Willard Gaylin's remarks about consent policy and health care for children: "Any position in a ... Hall, 1973, chapters 2, 3. 9. Goldstein J: Medical care for the child ...
Different Views Parental Rights, Child Welfare, and Public Health: The Case of PKU Screening RUTH R. FADEN, PHD, MPH, NEIL A. HOLTZMAN, MD,
AND
A. JUDITH CHWALOW, RN, DRPH
Abstract: Should there be legal requirements of parental consent and refusal for the screening of newborns for phenylketonuria (PKU)? There are cogent moral arguments against requirements of parental
consent for PKU screening. There are also compelling moral arguments against a policy of honoring parental refusals of PKU screening. (Am J Public Health 1982; 72:1396-1400.)
Informed consent has emerged as a mechanism for protecting individuals from potential violations of their autonomy by health professionals. Underlying the informed consent requirement is the position that it must be the individual patient and not the health professional who should ultimately determine what is in the patient's interest. Patients usually conclude that it is in their best interest to accept a recommended treatment but, sometimes, a patient's personal view of the situation leads the patient to reject medical advice. So long as there are no questions of competency, the essence of the obligation to obtain informed consent is respect for the patient's refusal of medical treat-
boundaries of this control, as in child abuse and child labor laws. These boundaries are often not well established, with public policies ill-formed, if formed at all. Elsewhere in this issue of the Journal, we report results of a study that evaluates the effects of a Maryland state regulation requiring parental consent for phenylketonuria (PKU) screening of neonates.' The results of this study refute two major moral and policy objections against the Maryland regulation: 1) meaningful parental choice would not be achieved, and 2) the program would cease to be costeffective if consent were required. Our empirical refutation of these important objections had an understandable impact on the public policy process. The state agency responsible for the regulation, the Maryland Commission on Hereditary Disorders, was heavily influenced by our results to continue with the parental consent regulation. The absence of any demonstrable cost to the public health appears to provide a good reason to continue with a regulation that has been reasonably effective in satisfying parental preferences to be informed about newborn testing. However, it does not necessarily follow from the empirical results of our evaluation study that parental consent and refusal requirements ought to be public policy for PKU screening. Although empirical data can play a significant role in the resolution of moral dilemmas about public policy, particularly when the dilemma turns on disagreements about the consequences of the policy, empirical data cannot always be determinative in the closure of policy disputes. Vexing moral concerns sometimes remain even after all issues of fact have been settled. In the case of PKU screening, to refute objections to the policy is not the same as providing a compelling ethical justification for the policy. The proper question is: "What policy of consent and refusal,
ment. When one moves from first-person consent to parental consent, however, the argument shifts. Questions of parental autonomy, or the right of parents to chart the development and life experiences of their children, are not nearly as compelling as questions of personal autonomy. Although there is a long-standing tradition granting parents and the family broad discretion and control over the care and welfare of children, there are established policies that restrict the From the Department of Health Services Administration, Division of Health Education, Johns Hopkins University School of Hygiene and Public Health; and the Department of Pediatrics, Johns Hopkins University School of Medicine. Address reprint requests to Dr. Ruth Faden, Johns Hopkins University School of Hygiene and Public Health, Division of Health Education, 615 N. Wolfe Street, Baltimore, MD 21205. This paper, submitted to the Journal December 23, 1981, was revised and accepted for publication May 20, 1982. Editor's Note: See also related article p 1347 and different view p 1401 this issue. © 1982 American Journal of Public Health
1 396
AJPH December 1982, Vol. 72, No. 12
DIFFERENT VIEWS
if any, ought we to adopt?" It prejudices any answer to cast the question as "Why shouldn't we have a parental consent policy?" In this paper we consider whether there exists a sufficiently compelling moral justification for parental consent and refusal requirements for PKU screening. The focus is entirely on questions of moral, not legal, justification, even though this justification is of legal requirements. Our central concern is the moral justifiability of the parental right to refuse PKU screening. We conclude that although a parental consent and refusal requirement is feasible practically, and may even be politically desirable, there is no compelling moral justification for such a policy, and there are powerful moral arguments against it.
normal infants and holds promise of the remote but important benefit of preventing mental retardation, which threatens not only the general well-being of infants, but also their future abilities to develop as autonomous agents. Thus, there appears to be no reasonable question or issue of judgment as to what is in the best interest of infants. Under these conditions, is a public policy that grants parents the right to consign their children to a state of irreversible mental retardation morally acceptable? We think not, and it is for this reason that we question the moral justification for requiring parental consent for PKU screening. To require parental consent entails an obligation to respect parental refusal, and it is the validity of such refusals that we question. If the principle consideration is the welfare of children, their welfare is best served in this case by a program of compulsory and exceptionless screening.
