LEARNING OBJECTIVES

After reading this article you will be able to:

• • discuss adolescent decision-making and how it compares with adult decision-making

• • better understand the ‘reach’ of capacity in young people and competence in children

• • understand the justification for the differential treatment of acceptance and refusal.

Consent to treatment in minors can be a complex and confusing issue in practice. The current position in England and Wales involving the ability of the capacitated minor to give consent to accept treatment, while having only limited ability to refuse, presents challenges to clinicians. This is the second of a pair of articles reviewing consent in minors. The first article (Hawkins 2023) explored the concept of autonomy, its relevance in clinical practice and factors that might interfere with its 'purity'. It then gave an overview of minors' rights, and the way that UK and European courts are representing those rights in their judgements. This second article builds on the first and explores the way minors make decisions, the common law and statutory legal position regarding consent, the reach of consent in minors, and the justification for the differential treatment of acceptance and refusal.

A contemporary treatment case

In 2019 a case was brought against the National Health Service (NHS) Gender Identity Development Service (GIDS) by a parent of a 15-year-old on the GIDS waiting list, along with a former nurse who had worked at a GIDS satellite site (Bell and Another v Tavistock and Portman NHS Foundation Trust [2020]). The complaint was that evidence regarding hormonal pubertal blockade was unclear and that true informed consent could not be given. In the process of filing the case, the nurse passed her role in the complaint to Keira Bell, then aged 23, who had been prescribed hormone blockers aged 16 by the GIDS, subsequently received testosterone and surgical intervention and later regretted her transition.

Keira Bell spoke to the BBC about her pathway, describing the way that ‘one step led to another’ and that she felt she ‘should have been challenged on the proposals or the claims that I was making for myself’ and that if she had been ‘that would have made a big difference’. She spoke of the way that she was ‘allowed to run with this idea that I had, almost like a fantasy as a teenager, and it has affected me in the long run as an adult’ (Holt Reference Holt2020).

The High Court (Bell v Tavistock [2020]) considered the fact that the vast majority of those prescribed hormone blockers proceed with cross-sex hormones, with resultant irrevocable physical effects. In addition, it reviewed the limited evidence base in relation to pubertal blockade and its efficacy. It found that ‘there will be enormous difficulties in a child under 16 understanding and weighing up this information and deciding whether to consent to the use of puberty blocking medication’. It concluded that it was ‘highly unlikely that a child aged 13 or under’ and ‘doubtful that a child aged 14 or 15’ could understand and weigh up the long-term risks and consequences of such a decision. It went on to say that in those over 16, ‘authorisation of the court should be sought’ (because of the presumption that they have the ability to consent, and in view of the potential serious consequences of the treatment (paras 151–152). In giving this judgment, the High Court departed from the seminal Gillick case (see discussion below), which had established that competent children could make their own decisions in relation to accepting treatment.

The case was taken to appeal (Bell and Another v Tavistock and Portman NHS Foundation Trust [2021]). The Court of Appeal recognised the ‘difficulties and complexities associated with the question of whether children are competent to give consent’. It also noted that ‘clinicians will inevitably take great care’ to be sure that ‘the consent obtained from both child and parent is properly informed by the advantages and disadvantages of the proposed course of treatment’ and its ‘long-term consequences’. However, it overturned the ruling of the High Court, effectively repositioning Gillick as the appropriate guidance for the under-16s (paras 91–94).

Notably, the Supreme Court refused the application to appeal in 2022, on the basis that it did ‘not raise an arguable point of law’. By so doing, as the appeal court of greatest authority in UK law, the Supreme Court gave its support to the existing law regarding consent in minors – which will be explored in this article (Supreme Court 2022).

Key questions posed by Bell v Tavistock are listed in Box 1.

