20th World Congress of Philosophy Logo

Bioethics and Medical Ethics

Second Best:A Case for Lowering Testing
Standards in Third World Countries

Dennis R. Cooley

bluered.gif (1041 bytes)

ABSTRACT: Recently, Sidney Wolfe, director of Public Citizen’s Health Research Group (PCHRG), charged the National Institute of Health (NIH) and Center for Disease Control (CDC) with sponsoring fifteen immoral HIV studies in sub-Saharan Africa. The trials are being conducted to determine if certain alternate medical procedures or a short course of treatment with AZT, zidovudine or other drugs prevent some mother-child HIV transmissions. (1) Since the control group receives only placebos rather than AZT, Wolfe claims that the tests give suboptimal treatment that will result in more children contracting HIV and AIDS. (2) Public Citizen’s Health Research Group and others are calling for an immediate cessation of these important experiments. Public Citizen raises an important moral question. (3) Is it morally permissible to lower testing standards for the Third World? Unlike PCHRG, I contend that the answer to this question is yes, if the trials meet certain conditions. I explain both the First Best and Second Best Method (FBM and SBM, respectively) of testing new drugs and then compare the two. Next, I show the FBM’s impracticality in developing countries releases researchers from the moral obligation to use it. I then propose a new set of criteria — the Second Best Criteria (SBC) — that show if a test is moral or not. Finally, I argue that imposing a developed country’s moral standards for clinical trials in the Third World is immorally insensitive to the needs and conditions present in the latter area.

bluered.gif (1041 bytes)

Section 1

In clinical trials there are two different standard protocols for testing new drugs. The First Best Method tests the effects of a known drug (a benchmark), rather than a placebo, against those of the research drug. Generally, the benchmark is the standard of care, i.e., the standard treatment in the medical field. Using this method guarantees that all the test subjects will receive some sort of treatment for their ailments.

The Second Best Method involves giving one half of the test group the research drug and the other half a placebo. Any beneficial results or side effects the former group exhibits that the control group does not can be attributed to the research drug. Since the control group merely receives a placebo, the members are unlikely to receive any benefits other than more personalized attention from the physician-researchers.

The greatest benefit of the First Best Method over the Second Best is that the test subjects, generally, receive more care. As a result, more people usually are prevented from being harmed than would occur if their diseases was left completely untreated. Based on the maxim that we ought to always minimize suffering when we easily can, obviously, if there is an option between the two testing protocols, FBM is to be preferred. Furthermore, by reducing pain for the test subjects we respect them as persons, rather than treating them as mere guinea pigs.

Section 2

Wolfe accuses the NIH and CDC of unethically performing SBM testing in situations where they could be using FBM. By making the charge, Wolfe implicitly assumes that the trials are being performed in countries where there is adequate access to the standard of care and financial resources with which to procure it. In other words, a place where FBM testing could be carried out, but is not.

According to Public Citizen, First Best experiments in Africa should use AZT, the standard of care, as a control rather than a placebo. In other trials, especially, ACTG 076, AZT has proven to reduce the mother-child transmission rate of the HIV virus from 23% to 8%. Conducting an SBM study with a placebo rather than AZT would be unethical; the HIV infection rate and subsequent loss of life in the placebo group would be unjustifiable on deontological and utilitarian grounds. In this situation, we ought to respect the test subjects as persons and try to minimize their suffering. If, at the same time, we can easily save more lives, then we certainly ought to do so.

Unfortunately, attempting to implement FBM trials using AZT for a benchmark in the sub-Sahara are doomed from the start for two reasons. First, the countries in which the trials are conducted cannot afford the expensive benchmarks, much less the treatments for all their infected citizens. One standard of care, AZT, is available for the experiments, but preventative treatments using the drug cost between $400 - $900 per person. (Brown, p. A 14) (4) In sub-Saharan Africa where the median per capita expenditure for all health care was $14 in 1992, the expense is prohibitive. (Mansergh, p. 8) (5)