The Function of Parental Consent Until recently, there has been little analysis of the function of a parental consent requirement when several competing interests are involved.2 Although case law links parental consent with the protection of parental self-interest and prerogatives,3 parental consent is generally not viewed as a mechanism for protecting the autonomy of parents or of parents' rights in medicine and biomedical research. Rather, parental consent is defended as a mechanism for protecting the welfare of children, on the theory that parents are their children's most conscientious advocates and have their children's best interest most accurately in focus. This analysis of the primary function of a parental consent requirement presupposes, but does not argue for, a well entrenched view in the literature of biomedical ethics, viz, that third-party consent is grounded in the principle of beneficence rather than the principle of autonomy. Even such starkly different positions as those defended by Paul Ramsey and the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research converge on this point.4 Even if we consider general theories of proxy consent that allegedly are rooted in the principle of autonomy, such as the theory of substituted judgment, the focus of the obligation is not to respect the autonomy of the proxy but rather to honor the proxy's ability to infer, based on prior knowledge of the compromised person's values and preferences, what the individual would choose, if he or she had been competent. It is very doubtful, however, whether substituted judgment can be made an adequate theory for justifying policies of parental consent, particularly in the case of newborns.5 If the primary function of a parental consent requirement is to protect the welfare of children, how does this apply to the case of PKU screening? In this form of screening, the intervention poses minimal risk* of harm to *Medical risks of PKU screening were discussed in Faden, et al: A Survey to Evaluate Parental Consent as Public Policy for Neonatal Screening.' More problematic is the risk of false positive test results. Unaffected infants with positive results could be started
AJPH December 1982, Vol. 72, No. 12
Child Welfare and the Harm Principle By making child welfare the overriding consideration in policy determinations about participation in child health programs, we are taking the position that in this context, any right of parents independent of their role as advocates for their children's welfare is subservient.** The general outlines of this position in contemporary ethical theory can be traced historically to John Stuart Mill's On Liberty, where he discusses what is now popularly referred to as the harm principle.*** This principle states that a person's liberty may justifiably be restricted if the person's (e.g., a parent's) action or negligence poses significant risk of harm to another
on therapy that is harmful to them. However, the chances of inappropriate treatment are reduced by adequate follow-up. It is our view that all states have an obligation to ensure adequate follow-up, particularly when screening is compulsory. It is also possible, although as yet undocumented, that false positive PKU results may lead to harmful social and psychological effects if parents falsely perceive the child to be sick or ill. In Maryland, parents are informed that a positive test does not mean that the baby is affected or sick, but rather that additional tests are needed. **Consider the position of Norman and Cindy Wingerden in a recent court case in Yates County, New York, as reported in the Rochester Democrat and Chronicle.6 The Wingerden's, who are members of the Church of the Firstborn Whose Names are Written in Heaven, were objecting to hospital compliance with a state law requiring administration of silver nitrate to newborns' eyes to prevent purulent conjunctivitis. They claimed that the administration of silver nitrate to their about-to-be born infant would violate church doctrine which requires that attempts to heal by intercession
through prayer precede medical intervention. The judge, Justice David 0. Boehm, ruled against the parents. However, his decision was confined to a consideration of whether the court could compel a hospital to violate the law, and not whether the law was itself constitutional (personal communication, Justice Boehm, July 6, 1981). From the perspective of the position proposed in this paper, the parents' rights to religious freedom would be overruled by the right of the infant to be protected from the severe and noncontroversial harm of blindness. For an analysis of First Amendment implications for such cases, see Capron.7 ***It is beyond the context of this article to address in detail this aspect of ethical theory. A useful interpretation of Mill's harm principle is found in Joel Feinberg's Social Philosophy.8 1 397
FADEN, ET AL.
