eMJA: Studdert & Brennan, Clinical trials in developing countries: scientific and ethical issues

Clinical trials in developing countries: scientific and ethical issues

David M Studdert and Troyen A Brennan

Since the 1994 finding that intensive zidovudine treatment of mothers and infants can dramatically reduce perinatal transmission of human immunodeficiency virus, this treatment has been widely adopted in developed countries. In developing countries, trials of less-intensive (and cheaper) regimens have gone ahead, many funded by foreign governments and the United Nations. Controversy has erupted over these trials, particularly over their use of placebo controls. Do differences in healthcare needs and budgets justify different ethical standards in the developed and the developing world?

MJA 1998; 169: 545-548
For editorial comment, see McNeill

Introduction - The arguments - Ethical principles in international human research - Efficacy and use of placebo - Standard of care: is it relative? - References - Authors' details
Make a comment - Register to be notified of new articles by e-mail - Current contents list - More articles on Ethics - ©MJA1998

Introduction
Controversy over AIDS clinical trials in developing countries has galvanised public attention and divided the medical research community.^1-4 The trials in question, in sub-Saharan Africa and Thailand, use randomised, placebo-controlled methods to test the effectiveness of interventions in preventing perinatal transmission of human immunodeficiency virus (HIV).^5,6 At least nine of these trials have received US government funding through the Centers for Disease Control (CDC) or National Institutes of Health (NIH), while five are funded by other foreign governments, and one by the United Nations Program on AIDS (UNAIDS). Background to the controversy is shown in the Box.
Debate was sparked in September 1997 by an article⁵ and an editorial¹³ in the New England Journal of Medicine. The editorial likened the trials to the notorious Tuskegee study of untreated syphilis (1938-1972),¹⁴ a US government study that continued to follow the natural progression of syphilis in a cohort of poor African American males long after penicillin became widely available. This comparison prompted two prominent members of the journal's editorial board to resign,¹⁵ and drew a rebuttal from the CDC and NIH.⁸
In October 1997, investigators discontinued use of placebos in an NIH-funded study of two short-course zidovudine regimens in Ethiopia.¹⁶ Then, in February 1998, the CDC and the Thai government announced that their placebo-controlled study of a short-course zidovudine regimen in Thailand had shown a 51% reduction in perinatal HIV transmission rates.¹⁷ At the same time, they announced that further use of placebos would cease in both this study and another in Cote d'Ivoire.¹¹ The CDC, NIH, UNAIDS, and France's National Agency for AIDS Research (ANRS) have called for international dialogue on the "far-reaching scientific and policy implications of these findings".¹²
Nonetheless, a range of other foreign-sponsored clinical trials continue in developing countries, as does the debate about their ethics. To date, this debate appears to have generated more heat than light: the lines of arguments are often unclear, and it is difficult to isolate the central points of disagreement. For example, both critics and defenders of the trials claimed that the Thai findings vindicated their respective positions.¹⁸ In this article, we outline the main arguments and propose a framework for analysing some of the scientific and ethical issues.
The arguments
The critics' main objection to the trials is that investigators withhold an effective therapy from women and children who are randomised into control arms. Deprivation of a therapy known to be effective, albeit in an intensive and expensive form, has been construed as "ethical relativism".^13,19 It is argued that this violates the principle that protections given to human subjects abroad should be "no less exacting" than those given in sponsoring countries.²⁰ Two similar studies under way in the US gave all participants access to zidovudine and other antiretroviral drugs.⁵
Arguments from defenders of the trials are interrelated, but may be divided roughly into three categories:
First, it is argued that significant differences in resources and health conditions between countries can alter the balance that must be struck between risks (to study participants) and benefits (to study participants and others).^8,21,22 In other words, a study may be acceptable in one country but not another because of differences in wealth or burdens of disease between the two.

Second, the technical superiority of placebo-controlled methods is argued to have ethical implications.^8,23 Faster answers are said to be possible with enrolment of fewer subjects. Defenders also attach moral weight to the fact that absolute, rather than marginal, benefits may be measured. At stake in this distinction, it is argued, is the difference between results that lead to life-saving policies, and results that simply raise new questions.

