After Helsinki: Unresolved Issues in International Research Ruth Macklin

Kennedy Institute of Ethics Journal, Volume 11, Number 1, March 2001, pp.17-36 (Article)

Published by Johns Hopkins University PressDOI:

For additional information about this article

No institutional affiliation (3 Jul 2018 20:46 GMT)

https://doi.org/10.1353/ken.2001.0005

https://muse.jhu.edu/article/18670

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 17 ]Kennedy Institute of Ethics Journal Vol. 11 No. 1, 17–36 © 2001 by The Johns Hopkins University Press

Ruth Macklin

After Helsinki: Unresolved Issues in InternationalResearch

ABSTRACT. Following a long process of revision, a new version of the Declara-tion of Helsinki was approved by the World Medical Association in 2000. Twoprovisions of the Declaration address ongoing international controversies regard-ing research sponsored by industrialized countries and conducted in developingcountries. Despite the issuance of the final version of the Declaration, opponentsremain locked in debate. Moreover, the Declaration remained silent on otherprominent controversies concerning international research. An analysis of thesecurrent controversies reveals reasons why they are not likely to be readily re-solved, despite apparent agreement by opponents on overarching ethical principles.

AREVISED VERSION OF THE Declaration of Helsinki was fi-nally approved by the World Medical Association in October2000, well over two years after the process of revision began.

But it would be a mistake to think that the newly revised Declaration laysto rest a spate of ongoing controversies and unresolved issues in interna-tional research, especially studies that an industrialized country sponsorsor conducts in a developing or resource-poor country. One reason whythe revised Declaration has not accomplished that task is that opponentsin the fiercest controversies over key provisions remain locked in debate.

Probably the two most contentious points are embodied in paragraph29: (1) “The benefits, risks, burdens and effectiveness of a new methodshould be tested against those of the best current prophylactic, diagnos-tic, and therapeutic methods.” And (2) “This does not exclude the use ofplacebo, or no treatment, in studies where no proven prophylactic, diag-nostic, or therapeutic method exists” (World Medical Association 2000).A close runner-up for contentiousness is the statement in paragraph 30:(3) “At the conclusion of the study, every patient entered into the study

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 18 ]

should be assured of access to the best proven prophylactic, diagnosticand therapeutic methods identified by the study.” In analyzing what is atstake in these ongoing debates, I argue that opponents are not likely toreach a resolution because of their adherence to fundamentally differentand incompatible premises.

A second reason why the Declaration, even in its current revision, can-not resolve ongoing controversies is that it simply does not address otheraspects of international research about which people disagree. Includedin this category are: (4) the question of what, precisely, is owed to thecommunity or country where the research is conducted after the trial isover; and (5) what specific mechanisms for prior review of research pro-tocols and monitoring of studies already in progress can best protect theresearch participants. In seeking to identify the underpinnings of the de-bates surrounding these points, I will argue that like the earlier items,these are unlikely to be easily resolved.

I intend to spare readers of this article any further analysis of the pla-cebo-controlled, mother-to-child HIV transmission studies conducted inThailand, Uganda, and other developing countries in the late 1990s (al-though of necessity, I will refer to that controversy in what follows). Thoseclinical trials may well have surpassed the infamous Tuskegee study in thenumber of articles that have described, analyzed, and argued about them(see, e.g., Angell 1997; Lurie and Wolfe 1997; Varmus and Satcher 1997;Annas and Grodin 1998; Crouch and Arras 1998; Grady 1998; Glantz etal. 1998; C. Levine 1998; Resnik 1998; Lie 1998; Schüklenk 1998; Tho-mas 1998; Del Río 1998; Brennan 1999; Benatar and Singer 2000; Greco2000; Schüklenk and Ashcroft 2000; London 2000; Rothman 2000). Iwill begin by identifying several premises on which opponents in thesevarious controversies appear to agree.

First, there is widespread agreement that research should be responsiveto the health needs of the population in the country or region where theresearch is conducted. Second, research is needed in developing countrieson common and serious diseases that rarely occur in industrialized coun-tries–e.g., malaria–and on those that pose a huge risk to life and healtheven if the same diseases do exist in industrialized countries–e.g., HIV/AIDS, tuberculosis. Third, it is unethical to exploit vulnerable popula-tions or individual participants in the conduct of research. Fourth, it isunacceptable to lower the ethical standards adopted in the industrializedworld when carrying out research in developing or resource-poor coun-tries. Despite the apparent consensus on these fundamental propositions,

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 19 ]

once we peel off the layer of harmony we confront unanswered ques-tions, as well as stark disagreements.

RESPONSIVENESS TO THE HEALTH NEEDS OF THE POPULATION

How should the first premise on which there is ostensible agreement beinterpreted? Is research “responsive” to the health needs of the popula-tion just so long as it addresses a health problem that is prominent in thecountry or region? Or must some steps be taken before the research isinitiated to seek to ensure that successful products are made available tothe population at the conclusion of the research? Opposing replies tothese question point to one of the gaps in the revised Declaration of Helsinkiidentified above: (4) the question of what is owed to the community orcountry where the research is conducted after the trial is over. The 2000revision of the Declaration contains a provision not included in the ear-lier versions. Paragraph 19 states: “Medical research is only justified ifthere is a reasonable likelihood that the populations in which the researchis carried out stand to benefit from the results of the research.” Althoughthe provision as stated is laudable, it leaves wide open what are the crite-ria for determining “likelihood,” and what degree of likelihood is neces-sary. This is precisely the point on which opponents disagree.

