Journal ofMedical Ethics 1997; 23: 181-185

Compensation for subjects of medicalresearch: the moral rights of patients andthe power of research ethics committeesStephen Guest University College, London

AbstractAwareness of the morally significant distinction betweenresearch and innovative therapy reveals serious gaps inthe legal provision for compensation in the UKforinjured subjects of medical research. Major problems arelimitations inherent in negligence actions and a culturethat emphasises indemnifying researchers beforecompensating victims. Medical research morally requirescompensation on a no-fault basis even where there isproper consent on the part of the research subject. Inparticular, for drug research, there is insufficientprovision in the current patient guidelines of theAssociation of the British Pharmaceutical Industry,since they make "no legal commitment" to payingcompensation for injury to patient subjects. There is aneedfor the provision of both adequate insurance andcontractual arrangements for making payments. Thesolution is for Local Research Ethics Committees(LRECs) to make use of their power to withholdapproval of medical research where compensation is notlegally enforceable.

Definition of research and the generalprinciplesTHE DEFINITION OF RESEARCHWe might begin with the Oxford EnglishDictionary's definition. "Research" is there definedas "An investigation directed to the discovery ofsome fact by careful study of a subject; a course ofcritical or scientific discovery". The importantpoint is that it is accompanied by an intention tofurther knowledge, irrespective of benefit to thesubject. This essentially captures the idea ofresearch used by the Royal College of Physicians(RCP).' The crucial component is the intentionwith which the research is done since there areclearly novel ways in which techniques (or drugs)may be used, especially in cases where there is ahigh risk in any case that a patient will die, and

Key wordsABPI guidelines; negligence actions; LRECs; drugresearch; compensation; innovative therapy; research onpatients;research on healthy volunteers; NHS indemnity.

where it is justifiable to use some less orthodoxtreatment; in such cases, the experimental spin-offis only secondary to the intention to benefit thepatient. Research therefore begins where benefit tothe patient becomes only secondary. It is thereforedifferent from innovative therapy, where the aim is"entirely for the benefit of a particular individualpatient".2 Although not clearly established inEnglish law, Canadian law clearly accepts the dis-tinction: Zimmer v Ringrose.3 As the RCP pointsout, innovative treatment may become part ofmedical practice without becoming the subject offormal research.

THE JUSTIFICATION OF THE BENEFIT TO OTHERSThis is an important justification for energeticallypursuing medical research. But it essentially meansthat research subjects are being used for the benefitof others; a point not to be overshadowed by the factthat they consent. It is unfair to them if they areexpected to bear a disproportionate loss in the caseof mishap. Here it is salutary to keep thalidomide-type tragedies, where the consequences are sodrastic, in mind. To employ sensible philosophicalterminology, research subjects possess rights whichcannot simply be balanced against the generalwelfare of society even where the subjects appear toconsent. This can be seen in the moral futility ofdenying duty to a severely injured research subjectby saying: "you consented to taking the drug; youknew it was a new drug and that its effects were notfully known".A general limitation to consent to harm to yourself

exists in the criminal law and although there is anexception for surgical interventions these have to bejustifiable.4 It would therefore beg the question tosay that all research where consent was obtained wasfor that reason legally justified. It is useful to notethe Helsinki Declaration of the World MedicalAssociation in 1964, speaking for the patient subjectof research on the point: "The physician cancombine medical research with professional care . ..only to the extent that medical research is justified byits potential diagnostic or therapeutic value for thepatient" and (speaking for the healthy subject ofresearch) "the interest of science and society should

group.bmj.com on December 4, 2014 - Published by http://jme.bmj.com/Downloaded from

182 Compensation for subjects of medical research: the moral rights of patients and the power of research ethicscommittees

never take precedence over considerations related tothe wellbeing of the subject". It follows that, becausethere is a risk that research may focus insufficientlyon a research subject's moral rights, injury causedthrough research raises special considerationsdistinct from those applying to ordinary patientcare.5Of course, medical research risks cover a wide

range down to the case where the harm and risk arede minimis. Many trials do not create risks of injury atall (measuring heartbeat or breathing rate), andcompensation will not be necessary where there iseven a high risk of negligible injury (minor short-term painless bruising). In these undramatic casesconsent really does remove any requirement to paycompensation.