Finally, defenders attach importance to satisfaction of process requirements.^8,24,25 For example, they highlight the support and participation of host country governments and the approval of ethics review committees in the host and sponsoring countries. Study subjects' consent to participate is also highlighted, although critics have questioned whether this consent is truly "informed".²⁶
More ardent defenders combine strands of difference and process arguments and caution against an "imperialism"²⁷ that would second-guess the willing participation of individuals in the developing world.
Ethical principles in international human research
A decade ago, Marcia Angell (now Executive Editor of the New England Journal of Medicine) asked whether ethical standards should be "substantially the same everywhere, or is it inevitable that they differ from region to region, reflecting local beliefs and custom?"²⁸ Today it is widely accepted that all research subjects are entitled to minimum guarantees that are transnational and non-negotiable.²⁹ Realisation of these entitlements is, of course, a separate matter. However, institutional mechanisms for promulgating and applying human subject protections have advanced considerably, hurried on by several scandals and major public inquiries.^30,31 Many countries, including Australia, New Zealand, and the United States, use formal ethics review committees and similar criteria to evaluate any government-funded study involving human subjects, wherever and however it is conducted.
Among the minimum guarantees, formalised in the World Health Organization's Declaration of Helsinki³² and reiterated in guidelines promulgated by governments and institutions, are:
The study must have a valid scientific design and be performed by qualified persons.

It must strike a reasonable balance between the predictable risks and foreseeable benefits to the subjects or others, with the proviso that the subjects' interests should never be subordinate to those of science and society.

All study subjects must consent to participate, a decision that must be made without duress or coercion and only after details of the study are provided.
When investigators from one country conduct research in another country, it is appropriate to add to this list of minimum guarantees the requirement that:
The research must hold the promise of direct, tangible, and significant benefit to the host country population, if not to the study subjects themselves.⁸
However, determining whether such benefits are likely to accrue can be problematic, because researchers and government officials in the host country will often have a vested interest in ensuring that the research proceeds.
Efficacy and use of placebo
Satisfaction of the above process requirements is a necessary, but not sufficient, basis for ethical research. A second inquiry is needed before accepting placebo-controlled studies: does the drug or intervention proposed for randomisation have proven therapeutic efficacy? If so, then a placebo-controlled study design is ethically suspect.³³ Investigators will not be in the state of "equipoise", or indeterminacy, necessary to avoid conscious provision of inferior treatment to study subjects.³⁴
The zidovudine trials might be said to present the "easy case" for such analysis. Findings from the ACTG 076 protocol in 1994 were compelling, and the recent Thai results must now deepen concern about continued use of placebos in any zidovudine study, whether in Thailand or elsewhere. However, there are important subtleties.
Firstly, efficacy may or may not be generalisable across countries with wide disparities in baseline levels of health status, healthcare services, and burdens of disease.^8,25,35 Whether this problem raises serious doubts about the efficacy of the ACTG 076 regimen in sub-Saharan Africa is debatable, especially after the Thai results. However, it does illustrate a more general point -- failure to separate the question of efficacy, at least initially, from considerations about whether a universal standard of care is applicable may obscure important scientific questions.
