Children and health-care research: best treatment,best interests and best practice

In this article, I consider aspects of the law and ethics surrounding the conduct of medical research in children. I discuss the ethical principles and common law applicable to all kinds of research and the legislation that pertains specifically to clinical trials of medicinal products. I focus specifically on issues related to obtaining parental consent for research involving minors who lack capacity to consent for themselves. Issues relating to obtaining consent from competent minors who can consent on their own behalf have been explored in a companion piece. ¹

There are ‘good reasons for conducting clinical research on children’ ² despite the fact that they are rightly regarded as a vulnerable group. But such research presents specific difficulties. It is well known, for instance, that drugs administered to children are frequently either not licensed for use in that population or are prescribed off-label. ^3,4 Often the dosages are calculated according to data generated in adult trials and simply scaled down. Furthermore, children's metabolism differs from that of adults, changes depending on the stage of development and many diseases are specific to children. This is not ideal but is unavoidable because, in the absence of research, the data on appropriate formulations and dosages for children are just not available. ⁵ The need to conduct research in this population is therefore firmly established and ethical guidelines indicate that in appropriate circumstances such research is permissible even if it is not intended to benefit the individual child participant. ⁶ The research may, for example, be designed to determine the effectiveness of a treatment in different age groups, or perhaps the palatability of specific formulations of a particular medication. Both are valid and necessary investigations but neither will necessarily benefit the individual children in the study. Consequently, conducting medical research involving children raises a plethora of ethical and legal issues concerning the ways in which their participation in research might be authorized. As with all health-care research, consent must be obtained to ensure that participation is legitimate, and in the case of children it is usually the parents who provide this consent. The assessment of best interests is thus a central consideration in this process and forms the basis of this article.

Best interests and parental consent

The General Medical Council's guidance for doctors who are considering conducting research involving children or young people explains that children ‘may be vulnerable because they cannot always recognise their best interests, express their needs or defend their rights’. ^7,8 In such circumstances, a child's interests would normally be safeguarded by their parents or another adult who has parental responsibility (PR), and this would be formalized through the process of giving consent. Unsurprisingly, however, with regard to health-care research there is evidence that parents are often reluctant to enter their offspring into clinical trials or are confused by the prospect. One study indicates that 41% of parents preferred the doctor to decide on their behalf rather than being responsible for the decision themselves, and 29% of parents who did enter a child into a study did so because they believed there was no alternative. ⁹ The responses also showed that 38% of those who agreed to their child's participation did so in the hope that the child would benefit from state-of-the-art treatment as a result of their involvement in the research. Other factors, such as wishing to help other children with similar conditions or giving something back in return for the care received, may also influence parents’ decisions. ¹⁰

Although it seems probable that parental decisions to enrol children in research are based on their perception of what is in the best interests of their child at the time, such decisions are likely to have been assessed on considerations much broader than those that would normally influence treatment decisions. Considerations such as altruism and benefit to the wider community are valid and appropriate when adults or parents are making such decisions about themselves, but the moral baseline is rather different when making proxy decisions for children. ¹¹ Thus, while it is acceptable, even laudable, for adults to enter research studies on this basis, it is questionable whether parents have the right to sanction such gestures on behalf of their children and in the process expose them to the associated risks, however minimal those risks may be. Indeed, the legal mechanisms associated with obtaining valid consent for children might even militate against this.

The common law and the best interests of the child

There is a legal presumption that children below the age of 16 lack the maturity to make health-care decisions. Under the Gillick criteria that is a rebuttable presumption, ¹² but this paper is concerned solely with those who are not competent to decide for themselves. In England and Wales the Children Act 1989 gives the adults who are responsible for a child's welfare, usually the parents, the power to make medical decisions. If there is no parent or PR has been lost, this role will be undertaken by another adult to whom PR has been assigned or, occasionally, the courts. In Scotland the Children (Scotland) Act 1995 operates according to the same principles, as does the Northern Ireland the Children (Northern Ireland) Order 1995 in that jurisdiction. The right is reinforced by the European Convention on Human Rights, ¹³ and the European Convention on Biomedicine ¹⁴ applies the same approach. Although the UK is not currently a signatory to the latter Convention, the provision is nevertheless informative but not binding.

Ordinarily the authority to consent to or refuse medical treatment for a child rests with her parents. But it is a right that exists only as long as what is proposed advances the welfare of the child and is demonstrably in the child's best interests. Judicial precedent in cases concerning medical treatment indicates that the courts generally construe ‘best interests’ to mean selecting the best treatment option for the child from a range of options available. ¹⁵ Where what is proposed is a medical treatment designed to benefit the child it is relatively easy to show that it is in the child's best interests. It may, however, be less clear-cut in relation to research. The very fact that research is being conducted implies that the outcome is either unknown or uncertain. ¹⁶ After all, if the result was known it would be unethical to do the research. Since it is uncertain whether an individual research participant will derive any benefit, especially where the purpose of the investigation is to add to the body of knowledge about a particular condition or treatment for the common good, this kind of assessment is inappropriate when authorizing a child's participation in medical research. Consequently, despite the obvious ethical imperative to encourage research involving children generally, the legal validity of parental consent to the involvement of a particular child in health-care research is clearly problematic.