The CIOMS International Ethical Guidelines for Biomedical Research(1993), also undergoing revision at the present time, contain two sepa-rate provisions that address this point:

Guideline 8: As a general rule, the sponsoring agency should ensurethat, at the completion of successful testing, any productdeveloped will be made reasonably available to inhabitants of the underdeveloped community in which the re-search was carried out. (CIOMS 1993, p. 26)

Guideline 15: As a general rule, the sponsoring agency should agree inadvance of the research that any product developed throughsuch research will be made reasonably available to the inhabitants of the host community or country at the comple-tion of successful testing. Exceptions . . . should be justi-fied and agreed to by all concerned parties before the re-search begins. (CIOMS 1993, p. 45)

A discussion paper issued by the Nuffield Council on Bioethics (1999)says of these provisions in the CIOMS Guidelines that “they may be dif-ficult to follow in practice.” The paper observes that such research isgenerally undertaken without any guarantee that the treatment in ques-

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 20 ]

tion would be provided to the participants in the event of a possible out-come. An obvious response to this observation is that the province ofethics is not simply to describe what generally occurs but rather, to pre-scribe what ought to take place. The Nuffield discussion paper was areport on a conference sponsored by the Council in February 1999 inLondon. At the conference, some participants contended that the “rea-sonable availability” provisions in the CIOMS Guidelines are too strong,as they would either prevent research from going forward or lead to in-tolerable delays if made a condition of initiation of the research. Otherparticipants took an opposing view, saying the provision is too weak as itfails to guarantee availability of the successful products of research to thepopulation. One researcher from Chile cited the example of Norplant, aproduct researched and developed in her country well over a decade ago,but that still is not generally available to Chilean women.1

One prevailing view holds that as long as research projects investigatehealth problems of the population from which subjects are drawn, and ifthe study adheres to proper ethical rules for the conduct of research in-volving human subjects, then sponsors or researchers have no additionalobligation to ensure availability of any resulting products. Others replythat failure to ensure in advance that products are made reasonably avail-able in a timely manner to the participants and others in the communityor country is a violation of distributive justice, and could constitute areason not to embark on a trial at all (Crouch and Arras 1998; Glantz etal. 1998; Page [unpublished] 2000). Opponents agree on the underlyingpremise that research must be responsive to the health needs of the popula-tion; yet they disagree in their interpretation of what that premise entails.

Opponents in the much discussed placebo-controlled, HIV maternal-to-child transmission studies also agreed on the fundamental premise yetdisagreed over whether an alternative trial design would be responsive tothe health needs of the population. Defenders of the placebo-controlledstudies argued that to compare the shorter, cheaper experimental regimenwith the treatment that is the “gold standard” in the United States andWestern Europe would fail the “responsiveness” test, since the costly, muchmore complex preventive regimen could never be implemented in resource-poor countries or those lacking the health care infrastructure to adminis-ter the treatment (Levine 1999). Critics could reply that the alternativeresearch design was nonetheless responsive since the “short-course” regi-men being tested was the one that could be made available (which has

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 21 ]

occurred in Thailand, but not in Uganda, for which even the cheapertreatment remained too costly).

It is evident that the substantive views held by opponents in both de-bates—what is owed to research subjects during a trial, and what is owedto the country or community after a trial is completed—determine justhow they interpret the uncontroversial premise that research must be re-sponsive to the health needs of the population in which the studies areconducted. Beyond that lies a deeper question about the nature of ethicalguidelines. Should they be “pragmatic” or “aspirational”? Adherents ofthe view that statements such as the Declaration of Helsinki and docu-ments like the CIOMS International Guidelines must be “pragmatic” arelikely to rely on current and past practices as a guide to what is possible.The pragmatists dismiss “aspirational” guidelines as too lofty and, there-fore, unrealistic (Grady 1998; Levine 1999). For their part, the“aspirationists” tend to be reformers who judge past or current practicesto be ethically insufficient to ensure that the highest standards for re-search apply everywhere, not just in wealthy, industrialized countries(Glantz et al. 1998; Greco 2000; Schüklenk 1998; Rothman 2000).

NEEDED RESEARCH IN DEVELOPING COUNTRIES

Agreement is virtually universal that research is needed in developingcountries on common and serious diseases that rarely occur industrializedcountries and on those that pose a huge risk to life and health even if thesame diseases do exist in industrialized countries. Yet despite that consen-sus, sharp disagreement exists on a range of specific questions: Must theresearch design of a clinical trial demonstrate that a proposed treatment isbetter than nothing in resource poor countries that lack a current, standardtreatment available in wealthier countries? Or would an equivalency trialbe satisfactory, showing that an experimental treatment is as good, oralmost as good, as the current treatment used in the wealthier country(Lurie and Wolfe 1997; Lie 1998; Brennan 1999)? If studies with provenbenefits have been conducted in one or more countries, when does it be-come ethically unacceptable to repeat those studies in still another coun-try, instead of providing the proven treatment to the population in thecountry where no trials have taken place? Are observation studies of the“natural history” of diseases ethically permissible in resource-poor coun-tries where effective treatments are not affordable to the majority of thepeople, even when effective treatments for that same condition are avail-able in the country sponsoring or conducting the research (Rothman 2000)?

