this book a wealth of experience and knowledge. The AIDSPandemic is based on a compilation of previously published essays byGostin, which have been rewritten and updated . The book is dividedinto five parts: I. AIDS in the courtroom; II. Rights and dignity; III.Policy, politics, and ethics; IV. Special populations; and V. AIDS in theworld. Only the last section adopts a global perspective.

Gostin divides the history of the AIDS epidemic in the UnitedStates into three phases: Denial, Blame, and Punishment (1981-87) ;Engagement and Mobilization (1987-97); and Complacency, Injustice,and Unfulfilled Expectations (1997-present) . It is the third phase thatis primarily the focus of this book. Gostin argues that America, andthe world, have failed to justly respond to the needs of people livingwith HIV/ AIDS. They have failed to eliminate discrimination on thegrounds of serological status, and have failed to capture the potential ofanti-retroviral therapy (ART) by removing barriers to access for themillions of poor people worldwide in desperate and immediate need ofthe drugs.

Double Standards in Medical Research in DevelopingCountries

The debate about international research ethics was sparked by aset of trials, sponsored by the NIH in America, and conductedthroughout Africa and Asia, which tested the effectiveness of low dosesof AZT in reducing perinatal transmission of HIV. Prior to these trialsthe '076 regime' of AZT had been adopted as the accepted standard ofcare in the USA and other developed countries. However, the cost ofthis regime, at US$800 per mother and child pair, was prohibitivelyexpensive for developing countries. Furthermore, social conditions(such as breast feeding due to lack of access to clean water for formula)and lack of health infrastructure (resulting in presentation to medicalservices late in pregnancy and the absence of facilities for providingintravenous drug doses during labour) in developed countries, madethe 076 method of treatment inappropriate for widespread use.Subsequent trials were therefore run to test the effectiveness of lower,and thus cheaper, doses of AZT.! The field of international researchethics was thus born in the context of the global HIV/ AIDS pandemic.

Macklin's book is the first to offer an in-depth analysis ofexploitation and justice in international research. While most of thematerial will not be new to those who are familiar with the debates inthis field, it does offer a thorough overview of the issues. In particular,Macklin provides an insider's perspective on the development ofinternational guidelines within organisations, including the Council forInternational Organisations of Medical Sciences (CIOMS), the WorldHealth Organisation (WHO), UNAIDS, the Nuffield Council and theWorld Medical Association (WMA) , as they struggle to establish a


position that is both ethically justified and practical. Having beendirectly involved with many of the negotiations undertaken by theseorganisations, Macklin is able to directly quote submissions anddiscussion documents that are not typically part of the public dialogueon these guidelines.

Macklin's book addresses the ethical issues of justice,exploitation, informed consent and human rights, and the practicalquestions of differential pricing, collaborative international researchand capacity building. She concludes that it is not ethical for sponsorsfrom rich countries to conduct research on populations in developingcountries that would not be approved in their own countries. Sheargues that this constitutes an unjustifiable double standard.

ConclusionOne of the interesting commonalities between the two texts is the

discussion of human rights and public health, drawing on the work ofthe late Jonathon Mann. Both acknowledge a human right to healthand the association between poverty and ill-health. They see therealisation of a global right to health as an essential step towards theimprovement of the health status of populations in developingcountries, and in particular, of turning the tide against the HIV/ AIDSpandemic.

However, as both authors recognise, a right to health is not apanacea to all difficult questions regarding HIV/ AIDS. The 3bySinitiate launched by WHO and UNAIDS in 2003, with which Macklinwas directly involved, demonstrates some of the challenging issues thatare not resolved by a simple right to health. In late 2003 it wasestimated that six million people in developing countries were inimmediate need of ART, yet only a few hundred thousand of thesepeople had access to the drugs. The initiative hopes to mobiliseresources to get three of the six million in need on ART by 2005. Thequestion arose of how to prioritise the competing demands of differentvulnerable groups. A right to health-care could not answer theseastonishingly difficult and complex questions of equity and justice.

I would recommend both books as insightful and illuminatingoverviews of the respective topics, both written by authors whothemselves have been actively involved in the international debates andwho both bring years of personal experience to their analyses.

ANGELA BALLANTYNECentre for Human BioethicsMonash University

Lurie P and Sidney W, 'Unethical trials of interventions to reduce the perinataltransmission of the human immunodeficiency virus in developing countries', NewEngland Journal ofMedicine, vol. 337, 1997, pp . 853-56.

payments to research subjects

Documents