HEALTHY (NON-PATIENT) SUBJECTS AND PATIENTSUBJECTSIt is important to draw this distinction for severalreasons. They involve different types of study, differ-ent types of recruitment and, importantly, differentrisks. With seriously ill patients, because of thehigher risks which it seems justifiable to take,research can join more closely with innovative treat-ment to the point where the two merge, perhapsindistinguishably.

In my view, all research must be subject to thesame principles and the fact that the subject is apatient only alters the degree of risk which theresearcher may reasonably request from the subject.See, for example, the statement by the Royal Collegeof Physicians that "the same general principles andarrangements should apply in clinical trials involvingpatients as those described . .. for healthy volunteerstudies".6 This wisely echoes the Pearson Report oncivil liability for personal injury which singled outmedical research as an area of particular concern andtook the view that a patient was in the same positionas a healthy person volunteering to take part inresearch.7

Compensating the injured subjectTHE SPECIAL POSITION OF RESEARCH WITH DRUGSEarly uses of drugs are a special case and remind usof the thalidomide experience. The Medicines Act1968 controls medicines for research on patients.Licences are not required for research on healthyvolunteers. The governing body is the MedicinesControl Agency (MCA) which licenses drugs. Themain categories are product licences, for which vari-ations for new uses may be granted, and clinical trialcertificates (CTCs) from which exemptions for newuses may also be granted (CTXs, or DDXs - exemp-tions granted to doctors and dentists). The CTXscheme is now the usual means of gaining drugclearance and was introduced in 1981 followingcomplaints from the pharmaceutical companiesabout complexity and time delay ("drug lag") in

granting variations to product licences, and inissuing CTCs. For a CTX only a summary of thedata requirements is needed and the fundamentaltest used by the CTX panel acting for the MCAappears to be the negative one of whether on theprovided information the testing of the medicine"would not involve a serious risk to the trial subject".

Nevertheless, drugs are not entirely within aspecial category and their experimental use shouldbe compared to novel surgical and other medicaltechniques. The increasing use of keyhole surgery,the developing use of lasers, sometimes combinedwith photo-sensitive drugs, ultrasound and radio-active isotopes, fall within the same general scope ofresearch. The Department of Health is presentlyconsidering a licensing body and scheme similar tothe MCA for this wider category.

DIFFICULTIES WITH THE TORT OF NEGLIGENCEA primary problem is access to justice through theexpensive and time-consuming adversarial process.But this is only the beginning. In the UK the plain-tiff must establish both that the researcher causedthe injury and that he/she was negligent in doing so.But causation is notoriously difficult to prove inmedical negligence actions because it is relativelyeasy to suggest that other causes were predominant.That is particularly so where a research subject diesshortly after an experiment. No-fault schemes incountries such as Sweden and New Zealand havebeen seriously undermined because under theschemes causation still has to be proved. Indeed,"no-fault" is an unhappy title, since finding fault isalso part of the necessary process of establishingcausation, the test being not only "but for" theresearcher's action the subject would not have beeninjured but that the injury was "reasonably foresee-able in the ordinary course of events". There is onedifference between the ordinary case and medicalresearch although perhaps it should not be overes-timated. It is that an injured subject cannot easilyclaim that a risk was "reasonably foreseeable in theordinary course of events", because the nature ofexperiment is such that a. unforeseeable events areto be expected and b. there is an inherent difficultyin establishing what the "ordinary course of events"actually is.