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 22 ]

Answers to the above-noted questions are provided in paragraph 29 ofthe 2000 revision of the Declaration of Helsinki: “The benefits, risks,burdens and effectiveness of a new method should be tested against thoseof the best current prophylactic, diagnostic, and therapeutic methods.This does not exclude the use of placebo, or no treatment, in studies whereno proven prophylactic, diagnostic, or therapeutic method exists.” As weshall see below, this answer is not without ambiguity. Moreover, despitethe apparent clarity with which this proposition is stated, critics havedetermined that it may warrant a rejection of the standing of the Declara-tion of Helsinki. Participants in a conference held at the National Institutesof Health (NIH) during the last week of November 2000 debated theDeclaration’s ban on the use of placebos in research except when no proventreatment exists. Robert J. Temple, director of medical policy at the Food andDrug Administration’s Center for Drug Evaluation and Research and along-standing defender of placebo-controlled trials, said that the revisedDeclaration of Helsinki takes an absolute position. “It would bar usingplacebos, for example, to evaluate a new drug for hay fever or migraine. Buttesting such a medicine against another drug approved to treat such condi-tions generally doesn’t measure its effectiveness as well as testing it against aplacebo” (Okie 2000, p. A03). Robert J. Levine, who has argued strongly forabandonment of the “best proven treatment” provision in the earlierDeclaration of Helsinki (Levine 1999), was quoted as saying: “Forbiddingthe use of placebos rules out development of all new therapies for condi-tions for which there are proven therapies. . . . If researchers had followedsuch rules in the past, he said, drugs currently used to treat high bloodpressure or stomach ulcers never would have been developed because ofthe existence of older, less-effective treatments” (Okie 2000, p. A03). Levineand Temple were countered by Kenneth J. Rothman, who defended theDeclaration of Helsinki’s position on placebos, saying the interests of theindividual patient always should supersede the goals of science or society.

In a telling comment, Levine said: “The United States and most othercountries have been ignoring the Declaration of Helsinki for years. I don’tknow why people think this revision . . . is going to bring about a greatchange in behavior.” Change of behavior aside, Levine’s comment promptstwo questions: first, why has the United States (presumably, U.S. researchersand IRBs) been “ignoring” the Declaration of Helsinki; and second, if thatis the case, should we conclude that the United States and other countrieshave been unethical in so doing or instead, agree with Levine and Templethat there must be something wrong with the Declaration of Helsinki?

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 23 ]

If it is true–and it probably is–that the United States has been ignoringthe Declaration of Helsinki, it may well be because the Federal Regula-tions governing research involving human beings are so detailed in theirprocedural aspects that researchers and IRBs see no need for another,different set of ethical rules for research. Furthermore, the U.S. regula-tions are just that–official regulations promulgated by a federal agency,with enforcement mechanisms and sanctions for noncompliance. The Dec-laration of Helsinki, on the other hand, is simply a “declaration,” issuedby an international association of medical professionals, which has noenforcement mechanisms and no sanctions for noncompliance. The UnitedStates has declined to become a signatory even to international instru-ments issued by the United Nations—human rights documents like theCovenant on Economic, Social, and Cultural Rights and the Conventionon the Elimination of All Forms of Discrimination Against Women. So itis hardly surprising that researchers, ethical review bodies, and govern-mental agencies do not consider the Declaration of Helsinki to be a neces-sary adjunct to the “official” Common Rule, which governs most feder-ally funded research in the United States. While the U.S. federal regulationsare almost entirely procedural in their provisions, the Declaration ofHelsinki tends toward a more substantive approach to ethical principlesgoverning human subjects research.

What, then, should we conclude about whether the United States isbehaving unethically in ignoring the Declaration of Helsinki? It is obvi-ous that critics of paragraph 29 believe that the “best current” provisionand the rejection of placebo controls for the vast majority of clinical trialsare unreasonable, if not hopelessly aspirational, and therefore, it is notunethical to ignore them. Defenders of those provisions would concludethat U.S. researchers and IRBs are remiss for not adhering to the Declara-tion of Helsinki in addition to complying with the U.S. federal regula-tions. Once more, people’s views on the substantive ethical issues at stakedetermine how they view the relevance of the Declaration of Helsinki toresearch conducted in the United States or elsewhere in the world.

Here again we face the question of whether ethical guidelines shouldbe “aspirational” or “pragmatic.” Those who refer to guidelines as“aspirational” imply that they are impossibly ideal, not able to be real-ized in practice; whereas “pragmatic” conveys the sense that the guide-lines are truly usable in the practical world. The connotations of theseterms implies the answer: pragmatic is better than aspirational. But thereis another way to describe the nature of guidelines. Are they prescriptive,

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 24 ]

stating what ought to be the case even if current practices fall quite short?Or are they merely descriptive of what is, in fact, done most of the time?The quotation cited earlier from the Nuffield Council on Bioethics dis-cussion paper illustrates the confusion between these rather obviouslydifferent activities of prescribing and describing.

In sum, then, agreement about the need for research in developing coun-tries is a thin reed compared to the wide disagreement about acceptableresearch designs in attempts to fulfill that need.

EXPLOITATION OF VULNERABLE POPULATIONS

The third point of clear agreement is on the proposition that is unethi-cal to exploit vulnerable populations or individual participants in the con-duct of research. No one is in favor of exploitation, so it is easy to cometo apparent agreement on this point. Yet, here again, agreement evapo-rates once we examine various allegations in specific cases. With regardto the obligation following successful trials, discussed earlier, some com-mentators are unequivocal in alleging that exploitation occurs if success-ful products are not made “reasonably available.” For example, “. . . ifthe results of a clinical trial are not made reasonably available in a timelymanner to study participants and other inhabitants of a host country, theresearchers might be justly accused of exploiting poor, undereducated sub-jects for the benefit of more affluent populations of the sponsoring coun-tries” (Crouch and Arras 1998, p. 29). And another, “[i]f developed coun-tries use inhabitants of underdeveloped countries to create new productsthat would be beneficial to both the developed and the underdevelopedcountry, but the underdeveloped country cannot gain access to the prod-uct because of expense, then the subjects in the underdeveloped countrieshave been grossly exploited” (Glantz et al. 1998, p. 39).

