- Email: [email protected]
Rethinking maternal-fetal conflict: gender and equality in perinatal ethics
10.
11.
12.
13.
human chorionic gonadotropin and risk of breast cancer. Cancer Epidemiol Biomarkers Prev 1995;4:437– 40. Alvarado MV, Alvarado NE, Russo J, Russo IH. Human chorionic gonadotropin inhibits proliferation and induces expression of inhibin in human breast epithelial cells in vitro. In Vitro Cell Dev Biol 1994;30A:4 – 8. Srivastava P, Russo J, Russo IH. Chorionic gonadotropin inhibits rat mammary carcinogenesis through activation of programmed cell death. Carcinogenesis 1997;18:1799 – 808. Huynh H. In vivo regulation of the insulin-like growth factor system of mitogens by human chorionic gonadotropin. Int J Oncol 1998;13:571–5. Huang Y, Bove B, Wu Y, Russo IH, Yang X, Russo J. Microsatellite instability during the immortalization and transformation of human breast epithelial cells in vitro. Mol Carcinogenesis 1999;24: 118 –27.
Rethinking maternal-fetal conflict: Gender and equality in perinatal ethics
Address reprint requests to:
Ch. V. Rao, PhD Department of Obstetrics and Gynecology University of Louisville Health Sciences Center 438 MDR Building Louisville, KY 40292 E-mail: [email protected]
Received March 9, 2000. Received in revised form May 30, 2000. Accepted June 22, 2000. Copyright © 2000 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
ing fetal needs perpetuates social inequalities. This model provides the ethical foundations for moving law and policy away from criminalization and toward prevention of prenatal harm. (Obstet Gynecol 2000;96:786 –91. © 2000 by The American College of Obstetricians and Gynecologists.)
Lisa H. Harris, MD Practitioners who care for pregnant women face dilemmas when their patients use illicit drugs, reject medical recommendations, or cause fetal harm. Many ethics scholars characterize those situations as maternal-fetal conflicts. In conflict-based models, maternal rights are considered to conflict with fetal rights, or moral obligations owed to pregnant women are considered to conflict with those owed to their fetuses. I offer an alternative model of pregnancy ethics by applying relational and equality-based moral theories to situations of fetal harm by pregnant women. In this model, clinicians faced with ethical dilemmas should attempt to understand pregnant women and their decisions within their broad social networks and communities, ask how the clinician’s personal standpoint influences outcomes judged to be ethical, and determine whether the clinician’s ethical formulations reduce or enhance existing gender, class, or racial inequality. This model focuses on the mutual needs of pregnant women and fetuses rather than on their mutually exclusive needs. It also avoids many pitfalls of traditional ethical formulations, specifically their tendency to neglect gender-specific modes of moral reasoning, their implicit assumptions that application of universal principles like autonomy and beneficence results in objective ethical solutions, and their failure to account for the ways that project-
From the University of Michigan Health System, Department of Obstetrics and Gynecology, Robert Wood Johnson Clinical Scholars Program, Ann Arbor, Michigan. This work was supported by the Robert Wood Johnson Foundation.
786 0029-7844/00/$20.00 PII S0029-7844(00)01021-8
How should caregivers for pregnant women respond when a patient causes fetal harm by ignoring medical advice or using illegal drugs or alcohol during pregnancy? When pregnant women engage in behaviors that are potentially harmful to their fetuses, clinicians find themselves in ethically, clinically, and sometimes legally challenging positions. Those situations have been characterized as so-called maternal-fetal conflicts. A conflict-based approach to ethical dilemmas in pregnancy is the only well-represented one in obstetric literature, despite the existence of alternative ways of framing ethical problems. In this commentary, the limitations of conflict-based perinatal ethics are elucidated, and an alternative is proposed in which the set of issues considered morally relevant to pregnancy is broadened to include substantive issues of gender, class, race, personal standpoint of ethical adjudicators, and social and political arrangements in which clinical care occurs. This perspective is a departure from usual ways of considering perinatal ethical dilemmas, in which moral principles, such as autonomy and beneficence, are viewed as the primary relevant factors. By broadening the set of issues considered morally relevant to prenatal fetal harm we can begin to focus on the mutual needs of pregnant women and their fetuses, rather than on their conflicting or mutually exclusive needs. This alternative approach guides law, policy, and medical practice to a clinically sound and socially egalitarian position.