(e.g., the parent's child). The point of this appeal for our purposes is that parental refusal of PKU screening unjustifiably poses a risk of harm to the child. Central to this argument for compulsory newborn screening for PKU is the absence of any reasonable question as to the welfare of children. To the extent that potential benefits to individual children may be present, a justification tor a parental consent requirement is enhanced. Thus, from the perspective of a general policy on parental consent in public health programs, a central issue is whether risk/ benefit judgments for individual children may sometimes differ from judgments about risks and benefits for children in general. If some children would be better off without the program, then parental consent may be required as a safeguard to protect the interests of those children.t However, only if there is a reasonable possibility that some children would be better off should a policy permitting parental refusal be adopted. It is important to clarify our position, which is based on the position that there is no reasonable question or issue of judgment as to the net beneficial character of the policy we propose. We do not mean that there can be no doubt in any parent's mind about the net benefit of the intervention. In the case of PKU, one could imagine parents opposing screening based on bizarre views about child welfare. For example, parents might believe that the screening test condemns children to eternal damnation. We would not interpret such metaphysical beliefs about damnation as providing a reasonable judgment as to the best interests of infants. Clearly, much turns here on what counts as "reasonable," a term which has defied adequate analysis in the courts as well as in the psychological and philosophical literatures. In the end, it may be best to resort to procedural mechanisms for resolving problems about reasonableness. For example, the question of reasonable doubt about the net benefit of a program might be left to the judgment of appropriately constituted panels, such as the Maryland Commission on Hereditary Disorders. This approach to child health programs will be difficult to interpret and apply in some cases. A hard case is found in childhood immunizations. Here there is substantial controversy concerning whether the benefits of certain immunizations, such as pertussis, exceed the risks for all children. In addition, there is controversy concerning whether, at least theoretically, an individual child would be better off not immunized. Certainly, there is a risk of contracting the disease or experiencing a complication from the immunization, and it is generally acknowledged that an individual child who is not immunized is afforded some protection from natural strains of the disease by the community's general level of immunity. On the other hand, unless sufficient numbers of children continue to be immunized, this limited protection disappears. Should a child contract any of these
tAlternative safeguards to parental consent include excluding certain groups of children from the program and a parental disclosure obligation (see following footnote). 1 398
diseases, the harm to the child can be substantial. For these reasons, many have argued that all children should be immunized.tt However, the interpretation of net benefit in this case of immunization is far more ambiguous than in PKU screening, where the net benefits are quite clear. The policy implications for parental consent and refusal are also less clear in immunization.
Parental Consent vs Parental Notification In arguing against a right of parents to refuse neonatal screening, we make a distinction between the obligation to obtain parental consent (and respect parental refusal) and the obligation to inform parents about the procedures or interventions that are performed on their children. That parents have a right to know, even if not a right to consent, can be justified on several grounds. Parents can often supply information about their children's medical history that is relevant to the proposed public health intervention. For example, in the case of PKU screening, if parents knew that screening was to be performed, they could alert providers to the presence of a family history of PKU or bleeding disorders. There is also the possibility of direct benefit to the parents in the form of relieved anxiety or a general sense of satisfaction about their child's health care and health status. As indicated by the mothers in our study,' parents have a strong desire to be involved in the medical care their children receive. Our data suggest that it is possible to respect parents' desires to be informed through a relatively efficient and non-intrusive mechanism. We advocate that a disclosure form be routinely distributed to all parents, even though their express consent will not be solicited. However, we should emphasize that this policy alternative was not empirically evaluated in our study, and thus we do not know whether the quality of information disclosed would diminish if a parental consent requirement were lifted.
Compulsory Screening and State Obligations If PKU screening is made compulsory, what is the state's obligation to afflicted infants and their parents? We see the state as obligated to assure that the medical resources necessary for effective treatment are made available. By "medical resources," we mean adequate follow-up and counseling for parents of babies with positive tests, as well as referral to pediatricians who are experienced in the management of PKU. Unless the state provides these re-
ttExceptions to this rule are those children for whom immunizations are medically contraindicated, as in the case of children with certain allergies or certain serious diseases. It is the responsibility of the provider administering the immunization to assure that no child with such a history is immunized. While it is true that requiring parental consent would provide an added safeguard for such children, this additional protection could be achieved as well through a professional obligation to question parents about contraindications. AJPH December 1982, Vol. 72, No. 12
DIFFERENT VIEWS
sources, it is doubtful that the promised benefits which justified compulsory screening will be achieved. By applying this same reasoning, it is possible to generate a state obligation to finance PKU treatment. Ideally, questions of financial responsibility would not be at issue. Given the current health care structure, the state should be obligated to provide treatment for infants for whom the financing of care can not be arranged through other mechanisms (e.g., Medicaid, crippled childrens' funds, third party payers, parental resources, etc.). Apart from providing financial resources, what ought the state to do if a parent of a PKU child detected through compulsory screening refuses to allow treatment of the infant? It remains our position that parents do not have a right to consign their children to irreversible mental retardation in situations where there is strong epidemiological evidence that the treatment is effective. We therefore recommend that legal steps be taken to compel parents to treat and, if necessary, to remove the infant from parental control for treatment through statutory provisions covering child neglect. Our position on the issue conflicts directly with Joseph Goldstein's position that parental autonomy validly overrides child welfare in all cases of clinical decisionmaking, unless the child's life is at stake.9 His position is based largely on deep concerns about the integrity and privacy of the parent-child relationship and about procedural and empirical difficulties in making someone other than the parent responsible for determining what is in an individual child's best interests. However, one of the most compelling features of the PKU case is that parental refusal of treatment is likely to rob infants of the future ability to be selfgoverning (autonomous), thus depriving them of their right to determine for themselves what is in their best interest. The importance of protecting this capacity for self-governance outweighs even the possibility that a state overriding of parental authority may result in rejection of the child by the parents and the child's subsequent dependence on the state social welfare system.