Minors’ decision-making and risk-taking

The literature on decision-making in minors remains dominated by conventional stage theories of child development. The archetypal stage theory, that of Piaget – which outlined discrete stages of intellectual development, through which a child must pass en route to adulthood – has been criticised for its bias towards European norms and the study of children in experimental rather than everyday settings (Lansdown Reference Lansdown2005). Separate cultural theories have been developed which claim that application of stage theories leads to a ‘consistent underestimation’ of children. However, such alternative theories have ‘not sufficiently permeated [ … ] to influence law, policy and practice impacting on children's lives’ (Lansdown Reference Lansdown2005). The presumptions of the dominant theories tend to ‘foster oppressive stereotypes’ and can propagate the myth of the ‘supposed chasm between foolish child and wise adult’ (Alderson Reference Alderson and Montgomery1996).

Such myths overlook the sophisticated ability to reason and understand which minors indisputably have. There is now no significant doubt that individuals show improvements in reasoning, information processing and expertise as they develop from childhood into middle adolescence (Steinberg Reference Steinberg2005). In so doing they ‘become more capable of abstract, multidimensional, planned and hypothetical thinking’ (Steinberg Reference Steinberg2005). An expansive literature, dating back some 40 years, demonstrates that the performance of 14-year-olds is broadly equivalent to that of two different adult age groups – aged 18 and 21 – on tests of legal competence (Weithorn Reference Weithorn and Campbell1982). This conclusion has remained unchallenged ‘in the past several decades’ (Steinberg Reference Steinberg2005). This might suggest that formal tests of legal competence could be aligned to those used in adulthood for any minor over the age of 14.

The literature is thus of one voice in relation to the ability of adolescents to pass tests of legal competence. However, there is an additional literature, which is more cautionary, about the influence of social and emotional context on adolescents’ decision-making. Steinberg (Reference Steinberg2005) noted that ‘adolescence is a heightened period of vulnerability specifically because of gaps between emotion, cognition and behaviour’. Adolescent egocentrism and a supposed concept of personal invulnerability have been put forward as reasons why decisions by minors are unstable and lead to negative outcomes. Studies have indeed found evidence of optimism bias and notions of invulnerability – but interestingly, no more so in adolescents than in adults. Indeed, it is confirmed that adolescents tend to engage in a decision-making process that involves a balancing of risks (Millstein Reference Millstein2002) and there is no evidence that either egocentrism (Rodham Reference Rodham, Brewer and Mistral2006) or perceptions of invulnerability have a specific effect (Goldberg Reference Goldberg, Halpern-Fisher and Millstein2002; Rodham Reference Rodham, Brewer and Mistral2006; Knoll Reference Knoll, Magis-Weinberg and Speekenbrink2015). These findings demonstrate just how inadequate have been the instinctive assumptions about adolescence and the theories they supported.

Although both adolescents and adults take more risks in groups of peers than when alone, that effect has been found to be more evident during middle and late adolescence than in adulthood (Gardner Reference Gardner and Steinberg2005). This has been confirmed, supporting the theory that as adolescents step away from the authority of adults, they place ‘higher value on the opinions of other teenagers’, with acceptance by their peer group having particular importance (Knoll Reference Knoll, Magis-Weinberg and Speekenbrink2015).

So, despite the previous theories about cognitive capability, it seems that when adolescents do engage in risky behaviour, it is ‘despite knowing and understanding the risks involved’ – instead, in ‘real life situations, their actions are largely affected by feelings and social influence’ (Steinberg Reference Steinberg2005). This understanding seems to coincide with the well-recognised central nervous system reorganisation during which ‘regulatory systems are gradually brought under the control of central executive functions, with a special focus on the interface of cognition and emotion’ (Steinberg Reference Steinberg2005). The development of this aspect of the adolescent mind is not, however, in step with the intensity of the ‘major emotionally laden life dilemmas’ that they confront (Steinberg Reference Steinberg2005).

This comparative vulnerability of adolescent decision-making to feelings and social influence, understood to be part of internal reorganisation of regulatory systems, appears to be settled by the development of an internal cerebral executor with command over the interface between cognition and emotion. Any discussion of the concept of adolescent autonomy must therefore broaden its reach beyond pure tests of cognitive skill. It must attend to this specific vulnerability. The pursuit of a legal understanding of a minor's capacity that considers this vulnerability is pivotal to their ability to self-determine.