Obstetrics & Gynecology
Conflict-Based Perinatal Ethics In its earliest iterations, potential fetal harm by pregnant women was conceptualized as an issue of conflicting rights—a maternal right to act freely versus a fetal right to life or well-being.1 Largely because there is no national or international consensus on the issue of fetal rights, this formulation led to intractable debate and brought little meaningful clinical resolution to these difficult situations. Perinatologist Frank Chervenak and ethicist Lawrence McCullough successfully moved perinatal ethics away from issues of rights to the more fruitful terrain of moral obligations of the clinician to the pregnant woman and fetus. Rather than claiming that the fetus has independent status from which rights come, Chervenak and McCullough2,3 argued that the fetus has a dependent moral status as a patient. In their formulation, the fetus is a patient because clinicians can care for it medically. Clinicians have moral obligations to fetal patients as they do to maternal patients. Chervenak and McCullough elaborated those moral obligations using the well-known principle-based bioethical model of Beauchamp and Childress.4 They applied the concepts of autonomy and beneficence in clinician-patient interactions to pregnancy and, in so doing, provided the first coherent framework for obstetric ethics.2– 4 The “most common and challenging moral conflicts” occur when a clinician’s obligation to respect a pregnant woman’s autonomy (her wish to act according to her own will or choice) conflicts with an obligation to act with beneficence towards (for the benefit of) the fetus.2 Examples of those conflicts of maternal autonomy-based obligations and fetal beneficence-based obligations include maternal refusal of tocolytics recommended to stop preterm labor and refusal of corticosteroids recommended to enhance fetal lung maturity. Chervenak and McCullough3 provided many strategies to prevent ethical conflicts. However, they acknowledged that moral crises will arise occasionally and advised that “successful clinical management requires identification of the component moral obligations of the conflict and determination of the relative weight of those obligations. . . . A morally justified decision is consistent with what can be shown to be the weightiest obligations.”2 For them, perinatal ethics involves identifying separate moral obligations that the clinician has to pregnant women and to fetuses and balancing those obligations when they appear to conflict. The term “maternal-fetal conflict” describes situations of conflicting maternal and fetal rights or conflicting moral obligations of the clinician to pregnant women and fetuses. The ACOG Ethics Committee introduced this description of situations of potential or actual prenatal harm by pregnant women in its 1987 committee opinion, and maternal-fetal conflict has persisted as a way of character-
VOL. 96, NO. 5, PART 1, NOVEMBER 2000
izing those situations.5 Conflict between clinicians’ moral obligations, not maternal and fetal rights, is the version of maternal-fetal conflict best represented in obstetric literature and will be the focus of the remainder of this commentary. The 1999 revision of the 1987 Ethics Committee opinion replaced the term maternal-fetal conflict with “maternal-fetal relationship.”6 The reasons for that change, although not made explicit, might indicate the committee’s discomfort with viewing pregnancy as an adversarial relationship between a pregnant woman and her fetus and perhaps with the legal and policy implications of such conflict. What considerations besides conflicting principles of autonomy and beneficence might be relevant to the consideration of ethical dilemmas in pregnancy? The obvious one is that pregnancy is unique to women. Most traditional work on biomedical ethics has not taken gender into account as a central analytic variable, despite ample psychologic data documenting gender differences in moral reasoning.7 Psychologist Carol Gilligan observed that on widely used scales of moral development, women scored consistently lower than men.7 She wondered whether the scales used to measure moral development, which were developed by and validated on boys and men, were systematically unable to capture the ways that women construct and negotiate their moral worlds. Her further investigation showed that men overwhelmingly applied universal principles and rules in their moral negotiations, whereas women resisted mediating the world through general principles. Instead, women negotiated moral dilemmas by considering their relationships and ties with others. Only the ability to use universal rules and principles was measured in traditional moral development scales; women’s “different voice” was not captured.7 A perinatal ethic based in general principles like autonomy and beneficence might not address women’s ways of perceiving and resolving moral dilemmas. A pregnant woman can represent her own autonomy interests, but a fetus obviously is unable to articulate its beneficence-based needs. Fetal needs exist only as a projection of a physician’s or another party’s determination of what is thought to be in the best interest of a fetus. Although protection of a vulnerable party is well within the mandate of ethical medical care, this projection is problematic. Medical predictions can be erroneous. Data on court-ordered obstetric interventions suggest that in almost one third of cases in which court authority was sought for a medical intervention, the medical judgment was wrong in retrospect.8 Projection of fetal interests has another important problem; it carries with it an enormous risk of reproducing existing conditions of social and racial inequality. For example, most court orders for obstetric inter-
Harris
Rethinking Perinatal Ethics
787
ventions on behalf of the fetus are sought against poor women of color.8 A 1990 study of drug and alcohol use during pregnancy in Florida showed that although the rate of positive urine toxicology for drug and alcohol use among pregnant women was slightly higher among white women compared with black women, black women were ten times more likely to be reported to health authorities.9 Although those findings can be attributed to racism and not to the maternal-fetal conflict scheme, per se, the conflict-based scheme is particularly vulnerable to racism. When we separate the moral obligations owed to the pregnant woman and fetus, racism might become masked as fetal protection. Racial and social prejudices might find their way into identification of fetal interests and so-called conflicts. Common sense tells us that if a fetus inhabits the body of a pregnant woman, they share a physiology, and so have many common needs. One could argue that the most important aspect of fetal well-being is maternal well-being. Is there an ethical model based on shared interests rather than conflicting ones?
Building an Alternative Model How do we build a perinatal ethic that attends to concerns about women’s forms of moral reasoning, the vulnerability of conflict-based models to social and racial inequality, and shared interests? Much of that work is under way. During the past decade, the principle-based ethics in which the conflict-based perinatal ethic is grounded has been criticized by numerous bioethical scholars. New models that avoid many of the pitfalls of principle-based schemes have been proposed as alternatives. Three important limitations of principle-based bioethics have been identified. First, principle-based schemes are difficult to use when negotiating moral dilemmas between intimates. As ethicist James Nelson states, “the impartialist, universalizing penchant of standard moral theories makes them clumsy at best when it comes to illuminating the moral contours of intimate relationships.”10 Partiality and emotion have important roles in relationships among intimates, and it is difficult and perhaps inappropriate to put those aspects of relationships aside in order to sort out moral problems by universal principles alone. That kind of critique is informed largely by Gilligan’s observations of the importance of intimate social networks in women’s moral decision making. Its usefulness for pregnancy ethics will be shown. From that critique arose a relational ethic or ethic of care.11,12 Philosopher Rosemarie Tong articulated its core idea when she said, “what makes the conduct of someone . . . morally wrong or not depends on the