Another complication is that judgements about efficacy can be problematic when the scientific evidence is equivocal. For example, several years ago our own human subjects committee at the Harvard School of Public Health considered a protocol for a placebo-controlled study of the effect of vitamin A and multivitamins on perinatal HIV transmission in Tanzania. The decision to approve the study was difficult because data on the prophylactic benefits of micronutrient supplementation for HIV-infected women were inconclusive, and remain so today. (However, recently published findings show other pregnancy-outcome benefits³⁶ which would enter the ethical calculation were a similar protocol reviewed today.) How is equivocal evidence on efficacy to be weighed?
Although the efficacy question has ethical implications, we believe it is a technical or scientific question at the outset. Hence, it may be useful for the ethics review committee to require investigators to summarise the evidence for efficacy of an intervention. If efficacy is established anywhere, the investigators should satisfy a threshold level of doubt about its applicability to their study site. However, this task subjects the investigators to competing pressures, requiring them to demonstrate to ethics committees and sponsors that their hypothesis is valid and that the intervention has some reasonable expectation of an effect, while simultaneously marshalling evidence for doubt. To guard against this conflict, ethics review committees may wish to convene independent scientific panels to evaluate evidence on efficacy. When efficacy is established or is in the "grey area", committees might also require investigators to report on potential alternative approaches -- such as "equivalence"³⁷ or "observational"³⁸ study designs.
In summary, given that fundamental requirements of human subjects research are met, poor evidence about efficacy should allay many ethical concerns. On the other hand, when the proposed intervention has known efficacy -- or there is a sound basis for inferring efficacy -- and no alternative study approach is feasible, a further question arises: is it ethically acceptable to confine subjects in the control arm to treatments which are considered substandard in the sponsoring country?
Standard of care: is it relative?
The current debate is devoting much energy to the question of whether different conditions abroad can justify lower standards of care, especially when the differences arise from resource inequities. This question is certainly raised by some of the AIDS trials under scrutiny, as the ACTG 076 regimen has become a standard treatment for seropositive pregnant women in the United States and many other developed countries.³⁹ However, the larger question is whether non-conformity with sponsoring-country standards is a sufficient basis for stopping any placebo-controlled study abroad, or whether clear-cut answers to the efficacy question in this particular case merely help to explain why the trials are objectionable.
Lurie and Wolfe note, "as a model of an ethically conducted study",⁵ a comparison in Thailand of three short zidovudine regimens with a regimen similar to the ACTG 076 regimen. A placebo-controlled design was considered and rejected by our human subjects committee at the Harvard School of Public Health.⁵ Although the recent Thai findings¹⁷ were not at hand, the committee and investigators agreed that an equivalence design could achieve the study's main goals. In addition, the committee was mindful of anecdotal evidence that a regimen similar to ACTG 076 was soon to be widely available in Thailand.
One interpretation of the committee's decision is that it underscores a commitment to an inviolable standard of care. However, a better account of deliberations is that the standard-of-care question was not directly considered, because an alternative study approach was feasible, and there was genuine doubt about whether the placebo treatment even met the host country's standard of care. Had neither of these factors been present, then the standard-of-care question would have been under consideration as the primary basis for approving or rejecting the study protocol. We hesitate to predict the outcome.
A careful, global dialogue is needed on how international medical research involving human subjects should deal with inevitable differences in standards of care. Rigid adherence to a universal standard of care will narrow, perhaps even eliminate, the opportunities for placebo-controlled studies in the developing world. It is not clear that the research community, or the majority of citizens in any country, are ready to embrace this consequence, particularly as the gap in resources and disease burden between developed nations and sub-Saharan Africa continues to widen.⁴⁰
References