The second reason FBM is not feasible for implementation is the inaccessibility of AZT in developing countries for general consumption. A trial conducted in Thailand by the Harvard School of Public Health was criticized for using AZT as a benchmark because only a few people had access to the drug outside of these tests. To obtain results that will best help those in the Third World, medicines and procedures that are actually commonly available to the regions must be found. Anything else is an unethical waste of time, resources, and lives because it ignores the fundamental realities of Third World situations. (6)

Given these conditions, it is obvious a different moral standard must be applied to certain research experiments in developing nations. A more pragmatic approach to medical testing in developing countries, such as the Second Best Criterion expounded in the next section, is called for which will satisfy morality while still allowing for testing of new drugs and procedures. Second Best, that is, the placebo method, in some situations, can be moral even though AZT or other standards of care are available.

Section 3

The Second Best Criterion, a mixture of utilitarian and deontological conditions, classifies some clinical trials performed in developing countries as moral even though they would be impermissible in developed nations.

SBC has three necessary conditions, the first of which is utilitarian in nature: an experiment may be performed only if it is perceived as likely to maximize utility, though it need not actually maximize utility. The second principle requires that all the test subjects be respected as persons. Finally, there must be a dire need on the part of the human subjects for the experiment to be performed-trivial experiments are not permissible.

Returning to SBC’s first condition, researchers must rationally believe that conducting the trial is probably better overall than the alternatives of doing nothing or acting in another way. An example will be helpful in illustrating this requirement.

Suppose that a researcher wants to conduct trials on an expensive new drug, neo-aspirin, which is medically identical to aspirin. In this case, trial experiments for neo-aspirin approval would not be justified, according to SBC’s first condition, because it is not rational to believe utility will be maximized. If aspirin does everything the other drug does at a cheaper price, then it would be a waste of time and resources to test the latter.

However, if neo-aspirin has additional properties to recommend it, (7) and there are no other experimental drugs with a greater chance of success, then neo-aspirin would be test worthy. The probable utilities of conducting the experiment would be better than any alternative. (8)

Returning to the sub-Saharan HIV clinical trials decried by the Public Citizen, utility would most likely be served, if the experiments were allowed to proceed. Wolfe and others such as Joep Lange fail to recognize that it is not a choice between AZT and other benchmark drugs. Though FBM is a possible option for researchers, since the developing countries are so poor, and unlikely to transfer money earmarked for projects considered to be more important to the nation’s interests, the proper funding is not available to conduct FBM trials. Therefore, the actual choice is between some drug treatment or none at all. Without the trials, the human subjects will receive little to no medical care; hence it is rational to believe that this alternative will be highly disvaluable, while producing little to no positive value.

On the other hand, if the trails are performed, then at least some children may be prevented from becoming HIV infected. More people, thus, may be saved from pain or death than probably would have been if no tests were conducted. In a situation where it is a choice between either saving some from unnecessary pain or doing nothing, we should choose the former. Even though there is a high disvalue in the control group’s suffering from SBM testing, the utility is likely to be higher than omitting to act. Though it is doubtful that, in the short term, the trials’ benefits will outweigh the costs, it does not follow that SBM testing ought not be done.

Other benefits are likely to be reaped by these tests besides treatment for one half of the test group. If the alternate procedures or cheaper drugs prove to be useful, then they can be administered to non-test subjects, once the study is completed. The clinical trials are intended to find economical drugs for Third World countries. If they are successful, then, in the long run, other children will be saved from the pain and suffering associated with HIV infection. By conducting SBM trials, NIH and CDC are merely choosing the lesser of two evils alternatives.

SBC’s uncontroversial second condition is borrowed from Public Citizen’s objection to SBM trials. All would agree that there must always be a recognition of and respect for the intrinsic value of the human persons being used as test subjects. One necessary feature of respecting persons is to obtain their informed consent to participate in clinical trials prior to using them for such purposes. To do otherwise would be to illicitly devalue their autonomy which is an essential part of their personhood.

