NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on the Ethical and Legal Issues Relating to the Inclusion of Women in Clinical Studies; Mastroianni AC, Faden R, Federman D, editors. Women and Health Research: Ethical and Legal Issues of Including Women in Clinical Studies: Volume 2: Workshop and Commissioned Papers. Washington (DC): National Academies Press (US); 1999.

Cover of Women and Health Research

Women and Health Research: Ethical and Legal Issues of Including Women in Clinical Studies: Volume 2: Workshop and Commissioned Papers.

Show details

Ethical Issues Related to the Inclusion of Pregnant Women in Clinical Trials (I)

John Robertson

Researchers, institutional review boards (IRBs) and others reviewing clinical research including pregnant women must assess the effect of proposed research on the pregnant woman, on the developing fetus, and on the child whom the fetus, if carried to term, will become. In most instances concern with fetal effects is not by virtue of the fetus's interests in its own right, but by virtue of the effect which prenatal interventions affecting the fetus will have on offspring.

A set of guidelines for such research was developed by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Science Research in 1974. These guidelines were incorporated into federal regulations for research with human subjects in 1975, and continue to apply today (45 C.F.R. §§ 46.201–46.211). They are generally sound with the specifications and modifications discussed below.

The Pregnant Woman as Subject

Both the National Commission and the federal regulations distinguish clinical research involving pregnant women on the basis of whether the woman or the fetus is the subject of the research. In each case they make a further distinction between research that is therapeutic—the purpose of the activity is to meet the "health needs of the mother" or "the health needs of the particular fetus"—and research that is nontherapeutic.1 The amount of risk which may be accepted depends on this set of distinctions.

Therapeutic: To Meet the Health Needs of the Mother

Pregnant women may participate in clinical research where the "purpose of the activity is to meet the health needs of the mother" regardless of the degree of risk to the fetus and offspring. If the purpose of the research is not to meet her health needs, she may participate only if "the risk to the fetus is minimal."

While this rule is generally sound, it conceals some problems. The main problem concerns the broad phrase "health needs of the mother." Consider an established treatment for a disease or condition that is safe and effective for women whenever it is given, but also has a very high risk of affecting future offspring if given during pregnancy. Ethical judgment of whether the woman should be able to have the treatment during pregnancy will depend not merely on whether the treatment will affect her "health," but also on the burdens and benefits to her of having treatment during pregnancy or after. The type of benefit to her alone is not determinative, but the magnitude is. The more minor the benefits the less discretion the woman should have to accept treatment, if there is any risk beyond minimal to offspring.

Such a standard requires weighing the importance to the woman of the health need in question versus the risk to offspring. Treating morning sickness or a cold during pregnancy is certainly a health need. But if the drug used to treat those conditions is teratogenic, it would be unethical to take it even though it is directed at treating her "health." If this is true about established therapies, then it is even more true about experimental therapies. If use of an experimental drug poses more than minimal risks to the fetus and offspring, a woman should have even less of a moral right to take such a drug to treat a cold, morning sickness, or any condition that is not life-threatening or very serious, where the primary purpose of the research is to meet her health needs. Thus a researcher, an IRB, or other review body should make a judgment about the degree of the benefits or burdens of taking or forgoing the experimental treatment relative to the harm to the fetus and offspring if it is given. A purpose of treating the "health needs" alone of the pregnant woman is not ethical when the benefits to her are greatly outweighed by the risks to fetus and offspring. The current federal regulations are overbroad to the extent that they would permit such research to occur.

Nontherapeutic Research: Not Meeting the Health Needs of the Pregnant Woman

Where the purpose of clinical research involving a pregnant woman is not to meet her health needs, the regulations limit such research only to instances where "the risk to the fetus is minimal." The implicit ethical assessment is that a pregnant woman may not harm expected offspring when there is no health benefit to her.

The first thing to note about this regulation is the ambiguity inherent in "risk to the fetus." Strictly speaking, "risk to the fetus" could be interpreted to mean only those risks that will prevent the fetus from being born alive, i.e., that might induce miscarriage. But that meaning does not make sense because women do not have moral duties to bring previable fetuses to term. Hence, they would be morally entitled to engage in activity which has a risk of inducing miscarriage, because the fetus itself lacks interests or rights. Except for persons who view the fetus as a person or moral subject in its own right, the moral concern with research or other impacts on fetuses arises because fetuses generally go to term and become offspring. More than minimal risk to a fetus is of ethical concern because of the impact which that risk will have on the resulting child. Thus it is necessary to understand ''risk to the fetus" as "risk to the fetus that will be carried to term." The only qualification to this understanding would arise with research involving viable fetuses. In those cases risk to the fetus might also be of concern because it prevented an entity with interests in itself from being born.2

Thus understood, the point of the regulation is to protect expected offspring from experimental prenatal harms that are not justified by important health needs of the woman. The woman is not free to sacrifice the interests of expected offspring by her interest in serving the needs of science or of other women. She is free to make a martyr of herself, but she is not free to make a martyr of her children, whether the martydom occurs by prenatal or postnatal conduct.