788 Harris
Rethinking Perinatal Ethics
‘particulars’ of her life— especially on her network of human relationships.”11 Stated another way, the ethic of care holds that “moral reasoning is not solely, or even primarily, a matter of finding rules to arbitrate between conflicting interests. Rather, moral wisdom and sensitivity consist, in the first instance, in focusing on how . . . interests are often interdependent.”12 In other words, rules are not everything; relationships count too. Again, that ethic of care can be applied to obstetric ethics in useful ways. Principle-based ethics has a second limitation, it is not simply the particularities of a pregnant woman’s life that are important in judging ethical dilemmas. The life particularities of the parties adjudicating a moral dilemma are important also. As philosopher Mary Mahowald points out, all human subjects are situated; that is, all people come to an understanding of what is ethical based on their own standpoint,13 which is informed by multiple factors, including age, sex, politics, and religion. Scholars like Mahowald observe that some perspectives or standpoints are privileged over others; it is conceivable that the perspective of a dominant or advantaged social group might be more likely to be deemed ethical than that of a disadvantaged group. Some scholars, including Chervenak and McCullough, argue that the virtue of selfeffacement, ie, blunting of self-interest, eliminates biases from social differences between patients and ethical adjudicators, but it is not certain that self-effacement can be accomplished in reality. Mahowald is not convinced that it is possible to shed our own standpoints and adopt an objective point of view. Therefore, she argues that proportionate representation is the most appropriate remedial strategy. For example, she suggests that ethics committees and other groups conducting ethical deliberations should have diverse composition and represent multiple possible standpoints. Still other scholars offer a third critique of principlebased ethical models; principle-based ethics neglects the broad social and political arrangements in which clinical care occurs and in which ethical dilemmas are negotiated.14 Specifically, that critique means that we need to consider how sex, race, and class inequalities influence decision making in ethics. Legal scholar Dorothy Roberts argues that “the implementation of any ethical model will be determined by the hierarchies of power in which the doctor-patient relationship is embedded.”14 She points out that, for example, poor uninsured women often lack a personal physician and rely on public emergency rooms for care. In this setting a clinician may have a diminished sense of loyalty to a patient compared with that felt between a private physician and a patient. She also points out that prior experiences of racism might lead women of color to take a “more oppositional stance to the judgements of their
Obstetrics & Gynecology
doctors than most white middle-class women.”14 Both those conditions might make it more likely that a physician will perceive an ethical breach (and might explain some of the findings of the Florida study).9 How might this critique of principle-based ethics inform our approach to moral dilemmas in pregnancy? Consider the case of a pregnant woman in preterm labor who refuses to be admitted to the hospital for bedrest or tocolytics. A conflict-based model, as we have seen, focuses attention narrowly on the pregnant woman-fetus pair. Clinicians then balance the relative moral weights of obligations to respect maternal autonomy (the pregnant patient’s wish to avoid therapy) and fetal beneficence (the fetal well-being achieved by therapy to speed fetal lung maturity). An ethical solution would be achieved when one set of interests was determined to outweigh the other. A clinician applying a model attentive to issues of relation and equality, on the other hand, would assume a much wider gaze. The clinician would focus attention on important social and family relationships, contexts or constraints that might come to bear on pregnant women’s decision making, such as her need to care for other children at home or to continue working to support other family members, or whatever life project occupied her, and attempt to provide relief in those areas. The ethical arbiters themselves would attempt to be self-conscious about how their own life conditions influence both the perception of the ethical dilemma and of its appropriate response. For example, it might be difficult for a physician with material resources or a trusted extended family to imagine a situation in which absolutely no childcare arrangements could be made for that pregnant woman’s other children. Ethical decision makers would ask whether social and racial inequalities were being dramatized within the ethical drama: do prior experiences of racism in medical care explain why that pregnant woman might not trust or believe recommendations made to her? If the particulars of a pregnant woman’s life are addressed along with social and cultural contexts in which the ethical dilemma occurs, a clinician might not need to balance the relative moral weight of obligations owed to her against those owed to a fetus. In that formulation, fetal well-being is achieved when maternal wellbeing is achieved. Of course, not all moral dilemmas will be neatly resolved by widening the lens under which pregnant women and fetuses are viewed. However, there is no denying that factors other than conventional moral principles are relevant to moral decision making in pregnancy.
VOL. 96, NO. 5, PART 1, NOVEMBER 2000
Why Care? Issues in Law and Policy Ethical frameworks have an important relationship with policy and law, not just with what is good or virtuous. This does not mean that a moral obligation is synonymous with a legal one. A pregnant woman may have moral or ethical duties to the fetus she carries without having any legally enforceable obligations, just as one might have a moral obligation to care for an aging relative but no legal requirement to do so. However, there is a close relationship between ethical and legal duties. In the case of pregnancy, whether one adopts a conflict-based or relation- and equality-driven ethical standard will influence conclusions about the appropriateness of criminal punitive measures in situations of potential or actual prenatal harm. If, for instance, we frame prenatal drug use as a conflict, addiction becomes a case of disregard for the fetus by a pregnant woman. The goal is then to protect the fetus from the pregnant woman, and she is important primarily as an instrument through which to achieve fetal safety. As such, policies that advocate nonvoluntary constraints or punitive measures as a means of ensuring compliance with drug abstinence could be considered ethically appropriate despite their curtailing a pregnant woman’s autonomy. If we frame prenatal substance abuse within the alternative model, the pregnant woman and fetus are considered a single unit within a social network, and fetal well-being is achieved by making maternal well-being the primary goal. In this formulation, the nature of addiction as chronic illness, and the histories of abuse and violence that often accompany addiction could be highlighted, for example. An ethical policy choice would be one that supports voluntary counseling and addiction treatment. There are serious problems with laws regulating pregnant women’s behavior in the name of fetal protection, ie, with encoding maternal-fetal conflict into law. If prenatal behavior is punished, pregnant women might avoid medical care. We would doubtless see erosion of trust in clinician-patient relationships if caregivers must also serve as agents of the legal system. In addition, such laws would impose restrictions on women to which men could never be subject. Can or should the law treat women differently because they are (or could be) pregnant? Consider again the case of drug use during pregnancy: for the general population, in the absence of drug possession or selling, courts have refused to endorse criminal sanctions for dependence on a drug so that drug-dependent men and women will seek treatment without fear of prosecution. Prosecuting pregnant woman for drug addiction in pregnancy, without drug selling or possession, would mark a gendered change to the law in this area and would