The third and final criterion for the Second Best Criterion is the Principle of Dire Need (PDN) which requires an urgency that the tests be done. Roughly, PDN allows testing standards to be lowered only if doing so will serve a critical need of the test subjects, and, perhaps, society, such as saving lives or preventing severe harm to them. Put more formally:

(PDN) E is a critically needed experiment iff E’s success likely will prevent at least some loss of human test subject life or incapacitating pain and suffering for the test subjects, and, perhaps, society, where incapacitating pain and suffering is either long or short term pain  — either physical or emotional — that makes it impossible to function in everyday life.

This definition of pain and suffering recognizes that there are certain everyday jobs that we must do in our choice of lifestyles. If, for example, I am a teacher, then I must, among other things, grade, lecture, and meet with students. If I cannot perform these tasks due to pain, then I am incapacitated from participating in every day life.

To show the distinction between incapacitating versus non-incapacitating pain, it is helpful to compare two different conditions. A migraine makes it impossible to go about doing even the most common of everyday chores. The nausea and pounding headaches causes interacting with students, etc. unmanageable tasks. All the migraine sufferer can do is usually lie down hoping for a quick, merciful death; thereby incapacitating him from most daily activities.

Slight headaches on the other hand, interfere with but do not make it impossible to accomplish every day tasks. One may still do the tasks required by the teaching profession, for example, albeit in a less energetic and pleasant state. The pain of slight headaches are just as real as that of migraines, but it is not so severe that it causes a complete cessation of activity as is characterized by the latter malady.

PDN is a necessary condition for ethical trials because it eliminates the possibility of trivial medical experiments being justified. (9) Only the most critical situations will abrogate the rule to always conduct FBM trials whenever possible.

The Principle of Dire Need recognizes that incapacitating pain, suffering, and the loss of life are extremely disvaluable and ought to be avoided if possible. Pain and suffering are not only bad because of their effects, they are bad in themselves. Death is evil for it is the loss of something valuable, an actual or potential human person.

The people who attack the use of Second Best Method testing in developing countries are rather near-sighted to the actual plight of people who inhabit these areas. In sub-Sahara Africa there are an estimated 13 million HIV infected people, (UNAIDS (2), p. 1) There is no time to waste trying to do something to help them; the situation is too dire to omit acting.

The mother-child HIV transmission research being done in the sub-Sahara passes the PDN test. If the trials were not done, then there would likely be a greater transmission of the HIV virus to the infants of the infected mothers. As a consequence, a greater number of AIDS cases would develop from both the infected offspring and anyone they may in turn infect. The cycle of infection and death has the potential of increasing instead of decreasing. Since these tests will probably lead to drugs that will help prevent loss of life and incapacitating pain and suffering, they satisfy PDN.

Summing up the Second Best Criterion for second best medical tests, a clinical trial in a developing country is moral if

1. In the short and long term, use of the new drug will likely maximize utility, and

2. The test subjects are respected as persons, which includes obtaining their informed consent, and

3. the test is of critical need for the likely prevention of loss of human life and incapacitating pain and suffering, especially for the test subjects. (PDN)

The HIV trials in the sub-Sahara pass the Second Best Criterion. First, there is a great deal of value in infants being saved from HIV infection, thereby eliminating one source of AIDS. Secondly, there is a possibility that newer, cheaper drugs can be discovered and made available to more consumers. Utility is potentially greater for this alternative than doing nothing, or perhaps, using AZT in experiments for unrealistic situations.

Also, the trial subjects have given their informed consent to the project. They can even leave the trial when and if they choose to do so. Finally, the loss of life and pain and suffering is reduced. By giving the pregnant mothers some treatment rather than none, there is a greater likelihood of fewer HIV infected infants being born, thereby satisfying PDN.

Section 4

Though Public Citizen denounced the sub-Saharan HIV experiments as analogous to those conducted by the Nazis, there are moral dissimilarities between the two. The latter trials do not pass the Second Best Criterion while the African ones do. First, unlike the US sponsored trials, the European test subjects were involuntary participants in the Nazi experiments.

Secondly, PDN is not violated by the HIV trials, while it is by the other set of cases. Nazi experiments, such as those conducted to discover how much pressure a human being could endure before dying, were not intended to benefit the human subjects. Furthermore, there was obviously not a dire need on the part of the subjects for the trial to be performed.