Annas GJ, Grodin MA. An apology is not enough. Boston Sunday Globe 1997 May 18; Sect C: 1.
Stolberg SG. U.S. AIDS research abroad sets off outcry over ethics. New York Times 1997 Sep 18; Sect A: 1.
AIDS in Africa. The Economist 1997; Sep 27: 19.
Stolberg SG. Job itself in doubt as new feud looms on surgeon general. New York Times 1998 February 5; Sect A: 1.
Lurie P, Wolfe SM. Unethical trials of interventions to reduce perinatal transmission of the human immunodeficiency virus in developing countries. N Engl J Med 1997; 337: 853-855.
Phanuphak P. Ethical issues in studies in Thailand of vertical transmission of HIV. N Engl J Med 1998; 338: 834-835.
Connor EM, Sperling RS, Gelber R, et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N Engl J Med 1994; 331: 1173-1180.
Varmus H, Satcher D. Ethical complexities of conducting research in developing countries. N Engl J Med 1997; 337: 1003-1005.
Waldholz M. AZT price cut for third world mothers-to-be. Wall Street Journal 1998 Mar 5: Sect B: 1.
Frenkel LM, Cowles MK, Shapiro DE, et al. Analysis of the maternal components of the AIDS clinical trial group 076 zidovudine regimen in the prevention of mother-to-infant transmission of human immunodeficiency virus type 1. J Infect Dis 1997; 175: 971-974.
Centers for Disease Control. Update on CDC Collaborative research studies on perinatal HIV prevention in the developing world: preliminary results find short-course AZT effective. 1998 Feb:<http://www.cdc.gov/nchstp/od/Perinatal>
Centers for Disease Control and Prevention (CDC), The Joint United Nations Programme on HIV/AIDS (UNAIDS), The National Institutes of Health (NIH), and The Agence Nationale de Recherche sur le SIDA (ANRS). Joint statement. 1998 Feb 18:<http://www.unaids.org/highband/press/cdcjs8>
Angell M. The ethics of clinical research in the third world. N Engl J Med 1997; 337: 847-849.
Jones JH. Bad blood: the Tuskegee syphilis experiment. New York: Free Press, 1993.
Altman LK. AIDS experts leave journal after studies are criticized. New York Times 1997 October 15; Sect A: 1.
Another HIV-1 trial loses placebo control. Lancet 1997; 350: 831.
Centers for Disease Control. Short-course regimen of AZT proven effective in reducing perinatal HIV transmission: offers hope for reducing mother-to-child HIV transmission in developing world. 1998 Feb 24. <http://www.cdc.gov/od/oc/media/ pressrel/r980224.html>
US ends placebo use in foreign AIDS study. Chicago Tribune 1998 Feb 19: Sect N: 14.
Angell M. Tuskegee revisited. Wall Street Journal 1997 Oct 28; Sect A: 22.
World Health Organization. International ethical guidelines for biomedical research involving human subjects. Geneva: Council for International Organizations of Medical Sciences, 1993.
Bagenda D, Muske-Mudido P. A look at ... ethics and AIDS. Washington Post 1997 Sep 28; Sect C: 3.
Saba J, Ammann A. A cultural divide on AIDS research. New York Times 1997 Sep 20; Sect A: 15.
Simonds RJ, Rogers MF, Dondero TJ. Ethics of placebo-controlled trials of zidovudine to prevent the perinatal transmission of HIV in the third world [letter]. N Engl J Med 1998; 338: 836-837.
Study volunteers support placebo-controlled AIDS trials; AZT to prevent vertical transmission of HIV in South Africa. AIDS Weekly Plus 1997 Oct 27: 24.
Mbidde EK. Ethics of placebo-controlled trials of zidovudine to prevent the perinatal transmission of HIV in the third world [letter]. N Engl J Med 1998; 338: 837.
French HW. AIDS research in Africa: juggling risks and hopes. New York Times 1997 Oct 9: Sect A: 1.
Cohen J. Ethics of AZT studies in poorer countries attacked. Science 1997; 276: 1022.
Angell M. Ethical imperialism? Ethics in international collaborative clinical research. N Engl J Med 1988; 319: 1081-1083.
Penslar RL. Research ethics: cases and materials. Bloomington: Indiana University Press, 1995.
Report of the Committee of Inquiry into Allegations Concerning the Treatment of Cervical Cancer at the National Women's Hospital and into Other Related Matters. Report. Auckland: Government Printing Office, 1988.
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. Belmont report: ethical principles for the protection of human subjects research. Washington, DC: Government Printing Office, 1988.
World Medical Association, Declaration of Helsinki IV. Adopted by the 18th World Medical Assembly, Helsinki 1964, as amended by the 41st World Medical Assembly, Hong Kong, 1989. <http://www.nih.gov/grants/oprr/irb_appendices.html#j6>
Chalmers TC. The clinical trial. Milbank Mem Fund Q 1981; 59: 324-339.
Freedman B. Equipoise and the ethics of clinical research. N Engl J Med 1987; 317: 141-145.
Merson MH. Ethics of placebo-controlled trials of zidovudine to prevent the perinatal transmission of HIV in the third world [letter]. N Engl J Med 1998; 338: 836.
Fawzi WW, Msamanga GI, Spiegelman D, et al. Randomized trial of effects of vitamin supplements on pregnancy outcomes and T cell counts in HIV-infected women in Tanzania. Lancet 1998; 351: 1477-1482.
Ware JH, Antman EM. Equivalence trials. N Engl J Med 1997; 337: 1159-1161.
Blanche S. Ethics of placebo-controlled trials of zidovudine to prevent the perinatal transmission of HIV in the third world [letter]. N Engl J Med 1998; 338: 837-838.
Recommendations of the US Public Health Service Task Force on the use of zidovudine to reduce perinatal transmission of human immunodeficiency virus. MMWR Morb Mortal Wkly Rep 1994; 43 (RR-11): 1-20.
United Nations Development Program. Human Development Report. New York: Oxford University Press, 1997.