Harris
Rethinking Perinatal Ethics
789
provide a profound disincentive for women to seek treatment. Those in favor of criminalization argue quite simply that pregnancy is a condition unique to women; because only women are able to cause fetal harm by drug use, it is appropriate to treat pregnant women differently from men. In response to arguments of that sort, attorney Lynn Paltrow asks a provocative and important question: “Can becoming pregnant be a crime?”15 This question illuminates what we already know, that addiction usually precedes pregnancy and not the reverse. It is the act of becoming pregnant in the setting of addiction that would be a new crime if prenatal drug use were made illegal. Clinically this is troubling because we know that no contraceptive method is perfect and that access to family planning services is not universally available. Legally this is troubling because both men and women are involved in creating a pregnancy. As addiction is overwhelmingly likely to have existed at the time of conception, is it appropriate that women alone be held criminally responsible for fetal harm caused by ongoing addiction during pregnancy? This is a sex equality issue. There is very little empiric work on the effect of criminalizing prenatal drug abuse or other behaviors on pregnancy outcomes. One of the only attempts to date to do this was enormously ethically compromising.16,17 In 1989, staff at the Medical University of South Carolina began screening the urine of pregnant women whom they suspected were drug users. Women who had positive results of urine toxicology screens were sent letters telling them that they would be arrested if they had further positive screens. Thirty arrests were made. In 1990, the clinicians involved reported that they found fewer positive drug screens after their intervention and that their intervention effectively reduced prenatal drug use. The study had no control groups, so there is no way to know with certainty whether fewer screens were positive because fewer women presented for care. Worse than the study’s methodologic defects were its egregious ethical flaws, which were terrible regardless of ethical framework. Women were not arrested for crimes of drug selling or possession; they were arrested for being pregnant at the time of their addiction, a crime that in reality did not exist according to state law. No consent for drug screening or for participation in a study was obtained, and no institutional review board approval was sought. All but one of the women arrested were black. In 1994, the National Institutes of Health found that the research violated laws governing human research subjects, and in 2000 the United States Supreme Court will consider whether urine testing violated those women’s constitutional protections. That case highlights a central question: how do we prevent existing biases, power structures, racial
790 Harris
Rethinking Perinatal Ethics
inequalities, and stereotypes about women’s unique responsibility of pregnancy from entering our perceptions of our moral obligations and apparent conflicts between them? It is not clear that we can. An ethical model that forgoes identifying distinct obligations owed to the fetus and pregnant woman, but instead considers the pregnant woman and fetus a single entity and broadens the range of factors considered morally relevant to pregnant women could solve this dilemma. What does this mean for the clinician facing an ethical dilemma regarding pregnancy? Application of the alternative perinatal ethical model framed here requires that clinicians do the following three things: attempt to understand the pregnant woman and her decisions within her broad social networks and communities; ask how the standpoint of clinicians framing the ethical dilemma is related to the outcome judged to be ethical, and recognize that opinions from a diverse group of people might best accomplish this; and attend to issues of sex and race equality by asking, “Does this approach reduce or enhance existing conditions of advantage and disadvantage?” Maternal-fetal conflict is not the only way to frame a situation of potential prenatal harm. It might not, for the philosophical, psychological, and social reasons elucidated, be the best way. All things being equal, appealing to universal principles like autonomy and beneficence to sort out ethical dilemmas might result in objective solutions. However, as things are not always equal, other substantive issues must be considered. For issues that affect women primarily or exclusively, it is unacceptable not to consider issues of gender and sex equality as morally relevant. We also must recognize the ways in which conflict-based schemes are particularly vulnerable to perpetuation of social inequalities. What is considered ethical public policy in pregnancy will depend on the kind of ethical model chosen to frame moral dilemmas in pregnancy. A care-based, equality-driven ethical and policy model could form the ethical foundations for prevention, rather than criminalization, of prenatal harm.
References 1. Leiberman JF, Mazor M, Chaim W, Cohen A. The fetal right to live. Obstet Gynecol 1979;53:515–7. 2. Chervenak FA, McCullough LB. Perinatal ethics: A practical method of analysis of obligations to mother and fetus. Obstet Gynecol 1985;66:442– 6. 3. McCullough LB, Chervenak FA. Ethics in obstetrics and gynecology. New York: Oxford University Press, 1994. 4. Beauchamp TL, Childress JF. Principles of biomedical ethics. 4th ed. New York: Oxford University Press, 1994. 5. Ethics Committee of the American College of Obstetricians and Gynecologists. Patient choice: Maternal-fetal conflict. Washington, DC: American College of Obstetricians and Gynecologists, 1987.
Obstetrics & Gynecology
6. Ethics Committee of the American College of Obstetricians and Gynecologists. Patient choice and the maternal-fetal relationship. Washington, DC: American College of Obstetricians and Gynecologists, 1999. 7. Gilligan C. In a different voice. Cambridge, Massachusetts: Harvard University Press, 1983. 8. Kolder VEB, Gallagher J, Parsons MT. Court-ordered obstetrical interventions. N Engl J Med 1987;316:1192– 6. 9. Chasnoff IJ, Landres HJ, Barrett ME. The prevalence of illicit-drug or alcohol use during pregnancy and discrepancies in mandatory reporting in Pinellas County, Florida. N Engl J Med 1990;332:1202–6. 10. Nelson J. Making peace in gestational conflicts. Theor Med 1992; 13:319 –28. 11. Tong R. Blessed are the peacemakers: Commentary on making peace in gestational conflicts. Theor Med 1992;13:329 –35. 12. Royal Commission on New Reproductive Technologies. Proceed with care: Final report of the Royal Commission on New Reproductive Technologies. Ottawa, Canada: Minister of Government Services Canada, 1993. 13. Mahowald M. On treatment of myopia: Feminist standpoint theory and bioethics. In: Wolf SM, ed. Feminism and bioethics— beyond reproduction. New York: Oxford University Press, 1996. 14. Roberts DE. Reconstructing the patient: Starting with women of color. In: Wolf SM, ed. Feminism and bioethics— beyond reproduction. New York: Oxford University Press, 1996. 15. Paltrow LM. When becoming pregnant is a crime. Criminal Justice Ethics 1990;5:41–7.
VOL. 96, NO. 5, PART 1, NOVEMBER 2000
16. Paltrow LM. Pregnant drug users, fetal persons and the threat to Roe v. Wade. Albany Law Rev 1999;62:999 –1055. 17. Horger EO, Brown SB, Condon CM. Cocaine in pregnancy: Confronting the problem. J S Carolina Med Assoc 1990;86:527–35.
Address reprint requests to:
Lisa H. Harris, MD University of Michigan Health System Robert Wood Johnson Clinical Scholars Program 6312 Medical Science Building I 1150 West Medical Center Drive Ann Arbor, MI 48109-0604 E-mail: [email protected]
Received February 15, 2000. Received in revised form June 5, 2000. Accepted July 7, 2000.