Section 5

Well intentioned, though misguided, critics of current HIV clinical tests in Africa, such as Public Citizen must recognize that the situation in developing countries is different from that found in developed areas of the world. By failing to appreciate situational realities there is an attempt to impose irrelevant, harmful rules on poor societies incapable of bearing them.

Ramachandra Guha, addressing the rather imperialistic efforts of some American deep ecologists, illustrates the problem of forcing inappropriate moral principles on developing Third World nations. (Guha, p. 126) Instead of taking into account the actual situation of developing countries, American deep ecologists attempted to implement environmental principles that were appropriate only in developed countries such as the United States. The Third World’s needs for sustainable fuel sources, soil conservation, clean water, etc., are usually ignored in the race to maintain or restore areas to a pristine condition. In one instance, farmers were forced off their land in order to enlarge a tiger habitat. As a result of this myopia, people were unjustifiably harmed in the attempt to make them follow costly principles that were beyond their meager resources.

Though Guha’s criticism focuses on environmental concerns, I think that the point translates nicely for medical research in developing countries. Rich nations can afford FBM research and the moral medical questions that accompany it, poor ones cannot. It is a mistake of the highest degree to assume that the moral standards of the rich should be forced upon the poor. People are dying in Africa, this is the immediate concern that must be addressed. Once this problem is rectified, there is adequate time to discuss the moral niceties that only those with sufficient resources can afford.

In the end of his article, Guha calls for deep ecology to cease its imperialistic efforts and start working toward a more holistic approach to the Third World’s environmental problems. I will borrow his sentiments and apply them to the medical trials in the sub-Sahara. Ethical researchers and government watchdogs ought to work toward an appropriate synthesis of technology and alternative lifestyle. (Guha, p. 131) Only by being sensitive to existing conditions will we be able to tailor moral medical experiments and treatments for diseases in the Third World.

bluered.gif (1041 bytes)


(1) Some tests include shorter courses of treatment with AZT than were studied in the landmark trial 076 Other experiments range from intravenous HIV immunoglobin (HIVIG) administered to infants to disinfecting the cervix and vagina immediately before childbirth. (Cohen (2), p. 625)

(2) Joep M A. Lange also condemns the practice of placebo-research drug tests when benchmark-research tests can be performed. (Lange, pp. 548 - 550)

(3) Public Citizen has likened the sub-Saharan experiments to those performed by the Nazis during their reign of terror Marcia Angell, executive director of the New England Journal of Medicine, like Public Citizen, compared the trials to the infamous Tuskegee experiment. (Goodman, D3)

(4) UNAIDS estimates the cost at $1,500 per mother-child pair (UNAIDS (3), p. 1)

(5) UNAIDS estimates of the three million children infected with HIV, 85% are from the sub- Sahara (UNAIDS (3), p. 1) If each mother had been treated with AZT, the total costs would ave been between $1 to $4.5 billion dollars. The expenditure would be prohibitive in those countries that only spend $2.00 a year per capita on health care. (Cohen (2), p. 626)

Developing countries require cheaper treatments that will help reduce the number of HIV infected babies. However, even the $20 (1994 dollars) a short course of treatment with AZT would cost may be beyond the reach of the health budgets of the sub-Sahara. Though there are those able to afford more expensive health care, most cannot. In some countries, health care expenditures are around $2 per capita. (Cohen (2), p. 626)

Even in wealthy nations such as the U.S., the cost of treatment can be astronomical. One treatment consisting of a drug "cocktail" of 14 medications per day costs $12,000/year -per-person. Alaska, Arizona, Arkansas, Florida, Kansas, Mississippi, Montana, Nebraska, New Hampshire, South Dakota, Tennessee, and Wyoming have decided to do little to nothing to save certain segments of the uninsured from AIDS. (Pederson, p. 60)

It should also be mentioned that drug companies are-at best-reluctant to drop the prices of their drugs, especially if a cheaper drug is being tested, even if it is in-house. Glaxo Wellcome PLC, the owner of AZT and 3TC, has been accused with delaying testing of the cheaper 1592 that may do the same job as the more expensive pharmaceuticals. (Waldholz, p. 1)

To be fair, Glaxo has agreed to donate AZT to the clinical trials that test its effectiveness against placebos or in short versus long courses of treatment. However, even if AZT is shown to be effective as a result of these trials, it still may be beyond the limited resources of most developing countries.