This understanding of the regulation is ethically sound. The only argument against it would be the claim made by some feminists that a pregnant woman should be free to do what she wants with her body, and that any restrictions on her behavior is an intolerable restriction of her freedom. The very issue being discussed shows that this position is unsound, even if one believes that coercive state interventions to prevent prenatal harm to offspring are rarely justified on policy grounds. The regulation, however, is ethically sound. No one, not even the pregnant woman, has a moral right to engage in experimental clinical research not necessary to meet her own substantial health needs when there will be a major impact on offspring.

The Fetus as Subject

Clinical research involving pregnant women may also be directed at the fetus as the subject of the research. Again, the major ethical distinction in this category is between therapeutic and nontherapeutic fetal research, the former being cases where the "purpose of the activity [is] to meet the health needs of the particular fetus" (45 C.F.R. § 46.208(a)).

Therapeutic: To Meet the Health Needs of the Fetus

The federal regulations permit research with the fetus as subject when "the purpose of the activity is to meet the health needs of the particular fetus and the fetus will be placed at risk only to the minimum extent necessary to meet such needs."

This standard is ethically unexceptional once the ambiguity mentioned earlier in "health needs of the fetus" is resolved. The term in this context would apply to procedures that will enable the fetus to survive, i.e., come to term, and survive in a healthy or undamaged way. Thus experimental procedures designed to prevent or treat handicap or disease in offspring would be permitted, because the health needs of the fetus include the health needs of the child that the fetus will become. Prenatal procedures on the fetus are necessary to safeguard the welfare of offspring. Thus experimental in utero fetal surgery to correct diaphragmatic hernia in the fetus may be done because of the impact which that condition will have on offspring.

Note that there is no obligation to include the fetus in experimental research. Parents have no duty to subject their fetuses and offspring to experimental procedures, even when there is no alternative treatment available, precisely because it is experimental and thus not clearly a benefit. On the other hand, parents should be free to have experimental in utero therapies used when they reasonably believe that the benefits of the procedure to offspring outweigh the risks.

Nontherapeutic Fetal Research: Not to Meet the Fetus's Health Needs

The federal regulations restrict research not directed to meet the health needs of the fetus to situations in which "the risk to the fetus imposed by the research is minimal and the purpose of the activity is the development of important biomedical knowledge which cannot be obtained by other means" (45 C.F.R. § 46.209(b)).

This rule is ethically sound—indeed, is morally obligatory—in situations in which the pregnancy may or will go to term. In that case research not designed to benefit offspring would occur that has more than minimal risk of harming offspring. Because parents have no right to harm their offspring, whether by prenatal or postnatal conduct, they have no more right to include their offspring in prenatal experiments that carry a risk of harm than they do to include them in postnatal research. Note, however, that they would have the right to include them in minimally risky research on the ground that no ethical or legal duty would be violated in doing so.

However, this regulation is not justified in situations where the pregnancy will not go to term. In such cases, strictly speaking, there is no risk of harm to the fetus, because a previable fetus is insufficiently developed to have interests in its own right, and thus cannot be harmed. The National Commission, however, took the position that all fetuses should be treated equally—those going to be aborted should be treated the same as those going to term. Under the Commission's understanding no research could be done on fetuses going to be aborted that could not be done on fetuses going to term. Treating all fetuses the same overlooks the fact that nontherapeutic research on fetuses going to term could affect the interests of offspring, whereas research on fetuses to be aborted cannot hurt future offspring, much less previable fetuses, which are nonsentient and do not have interests.

There is one possible risk with nontherapeutic research on fetuses going to be aborted that is of ethical concern. That risk is that the woman who consents to that research might change her mind about abortion after the experimental procedure has begun. If so, research begun with no intention of harming offspring could end up harming children who are later born. Of course, once the experimental procedure has begun, the woman might be reluctant to change her mind precisely because of risk of harm to offspring. To make research ethically acceptable on fetuses going to be aborted, the experimental procedure should be administered shortly before the abortion or in other circumstances in which it is very clear that the pregnancy will in fact be terminated, and that the woman has had sufficient opportunity to contemplate that decision. Researchers and reviewers should assure that this condition is met.


1. § 46.207(a); § 46.208(a).

2. This statement assumes a certain view of why viable fetuses are protected. If protection is based on sentience alone—an interest in avoiding pain and suffering—they may not also have an interest in coming to term.

Copyright 1994 by the National Academy of Sciences. All rights reserved.
Bookshelf ID: NBK236571


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.3M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...