The specific example of exploitation cited by the latter commentatorsis that of the short-course AZT studies in Africa. Unlike the situation inThailand, where the Thai government had made a commitment to pro-vide the AZT before the trials began, no such discussion apparently tookplace in the African countries where even the short-course regimen turnedout to be too costly to provide in Uganda. Yet those who maintained that theUganda trials were ethically defensible surely would reject the contentionthat any exploitation was involved. The defense against this allegation ofexploitation turns out to be a procedural one: Ugandans reviewed andapproved the studies so they must have been ethically acceptable to Ugandaand, therefore, no exploitation was involved. Drs. Danstan Bagenda and

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 25 ]

Phillipa Musoke-Mudido of Uganda were quoted in the Washington Postas saying: “The ethical issues in our studies are complicated, but theyhave been given thought by the local community, ethicists, physicians,and activists. . . . We are suspicious of those who claim to speak for ourpeople, yet have never worked with them” (Grady 1998, p. 37).

There remains a profound uncertainty about the ethical sufficiency ofthe procedural steps that involve in-country involvement in the design,approval, and implementation of research. One article defending the pla-cebo-controlled AZT trials quoted from a letter written to NIH by Ed-ward K. Mbidde, chairman of the AIDS Research Committee of the UgandaCancer Institute. In his letter, Mbidde vigorously defended the IRB hechairs in Uganda, stating that “It is a wrong assumption that we do nothave the vision to deal with such issues” (Letter from Edward K. Mbiddeto Dr. Harold Varmus, then Director of NIH, 8 May 1997).2 Pulling out atrump card in ethical argumentation, Mbidde charged that it is “ethicalimperialism” for outsiders to dictate to Ugandan researchers and IRBswhat sort of research is ethical or unethical for Ugandans to carry out ontheir own people. Despite the unfortunate use of this inflammatory accu-sation, Mbidde raises a crucial issue for international collaborative re-search: How adequate are the protections for research subjects in devel-oping countries? A procedural defense prompts the obvious questionwhether a charge of exploitation can be defeated simply by pointing outthat persons from the developing country where the research took placehad a role in designing, approving, and conducting it.

Although relatively little hard data exists, there appears to be a widevariation among developing countries, and even within the same country,regarding both the existence and application of ethical laws, regulations,or guidelines, and the existence and quality of ethical review committeesat the institutional or national levels. Among the countries known to haveregulations or guidelines requiring prior ethical review of research by anindependent committee are Uganda, India, Nepal, Thailand, Zimbabwe,Zambia, and South Africa. Less is known about the actual operation ofthese committees—their membership requirements, terms of reference,and operating procedures. A report commissioned by the National Bio-ethics Advisory Commission (NBAC) included a recommendation thatethics boards in countries where the U.S. sponsors or conducts researchshould gain additional experience in ethics. The report states that “[m]anyU.S. researchers voiced concerns that host country boards had little fa-miliarity with ethics. Thus, when ethics boards convened, they focused

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 26 ]

on other issues where they felt more comfortable, such as the scientificdesign or the budget” (Kass and Hyder 2000, p. 7).

Review and approval of a study provides no guarantee of its ethicalsoundness, whether the review body is a U.S. IRB or a research ethicscommittee in another country. In my role as consultant to the Joint UnitedNations Programme on HIV/AIDS (UNAIDS) and in other review capaci-ties, I have studied an admittedly relatively small sample of protocols,approved by U.S. IRBs, for research to be conducted in developing coun-tries. Some of these approved protocols revealed that the IRB did not noticeor ignored serious risks to research subjects, that it failed to take account ofsignificant psychosocial risks to subjects, overlooked the need for confi-dentiality protections during the course of research, or approved informedconsent documents that had grossly inadequate disclosures to the subjectsof the procedures to be undertaken and contained myriad other shortcom-ings. In addition, as Rothman (2000) notes, U.S. IRBs have little familiar-ity with developing countries. Nor is it the case that involvement of host-country researchers in the design and implementation of a study can ensurethat all features of the research comply with the highest ethical standards,or even minimally acceptable ones. Adherence to correct procedures is anecessary condition for ethical conduct but surely not a sufficient condition.

So, although everyone agrees that exploitation is a serious moral wrongthat must be avoided, there is no consensus on what constitutes exploita-tion and little in the way of sustained analysis of the concept in the con-text of international research.3 One commentator offers the following,almost tautologous account: “Exploitation occurs when an individual orgroup use [sic] unethical means to obtain a benefit from another indi-vidual or group” (Resnik 1998, p. 306). Another account, slightly moreilluminating, observes that all research subjects face some risk of exploi-tation: “They face the possibility that researchers may regard them purelyas a means to benefit society or, more subtly, enroll them when the soci-etal benefits to be gained from the research do not justify the risks thatsubjects face” (Wendler 2000, p. 312). In the absence of a fuller accounton which opponents in these different controversies can agree, an appealto exploitation will accomplish no more than to identify the chief groundsfor criticizing a particular study.