There is no specific case law in the UK on the legal authority of parental consent to research on children with regard to the assessment of best interests, but a case known as the ‘Kennedy Krieger lead paint study’ in the USA did test this point. ¹⁷ The study was constructed to compare different methods of reducing the content of lead in social housing in an area of Baltimore where there was a high incidence of lead poisoning among resident children. Local landlords were given incentives to enter the study and encouraged to rent their properties to families with young children. Although it seemed likely that some children would have been exposed to lead poisoning regardless of the study, on discovering the details of the study some parents complained that they were not adequately informed about the levels of lead to which their children were exposed or the risks of participating in the research. In the subsequent court case it was held that:

It is not in the best interests of a specific child, in a non-therapeutic research project, to be placed in an environment, which might … be, hazardous to the health of the child… the ‘best interests of the child’ is the overriding concern of this court in matters relating to children. ¹⁸

Mason and Laurie rightly assert that ‘the same sentiment underlies the legal framework in the United Kingdom’, ¹⁹ raising concerns about the legality of parental consent to the inclusion of children in research in this jurisdiction.

Courts in the UK have repeatedly discussed the importance of best interests, the ways in which they might be construed, and the factors that ought to be considered in deciding on the best interests of a child in relation to medical interventions and treatment provision. A notable early case concerned the question of whether parents could authorize a blood test on their child in order to determine its paternity. ²⁰ Rather than focusing on the invasive nature of the test, the judges in this case considered matters of justice and the possible benefits, or otherwise, to the child in relation to the potential outcomes of the test. Thus, best interests was very broadly defined and ultimately it was decided that the blood tests could be authorized ‘unless satisfied that it would be against the child's interests’. ²¹ Assessing best interests in this way would help to legitimate the inclusion of minors in many forms of research. Their participation would be permissible so long as the procedures were not so harmful as to be obviously contrary to the individual child's interests. However, more recent cases indicate that the extent to which parental consent can be relied upon to render medical interventions lawful is more limited. For example, in relation to surgery the courts have stated that while parental consent can authorize interventions designed to benefit the specific child, this excludes interventions aimed at ‘the interests of those responsible for the care of the child or in the interests of society in general’. ²²

While the uncertain status of parental consent in these circumstances is problematic in and of itself, the unease is heightened in situations where parents, or those with PR, disagree between themselves as to whether the child should participate in research. In relation to the beneficial medical treatment, the refusal of one parent is not a bar to therapeutic intervention so long as one authorized person gives consent and the proposed treatment is in the best interests of the child. ²³ This implies that the welfare interests of the non-competent child ought properly to take primacy over all others in relation to research and, as Lord Fraser has claimed, ‘parental rights to control a child exist not for the benefit of the parent but for the child’. ²⁴ Accordingly, justifying non-therapeutic research that cannot be regarded as potentially beneficial to the child in the face of a parent's determined refusal would seem to be legally foolhardy, even if the other parent agreed. By contrast, the situation in relation to clinical trials of medicinal products is rather more certain.

Statute law and clinical trials

Research that involves medicinal products in the UK falls under the statutory auspices of the Clinical Trials Regulations 2004, ²⁵ which transpose the European Clinical Trials Directive 2001 into domestic law. ²⁶ The Regulations operate explicitly on the premise that the parent or another person with PR will authorize the child's involvement through giving their consent. Both legal instruments specify that before a child's involvement in a trial commences it must be shown that ‘the informed consent of the parents or legal representative has been obtained’ and that this ‘consent must represent the presumed will of the minor’. ²⁷ There is no explicit requirement to assess whether such inclusion is in the child's best interests. Nevertheless, both the European Directive and the UK's Clinical Trials Regulations incorporate four general guiding principles applicable to all research that are relevant to the assessment of best interests.

First, the interests of the patient (or participant) should always prevail over those of society and science; secondly, the design of the trial must minimize pain, discomfort, fear and foreseeable risks of the disease. In addition, the risk threshold and degree of distress must be clearly defined and monitored. Furthermore, under Schedule 1 Part 4 of the Regulations, the validity of consent from the parent or legal guardian is contingent upon a number of additional requirements. Specifically, before deciding whether to enrol the child, the parent or person with PR must be given all necessary information relevant to their decision. They must be allowed sufficient time to consider the information and to assess any risks and benefits. They must also be notified of their right to revoke their consent to participation at any time and thereby withdraw the child from the research. The bestowing of incentives to the parent, child or the person with PR is strictly forbidden. Consent should be confirmed by an iterative process throughout the study rather than being a one-off procedure, and the parent or person with PR should be provided with details of how and where to obtain further information about the trial if required. These measures are designed to prevent coercion and ensure that parents can act according to their understanding of the best interests of their child.