Some Caveats It could be argued that a policy of compulsory PKU screening secures the short-term and obvious interests of infants at the expense of the long-term and more fundamental interests of children to be raised in secure and responsible family units. To this objection it should be noted that our application of the harm principle is not restricted in its analysis to short-term or immediate risks of harm, nor is it intended to handle traditional cases in which the state overrides parental authority. At this point, we are not prepared to generalize from our arguments about how a state ought to respond where parents' refuse PKU treatment to all cases of parental refusal where a child's welfare may be at stake. Our position in this particular case is heavily influenced by the obligation to protect the future autonomy of the child. Thus anyone who argues against our position by appealing to parental rights bears the powerful burden of AJPH December 1982, Vol. 72, No. 12
showing that parental autonomy in this case overrides the autonomy interests of the child. Our central concern in this paper is not with traditional cases of conflict over medical treatment for a particular child, however, but with the application of the harm principle to public health programs where the state is already and inevitably involved. The policy issue in this instance is not whether a parental refusal should be overridden but rather whether parental authorization should be solicited in the first place. Moreover, we are applying the principle to programs that are applicable to all children. In such contexts, it is extremely unlikely that the integrity and privacy of the family unit would be seriously threatened. These contexts should be distinguished from situations in which parents have made a particular decision about their child which the medical community or the state now wishes to override. To apply a distinction used by Robert Bennett,'0 we have proposed the harm principle for public health cases where intervention focuses not on any individual child but on some defined disease or condition (PKU) as it affects children as a class. Decisions about the beneficial character of an intervention in a public health context are thus judgments which are applicable to children in general-meaningful judgments which can be made without regard to individual differences. Our position obviously presupposes that there are cases in which it is possible to make an objective determination about what is in the best interests of children, an assumption that not all commentators are prepared to grant.5 In contrast to the public health contexts we have in mind, state overriding of parental authority often requires a more subjective judgment concerning characteristics that are unique to the particular child, and where threats to the integrity of the parent-child relationship are salient. These cases typically involve a disagreement between parents and medical or state authorities about which of several alternatives best serves the child's interests-for example, whether a child with scoliosis should be treated surgically or with a brace, or whether a child with a cleft palate should have plastic surgery. Finally, we wish to emphasize that, with regard to genetic services, we are here questioning parental consent for PKU only. Our argument is not proposed as extending even to any other form of genetic screening or counseling. For example, our position does not apply to sickle-cell screening, where many questions persist about both potential benefits and potential harms. Even in the case of neonatal screening, we have been careful not to extend our suggestions to screening for other diseases. In many states, including Maryland, newborns are screened not only for PKU but for other diseases such as hypothyroidism, galactosemia, and branched chain ketoaciduria. As this is usually accomplished by a single heel prick, there may be serious practical obstacles to permitting compulsory screening for some of these diseases and parental discretion for others. However, the benefits and risks of screening may differ in important ways for each of these conditions and it may be necessary to inform parents that certain screening tests are optional while others are not. Before a policy of compulsory neonatal screening is adopted, each condition for which 1 399
FADEN, ET AL.
screening is contemplated must be considered individually against harm principle considerations, the role of parental expertise, the value of family privacy, and other factors. For example, in the case of branched chain ketoaciduria (BCK), the consequence of parental refusal of screening and thus of treatment is death. However, approximately 30 per cent of treated BCK infants will have significant chronic disability. Many of these infants will be mentally retarded, and all will have an increased chance of death from intercurrent infections. Those who uphold the right of parents to refuse lifesaving treatment for defective newborns may wish to champion a parental right to refuse BCK screening. In that case, the disclosure should indicate, as it currently does not, the variable response to BCK treatment.