Unless AZT is donated to them, the poorer countries will be unable to use the clinical test results. Help for the poor is unlikely as Andrew Revell, the project manager for AZT at Glaxo says, "...I very much doubt that it will be Glaxo Wellcome giving unlimited amounts of AZT." (Cohen (2), p. 626)

(6) Using the 076 protocol of long term treatment ignores important facts about pregnant women in sub-Saharan countries First, women do not "visit medical clinics until they are in labor-if then. What’s more, these women often don’t know they are infected with HIV." (Cohen (2), p.626)

What is needed is a short term treatment that will be available to women which takes the facts of the situation into account. Hence trials should try to find drugs or procedures that will limit the mother-child transmission of HIV that are inexpensive, available in the region, and effect in short term treatment.

(7) For example, unlike aspirin, it does not cause the side effect of stomach upset with regular use to help prevent heart attacks

(8) The method of determining the best outcome is based on probability statistics If the consequences of introducing a drug are likely to have greater utility than any alternative, then testing ought to be done. Suppose, for example, that neo-aspirin has 36% probability of producing the best outcome; a second experimental drug, not-aspirin, has a 30% probability; and not testing either in favor of aspirin has 34%. In this situation, SBC’s first condition says to test the neo-aspirin.

(9) Suppose, for instance, that a pharmaceutical company wants to test a new type of toothpaste that is likely to brighten teeth to their whitest The corporation does not want to supply an alternative brand for a benchmark, so they use a placebo. Eventually, the toothpaste is approved for mass consumption.

The experiment passes the first criterion for it is likely, in the long run, to produce the greater utility than alternatives. Whiter teeth makes people happier. By getting the informed consent of the test subjects, the company satisfies the second criterion. Everyone’s autonomy is respected. However, the experiment is unethical for it is not necessary. The stakes where not high enough to forgo FBM; only drastic situations may require drastic actions. Discolored teeth are trivial.


Brown, David, "Medical Group Condemns U.S. AIDS Drug Tests in Africa for Using Placebo", The Washington Post, 23 April 1997, A 14.

Brown, Phyllida, "Uganda’s AIDS Agony Revealed", New Scientist, 12 June 1993, p. 6.

Cohen, Jon, "AIDS Trials Questioned", Science, vol. 276, 25 April 1997, pp. 520-2.

——"Bringing AZT to Poor Countries", Science, vol. 269, 4 August 1995, pp. 624-626.

DeNoon, Daniel J,, Senior Editor, "(AW) Ethics ‘Disaster’ Seen in AIDS Drug Trials Using Suboptimal Therapy", AIDSWEEKLY Plus, 12 May 1997, http:/www.aegis.com/aegis/aidswkly/aw1997/AW970505.html, pp. 1-5.

Esparza, Jose, William L. Heyward, and Saladin Osmanov, "HIV Vaccine Development: From Basic Research to Human Trials", From the Joint United Nations Program on HIV/AIDS (UNAIDS), http:/www.us.unaids.org/highband/document/vaccines/aids8.html, pp. 1-20.

Falk, Geoffrey T., "Review of "The African AIDS Epidemic," by J.C. Caldwell and Pat Caldwell, Scientific American, March 1996", Circumcision Information and Resource Pages, http:/www.cirp.org/CIRP/library/disease/HIV/falk/, pp. 1-3.

Goodman, Ellen, "Double Ethical Standard being Imposed on AIDS Infected Mothers", The Daily Reflector, 28 September 1997, p. D3.