ETHICAL STANDARDS AND “STANDARD OF CARE”

What substantive ethical standards ought to apply to research con-ducted in developing countries? This question brings us to the fourth point

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 27 ]

on which there is apparent consensus: It is unacceptable to lower theethical standards adopted in the industrialized world when carrying outresearch in developing or resource-poor countries. Some commentatorshave maintained that exactly the same standards should be employed theworld over, whereas others contend that different standards are requiredbecause different circumstances obtain. But does “different” necessarilymean “lower”? What is the relevance of “standard of care” argumentsused to defend or criticize the care and treatment research participantsreceive during a trial? The conflicting ways in which the terms “stan-dard” and “standard of care” have been interpreted demonstrate thatagreement on the basic principle can nonetheless yield conflicting judg-ments about which research is ethically acceptable and which is not. Inanswer to the question “Should there be one ethical standard for researchconducted in industrialized countries and another for resource-poor coun-tries?,” a variety of intriguing and apparently contradictory answers havebeen proposed; they are quoted and numbered below for ease of refer-ence in the discussion that follows.

(1) Opponents [of placebo-controlled trials] contend that reference to the“standard of care” in developing countries to justify placebo-controlledtrials is ethically suspect. When people receive no treatment at all, therecan be no “standard” of care. Defenders say that women in the trialwho receive placebo are not being made worse off than they would beif they were not in the trial at all. That is what is meant by “standard ofcare” in this context. (C. Levine 1998, p. 47)

(2) In developing countries, the standard of care . . . is not based on aconsideration of alternative treatments or previous clinical data, but isinstead an economically determined policy of governments that cannotafford the prices set by drug companies. . . . Acceptance of a standardof care that does not conform to the standard in the sponsoring coun-try results in a double standard in research. Such a double standard,which permits research designs that are unacceptable in the sponsoringcountry, creates an incentive to use as research subjects those with theleast access to health care. (Lurie and Wolfe 1997, p. 855)

(3) “Standard of care” is a concept borrowed from the medical-legal con-text, denoting the level of conduct against which a physician’s or healthprovider’s treatment of a patient will be judged in determining whethersuch conduct constitutes negligence. It generally means: “what a rea-sonably prudent physician (or specialist) would do in the same or simi-

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 28 ]

lar circumstances” (Annas 1993, p. 4). Defined in this way, it can mean-ingfully describe the types or level of treatments provided to patients inthe clinical setting, but it might not serve as a justification for whatought to be provided to participants in research. (NBAC 2000, p. 13)

(4) Firstly, there has been a failure to define adequately the “standard ofcare.” Secondly, it has been incorrectly assumed that the standard setby developed countries can be considered the norm. . . . It is arbitraryand not justifiable to select only one [requirement] . . .–for example,which drugs are used–to compare the standard of care in developedand developing countries. . . .The United States’s standard of care shouldnot be emulated throughout the world. (Benatar and Singer 2000, pp.824-25)

(5) [T]his debate has been complicated by some unrecognized ambiguitiesin the notion of a standard of care. . . . [T]his concept is ambiguousalong two different axes, with the result that there are at least fourpossible standard of care arguments that must be clearly distinguished.(London 2000, p. 381)

(6) There are strong practical as well as principled reasons for Americansto follow American ethical standards when they do research abroad.(Rothman 2000, no page number)

(7) We can also see why we should not expect a single standard of researchto govern all study designs. There are a variety of ethical principles thatapply to research on human subjects, and they sometimes conflict. . . .In order to achieve an optimal balance of these different ethical stan-dards, we need to take into account various social, cultural economic,political, as well as scientific factors. . . . One might even argue that it isunjust, unfair, and insensitive to demand that the exact same standardsof research that govern study designs in developed nations should alsobe implemented in developing countries. (Resnik 1998, pp. 304-5)

(8) When Helsinki calls for the “best proven therapeutic method” does itmean the “best therapy available anywhere in the world”? Or does itmean the standard that prevails in the country in which the trial isconducted? . . . [T]he best proven therapy standard must necessarilymean the standard that prevails in the country in which the clinicaltrial is carried out. (R. Levine 1998, p. 6)

(9) It is clear that the pressures to lower the ethical standards set by theDoH [Declaration of Helsinki] are primarily economic–it costs less to

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 29 ]

run a trial where you do not have to provide for medical care. . . . So letus push to keep the highest ethical standards applied everywhere . . . .(Greco 2000, p. 97)

(10) Just because the obligation to make research products available to thehost country after the trial is over is not the prevailing internationalstandard does not mean it is not an ethical standard to which we shouldaspire. (Page 2000, p. 20)

What are we to make of these assorted claims? Is the question of whatethical standard should prevail in international research a hopeless sea ofconfusion? Is it an ethical domain like some others, in which reasonablepeople can disagree? Or are some of the opponents in this debate simplyunreasonable? One good starting point would be to identify and seek toclear up ambiguities that lurk in the terms of these debates. Three of thesources of the statements quoted above strive to do just that.