Copyright © 2000 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Harris
Rethinking Perinatal Ethics
791
11.
12.
13.
human chorionic gonadotropin and risk of breast cancer. Cancer Epidemiol Biomarkers Prev 1995;4:437– 40. Alvarado MV, Alvarado NE, Russo J, Russo IH. Human chorionic gonadotropin inhibits proliferation and induces expression of inhibin in human breast epithelial cells in vitro. In Vitro Cell Dev Biol 1994;30A:4 – 8. Srivastava P, Russo J, Russo IH. Chorionic gonadotropin inhibits rat mammary carcinogenesis through activation of programmed cell death. Carcinogenesis 1997;18:1799 – 808. Huynh H. In vivo regulation of the insulin-like growth factor system of mitogens by human chorionic gonadotropin. Int J Oncol 1998;13:571–5. Huang Y, Bove B, Wu Y, Russo IH, Yang X, Russo J. Microsatellite instability during the immortalization and transformation of human breast epithelial cells in vitro. Mol Carcinogenesis 1999;24: 118 –27.
Rethinking maternal-fetal conflict: Gender and equality in perinatal ethics
Address reprint requests to:
Ch. V. Rao, PhD Department of Obstetrics and Gynecology University of Louisville Health Sciences Center 438 MDR Building Louisville, KY 40292 E-mail: [email protected]
Received March 9, 2000. Received in revised form May 30, 2000. Accepted June 22, 2000. Copyright © 2000 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
ing fetal needs perpetuates social inequalities. This model provides the ethical foundations for moving law and policy away from criminalization and toward prevention of prenatal harm. (Obstet Gynecol 2000;96:786 –91. © 2000 by The American College of Obstetricians and Gynecologists.)
Lisa H. Harris, MD Practitioners who care for pregnant women face dilemmas when their patients use illicit drugs, reject medical recommendations, or cause fetal harm. Many ethics scholars characterize those situations as maternal-fetal conflicts. In conflict-based models, maternal rights are considered to conflict with fetal rights, or moral obligations owed to pregnant women are considered to conflict with those owed to their fetuses. I offer an alternative model of pregnancy ethics by applying relational and equality-based moral theories to situations of fetal harm by pregnant women. In this model, clinicians faced with ethical dilemmas should attempt to understand pregnant women and their decisions within their broad social networks and communities, ask how the clinician’s personal standpoint influences outcomes judged to be ethical, and determine whether the clinician’s ethical formulations reduce or enhance existing gender, class, or racial inequality. This model focuses on the mutual needs of pregnant women and fetuses rather than on their mutually exclusive needs. It also avoids many pitfalls of traditional ethical formulations, specifically their tendency to neglect gender-specific modes of moral reasoning, their implicit assumptions that application of universal principles like autonomy and beneficence results in objective ethical solutions, and their failure to account for the ways that project-
From the University of Michigan Health System, Department of Obstetrics and Gynecology, Robert Wood Johnson Clinical Scholars Program, Ann Arbor, Michigan. This work was supported by the Robert Wood Johnson Foundation.
786 0029-7844/00/$20.00 PII S0029-7844(00)01021-8
How should caregivers for pregnant women respond when a patient causes fetal harm by ignoring medical advice or using illegal drugs or alcohol during pregnancy? When pregnant women engage in behaviors that are potentially harmful to their fetuses, clinicians find themselves in ethically, clinically, and sometimes legally challenging positions. Those situations have been characterized as so-called maternal-fetal conflicts. A conflict-based approach to ethical dilemmas in pregnancy is the only well-represented one in obstetric literature, despite the existence of alternative ways of framing ethical problems. In this commentary, the limitations of conflict-based perinatal ethics are elucidated, and an alternative is proposed in which the set of issues considered morally relevant to pregnancy is broadened to include substantive issues of gender, class, race, personal standpoint of ethical adjudicators, and social and political arrangements in which clinical care occurs. This perspective is a departure from usual ways of considering perinatal ethical dilemmas, in which moral principles, such as autonomy and beneficence, are viewed as the primary relevant factors. By broadening the set of issues considered morally relevant to prenatal fetal harm we can begin to focus on the mutual needs of pregnant women and their fetuses, rather than on their conflicting or mutually exclusive needs. This alternative approach guides law, policy, and medical practice to a clinically sound and socially egalitarian position.
Obstetrics & Gynecology
Conflict-Based Perinatal Ethics In its earliest iterations, potential fetal harm by pregnant women was conceptualized as an issue of conflicting rights—a maternal right to act freely versus a fetal right to life or well-being.1 Largely because there is no national or international consensus on the issue of fetal rights, this formulation led to intractable debate and brought little meaningful clinical resolution to these difficult situations. Perinatologist Frank Chervenak and ethicist Lawrence McCullough successfully moved perinatal ethics away from issues of rights to the more fruitful terrain of moral obligations of the clinician to the pregnant woman and fetus. Rather than claiming that the fetus has independent status from which rights come, Chervenak and McCullough2,3 argued that the fetus has a dependent moral status as a patient. In their formulation, the fetus is a patient because clinicians can care for it medically. Clinicians have moral obligations to fetal patients as they do to maternal patients. Chervenak and McCullough elaborated those moral obligations using the well-known principle-based bioethical model of Beauchamp and Childress.4 They applied the concepts of autonomy and beneficence in clinician-patient interactions to pregnancy and, in so doing, provided the first coherent framework for obstetric ethics.2– 4 The “most common and challenging moral conflicts” occur when a clinician’s obligation to respect a pregnant woman’s autonomy (her wish to act according to her own will or choice) conflicts with an obligation to act with beneficence towards (for the benefit of) the fetus.2 Examples of those conflicts of maternal autonomy-based obligations and fetal beneficence-based obligations include maternal refusal of tocolytics recommended to stop preterm labor and refusal of corticosteroids recommended to enhance fetal lung maturity. Chervenak and McCullough3 provided many strategies to prevent ethical conflicts. However, they acknowledged that moral crises will arise occasionally and advised that “successful clinical management requires identification of the component moral obligations of the conflict and determination of the relative weight of those obligations. . . . A morally justified decision is consistent with what can be shown to be the weightiest obligations.”2 For them, perinatal ethics involves identifying separate moral obligations that the clinician has to pregnant women and to fetuses and balancing those obligations when they appear to conflict. The term “maternal-fetal conflict” describes situations of conflicting maternal and fetal rights or conflicting moral obligations of the clinician to pregnant women and fetuses. The ACOG Ethics Committee introduced this description of situations of potential or actual prenatal harm by pregnant women in its 1987 committee opinion, and maternal-fetal conflict has persisted as a way of character-