Young people and capacity – statutory direction and common law default

Young people (those 16 and 17 years of age) fall under the auspices of the Family Law Reform Act 1969. By its authority conferred under section 8(1): ‘The consent of a minor who has attained the age of sixteen years [ … ] shall be as effective as it would be if he were of full age’. Issues related to decision-making capacity thus fall under the jurisdiction of the Mental Capacity Act 2005 (MCA).

Capacity is assessed in line with the tests in sections 2 and 3 of the MCA. The draft new Code of Practice for the MCA (currently under consultation) that supports the Liberty Protection Safeguards (LPS) makes it clear that section 3 should be applied before section 2, to be sure that decision-making capacity is assessed before a diagnostic assessment (HM Government 2022, para 4.12) (Box 2).

In fact, the MCA applies to young people (over the age of 16) in its entirety, apart from in four important respects: three statutory (lasting powers of attorney, advanced directives and the making of a statutory will), and one provided by its own Code of Practice (Department for Constitutional Affairs 2007), which is of interest here. The Code throws a line to a different way of thinking by saying that if young people are ‘unable to make a decision for some other reason, for example because they are overwhelmed by the implications of the decision, the Act will not apply and the legality of any treatment should be assessed under common law principles’ (para. 12.13). The common law principles in question are those derived from the seminal case of Gillick (see the section ‘Children and competence’ below). As will be shown, this reverts to a different conception of decision-making capacity and allows clinicians to take a broader view beyond MCA capacity towards competence. This position is confirmed in the draft new Code (HM Government 2022, paras 21.19 and 21.20). We contend that the draft Code may be referring (either deliberately or inadvertently) to the specific vulnerability in decision-making that is present in all minors and highlighted by the literature on decision-making that we noted in the previous section.

Interestingly, the draft code makes mention of the United Nations Convention on the Rights of the Child (UNCRC) (HM Government 2022, para 21.3). The proposed Liberty Protection Safeguards, through this draft Code, establish that if a young person lacks capacity as defined in the MCA, before establishing best interests in line with section 4 of the Act, ‘professionals may [ … ] choose to seek consent from those with parental responsibility’, provided that the person with parental responsibility is deciding ‘based upon what is in the young person's best interests’ (HM Government 2022, para. 21.16). But those with parental responsibility, in giving consent to treatment, ‘cannot consent to arrangements that amount to a deprivation of liberty on the young person's behalf’ (para. 21.75). In such a situation a Liberty Protection Safeguards authorisation will be required (para 21.63) (Box 3).

Children and competence – the common law position

For the sake of clarity, we contend that the contemporary debate regarding competence in children (those under the age of 16) changes orientation between a narrow and a broad conception: narrow, in that the focus should align with decision-making capacity – as per the MCA above; and broad, in that, alongside decision-making capacity, complex considerations of maturity need exploration before endowing the child with the authority to decide. This distinction is seen in the different emphases of the lead judgments in the Gillick case (Gillick v West Norfolk and Wisbech Area Health Authority [1986]). Lord Fraser seemed to move towards a narrow conception. Lord Scarman, however, read a broader, developmental maturity into the concept of competence (Box 4).

There is no real doubt that the Lords, despite their differing approaches, were attempting to capture something of the individual developmental context and, in so doing, were pursuing the concept of maturity. If there remains any doubt that Lord Scarman was attempting to provide a principle of solid utility, but with sufficient individual flexibility, it is settled by the following: ‘If the law should impose on the process of “growing up” fixed limits where nature knows only a continuous process, the price would be artificiality and a lack of realism in an area where the law must be sensitive to human development and social change’.

In the first of the so-called refusal cases (Box 5), Re R (A Minor) (Wardship: Medical Treatment) [1991], Lord Donaldson refined Lord Scarman's definition and moved further from Lord Fraser's position. The settled and accepted legal definition of competence in children lies in an amalgam of Lord Scarman from Gillick and Lord Donaldson from Re R (Box 6).

