Ethical Issues Related to the Inclusion of Pregnant Women in Clinical Trials (II)
Bonnie Steinbock
More than a billion drug prescriptions are written every year, there is unlimited self-administration of "over-the-counter" drugs, and approximately 500 new pharmaceutical products are introduced annually (Briggs et al., 1983, cited in Elias & Annas, 1987, p. 196). Moreover, a surprisingly high number of pregnant women use legal drugs; 40 percent in the first trimester, according to one study (Heinonen et al., 1977, cited in Elias and Annas, 1987, p. 196). These facts lead to the conclusion that "the potential for drug teratogenicity is thus truly remarkable" (Elias and Annas, 1987, p. 196).
Much information about the pharmacology of the maternal—fetal unit has been derived from animal studies, but it is extremely difficult to predict whether observations made in animals will have relevance to human beings. For example, preliminary testing of the rubella vaccine in monkeys indicated that the vaccine did not cross the placenta. However, when human studies were undertaken with women about to undergo abortions, it was found that the vaccine virus did cross the placenta and infect the fetus. Thalidomide is another dramatic example that negative animal data do not prove that a drug is innocuous to humans. This presents a dilemma. If we include pregnant women in clinical trials, we risk exposing fetuses to the risk of teratogenicity. If we exclude pregnant women from clinical trials, we will not have information about the effects of various drugs on the maternal/placental/fetal unit. We must therefore steer between Scylla and Charybdis, and we need appropriate guidelines to help.
This issue was addressed by the National Commission for the Protection of
Human Subjects of Biomedical and Behavioral Research, the first of whose mandates was to review and report on research involving living fetuses. The result was a report, Research on the Fetus. Among its recommendations were the following: nontherapeutic research on the pregnant woman or on the fetus in utero may be conducted or supported, provided it will impose minimal or no risk to the fetus, the woman's informed consent has been obtained, and the father has not objected (Research on the Fetus, pp. 73–76).
Several key concepts are included in this recommendation. The first is nontherapeutic research, that is, research that does not benefit the research subject, in this case, either the pregnant woman or the fetus. Placing restrictions on the use of pregnant women in nontherapeutic research limits their freedom of choice, but it cannot be said to harm them as individuals. Women taken as a class may be harmed by the exclusion of women from clinical trials. Indeed, such exclusion is likely to affect adversely society as a whole, as important knowledge that might have been acquired may not be gained. The situation is quite different for therapeutic research, to which I will return shortly.
The next key concept is that of risk to the fetus. The National Commission required that the risk to the fetus from the research be minimal or nonexistent. It maintained that all fetuses should be protected from potentially harmful research, regardless of whether they were going to be aborted or going to be born: ". . . the same principles apply whether or not abortion is contemplated; in both cases, only minimal risk is acceptable" (Research on the Fetus, p. 66). This requirement was referred to as "the principle of equality."
I disagree. In my view, because of the difference between children and early-gestation fetuses, it is crucially important whether the woman is going to abort or going to term. Early-gestation fetuses are not sentient or conscious or aware of anything. No matter what is done to them, they feel nothing. Nonsentient fetuses cannot be harmed in the way that sentient beings can be harmed; that is, they can't be hurt or made to suffer. Treatment that would cause a sentient being to experience pain is not necessarily harmful to nonsentient fetuses.
However, pain isn't the only way in which a being can be harmed. What if a fetus is exposed to substances that prevent it from developing normally, such as the rubella virus, thalidomide, alcohol, and so forth. Here, however, the harm is not to the fetus, but to the born child. It is the child who must go through life deaf and mentally retarded when the fetus has been harmed by prenatal exposure to rubella. It is the child who must go through life without limbs when the fetus has been harmed by thalidomide. It is the child who must go through life with learning disabilities when the fetus has been harmed by prenatal exposure to alcohol. If the woman aborts in the first trimester, before the fetus becomes sentient or conscious, there is no one who can be harmed. That is why a woman who plans to abort has only her own health to consider regarding drinking or
smoking, while the woman who plans to go to term has the health of her future child to consider, as well as her own health.
If this is right, then it makes no sense to insist, as did most of the Commissioners, that no procedures should be applied to a fetus-to-be-aborted that would not be applied to a fetus-going-to-term. The reason for banning potentially harmful nontherapeutic research on fetuses-going-to-term is not to protect the fetus per se, but rather to protect the future child. If the woman is going to abort, there won't be any future child, and literally no one who can be harmed or protected. Moreover, if women who are scheduled to abort are willing to participate in clinical trials, and give their informed consent, much useful information that will serve to protect future children may be gained. What if the woman is going to term? In this case, the interests of the surviving child must be considered. Could there be any objection if there are only minimal or no risks to the future child? Paul Ramsey opposed all nontherapeutic research on children, on the ground that they have not given informed consent (Ramsey, 1976). Richard McCormick thinks that some nontherapeutic research on children can be justified, and that parents can give proxy consent for their children where there is no discernible risk or undue comfort. Proxy consent is morally legitimate insofar as it represents what the child ought to choose—and everyone ought to be willing to participate in experiments that benefit the human community (McCormick, 1974). Both Ramsey and McCormick regard informed consent, either given directly or through a proxy, as morally required. However, it is hard to see the point of requiring informed consent in situations when it is literally impossible. Surely the important point is whether the research is likely to harm the children, either after or before birth. I am assuming that the question of whether research will impose more than minimal risks upon offspring is an objective and scientific matter. If so, then this is not a matter for potential participants in nontherapeutic research to assess. Rather, it is the duty of researchers to determine if the research poses more than minimal risks to offspring. If it does not, then there doesn't seem to be any objection to it.