Authors' details

Department of Health Policy and Management, Harvard School of Public Health, Boston, MA, USA.
David M Studdert, LLB, ScD, MPH, Instructor; and Fellow in Medical Ethics, Harvard Medical School, Boston, MA; currently, Social Scientist, RAND Corporation, Santa Monica, CA, USA.
Troyen A Brennan, MD, JD, MPH, Professor of Law and Public Health; and Professor of Medicine, Harvard Medical School, Boston, MA.

Reprints: Dr D M Studdert, RAND Corporation, 1700 Main Street, Santa Monica, CA 90407-2138 USA.
E-mail: studdertATrand.org

The Medical Journal of Australia eMJA

Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company
Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>".
<URL: http://www.mja.com.au/> © 1998 Medical Journal of Australia.
We appreciate your comments.


Introduction	Controversy over AIDS clinical trials in developing countries has galvanised public attention and divided the medical research community.^1-4 The trials in question, in sub-Saharan Africa and Thailand, use randomised, placebo-controlled methods to test the effectiveness of interventions in preventing perinatal transmission of human immunodeficiency virus (HIV).^5,6 At least nine of these trials have received US government funding through the Centers for Disease Control (CDC) or National Institutes of Health (NIH), while five are funded by other foreign governments, and one by the United Nations Program on AIDS (UNAIDS). Background to the controversy is shown in the Box. Debate was sparked in September 1997 by an article⁵ and an editorial¹³ in the New England Journal of Medicine. The editorial likened the trials to the notorious Tuskegee study of untreated syphilis (1938-1972),¹⁴ a US government study that continued to follow the natural progression of syphilis in a cohort of poor African American males long after penicillin became widely available. This comparison prompted two prominent members of the journal's editorial board to resign,¹⁵ and drew a rebuttal from the CDC and NIH.⁸ In October 1997, investigators discontinued use of placebos in an NIH-funded study of two short-course zidovudine regimens in Ethiopia.¹⁶ Then, in February 1998, the CDC and the Thai government announced that their placebo-controlled study of a short-course zidovudine regimen in Thailand had shown a 51% reduction in perinatal HIV transmission rates.¹⁷ At the same time, they announced that further use of placebos would cease in both this study and another in Cote d'Ivoire.¹¹ The CDC, NIH, UNAIDS, and France's National Agency for AIDS Research (ANRS) have called for international dialogue on the "far-reaching scientific and policy implications of these findings".¹² Nonetheless, a range of other foreign-sponsored clinical trials continue in developing countries, as does the debate about their ethics. To date, this debate appears to have generated more heat than light: the lines of arguments are often unclear, and it is difficult to isolate the central points of disagreement. For example, both critics and defenders of the trials claimed that the Thai findings vindicated their respective positions.¹⁸ In this article, we outline the main arguments and propose a framework for analysing some of the scientific and ethical issues.
The arguments	The critics' main objection to the trials is that investigators withhold an effective therapy from women and children who are randomised into control arms. Deprivation of a therapy known to be effective, albeit in an intensive and expensive form, has been construed as "ethical relativism".^13,19 It is argued that this violates the principle that protections given to human subjects abroad should be "no less exacting" than those given in sponsoring countries.²⁰ Two similar studies under way in the US gave all participants access to zidovudine and other antiretroviral drugs.⁵ Arguments from defenders of the trials are interrelated, but may be divided roughly into three categories: First, it is argued that significant differences in resources and health conditions between countries can alter the balance that must be struck between risks (to study participants) and benefits (to study participants and others).^8,21,22 In other words, a study may be acceptable in one country but not another because of differences in wealth or burdens of disease between the two. Second, the technical superiority of placebo-controlled methods is argued to have ethical implications.^8,23 Faster answers are said to be possible with enrolment of fewer subjects. Defenders also attach moral weight to the fact that absolute, rather than marginal, benefits may be measured. At stake in this distinction, it is argued, is the difference between results that lead to life-saving policies, and results that simply raise new questions. Finally, defenders attach importance to satisfaction of process requirements.^8,24,25 For example, they highlight the support and participation of host country governments and the approval of ethics review committees in the host and sponsoring countries. Study subjects' consent to participate is also highlighted, although critics have questioned whether this consent is truly "informed".²⁶ More ardent defenders combine strands of difference and process arguments and caution against an "imperialism"²⁷ that would second-guess the willing participation of individuals in the developing world.