In two more recent cases involving medical treatment and consent, the judges were, however, inclined to revise this settled conception and definition of competence as above and instead to align it with the test of decision-making capacity (and hence the MCA) applied to young people – in Re S (A Child) (Adoption: Consent of Child Parent) [2017] the judge regarded it as ‘appropriate, and indeed helpful, to read across to, and borrow from, the relevant concepts and language of the Mental Capacity Act 2005’; the High Court judgment in Bell v Tavistock [2020] took a similar approach. However, at the crucial point of applying the capacity test, both reverted to considerations of maturity which needed to be made before endowing the minor with decision-making authority. In these cases, the lure of a simple test of decision-making capacity could be powerfully felt. But when attempts were made to apply it, with bold assertions about its reach, aspects of maturity and future orientation were felt to be crucial. Such considerations would not be expected when using the capacity test in young people as per the MCA as above.

So, each time the UK courts have tried to redefine child competence within the boundaries of decision-making capacity as per the MCA, they pause and note the need for an elusive ingredient that attends to concerns about developmental maturity. It seems that the bar of Gillick competence is set higher than that of decision-making capacity, since decision-making capacity is a necessary but not sufficient component of competence. Indeed, one commentator argued that that the test of child competence ‘is sufficiently exacting that many adults might fail it’ (Bainham Reference Bainham1992). Various tools have been designed that assess decision-making capacity in children and young people. These include the Competency Questionnaire – Child Psychiatric version (CQ-ChP) (Billick Reference Billick, Edwards and Burgert1998), the Competency Questionnaire – Pediatric version (CQ-Peds) (Billick Reference Billick, Burgert and Friberg2001), the Hopkins Competency Assessment Test (HCAT) (McAliley Reference McAliley, Hudson-Barr and Gunning2000) and the MacArthur Competence Assessment Tools for clinical research (MacCAT-CR) (Appelbaum Reference Appelbaum and Grisso2001) and for treatment (MacCAT-T) (Grisso Reference Grisso and Appelbaum1998). These are based on the four-stage model of decision-making capacity as per the MCA (Box 2). They all have reasonable interrater and test–retest reliability, but their clinical validity is unknown as there is no gold standard with which to compare. Indeed, a Cochrane review of the available tools was published in 2014 (Hein Reference Hein, Daams and Troost2014) but withdrawn in 2015 (Hein Reference Hein, Daams and Troost2015) on the basis that there was no clear diagnostic gold standard against which to compare the reviewed tests. In being based on the four-stage model, they are more likely to hold some validity in relation to young people, whose capacity is examined through the same lens, rather than children, whose test of Gillick competence requires additional ingredients (see below).

The elusive ingredient – maturity – and its definition

Lord Scarman, in Gillick, explored the concept of the requisite maturity inherent in judgement of competence (Boxes 4 and 6). Other international jurisdictions have also turned their mind to maturity. A sideways glance to these provides profitable cross-reference.

The Canadian Supreme Court, in AC and others v Manitoba (Director of Child and Family Services) [2009], a case of refusal of life-sustaining treatment by a minor, drew heavily on Gillick as an exemplar. It noted the tension between rising adolescent autonomy and the state's protective duty. In reviewing decision-making, the judgment noted that ‘while many adolescents may have the technical ability to make complex decisions, this does not always mean that they will have the necessary maturity and independence of judgement to make truly autonomous choices’. In relation to maturity, it further observed ‘there is no simple and straightforward means of definitely evaluating – or discounting – the myriad of subtle factors that may affect an adolescent's ability to make mature, stable and independent choices in the medical treatment context’. It accepted that in most day-to-day clinical scenarios, the decisions are not grave, and painstaking assessment is not required. But when grave and irreversible decisions are faced, careful evaluation of maturity is needed.

In Belgium, euthanasia law allows, in certain circumstances, minors (with support from their families) to shorten their potential suffering prior to inevitable death from terminal illness. It demands assessment of developmental maturity as part of its framework. It names the requisite ability as the ‘capacity for discernment’. This ‘relates to the ability of the minor to understand the real implications of his euthanasia request and its consequences’ (van Assche Reference van Assche, Raus and Vanderhaegen2018). The pursuit of maturity is central to it.