What if the risks are either significant or unknown? Should a woman be allowed to expose her not-yet-born child to such risks? It is difficult to imagine a situation in which a woman would want to expose her future child to risks, when there is no benefit either to herself or to the child. But imagine a woman with a Mother Theresa complex. She wants to volunteer for medical research to help humanity, and she's willing to take the risk that it might harm either her or her baby. It seems entirely reasonable for us to tell her that while she is permitted to take such risks on her own behalf, she is not entitled to impose such risks on her not-yet-born child. After all, preventing her from participating in an experiment isn't infringing her bodily integrity. It isn't monitoring her lifestyle. So I see no objection to regulations preventing pregnant women who plan to go to term from participating in risky nontherapeutic research.
Restrictions are harder to justify where the research offers a potential benefit to the pregnant woman. Experimental therapy may offer the only hope to individuals who are sick and cannot be helped by tested methods, such as people who have AIDS. They have a direct personal interest in being included in clinical trials. Not allowing them to participate does not merely infringe their autonomy and right to decide for themselves; it may foreclose the only hope they have of survival. It seems, therefore, that it would be wrong to exclude pregnant women who are not going to term from experimental trials that might benefit them.
What about women who wish to continue their pregnancies? I don't think it matters much if the therapy is experimental or conventional. The question is the same: does a woman who is planning on-going to term have the right to undergo therapy that poses a risk to her fetus?
A recent story in the New York Times described an Italian woman who refused cancer therapy out of concern that it would harm the fetus she was carrying. She was willing to die in order to avoid harming her fetus. If one views the fetus as having the same status as a born child, then this may seem like a noble act of self-sacrifice. (This is how the Vatican regards it. I believe that they are taking steps to canonize her.) My own view is that her refusal of therapy is certainly permissible, but not morally required. No one is morally required to sacrifice her own life or health to sustain the life of a fetus (Thomson, 1971).
But what if the therapy isn't likely to be lethal to the fetus, but rather risks causing it to be born with severe handicaps? If the risk is great enough, and the handicaps severe enough, terminating the pregnancy might be morally required. For abortion is not a harm to the nonconscious fetus, but being born with very severe impairments may be unfair to the child (Steinbock and McClamrock, in press).
What if the potential benefit is to the fetus, that is, the surviving child? In general, parents have the responsibility for deciding whether to impose experimental treatment on their minor children. Similarly, the prospective parents should be allowed to decide, within comparable limits, whether the potential benefits to the fetus outweigh the risks. However, there is one glaring difference between the two situations. Prenatal treatment of a fetus can be done only through the body of its mother. So the risks to her are an important part of the decision. In recent years, fetal therapy and surgery has grown by leaps and bounds. In one dramatic case (which by now has no doubt been repeated several times) a surgeon removed a previable fetus from the uterus, repaired his diaphragmatic hernia, put the fetus back in the womb, and delivered him six weeks later by cesarean section (Kolata, 1990). The mother had no obligation to try the therapy, given the risks and burdens to her from two cesareans and six weeks of enforced bed rest, especially since it was very experimental and carried
no guarantee of success. Even if such therapy should become ''routine," it still should never be compulsory. But neither should anyone deny a pregnant woman the chance to save her baby's life.
Finally, I'd like to consider the role of the woman's partner in making these decisions. By partner, I mean the man who is not only the genetic father, but who also intends to be a rearing parent. It seems to me that if the woman is planning to abort, the man should have no say in whether she participates in a clinical trial. For while a man has a legitimate interest in the well-being of his offspring, if the woman decides to abort, there won't be any offspring. The decision to participate in a clinical trial belongs solely to the pregnant woman.
A man would have a legitimate interest in preventing a woman who did not plan to abort from participating in nontherapeutic research that posed some risk to the not-yet-born child. However, there's a strong case for society's banning pregnant women who plan to go to term from such clinical trials, whether or not the father objects.
Men have legitimate interests in the health of their not-yet-born children. It is not unreasonable for them to be concerned if their pregnant wives smoke or abuse alcohol or drugs. It seems unfair that a man who intends to parent a child should have to stand by and watch behavior that risks harming his future child. He is certainly justified in trying to persuade his wife to get treatment, for the sake of their baby. He might even be justified in coercing her to get treatment, since this will benefit both her and the baby. But he would not be justified in preventing his pregnant wife from getting therapy necessary for her own life and health, to protect the future child. Being a Good—or Splendid—Samaritan may be noble and praiseworthy; it is not something one individual has any right to demand of another.
REFERENCES
Briggs, G. G., Bodendorter T. W., Freeman R. K., et al. 1983. Drugs in Pregnancy and Lactation: A Reference Guide to Fetal and Neonatal Risk. Baltimore: Williams & Wilkins.
Elias, S., and Annas, G. 1987. Reproductive Genetics and the Law. Chicago: Year Book Medical Publishers.
Heinonen, O. P., Slone, D., and Shapiro, S. 1977. Birth Defects and Drugs in Pregnancy. Littleton, Mass.: Publishing Science Group.
Kolata, G. 1990. Lifesaving surgery on a fetus works for the first time. The New York Times, Thursday, May 31, A1.
McCormick, R. A. 1974. Proxy consent in the experimentation situation. Perspectives in Biology and Medicine 18:2–20.
The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. 1975. Research on the Fetus: Report and Recommendations (cited as Research on the Fetus), DHEW Pub. No. (OS) 76-127.