As far as proving negligence goes, the standard ofcare required causes particular problems in the caseof research into alternative means of treatment; afterall, the researcher is trying by his research to establishnew standards of care. Of more concern since it isthe situation that most frequently arises in practice,is the standard of care relating to the obtaining ofconsent. Without doubt, full consent should beobtained. Even the "therapeutic" privilege in notobtaining full consent, justified on the ground thatthe patient's anxiety might exacerbate hiscondition,8 cannot apply to research subjects. Ifthe patient is in too delicate a state to be told the

group.bmj.com on December 4, 2014 - Published by http://jme.bmj.com/Downloaded from

Stephen Guest 183

consequences of medical intervention he should notbe a subject of research.9The availability of compensation in practice arises

predominantly through two sources; the so-calledNational Health Service (NHS) indemnity andthrough medical insurance associations. As far as theNHS is concerned, recent arrangements protectNHS employees from claims in negligence in per-forming local health authority duties of which, formany, research is one.'0 It should be noted that thetrusts are required to pay any claims out of their ownresources, although there is provision for trusts to begiven loans from the regional health authority forthis purpose. But the NHS does not cover the non-negligent cases other than through the unwieldy anduncertain ex gratia payments which the NHS isempowered to make subject to DoH and Treasuryapproval." It is important to see how the indemnityis viewed by the DoH. The guidelines which intro-duce the new arrangements speak exclusively interms of how payments for claims are to be dividedbetween the health authority and other bodies suchas sponsoring institutional or corporate bodies andnot in terms of compensation to an injured party.Indeed, the culture of thinking about compensationis weighted towards the protection of the researcher,etc, rather than towards the protection of the injuredparty. Evidence, for example, is in the DoH's 1991guidance for LRECs which contemplates thatmembers should not be indemnified for "gross neg-ligence".'2 This is a clear indication that the circulardid not consider compensation for the injuredsubject as the paramount principle because it wouldfollow that if the LREC were grossly negligent therewould not be DoH financial backing and so nothingby way of compensation.

Outside the NHS, institutions such as theMedical Defence Union and a number of other suchassociations (nurses, physiotherapists, dentists, etc)protect their members against events widely charac-terised as "legal liability". The question of compen-sation thereby centres naturally on the extent of theresearchers' professional obligations to their patientsand, where applicable, to their research subjects. Inpractice, the MDU and other associations assumethe responsibility for a member's legal liability andindemnify the member for the costs. As with theNHS, although perhaps more understandably, theemphasis lies on indemnifying the member ratherthan protecting an injured party.

THE CONSUMER PROTECTION ACT 1987The notorious difficulty with establishing liability isthe "state of the art" defence which the act incorpo-rates,'3 a particularly pressing problem when thewhole point of the medical research is to determinefacts which necessarily go beyond what is currentlypart of "the state of the art". In the case where theproduct is "defective", and has caused harm, there isentitlement to compensation under the act. An

advantage is that the liability is strict, that is, is not atfirst sight dependent on fault on the part of theproducer although there is perhaps a fault element inproving that a product is defective. But the well-known drawback with the remedy is the producer'sdefence of development risk. Where the current stateof scientific knowledge at the time the product wasproduced would not have revealed the defect theproducer is not liable. Given that research attemptsto discover what is beyond the current state ofknowledge, the injured research subject is not in astrong position.

COMPENSATION RIGHTS ARISING THROUGHCONTRACTS OR INSURANCEA difficulty is that there is no general requirementupon any person or body to take out insurance toprotect research subjects and the DoH made it clearin 1991 that NHS bodies are not empowered to offeradvance indemnity to participants in researchprojects.'o Some institutions, such as a number ofuniversities and research centres do take out insur-ance but the practice is not by any means universal.For drug research on healthy volunteers, theAssociation of the British Pharmaceutical Industry(ABPI) recommends to its members that no-faultliability "in the event of injury" and on proof ofcausation, should be accepted by the sponsoringcompany as part of a legally binding contract betweenthe researcher (acting as agent for the company) andthe subject.'4