In the first of these (3), the National Bioethics Advisory Commission’sreport distinguishes between the original (and typical) context in which“standard of care” is invoked–the clinical setting–and questions whetherthe concept can be uncritically imported into the research setting. TheNBAC report makes a further observation:

. . . an ambiguity lurks in the term “standard,” which sometimes means“what is normally done,” as in “standard practice.” In this meaning, “stan-dard practice” in some countries—such as reusing syringes or other dispos-able equipment—would not be acceptable to U.S. researchers and wouldnot constitute a justification to employ the local, unsafe practice. But “stan-dard” can also refer to a level that must be attained, as in “a standard foradmission to medical school,” or “the standard for maintaining hygienicpractice in treatment and research.” In this latter sense, U.S. researcherswould be bound by the proper medical standard that prohibits the reuse ofdisposable equipment, even if reuse is standard practice in some countries.(NBAC 2000, p. 13)

In the second source (4), the authors propose a new analysis of “stan-dard of care,” arguing that equal standards of medical care during re-search could be taken to mean any one or a combination of several re-quirements, encompassing more than simply medications or other“proven” therapies. Among the five elements specified in this “expanded”concept for standard of care are “[r]esearch undertaken by a team of thesame culture and language group as the subjects, so that the same degreeof effective communication, trust, and genuine informed consent is

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 30 ]

achieved through a legitimate informed decision making process;” and“[c]are provided by a research team with equivalent qualifications, train-ing, and expertise” (Benatar and Singer 2000, p. 824). The authors alsopropose an “alternative” concept of standard of care, one they contendcould be achieved globally despite existing or future economic inequali-ties. This alternative concept is geared toward clinical trials in developingcountries, and includes undertaking only such research that will be ofbenefit to the community being researched, and translating research find-ings into components of accessible care in the community being researched(Benatar and Singer 2000, p. 825).

The third and most sustained analysis (5) identifies four different inter-pretations of “standard of care” culled from the literature. The first twointerpretations arise as a result of two different relevant reference points,harking back to the quotation cited above in (8): Does “best proventherapy” mean the “best therapy available anywhere in the world”? Ordoes it mean the standard that prevails in the country in which the trial isconducted? These two reference points are termed “global” and “local”and give rise to two different judgments of what “standard of care” means.However, limiting the debate to these two reference points is simplisticand fails to capture a more subtle yet crucial distinction found in thesecond pair of interpretations. This latter pair is termed “de facto” and“de jure.” The de facto interpretation uses “standard of care” to meanthat the standard of medical practices in a community is determined bythe actual medical practices in that community. In other words, the stan-dard of care is no more than a description of what, in fact, occurs. The dejure concept, on the other hand, interprets the “standard of care” as thatdetermined by the judgment of experts in the medical community regard-ing which diagnostic and therapeutic practices have proven most effective.

Full details of this complex analysis cannot be reproduced here. Theauthor concludes by opting for the “local, de jure” interpretation of “stan-dard of care” because it rejects the “bad, if not perfidious” local, de factostandard while still permitting research to go forward that may not meetthe stringent requirements of the “global de jure” standard. However, thelocal, de jure standard would not permit researchers knowingly to denysubjects care that has proven effective for their illness in their population,and thus ensures that subjects of clinical research in developing countriesare not exploited. At the same time, this standard requires attention tosubstantive differences in social, cultural, and economic contexts and theirimpact on the permissibility of international research.

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 31 ]

SEEKING RESOLUTIONS

Armed with these proposed new definitions and multiple interpreta-tions of the terms “standard” and “standard of care,” are opponents inthe controversies over international research likely to resolve their differ-ences? Alas, I fear not. My pessimism stems from several factors. Firstand foremost is the prevalence of “the method of tenacity” (borrowedfrom the philosopher, Charles Sanders Peirce (1955))4 in maintaining abelief system. Defenders of the view that the local, de facto standard ofcare is ethically appropriate because subjects are not being made worseoff are not likely to alter their position when confronted with a de jurestandard that may require more by way of care or treatment than re-search subjects would receive outside a trial. What counts for the de factogroup is simply that research subjects are not made worse off as a resultof their participation than they were before the research began. Similarly,those who interpret the Declaration of Helsinki in terms of the global, dejure standard–requiring the best current methods “anywhere in the world”–are not likely to be persuaded by the subtlety of arguments that wouldpermit the local de jure standard in some circumstances.

A second factor that makes resolution of disagreements unlikely is thelimited power of persuasive definitions. Those who understand “stan-dard of care” in its narrowest meaning—referring only to which drugsare used for a control group in a clinical trial—are not likely to abandonthat meaning in favor of an expanded or alternative concept proposed byeven the most thoughtful bioethicists who argue both for the need toattend to morally relevant considerations of context and for the equallyimportant need to reduce the huge inequities in global health (Benatarand Singer 2000).

Still another reason why opponents in at least some debates over theethics of international research will not come to agree is that they accordhigher priority to different ethical principles that come into conflict. Resnik(1998, p. 305) says of the disputed placebo-controlled perinatal trans-mission studies that “the conflict can be viewed as beneficence and in-formed consent versus scientific rigor, justice and social utility.” Brennan(1999) also identifies a conflict of ethical principles, this time the conflictis utilitarianism (understood as efficiency) versus distributive justice and“the investigator’s moral duty to the research subject.” However, sinceResnik comes down on the side of defending those trials, he considersthat justice supports them. Brennan, on the other hand, comes down on

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 32 ]

the side of opposing the disputed trials, and finds justice to be on his sideof the controversy. Obviously, people are not willing to concede that jus-tice lies anywhere but on the side they support.

This result is not surprising, since “justice” is not an unambiguousconcept and there is not a single, unequivocal principle of justice that canbe applied in assessing the ethics of international research (Macklin 1998).According to one interpretation, it is unjust to use a particular study de-sign in a developing country when that same study design could not beused in the industrialized country sponsoring the research because it vio-lates the precept, “treat like cases alike, different cases differently.” Analternative interpretation contends that the cases are not alike since factsand circumstances differ in the developed and developing countries. Thisis an interesting philosophical debate, one that involves a deeper analysisthan can be provided here (Macklin 1998).