The UK's Clinical Trials Regulations make no explicit provision for situations where parents disagree about their child's involvement in a research project. However, in limited circumstances, such as where the research will be conducted in an emergency setting and it is impossible to obtain contemporaneous parental consent, provision is included in paragraph 1, Part 4 of Schedule 1 for making alternative arrangements to obtain consent. For instance, where necessary, a legal representative, who should be a person who has knowledge and understanding of the trial and its aims and objectives, including ‘the conditions under which it is to be conducted’ and any associated ‘risks and inconveniences’, will represent the child's interests. Taken together, all of these requirements tend towards safeguarding the best interests of the child participant but in practice they may not always be effective.

Best interests in practice

There is no doubting that children are a particularly vulnerable group, or that they have been exploited in the past in the name of research. ²⁸ However, whereas previous abuses were probably founded on the inherent exploitability of particularly vulnerable children, such as those in children's homes or from other disadvantaged communities, the current calls for increased research involving children are based upon more wholesome motives. Perhaps it is for this reason that parents may be more willing to agree to their children's participation in research than might be expected. ²⁹ Nevertheless, in practice, when the health of a child is at stake parental attitudes to best interests can be difficult to determine.

Although many of the treatments available to children have not been systematically researched in the childhood population, they are routinely used, and generally without mishap. Doctors therefore use this experience to offer child patients treatments that they believe will promote their best interests. Parental informed consent is required to authorize the treatment. However, in this context, it is reported that 86% of parents are not aware that unlicensed and off-label prescribing to children occurs. ³⁰ Once informed of this fact, one study found that 20% of parents of healthy children said they would refuse to permit their children to be treated with off-label medications, and a further 9% of parents with chronically sick children claimed they would do the same. ³¹ These parents presumably believe they would be acting in the best interests of their children by refusing treatments that have not been specifically tested in children despite the fact the doctors recommending the treatments clearly regard the treatment as being beneficial in the child's best interests. Moreover, while there may be a lack of formal research in many if not most cases, there is a wealth of practical clinical evidence as to how the medications perform in situ which validates their safety and efficacy in practice. Yet parents are clearly concerned that exposing their offspring to such medications may be contrary to their best interests, a fact which seems to be at odds with the attitude of those parents who agree to the participation of their children in research. ³² Parental assessments of what is in the best interests of their children may therefore be inherently unreliable.

Conclusion

The law requires that parents or others with PR give consent to the involvement of children in health-care research. More generally, it is expected that parental consent will be based upon the child's best interests and case law demands that such consent will only be valid if it is. This raises two distinct anxieties. The first concerns whether it is ever possible to argue that involving a child in medical research is in her best interests. In the limited circumstances where a child is recruited to the study of a medicine or other intervention designed to treat the condition from which she suffers, where the intervention proves to be effective and the child was in the active treatment arm, it might be possible to construct an argument that participation in the study was beneficial. But this would only be known with hindsight and the likelihood is necessarily limited; some of the participants will undoubtedly be in a control or even placebo arm and therefore derive no benefit at all. Also, in many cases the study design dictates that the data generated will benefit only future child patients rather than those participating in the research. Given the level of uncertainty about whether any particular child will derive any benefit, it is difficult to argue that involvement in research is demonstrably in the individual child's best interests, making the legitimacy of parental consent to the child's participation in the research highly questionable.

Secondly, the validity of parental consent itself is questionable if parental understandings of best interests are as inconsistent as the research cited here suggests. The evidence seems to indicate that based on best interests some parents are unwilling to subject their children to risks associated with receiving off-label treatment, while others are content that it may be in their child's best interests to participate in a clinical trial. The reasons behind this are worthy of further research. In the meantime the subjective nature of parental understandings of best interests are worrying because the law expects that parental consent will reflect the child's best interests. Greater legal clarity is therefore also urgently required. ³³

Research must be performed if the efficacy and safety of new and existing treatments is to be firmly established. This might involve subjecting some individual child participants to some risks and it is for the parents to assess that risk on their child's behalf before consenting to their child's participation. However, this is at odds with the legal presumption that parents will give consent to interventions that are in their child's best interests. Nevertheless, if research is to be conducted for the benefit of children as a whole risks have to be taken, and who better to assess that risk on their child's behalf than the parent?