VOL. 96, NO. 5, PART 1, NOVEMBER 2000
izing those situations.5 Conflict between clinicians’ moral obligations, not maternal and fetal rights, is the version of maternal-fetal conflict best represented in obstetric literature and will be the focus of the remainder of this commentary. The 1999 revision of the 1987 Ethics Committee opinion replaced the term maternal-fetal conflict with “maternal-fetal relationship.”6 The reasons for that change, although not made explicit, might indicate the committee’s discomfort with viewing pregnancy as an adversarial relationship between a pregnant woman and her fetus and perhaps with the legal and policy implications of such conflict. What considerations besides conflicting principles of autonomy and beneficence might be relevant to the consideration of ethical dilemmas in pregnancy? The obvious one is that pregnancy is unique to women. Most traditional work on biomedical ethics has not taken gender into account as a central analytic variable, despite ample psychologic data documenting gender differences in moral reasoning.7 Psychologist Carol Gilligan observed that on widely used scales of moral development, women scored consistently lower than men.7 She wondered whether the scales used to measure moral development, which were developed by and validated on boys and men, were systematically unable to capture the ways that women construct and negotiate their moral worlds. Her further investigation showed that men overwhelmingly applied universal principles and rules in their moral negotiations, whereas women resisted mediating the world through general principles. Instead, women negotiated moral dilemmas by considering their relationships and ties with others. Only the ability to use universal rules and principles was measured in traditional moral development scales; women’s “different voice” was not captured.7 A perinatal ethic based in general principles like autonomy and beneficence might not address women’s ways of perceiving and resolving moral dilemmas. A pregnant woman can represent her own autonomy interests, but a fetus obviously is unable to articulate its beneficence-based needs. Fetal needs exist only as a projection of a physician’s or another party’s determination of what is thought to be in the best interest of a fetus. Although protection of a vulnerable party is well within the mandate of ethical medical care, this projection is problematic. Medical predictions can be erroneous. Data on court-ordered obstetric interventions suggest that in almost one third of cases in which court authority was sought for a medical intervention, the medical judgment was wrong in retrospect.8 Projection of fetal interests has another important problem; it carries with it an enormous risk of reproducing existing conditions of social and racial inequality. For example, most court orders for obstetric inter-
Harris
Rethinking Perinatal Ethics
787
ventions on behalf of the fetus are sought against poor women of color.8 A 1990 study of drug and alcohol use during pregnancy in Florida showed that although the rate of positive urine toxicology for drug and alcohol use among pregnant women was slightly higher among white women compared with black women, black women were ten times more likely to be reported to health authorities.9 Although those findings can be attributed to racism and not to the maternal-fetal conflict scheme, per se, the conflict-based scheme is particularly vulnerable to racism. When we separate the moral obligations owed to the pregnant woman and fetus, racism might become masked as fetal protection. Racial and social prejudices might find their way into identification of fetal interests and so-called conflicts. Common sense tells us that if a fetus inhabits the body of a pregnant woman, they share a physiology, and so have many common needs. One could argue that the most important aspect of fetal well-being is maternal well-being. Is there an ethical model based on shared interests rather than conflicting ones?
Building an Alternative Model How do we build a perinatal ethic that attends to concerns about women’s forms of moral reasoning, the vulnerability of conflict-based models to social and racial inequality, and shared interests? Much of that work is under way. During the past decade, the principle-based ethics in which the conflict-based perinatal ethic is grounded has been criticized by numerous bioethical scholars. New models that avoid many of the pitfalls of principle-based schemes have been proposed as alternatives. Three important limitations of principle-based bioethics have been identified. First, principle-based schemes are difficult to use when negotiating moral dilemmas between intimates. As ethicist James Nelson states, “the impartialist, universalizing penchant of standard moral theories makes them clumsy at best when it comes to illuminating the moral contours of intimate relationships.”10 Partiality and emotion have important roles in relationships among intimates, and it is difficult and perhaps inappropriate to put those aspects of relationships aside in order to sort out moral problems by universal principles alone. That kind of critique is informed largely by Gilligan’s observations of the importance of intimate social networks in women’s moral decision making. Its usefulness for pregnancy ethics will be shown. From that critique arose a relational ethic or ethic of care.11,12 Philosopher Rosemarie Tong articulated its core idea when she said, “what makes the conduct of someone . . . morally wrong or not depends on the
788 Harris
Rethinking Perinatal Ethics
‘particulars’ of her life— especially on her network of human relationships.”11 Stated another way, the ethic of care holds that “moral reasoning is not solely, or even primarily, a matter of finding rules to arbitrate between conflicting interests. Rather, moral wisdom and sensitivity consist, in the first instance, in focusing on how . . . interests are often interdependent.”12 In other words, rules are not everything; relationships count too. Again, that ethic of care can be applied to obstetric ethics in useful ways. Principle-based ethics has a second limitation, it is not simply the particularities of a pregnant woman’s life that are important in judging ethical dilemmas. The life particularities of the parties adjudicating a moral dilemma are important also. As philosopher Mary Mahowald points out, all human subjects are situated; that is, all people come to an understanding of what is ethical based on their own standpoint,13 which is informed by multiple factors, including age, sex, politics, and religion. Scholars like Mahowald observe that some perspectives or standpoints are privileged over others; it is conceivable that the perspective of a dominant or advantaged social group might be more likely to be deemed ethical than that of a disadvantaged group. Some scholars, including Chervenak and McCullough, argue that the virtue of selfeffacement, ie, blunting of self-interest, eliminates biases from social differences between patients and ethical