Ethical principles in international human research	A decade ago, Marcia Angell (now Executive Editor of the New England Journal of Medicine) asked whether ethical standards should be "substantially the same everywhere, or is it inevitable that they differ from region to region, reflecting local beliefs and custom?"²⁸ Today it is widely accepted that all research subjects are entitled to minimum guarantees that are transnational and non-negotiable.²⁹ Realisation of these entitlements is, of course, a separate matter. However, institutional mechanisms for promulgating and applying human subject protections have advanced considerably, hurried on by several scandals and major public inquiries.^30,31 Many countries, including Australia, New Zealand, and the United States, use formal ethics review committees and similar criteria to evaluate any government-funded study involving human subjects, wherever and however it is conducted. Among the minimum guarantees, formalised in the World Health Organization's Declaration of Helsinki³² and reiterated in guidelines promulgated by governments and institutions, are: The study must have a valid scientific design and be performed by qualified persons. It must strike a reasonable balance between the predictable risks and foreseeable benefits to the subjects or others, with the proviso that the subjects' interests should never be subordinate to those of science and society. All study subjects must consent to participate, a decision that must be made without duress or coercion and only after details of the study are provided. When investigators from one country conduct research in another country, it is appropriate to add to this list of minimum guarantees the requirement that: The research must hold the promise of direct, tangible, and significant benefit to the host country population, if not to the study subjects themselves.⁸ However, determining whether such benefits are likely to accrue can be problematic, because researchers and government officials in the host country will often have a vested interest in ensuring that the research proceeds.
Efficacy and use of placebo	Satisfaction of the above process requirements is a necessary, but not sufficient, basis for ethical research. A second inquiry is needed before accepting placebo-controlled studies: does the drug or intervention proposed for randomisation have proven therapeutic efficacy? If so, then a placebo-controlled study design is ethically suspect.³³ Investigators will not be in the state of "equipoise", or indeterminacy, necessary to avoid conscious provision of inferior treatment to study subjects.³⁴ The zidovudine trials might be said to present the "easy case" for such analysis. Findings from the ACTG 076 protocol in 1994 were compelling, and the recent Thai results must now deepen concern about continued use of placebos in any zidovudine study, whether in Thailand or elsewhere. However, there are important subtleties. Firstly, efficacy may or may not be generalisable across countries with wide disparities in baseline levels of health status, healthcare services, and burdens of disease.^8,25,35 Whether this problem raises serious doubts about the efficacy of the ACTG 076 regimen in sub-Saharan Africa is debatable, especially after the Thai results. However, it does illustrate a more general point -- failure to separate the question of efficacy, at least initially, from considerations about whether a universal standard of care is applicable may obscure important scientific questions. Another complication is that judgements about efficacy can be problematic when the scientific evidence is equivocal. For example, several years ago our own human subjects committee at the Harvard School of Public Health considered a protocol for a placebo-controlled study of the effect of vitamin A and multivitamins on perinatal HIV transmission in Tanzania. The decision to approve the study was difficult because data on the prophylactic benefits of micronutrient supplementation for HIV-infected women were inconclusive, and remain so today. (However, recently published findings show other pregnancy-outcome benefits³⁶ which would enter the ethical calculation were a similar protocol reviewed today.) How is equivocal