ABPI guidelines for patientsOn the other hand, the ABPI recommends membersto compensate patient subjects of research for "moreserious injury of an enduring and disabling character(including exacerbation of an existing condition)and not for temporary pain or discomfort or lessserious or curable complaints", on a no-fault basisalthough "without legal commitment".'5 Perhapsreasonably they restrict liability to pre-marketeddrugs or marketed medical medicines which arebeing tested for use other than that in the productlicence. Clearly, these guidelines are intended(unlike the healthy volunteer guidelines) not to havelegal force and it would be difficult to establish acontract in the absence of an express agreement withthe patient. Particularly, it would be almost impos-sible to establish the necessary "contractual inten-tion" because of the "without legal commitment"wording. The ABPI has met with criticism of thisprovision but maintains that its members would inreality pay out; that the cases are rare; and that thereare problems inherent in establishing a legal contractwith a patient, such as the compos mentis of thepatient and the special doctor/patient relationship.'6

These three arguments are not particularly strong.Drug injuries can be very expensive to compensate,as the thalidomide tragedy proved, and it is anunsatisfactory position for the injured subject that

group.bmj.com on December 4, 2014 - Published by http://jme.bmj.com/Downloaded from

184 Compensation for subjects of medical research: the moral rights of patients and the power of research ethicscommittees

the research sponsor be faced with balancing costagainst breaching a legally non-enforceable albeitmoral undertaking. It seems conceivable to me that,in some events, a sponsoring company, or itsoverseas controlling company, might well decidethat the costs are simply not worth it. The argumentthat the cases would be rare carries no weight since ifinjury occurs compensation is required; in any case,since the guidelines envisage paying compensation,they must, too, envisage the event of injury. If thethird argument means that there might be patientstoo ill to make a contract, then they will almost cer-tainly be too ill to consent to research. Perhapscarrying more weight is the argument that makingthe researcher a contractor would intrude inappro-priately into the doctor-patient relationship. Butwhy? The doctor must have the patient's bestinterests at heart as part of his general duty toprovide patient care, and it is clearly in the patient'sinterest to have adequate compensation arrange-ments in place. In any case, what is necessary wouldnot be a formal document like a will or a mortgagedeed but a relatively simple addition to the patient'sconsent form.'7

The powers of Local Research EthicsCommitteesThese committees now substantially controlresearch. There are a large number, all of which areguided by a 1991 DoH circular.'8 But the circulardoes not have legal force. Local Research EthicsCommittees broadly cover NHS research withintheir geographical area, and non-NHS research sub-mitted by agreement, and they are to be regarded asindependent ("not beholden to" the NHS bodies).The important (indeed, crucial) point is that "NoNHS body should agree to a research proposalwithout the approval of the relevant LREC" and soapproval can be made conditional upon compensa-tion. 9 It is within the power of a LREC, therefore, tomake its consent to research conditional uponadequate compensation arrangements being in place.The statement in the RCP guidelines that "there islittle that RECs can do" about the unsatisfactory sit-uation regarding compensation is therefore false.20The power to require compensation, too, is fre-quently exercised and evidence for this is thecommon practice ofLRECs to make it a requirementthat sponsoring companies which are not members ofthe ABPI should comply with the ABPI guidelines.

Should there be positive vetting by the LREC ofthe safety of drugs? There is anecdotal evidence thatthe CTX panel takes the view that the onus lies onthe LREC because of the speeded-up test forgranting a CTX certificate.2' Further, the RCPstates that the CTX (or CTC or DDX) status "doesnot imply that the ethics of the investigation havebeen subject to significant review and in no wayreduces the duties of RECs".22

ConclusionsThere should be explicit recognition by all those incharge of medical research of the moral impor-tance of treating compensation of the injuredresearch subject as paramount. In effect, thismeans adopting a principle of no-fault compensa-tion and recognition would veer the indemnityculture towards a compensation culture. AllLRECs should lobby the ABPI for their patientguidelines to be made identical to its healthy vol-unteer guidelines. In the meantime, LRECs shouldrequire that research information sheets containclear information concerning lack of legal rights tocompensation so that the issue becomes clearlypart of the research subject's overall consent.Finally, LRECs should be fully prepared, whererisks are more than negligible, to refuse consentwhen legally enforceable no-fault compensation islacking.