adjudicators, but it is not certain that self-effacement can be accomplished in reality. Mahowald is not convinced that it is possible to shed our own standpoints and adopt an objective point of view. Therefore, she argues that proportionate representation is the most appropriate remedial strategy. For example, she suggests that ethics committees and other groups conducting ethical deliberations should have diverse composition and represent multiple possible standpoints. Still other scholars offer a third critique of principlebased ethical models; principle-based ethics neglects the broad social and political arrangements in which clinical care occurs and in which ethical dilemmas are negotiated.14 Specifically, that critique means that we need to consider how sex, race, and class inequalities influence decision making in ethics. Legal scholar Dorothy Roberts argues that “the implementation of any ethical model will be determined by the hierarchies of power in which the doctor-patient relationship is embedded.”14 She points out that, for example, poor uninsured women often lack a personal physician and rely on public emergency rooms for care. In this setting a clinician may have a diminished sense of loyalty to a patient compared with that felt between a private physician and a patient. She also points out that prior experiences of racism might lead women of color to take a “more oppositional stance to the judgements of their
Obstetrics & Gynecology
doctors than most white middle-class women.”14 Both those conditions might make it more likely that a physician will perceive an ethical breach (and might explain some of the findings of the Florida study).9 How might this critique of principle-based ethics inform our approach to moral dilemmas in pregnancy? Consider the case of a pregnant woman in preterm labor who refuses to be admitted to the hospital for bedrest or tocolytics. A conflict-based model, as we have seen, focuses attention narrowly on the pregnant woman-fetus pair. Clinicians then balance the relative moral weights of obligations to respect maternal autonomy (the pregnant patient’s wish to avoid therapy) and fetal beneficence (the fetal well-being achieved by therapy to speed fetal lung maturity). An ethical solution would be achieved when one set of interests was determined to outweigh the other. A clinician applying a model attentive to issues of relation and equality, on the other hand, would assume a much wider gaze. The clinician would focus attention on important social and family relationships, contexts or constraints that might come to bear on pregnant women’s decision making, such as her need to care for other children at home or to continue working to support other family members, or whatever life project occupied her, and attempt to provide relief in those areas. The ethical arbiters themselves would attempt to be self-conscious about how their own life conditions influence both the perception of the ethical dilemma and of its appropriate response. For example, it might be difficult for a physician with material resources or a trusted extended family to imagine a situation in which absolutely no childcare arrangements could be made for that pregnant woman’s other children. Ethical decision makers would ask whether social and racial inequalities were being dramatized within the ethical drama: do prior experiences of racism in medical care explain why that pregnant woman might not trust or believe recommendations made to her? If the particulars of a pregnant woman’s life are addressed along with social and cultural contexts in which the ethical dilemma occurs, a clinician might not need to balance the relative moral weight of obligations owed to her against those owed to a fetus. In that formulation, fetal well-being is achieved when maternal wellbeing is achieved. Of course, not all moral dilemmas will be neatly resolved by widening the lens under which pregnant women and fetuses are viewed. However, there is no denying that factors other than conventional moral principles are relevant to moral decision making in pregnancy.
VOL. 96, NO. 5, PART 1, NOVEMBER 2000
Why Care? Issues in Law and Policy Ethical frameworks have an important relationship with policy and law, not just with what is good or virtuous. This does not mean that a moral obligation is synonymous with a legal one. A pregnant woman may have moral or ethical duties to the fetus she carries without having any legally enforceable obligations, just as one might have a moral obligation to care for an aging relative but no legal requirement to do so. However, there is a close relationship between ethical and legal duties. In the case of pregnancy, whether one adopts a conflict-based or relation- and equality-driven ethical standard will influence conclusions about the appropriateness of criminal punitive measures in situations of potential or actual prenatal harm. If, for instance, we frame prenatal drug use as a conflict, addiction becomes a case of disregard for the fetus by a pregnant woman. The goal is then to protect the fetus from the pregnant woman, and she is important primarily as an instrument through which to achieve fetal safety. As such, policies that advocate nonvoluntary constraints or punitive measures as a means of ensuring compliance with drug abstinence could be considered ethically appropriate despite their curtailing a pregnant woman’s autonomy. If we frame prenatal substance abuse within the alternative model, the pregnant woman and fetus are considered a single unit within a social network, and fetal well-being is achieved by making maternal well-being the primary goal. In this formulation, the nature of addiction as chronic illness, and the histories of abuse and violence that often accompany addiction could be highlighted, for example. An ethical policy choice would be one that supports voluntary counseling and addiction treatment. There are serious problems with laws regulating pregnant women’s behavior in the name of fetal protection, ie, with encoding maternal-fetal conflict into law. If prenatal behavior is punished, pregnant women might avoid medical care. We would doubtless see erosion of trust in clinician-patient relationships if caregivers must also serve as agents of the legal system. In addition, such laws would impose restrictions on women to which men could never be subject. Can or should the law treat women differently because they are (or could be) pregnant? Consider again the case of drug use during pregnancy: for the general population, in the absence of drug possession or selling, courts have refused to endorse criminal sanctions for dependence on a drug so that drug-dependent men and women will seek treatment without fear of prosecution. Prosecuting pregnant woman for drug addiction in pregnancy, without drug selling or possession, would mark a gendered change to the law in this area and would
Harris
Rethinking Perinatal Ethics
789