Concluding that a resolution to these controversies might require philo-sophical analysis is not without its dangers, however. To appreciate thecynicism with which philosophers and bioethicists are viewed in somequarters in mainstream medicine, one has only to look at the commentsin an editorial entitled “The Ethics Industry” that appeared in The Lan-cet. The editorial accuses “the ethics industry” (that’s us) of being “oddlyparochial” and “rooted in armchair moral philosophy.” And further: “De-partments of ethics that are divorced from the medical profession, wal-lowing in theory and speculation, are quaintly redundant” (Editorial 1997,p. 897). Most interesting, however, is The Lancet’s quick and certain judg-ment of the most controversial studies of the past decade:

Most of the dilemmas upon which ethicists thrive can be solved by appealto principles fundamental to medical practice worldwide—that doctors dono harm to patients, that doctors do their best for patients. . . . It does notrequire recourse to the opposing arguments of 18th century philosophersto work out that the placebo-controlled trials of antiretroviral drugs toprevent perinatal transmission of HIV infection…infringe the first prin-ciple. Women in the placebo-treated group were harmed by receiving noeffective treatment to prevent transmission of HIV to their children (Edito-rial 1997, p. 897).

Without recourse to novel interpretations of “standard of care” or pro-posed new definitions, without appeal to debates about the proper role ofefficiency and economics in international research, and surely withoutrecourse to philosophical principles of justice, The Lancet drew its con-

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 33 ]

clusion about the ethics of the placebo-controlled trials. All it needed wasan appeal to the “first principle” of medical practice worldwide—“thatdoctors do no harm to patients.”

To end on a more positive note, let me return to the earlier discussion ofexploitation. It serves as a reminder of why research in developing countriesraises somewhat different (or at least more serious) problems than researchconducted in the United States or other industrialized countries. The earlierdiscussion drew the unsatisfying conclusion that what is required is a fullaccount of exploitation in the context of international research, one thatopponents in these various disputes could be likely to agree upon. But theproblem noted there is that the very factors that give rise to the dispute arethe ones that will lead critics of the research to make the charge of exploita-tion and defenders to reject that allegation. A different approach focuses notso much on what constitutes exploitation, but instead on criteria for deter-mining the vulnerability of subject populations or individual subjects.

A UNAIDS Guidance Document takes this approach in two of its guid-ance points (UNAIDS 2000). Guidance Point 3 lists factors that may in-crease vulnerability to exploitation of host countries and communities,including: the level of the proposed community’s economic capacity; acommunity’s experience with and/or understanding of scientific research;a community’s level of experience and capacity for conducting ethicaland scientific review; and the local infrastructure, personnel, and techni-cal capacity for conducting the proposed research (UNAIDS 2000, pp.15-16). Guidance Point 7 lists additional factors that create conditions forpossible exploitation or increased vulnerability among research participants.These include social and legal marginalization of groups from which partici-pants might be drawn; limited availability, accessibility, and sustainabilityof health care and treatment options; limited ability of individuals to un-derstand the informed consent process; and limited ability of individualsto be able to give freely their informed consent in the light of prevailingclass, gender, and other social and legal factors (UNAIDS 2000, p. 23).

If researchers, sponsors, and ethical review bodies were to undertake asystematic review of these conditions prior to preparing a research proto-col to implement in developing countries, it might provide a better way toassess vulnerability to exploitation in advance of initiating the researchrather than opening it up to allegations once the research is underway.This type of prior investigation of the context in which the research is tobe conducted will not appeal to those who seek to begin research as quicklyas possible, either in order to reach scientific conclusions about safety

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 34 ]

and efficacy or, in the case of industry, to reap the maximum possibleprofits. As one thoughtful researcher has put it:

The proclaimed and real urgency of many countries is for an efficacious vac-cine, drug, preventive or diagnostic method and not for a clinical trial—thesetrials can be done in other countries/regions/communities with similar inci-dences of infection or disease, but without the vulnerability of the alreadydestitute. In certain situations even the government and the investigators maybe considered vulnerable—e.g., the possibility of getting research money andof enhancing one’s career may act as inducers for doing trials” (Greco 2000).

If some resolution of these controversies in international research comesdown to the age-old battle between economics and efficiency on one side,and ethics on the other, then let ethics win out.

NOTES

1. I was a participant in the conference and these observations are taken frommy notes. The comments of the Chilean researcher were not reflected in thepublished Discussion Paper.

2. Dr. Mbidde’s letter was sent to Dr. Varmus in response to the Public CitizenNews Release of 22 April 1997. Public Citizen subsequently faxed copies ofDr. Mbidde’s letter to individuals on their mailing list.

3. For a full analysis of the general concept of exploitation, see Alan Wertheimer(1996).

4. Although the phrase, “the method of tenacity” is borrowed from Peirce, hisexplanation of the method is somewhat different from my description here.

REFERENCES

Angell, Marcia. 1997. The Ethics of Clinical Research in the Third World. NewEngland Journal of Medicine 337: 847-49.

Annas, George J. 1993. Standard of Care. New York: Oxford University Press.———, and Grodin, Michael A. 1998. Human Rights and Maternal-Fetal HIV

Transmission Prevention Trials in Africa. American Journal of Public Health88: 560-63.

Benatar, Solomon R., and Singer, Peter A. 2000. A New Look at InternationalResearch Ethics. British Medical Journal 321: 824-26.