Stephen Guest, BA, LLB, BLitt, PhD, is Reader inLegal Theory in the Faculty ofLaws, University College,London and a Legal Member of University College andHospital Committee (Alpha) on the Ethics of MedicalResearch.

AcknowledgementI thank members of the Medical Ethics and LawConference at University College London inFebruary, 1995, for helpful comments.

References and notes1 Royal College of Physicians. Guidelines on the practice of

ethics committees in medical research involving humansubjects [3rd ed]. London: Royal College of Physicians,1996: para. 6.4.

2 See reference 1: para. 6.5.3 (1981) 124 DLR (3d) 215 (Alberta CA).4 See R v Brown [1994] 1 AC 212: 282 and 266.5 See reference: 1. Also see: Royal Commission on Civil

Liability and Compensation for Personal Injury (the"Pearson report"), cmnd 7054 1978, para 1339; andsee: Neuberger J. Ethics and health care. King's FundInstitute Research Report 13, 1992.

6 See reference 1: para 9.15.7 See reference 5: para 1340.8 See Lord Scarman in Sidaway v Board of Governors of

the Bethlem Royal Hospital, etc [1985] 1 AC 871: 888-9.9 Note the admirably expressed statement of principle in

s 10 of the New Zealand Bill of Rights 1990: "Everyperson has the right not to be subjected to medical orscientific experimentation without that person'sconsent".

10 Department of Health. Local Research EthicsCommittees. London, 1991. OHSG(91)5.

11 See reference 10: para 3.12 See reference 10: para 2.11.13 S4(1)(e) states that it is a defence .. . that the state of

scientific and technical knowledge at the relevant timewas not such that a producer of products of the samedescription as the product in question might be

group.bmj.com on December 4, 2014 - Published by http://jme.bmj.com/Downloaded from

Stephen Guest 185

expected to have discovered the defect if it had existedunder his products if they were under his control . . .".

14 Association of the British Pharmaceutical Industry.Guidelines for medical experiments in non-patient humanvolunteers. London: ABPI, 1990. (1988, revised 1990).

15 Association of the British Pharmaceutical Industry.Clinical trial compensation guidelines. London: ABPI, 1991.

16 In correspondence with the UCL Research EthicsCommittee.

17 I am grateful to Professor Aubrey Diamond QC for thispoint.

18 Following an EC directive: EC (1991). Updated stan-dards for the testing of medicines for human use.91/507/EEC.

19 But see Marshall T, Moodie P. Scrutinising research

ethics committees. British Medical Jrournal 1988; 297:753. They say that there is no "absolute" requirement.In the sense that there is no statutory requirement, thisis true, but there are duties of employment arising fromDoH circulars, letters of intent and other evidence ofestablished practices which are reasonably sufficient toimpose the requisite legal duty. It is that those hospitaladministrators who are ultimately responsible for givingpermission for medical experiments, especially in allo-cating resources, should first ensure that approval hasbeen given by the local ethics committee.

20 See reference 1: para 9.3.21 I thank my research student, Caroline Wilson, for this

point.22 See reference 1: para Al.1 5.

group.bmj.com on December 4, 2014 - Published by http://jme.bmj.com/Downloaded from

committees.the power of research ethics moral rights of patients andof medical research: the Compensation for subjects

S Guest

doi: 10.1136/jme.23.3.1811997 23: 181-185 J Med Ethics 

http://jme.bmj.com/content/23/3/181found at: Updated information and services can be

serviceEmail alerting

right corner of the online article. cite this article. Sign up in the box at the top Receive free email alerts when new articles

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:

group.bmj.com on December 4, 2014 - Published by http://jme.bmj.com/Downloaded from