evidence on efficacy to be weighed? Although the efficacy question has ethical implications, we believe it is a technical or scientific question at the outset. Hence, it may be useful for the ethics review committee to require investigators to summarise the evidence for efficacy of an intervention. If efficacy is established anywhere, the investigators should satisfy a threshold level of doubt about its applicability to their study site. However, this task subjects the investigators to competing pressures, requiring them to demonstrate to ethics committees and sponsors that their hypothesis is valid and that the intervention has some reasonable expectation of an effect, while simultaneously marshalling evidence for doubt. To guard against this conflict, ethics review committees may wish to convene independent scientific panels to evaluate evidence on efficacy. When efficacy is established or is in the "grey area", committees might also require investigators to report on potential alternative approaches -- such as "equivalence"³⁷ or "observational"³⁸ study designs. In summary, given that fundamental requirements of human subjects research are met, poor evidence about efficacy should allay many ethical concerns. On the other hand, when the proposed intervention has known efficacy -- or there is a sound basis for inferring efficacy -- and no alternative study approach is feasible, a further question arises: is it ethically acceptable to confine subjects in the control arm to treatments which are considered substandard in the sponsoring country?
Standard of care: is it relative?	The current debate is devoting much energy to the question of whether different conditions abroad can justify lower standards of care, especially when the differences arise from resource inequities. This question is certainly raised by some of the AIDS trials under scrutiny, as the ACTG 076 regimen has become a standard treatment for seropositive pregnant women in the United States and many other developed countries.³⁹ However, the larger question is whether non-conformity with sponsoring-country standards is a sufficient basis for stopping any placebo-controlled study abroad, or whether clear-cut answers to the efficacy question in this particular case merely help to explain why the trials are objectionable. Lurie and Wolfe note, "as a model of an ethically conducted study",⁵ a comparison in Thailand of three short zidovudine regimens with a regimen similar to the ACTG 076 regimen. A placebo-controlled design was considered and rejected by our human subjects committee at the Harvard School of Public Health.⁵ Although the recent Thai findings¹⁷ were not at hand, the committee and investigators agreed that an equivalence design could achieve the study's main goals. In addition, the committee was mindful of anecdotal evidence that a regimen similar to ACTG 076 was soon to be widely available in Thailand. One interpretation of the committee's decision is that it underscores a commitment to an inviolable standard of care. However, a better account of deliberations is that the standard-of-care question was not directly considered, because an alternative study approach was feasible, and there was genuine doubt about whether the placebo treatment even met the host country's standard of care. Had neither of these factors been present, then the standard-of-care question would have been under consideration as the primary basis for approving or rejecting the study protocol. We hesitate to predict the outcome. A careful, global dialogue is needed on how international medical research involving human subjects should deal with inevitable differences in standards of care. Rigid adherence to a universal standard of care will narrow, perhaps even eliminate, the opportunities for placebo-controlled studies in the developing world. It is not clear that the research community, or the majority of citizens in any country, are ready to embrace this consequence, particularly as the gap in resources and disease burden between developed nations and sub-Saharan Africa continues to widen.⁴⁰