Establishing a minor's capacity for discernment involves consideration of:

• • the nature, purpose and efficacy of the intervention

• • risks and benefits

• • the sophistication required to understand the information and assess potential consequences

• • the intellectual and emotional characteristics of the minor

• • family and social issues

• • stability of the minor's views.

It is evident that, in grave situations, domestic courts wish to add something to decision-making capacity to attend to the issues of maturity. Some attempt to provide guidance, and some sidestep the debate, preferring to suggest that the capacity test from the MCA is commensurate with the task. It is equally clear that capacity as per the MCA neither aligns with the settled common law position for the assessment of competence in children (Boxes 4 and 6) nor even vaguely approaches the kind of considerations of international courts.

Box 7 outlines a clinical toolkit for assessment of Gillick competence in practice.

The reach of capacity and competence in minors

The potential for Parliament to intervene

Domestic (Gillick [1986] – comments of Lord Scarman in the previous section) and international courts (AC v Manitoba [2009] – ‘rigid statutory distinction [ … ] would fail to reflect the realities of childhood’) have expressed concern about the imposition of fixed statutory age limits on the reach of capacity. However, there are also references to the potential for Parliament to intervene. The judgment in Re W (A Minor) (Medical Treatment) [1993] commented that to confer ‘complete’ autonomy on minors was ‘a matter of social policy with which Parliament can deal [ … ] if it wishes to’. Re X [2021] noted that until Parliament chooses to step in, the minor ‘whether Gillick competent or 16 or over, is not, as a matter of our domestic law autonomous in the same way as an autonomous adult’.

Proposed amendments to the Mental Health Act 1983 regarding minors’ admission to psychiatric units under parental responsibility, although welcomed at the joint parliamentary committee stage, were not enacted in the amended Act (Department of Health 2004; Joint Committee on the Draft Mental Health Bill 2005). These included:

• • see children as a special group

• • protect them with additional safeguards

• • extend those safeguards to young people.

Instead, the choice was made to give guidance in its accompanying Code of Practice (Department of Health 2015), conceptualised as ‘the scope of parental responsibility’. The term replaced the term ‘zone of parental control’ used in the previous Code of Practice from 2008. It has been suggested that the concept of the scope of parental responsibility was an attempt to formalise the rising autonomy of minors (Hawkins Reference Hawkins, Player and Curtice2011) and to provide safeguards demanded by European case law (Akerele Reference Akerele2014). It is, however, fraught with nebulous statements and fails to enunciate the legal subtlety in this area. The revised Mental Health Act, notably, was not silent in relation to such matters in minors: see, for example, guidance in section 131(4) on preventing parental consent to admission for a young person aged 16–17. However, it did not offer a root and branch review of the autonomy and rights of minors.

A sliding scale of capacity or moving the goalposts?

Some academic commentators are unswerving in their pursuit of a single, operationally practicable definition of decision-making capacity in minors. In their opinion, when the requirements are met, minors will be entitled to self-determination. Their avid wish is met in the domestic courts, with a caution about the implications of investing children with unfettered rights to self-determination. Lyons (Reference Lyons2010) noted that there had been seven cases of minors before the courts involving refusal of life-saving treatment. The courts invoked two sorts of legal device to override the refusal (Box 8) – specific incapacity and welfare. What shines through in these cases is the prohibitive reticence of the domestic courts to allow minors to make decisions that would shorten their lives. This is not a ‘never’ but simply a ‘not now’ argument. All note the importance of taking risks, but all feel a moral obligation to step in to prevent ultimate harm. Examples of this come from conclusions of judgments in life-threatening self-determination cases:

• • Lord Donaldson (Re W [1993]): ‘for it is only by making decisions and experiencing the consequences that decision-making skills will be acquired’, a process which involves ‘giving minors as much rope as they can handle without an unacceptable risk that they will hang themselves’;

• • Justice Ward (Re E (A Minor) (Wardship: Medical Treatment [1993]): ‘There is compelling [ … ] force in the submission [ … ] that this court [ … ] should be very slow to allow an infant to martyr himself’.