Brennan, Troyen A. 1999. Proposed Revisions to the Declaration of Helsinki–Will They Weaken the Ethical Principles Underlying Human Research? NewEngland Journal of Medicine 341: 527-31.

MACKLIN • UNRESOLVED ISSUES IN INTERNATIONAL RESEARCH

[ 35 ]

CIOMS. Council for International Organizations of Medical Sciences. 1993. In-ternational Ethical Guidelines for Biomedical Research Involving HumanSubjects. Geneva: CIOMS.

Crouch, Robert A., and Arras, John D. 1998. AZT Trials and Tribulations.Hastings Center Report 28 (6): 26-34.

Del Río, Carlos. 1998. Is Ethical Research Feasible in Developed and DevelopingCountries? Bioethics 12: 328-30.

Editorial. 1997. The Ethics Industry. The Lancet 350: 897.Glantz, Leonard H.; Annas, George J.; Grodin, Michael A.; and Mariner, Wendy

K. 1998. Research in Developing Countries: Taking “Benefit” Seriously.Hastings Center Report 28 (6): 38-42.

Grady, Christine. 1998. Science in the Service of Healing. Hastings Center Re-port 28 (6): 34-38.

Greco, Dirceu B. 2000. Revising the Declaration of Helsinki: Ethics vs. Econom-ics or the Fallacy of Urgency. Canadian HIV/AIDS Policy & Law Newslet-ter 5 (4): 94-97.

Kass, Nancy, and Hyder, Adnan. 2000. Attitudes and Experiences of U.S. andDeveloping Country Investigators Regarding U.S. Human Subjects Regula-tions. In Ethical and Policy Issues in International Research, Part F, Na-tional Bioethics Advisory Commission. Paper prepared for NBAC and avail-able in Vol. II.

Levine, Carol. 1998. Placebos and HIV: Lessons Learned. Hastings Center Re-port 28 (6): 43-48.

Levine, Robert J. 1998. The “Best Proven Therapeutic Method” Standard inClinical Trials in Technologically Developing Countries. IRB: A Review ofHuman Subjects Research 20 (1): 5-9.

———. 1999. The Need to Revise the Declaration of Helsinki. New EnglandJournal of Medicine 341: 531-34.

Lie, Reidar K. 1998. Ethics of Placebo-Controlled Trials in Developing Nations.Bioethics 12: 307-11.

London, Alex John. 2000. The Ambiguity and the Exigency: Clarifying “Stan-dard of Care” Arguments in International Research. Journal of Medicineand Philosophy 25 (4): 379-97.

Lurie, Peter, and Wolfe, Sidney. 1997. Unethical Trials of Interventions to Re-duce Perinatal Transmission of the Human Immunodeficiency Virus in De-veloping Countries. New England Journal of Medicine 337: 853-56.

Macklin, Ruth. 1998. Justice in International Research. In Beyond Consent: Seek-ing Justice in Research, ed. Jeffrey Kahn, Anna Mastroianni, and JeremySugarman, pp. 131-46. New York: Oxford University Press.

KENNEDY INSTITUTE OF ETHICS JOURNAL • MARCH 2001

[ 36 ]

NBAC. National Bioethics Advisory Commission. 2000. Ethical and Policy Is-sues in International Research. Washington, DC: NBAC.

Nuffield Council on Bioethics. 1999. Ethical Issues in Research In DevelopingCountries. Discussion paper.

Okie, Susan. 2000. Health Officials Debate Ethics of Placebo Use. WashingtonPost (24 November): A03.

Page, Alice. Unpublished (2000). Prior Agreements in International Clinical Tri-als: Ensuring the Provision of Drugs and Other Research Benefits to Devel-oping Countries. Background paper prepared for the National BioethicsAdvisory Commission.

Peirce, Charles Sanders. 1955. The Fixation of Belief. In Philosophical Writingsof Peirce, ed. Justus Buchler, pp. 5-22. New York: Dover Publications.

Resnik, David B. 1998. The Ethics of HIV Research in Developing Nations. Bio-ethics 12: 286-306.

Rothman, David J. 2000. The Shame of Medical Research. New York Review ofBooks (30 November): PDF file, unpaginated. Available at: http://www.nybooks.com/nyre/WWWfeatdisplay.cgi?20001130060F.

Schüklenk, Udo. 1998. Unethical Perinatal HIV Transmission Trials EstablishBad Precedent. Bioethics 12: 312-19.

———, and Ashcroft, Richard. 2000. International Research Ethics. Bioethics14: 158-72.

Thomas, Joe. 1998. Ethical Challenges of HIV Clinical Trials in DevelopingCountries. Bioethics 12: 320-27.

Varmus, Harold, and Satcher, David. 1997. Ethical Complexities of ConductingResearch in Developing Countries. New England Journal of Medicine337:1003-5.

UNAIDS. 2000. Ethical Considerations in HIV Preventive Vaccine Research:UNAIDS Guidance Document (May). Geneva, Switzerland: Joint UnitedNations Programme on HIV/AIDS.

Wendler Dave. 2000. Informed Consent, Exploitation and Whether It Is Possibleto Conduct Human Subjects Research Without Either One. Bioethics 14:310-39.

Wertheimer, Alan. 1996. Exploitation. Princeton, NJ: Princeton University Press.World Medical Association. 2000 [1964]. Declaration of Helsinki, as amended

by the WMA 52nd General Assembly, Edinburgh, Scotland, October 2000.