References	Annas GJ, Grodin MA. An apology is not enough. Boston Sunday Globe 1997 May 18; Sect C: 1. Stolberg SG. U.S. AIDS research abroad sets off outcry over ethics. New York Times 1997 Sep 18; Sect A: 1. AIDS in Africa. The Economist 1997; Sep 27: 19. Stolberg SG. Job itself in doubt as new feud looms on surgeon general. New York Times 1998 February 5; Sect A: 1. Lurie P, Wolfe SM. Unethical trials of interventions to reduce perinatal transmission of the human immunodeficiency virus in developing countries. N Engl J Med 1997; 337: 853-855. Phanuphak P. Ethical issues in studies in Thailand of vertical transmission of HIV. N Engl J Med 1998; 338: 834-835. Connor EM, Sperling RS, Gelber R, et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N Engl J Med 1994; 331: 1173-1180. Varmus H, Satcher D. Ethical complexities of conducting research in developing countries. N Engl J Med 1997; 337: 1003-1005. Waldholz M. AZT price cut for third world mothers-to-be. Wall Street Journal 1998 Mar 5: Sect B: 1. Frenkel LM, Cowles MK, Shapiro DE, et al. Analysis of the maternal components of the AIDS clinical trial group 076 zidovudine regimen in the prevention of mother-to-infant transmission of human immunodeficiency virus type 1. J Infect Dis 1997; 175: 971-974. Centers for Disease Control. Update on CDC Collaborative research studies on perinatal HIV prevention in the developing world: preliminary results find short-course AZT effective. 1998 Feb:<http://www.cdc.gov/nchstp/od/Perinatal> Centers for Disease Control and Prevention (CDC), The Joint United Nations Programme on HIV/AIDS (UNAIDS), The National Institutes of Health (NIH), and The Agence Nationale de Recherche sur le SIDA (ANRS). Joint statement. 1998 Feb 18:<http://www.unaids.org/highband/press/cdcjs8> Angell M. The ethics of clinical research in the third world. N Engl J Med 1997; 337: 847-849. Jones JH. Bad blood: the Tuskegee syphilis experiment. New York: Free Press, 1993. Altman LK. AIDS experts leave journal after studies are criticized. New York Times 1997 October 15; Sect A: 1. Another HIV-1 trial loses placebo control. Lancet 1997; 350: 831. Centers for Disease Control. Short-course regimen of AZT proven effective in reducing perinatal HIV transmission: offers hope for reducing mother-to-child HIV transmission in developing world. 1998 Feb 24. <http://www.cdc.gov/od/oc/media/ pressrel/r980224.html> US ends placebo use in foreign AIDS study. Chicago Tribune 1998 Feb 19: Sect N: 14. Angell M. Tuskegee revisited. Wall Street Journal 1997 Oct 28; Sect A: 22. World Health Organization. International ethical guidelines for biomedical research involving human subjects. Geneva: Council for International Organizations of Medical Sciences, 1993. Bagenda D, Muske-Mudido P. A look at ... ethics and AIDS. Washington Post 1997 Sep 28; Sect C: 3. Saba J, Ammann A. A cultural divide on AIDS research. New York Times 1997 Sep 20; Sect A: 15. Simonds RJ, Rogers MF, Dondero TJ. Ethics of placebo-controlled trials of zidovudine to prevent the perinatal transmission of HIV in the third world [letter]. N Engl J Med 1998; 338: 836-837. Study volunteers support placebo-controlled AIDS trials; AZT to prevent vertical transmission of HIV in South Africa. AIDS Weekly Plus 1997 Oct 27: 24. Mbidde EK. Ethics of placebo-controlled trials of zidovudine to prevent the perinatal transmission of HIV in the third world [letter]. N Engl J Med 1998; 338: 837. French HW. AIDS research in Africa: juggling risks and hopes. New York Times 1997 Oct 9: Sect A: 1. Cohen J. Ethics of AZT studies in poorer countries attacked. Science 1997; 276: 1022. Angell M. Ethical imperialism? Ethics in international collaborative clinical research. N Engl J Med 1988; 319: 1081-1083. Penslar RL. Research ethics: cases and materials. Bloomington: Indiana University Press, 1995. Report of the Committee of Inquiry into Allegations Concerning the Treatment of Cervical Cancer at the National Women's Hospital and into Other Related Matters. Report. Auckland: Government Printing Office, 1988. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. Belmont report: ethical principles for the protection of human subjects research. Washington, DC: Government Printing Office, 1988. World Medical Association, Declaration of Helsinki IV. Adopted by the 18th World Medical Assembly, Helsinki 1964, as amended by the 41st World Medical Assembly, Hong Kong, 1989. <http://www.nih.gov/grants/oprr/irb_appendices.html#j6> Chalmers TC. The clinical trial. Milbank Mem Fund Q 1981; 59: 324-339. Freedman B. Equipoise and the ethics of clinical research. N Engl J Med 1987; 317: 141-145. Merson MH. Ethics of placebo-controlled trials of zidovudine to prevent the perinatal transmission of HIV in the third world [letter]. N Engl J Med 1998; 338: 836. Fawzi WW, Msamanga GI, Spiegelman D, et al. Randomized trial of effects of vitamin supplements on pregnancy outcomes and T cell counts in HIV-infected women in Tanzania. Lancet 1998; 351: 1477-1482. Ware JH, Antman EM. Equivalence trials. N Engl J Med 1997; 337: 1159-1161. Blanche S. Ethics of placebo-controlled trials of zidovudine to prevent the perinatal transmission of HIV in the third world [letter]. N Engl J Med 1998; 338: 837-838. Recommendations of the US Public Health Service Task Force on the use of zidovudine to reduce perinatal transmission of human immunodeficiency virus. MMWR Morb Mortal Wkly Rep 1994; 43 (RR-11): 1-20. United Nations Development Program. Human Development Report. New York: Oxford University Press, 1997.