provide a profound disincentive for women to seek treatment. Those in favor of criminalization argue quite simply that pregnancy is a condition unique to women; because only women are able to cause fetal harm by drug use, it is appropriate to treat pregnant women differently from men. In response to arguments of that sort, attorney Lynn Paltrow asks a provocative and important question: “Can becoming pregnant be a crime?”15 This question illuminates what we already know, that addiction usually precedes pregnancy and not the reverse. It is the act of becoming pregnant in the setting of addiction that would be a new crime if prenatal drug use were made illegal. Clinically this is troubling because we know that no contraceptive method is perfect and that access to family planning services is not universally available. Legally this is troubling because both men and women are involved in creating a pregnancy. As addiction is overwhelmingly likely to have existed at the time of conception, is it appropriate that women alone be held criminally responsible for fetal harm caused by ongoing addiction during pregnancy? This is a sex equality issue. There is very little empiric work on the effect of criminalizing prenatal drug abuse or other behaviors on pregnancy outcomes. One of the only attempts to date to do this was enormously ethically compromising.16,17 In 1989, staff at the Medical University of South Carolina began screening the urine of pregnant women whom they suspected were drug users. Women who had positive results of urine toxicology screens were sent letters telling them that they would be arrested if they had further positive screens. Thirty arrests were made. In 1990, the clinicians involved reported that they found fewer positive drug screens after their intervention and that their intervention effectively reduced prenatal drug use. The study had no control groups, so there is no way to know with certainty whether fewer screens were positive because fewer women presented for care. Worse than the study’s methodologic defects were its egregious ethical flaws, which were terrible regardless of ethical framework. Women were not arrested for crimes of drug selling or possession; they were arrested for being pregnant at the time of their addiction, a crime that in reality did not exist according to state law. No consent for drug screening or for participation in a study was obtained, and no institutional review board approval was sought. All but one of the women arrested were black. In 1994, the National Institutes of Health found that the research violated laws governing human research subjects, and in 2000 the United States Supreme Court will consider whether urine testing violated those women’s constitutional protections. That case highlights a central question: how do we prevent existing biases, power structures, racial
790 Harris
Rethinking Perinatal Ethics
inequalities, and stereotypes about women’s unique responsibility of pregnancy from entering our perceptions of our moral obligations and apparent conflicts between them? It is not clear that we can. An ethical model that forgoes identifying distinct obligations owed to the fetus and pregnant woman, but instead considers the pregnant woman and fetus a single entity and broadens the range of factors considered morally relevant to pregnant women could solve this dilemma. What does this mean for the clinician facing an ethical dilemma regarding pregnancy? Application of the alternative perinatal ethical model framed here requires that clinicians do the following three things: attempt to understand the pregnant woman and her decisions within her broad social networks and communities; ask how the standpoint of clinicians framing the ethical dilemma is related to the outcome judged to be ethical, and recognize that opinions from a diverse group of people might best accomplish this; and attend to issues of sex and race equality by asking, “Does this approach reduce or enhance existing conditions of advantage and disadvantage?” Maternal-fetal conflict is not the only way to frame a situation of potential prenatal harm. It might not, for the philosophical, psychological, and social reasons elucidated, be the best way. All things being equal, appealing to universal principles like autonomy and beneficence to sort out ethical dilemmas might result in objective solutions. However, as things are not always equal, other substantive issues must be considered. For issues that affect women primarily or exclusively, it is unacceptable not to consider issues of gender and sex equality as morally relevant. We also must recognize the ways in which conflict-based schemes are particularly vulnerable to perpetuation of social inequalities. What is considered ethical public policy in pregnancy will depend on the kind of ethical model chosen to frame moral dilemmas in pregnancy. A care-based, equality-driven ethical and policy model could form the ethical foundations for prevention, rather than criminalization, of prenatal harm.
References 1. Leiberman JF, Mazor M, Chaim W, Cohen A. The fetal right to live. Obstet Gynecol 1979;53:515–7. 2. Chervenak FA, McCullough LB. Perinatal ethics: A practical method of analysis of obligations to mother and fetus. Obstet Gynecol 1985;66:442– 6. 3. McCullough LB, Chervenak FA. Ethics in obstetrics and gynecology. New York: Oxford University Press, 1994. 4. Beauchamp TL, Childress JF. Principles of biomedical ethics. 4th ed. New York: Oxford University Press, 1994. 5. Ethics Committee of the American College of Obstetricians and Gynecologists. Patient choice: Maternal-fetal conflict. Washington, DC: American College of Obstetricians and Gynecologists, 1987.
Obstetrics & Gynecology
6. Ethics Committee of the American College of Obstetricians and Gynecologists. Patient choice and the maternal-fetal relationship. Washington, DC: American College of Obstetricians and Gynecologists, 1999. 7. Gilligan C. In a different voice. Cambridge, Massachusetts: Harvard University Press, 1983. 8. Kolder VEB, Gallagher J, Parsons MT. Court-ordered obstetrical interventions. N Engl J Med 1987;316:1192– 6. 9. Chasnoff IJ, Landres HJ, Barrett ME. The prevalence of illicit-drug or alcohol use during pregnancy and discrepancies in mandatory reporting in Pinellas County, Florida. N Engl J Med 1990;332:1202–6. 10. Nelson J. Making peace in gestational conflicts. Theor Med 1992; 13:319 –28. 11. Tong R. Blessed are the peacemakers: Commentary on making peace in gestational conflicts. Theor Med 1992;13:329 –35. 12. Royal Commission on New Reproductive Technologies. Proceed with care: Final report of the Royal Commission on New Reproductive Technologies. Ottawa, Canada: Minister of Government Services Canada, 1993. 13. Mahowald M. On treatment of myopia: Feminist standpoint theory and bioethics. In: Wolf SM, ed. Feminism and bioethics— beyond reproduction. New York: Oxford University Press, 1996. 14. Roberts DE. Reconstructing the patient: Starting with women of color. In: Wolf SM, ed. Feminism and bioethics— beyond reproduction. New York: Oxford University Press, 1996. 15. Paltrow LM. When becoming pregnant is a crime. Criminal Justice Ethics 1990;5:41–7.
VOL. 96, NO. 5, PART 1, NOVEMBER 2000
16. Paltrow LM. Pregnant drug users, fetal persons and the threat to Roe v. Wade. Albany Law Rev 1999;62:999 –1055. 17. Horger EO, Brown SB, Condon CM. Cocaine in pregnancy: Confronting the problem. J S Carolina Med Assoc 1990;86:527–35.
Address reprint requests to:
Lisa H. Harris, MD University of Michigan Health System Robert Wood Johnson Clinical Scholars Program 6312 Medical Science Building I 1150 West Medical Center Drive Ann Arbor, MI 48109-0604 E-mail: [email protected]
Received February 15, 2000. Received in revised form June 5, 2000. Accepted July 7, 2000.
Copyright © 2000 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
Harris
Rethinking Perinatal Ethics
791