Conclusion We have argued against parental consent and refusal policies for PKU screening. Our position is based largely on two arguments: * First, the primary function and justification of a parental consent mechanism is to protect the welfare of children; yet, as regards PKU, there are no relevant distinctions between children and no reasonable doubt about their best interests. Thus, any unique expertise parents might have to make judgments about their children's interests does not apply. Other functions of parental consent requirements, such as respecting family integrity or privacy, are only minimally involved, and notification requirements should suffice to relieve parental anxiety and encourage parental involvement. * Second, the harm principle effectively overrides any parental right to refuse PKU screening rooted in a principle of parental autonomy. Parental refusal of screening or treatment unjustifiably poses a grave, if remote, risk of harm to the child. Perhaps the most compelling argument against our position involves a combination of political, pragmatic, and moral considerations: The likelihood of harm seems quite remote in PKU screening, because of both the relative rarity of the disease and the empirically demonstrated low rate of parental refusal; might it not therefore be reasonable to ask, "why bother?" That is to say, why bother to buck the prevailing social trend favoring policies of parental choice over compulsory programs? To this concern might be added the standard "slippery slope" argument that even if a policy of parental consent and refusal may be morally questionable in the particular case of PKU screening, legislation disallowing parental choice in this instance opens the door to compulsory programs in areas where such a policy would be morally reprehensive. The central problem here is whether it is possible to identify the proper conditions for distinguishing situations where parental consent ought to be required from those where it ought not or need not be obtained-
1 400
especially when convenience to the state is bound to be an important practical consideration. Whether the conditions we propose here are equal to the difficult task of adequately restraining control by state authorities is certainly debatable. Despite these "slippery slope" concerns and the relatively low likelihood that harm will result, we remain deeply skeptical about the moral justifiability of any increase in risk attributable to parental refusal of PKU screening. The harm principle is deeply entrenched at the center of our system of moral beliefs, and we think it clearly, applies in a most compelling way to those infants who would be victimized by parental refusals. Whether our position is sufficiently compelling to override political, practical, and moral arguments to the contrary must, of course, remain an open question. In the end, we merely note our thorough agreement with Willard Gaylin's remarks about consent policy and health care for children: "Any position in a complex area such as this one that does not produce anxiety is probably too simple to be argued, too unworthy to be held. Similarly conclusions must always be tentative, of a time, and receptive to immediate modification and constant reexamination."
REFERENCES I. Faden R, Chwalow AJ, Holtzman NA, Horn SD: A survey to evaluate parental consent as public policy for neonatal screening. Am J Public Health 1982; 72:1347-1352. 2. Gaylin W, Macklin R (eds): Who Speaks for the Child: The Problems of Proxy Consent. New York: Plenum, 1982. 3. Annas GJ, Glantz L, Katz BF: Informed Consent to Human Experimentation. Cambridge: Ballinger, 1977. 4. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research: Report and Recommendations: Research Involving Children. DHEW Pub. No. (OS)770004, Washington, DC, 1977. 5. Macklin R: Return to the best interests of the child. In: Gaylin W, Macklin R: Who Speaks for the Child: The Problems of Proxy Consent. New York: Plenum, 1982. 6. Monoghan N: Couple Fights Eyedrop Law for Newborn. The Rochester Democrat and Chronicle, March 5, 1980, pp IB, 2B. 7. Capron AC: The authority of others to decide about biomedical interventions with incompetents. In: Gaylin W, Macklin R: Who Speaks for the Child: The Problems of Proxy Consent: New York: Plenum, 1982. 8. Feinberg J: Social Philosophy. Englewood Cliffs, NJ: Prentice Hall, 1973, chapters 2, 3. 9. Goldstein J: Medical care for the child at risk: on state supervention of parental autonomy. Yale Law J 1977; 86:645-670. 10. Bennett R: Allocation of child medical care decision-making authority: a suggested interest analysis. VA Law Review 1976; 62:285-330. 11. Gaylin W: Who Speaks for the Child? In: Gaylin W, Macklin R: Who Speaks for the Child: The Problems of Proxy Consent. New York: Plenum, 1982.
ACKNOWLEDGMENTS The authors are indebted to Tom L. Beauchamp and Stephen Teret for their very helpful comments on this paper. One of the authors (R.R.F.) is also indebted to students in her Ethics and Health Policy course whose comments on this case over the years have contributed significantly to the current effort.
AJPH December 1982, Vol. 72, No. 12