Gorovitz, Samuel, "Informed Consent and Patient Autonomy," Ethical Issues in Professional Life, Joan C. Callahan (ed.) (Oxford University Press: New York, NY, 1988) pp. 182-188.

Heyward, William L., Saladin Osmanov, and Jose Esparza, "Establishment of WHO-Sponsored Field Sites for HIV Vaccine Evaluation in Developing Countries", Development and Application of Vaccines and Gene Therapy in AIDS (International Workshop, Naples, 15-16 June 1995), http:/www.us.unaids.org/highband/documnet/vaccines/naples.html, pp. 1-5.

Kant, Immanuel, "Crimina Carnis" Philosophical Perspectives on Sex and Love, Robert M. Stewart (ed.), (Oxford University Press: New York, NY, 1995) pp. 144-5.

Kaiser, Jocelyn, "Group Brands AIDS Trials Unethical" Science, vol. 276, 25 April 1997, p. 519

Lange, Joep M. A., "Current Problems and the Future of Antiretoviral Drug Trials", Science, vol. 276, 25 April 1997, pp. 548-550.

Mansergh, Gordon, Anne Haddix, et. al., "Cost-effectiveness of Short-Course Zidovudine to Prevent Perinatal HIV Type 1 Infection in a Sub-Saharan African Developing Country Setting", JAMA: The Journal of the American Medical Association, Special Communication, 10 July 1996, http:/www.ama-assn.org/special/hiv/library/jama96/sc6200.htm#tab3, pp. 1-11.

McKinley, James C. Jr., "A Ray of Light in Africa’s Struggle with AIDS" The New York Times, April 7, 1996, Sec. 1, p. 1, col.2

National Institutes of Health: "Guidelines for the Conduct of Research in the Intramural Research Programs at NIH", 3rd Edition, January 1997, http:/www.nih.gov/news/irnews/guidelines.htm#anchor130083, pp. 1-7.

—— "NIAID Funds Adult AIDS Clinical Trials Group", News Release, 30 November, 1995, http:/www.niaid.nih.gov/newsroom/actg_nr.htm, pp. 1-5.

—— "NIH Research is an Investment...In Tomorrow’s Health Care", Ad Hoc Group for Medical Research Funding, 3 July 1997, http:/www.aamc.org./research/, adhocgp/tomorrow.htm, pp. 1-4.

—— "Summary of the Meeting of a Panel to Review Studies of Transplacental Toxicity of AZT", Fact Sheet from the NIH, 3 July 1997, http:/www.niaid.nih.gov/factsheets/aztsumm.htm, pp. 1-4.

Papadopulos-Eleopulos, E., Valendar F. Turner, John M. Papadimitriou, and Harvey Bialy,   "AIDS in Africa: distinguishing fact and fiction", World Journal of Microbiology & Biotechnology, vol. 11, pp. 135-143.

Pederson, Daniel and Eric Larson, "Too Poor to Treat" Newsweek, 28 July 1997, p. 60.

Rath, Richard, "Horton Hears the W.H.O." Dissonance, AIDS and Africa Resource page, 3 June 1996, http:/www.way.net/dissonance/nyrb_ltr.html, pp. 1-3.

UNAIDS "HIV/AIDS: The Global Epidemic", Press Release, 28 November 1996, http:/www.us.unaids.org/highband/document/epidemic/situat96.html, pp. 1-15.

—— "HIV/AIDS epidemiology in sub-Saharan Africa", 3 July 1997, http:/www.hiv.unaids.org/unaids/press/factafr.htm, pp. 1-3.

—— "UNAIDS Announces New Clinical Trials for the Prevention of Mother-To-Child Transmission of HIV", Press Release, 9 July 1996, http:/www.hiv.unaids.org/unaids/press/mtctpren.html, pp. 1-2

No Author Listed: "AIDS in Africa", The Guide, 9 May 1996, http://www.guidemag.com/newsslant/aids.africa.html. pp.1-2

bluered.gif (1041 bytes)


Back to the Top

20th World Congress of Philosophy Logo

Paideia logo design by Janet L. Olson.
All Rights Reserved


Back to the WCP Homepage