This reticence in grave situations sits well with the well-established case-law principle that the ‘amount’ of capacity must rise to be ‘commensurate with the gravity of the decision [ … ]. The more serious the decision, the greater the capacity required’, i.e. a sliding scale (Re T (Adult Refusal of Treatment) [1992]). The principle was extended to minors in Re R [1991] and updated in Bell v Tavistock [2020]: ‘where the consequences of the treatment are profound [ … ] it may be that Gillick competence cannot be achieved’.

The derived position at law implies a belief that children should survive at all costs, and that it is not until after the age of majority (age 18) that the protective parental role (including the proxy parental role of the state) can be fully released. This is reminiscent of the concept of curtailing current autonomy to safeguard the potential for future autonomy described in the first article (Hawkins 2023). Perhaps societal conscience is better able to tolerate the occasional tragic case of treatment refusal and death after the age of majority. But the courts, in representing that conscience, are abundantly clear that minors should not be allowed self-determination if it would have irreversible consequences.

In such decisions, beneficent welfarism and paternalistic protectionism prevail. At law, incontrovertibly, acceptance and refusal are treated differently and almost any legal device is used to justify paternalistic interference when refusal has grave consequences.

The tension between autonomy and the right to life

The UK courts exert a prohibitive reticence to allow minors to shorten their lives. Any available mechanism is found at law to justify that position (Box 8). This inherent position limits minors’ autonomy. The UK courts are also clearer in examining decisions from the perspective of the rights of the minor (see part 1 (Hawkins 2023)).

When the domestic courts adopt the derived position, they create tension between individual articles of both the European Convention on Human Rights (ECHR) and the United Nations Convention on the Rights of the Child (UNCRC), notably:

• • the tension in the ECHR between Article 2 (the right to life) and Article 8 (the right to respect for private and family life)

• • the tension in the UNCRC between Article 6 (the recognition that every child has the inherent right to life) and Article 12 (the assurance that the views of a child will be given due weight in accordance with the child's age and maturity).

Such tensions were most clearly examined in Re X [2021]. The tension created between rights was noted, and preservation of life was observed to override concerns about infringement of both Article 8 of the ECHR and Article 12 of the UNCRC.

We are reminded that the rights of the ECHR are ‘formulations [ … ] of aspects of the good life, not the bad, and should be interpreted in a way that enhances a person's life’ (Fortin Reference Fortin2006). Furthermore, to endorse minors’ rights on an equal footing with those of adults would mean ‘dissolution of the child labor, mandatory education, statutory rape laws, and child neglect statutes’ and in doing so it would ‘give children rights for which they are ill prepared’ and ‘leave them more vulnerable than they presently are’ (Ross Reference Ross1997).

The derived legal position – justifiable intrusion or ‘palpable nonsense’?

Harris (Reference Harris2003) determinedly attacked the notion of differential treatment of acceptance and refusal, referring to it as ‘palpable nonsense’. In his view, to be informed sufficiently to decide to accept treatment is to understand the consequences of refusal: ‘consents and refusals are the Janus Faces of autonomous capacity’ in all cases, ‘including that of life and death.’ By overriding a competent child's refusal of treatment, ‘we are behaving as good adults should towards incompetent children’ but ‘are not [ … ] obtaining consent or respecting autonomy [ … ]. We are securing acquiescence – quite another thing’. Such academic argument is important in allowing robust scrutiny of the issue, but there has been caution expressed about the those who live ‘amongst the dreaming spires of the Academy’ rather than the ‘robust and ultimately pragmatic world of the court room’ (Re X [2021]).

In contrast, Gilmore & Herring (Reference Gilmore and Herring2011) offered an elegant and delicately crafted blend of law and ethics, which drew on Coggon's tripartite model of autonomy (see part 1 (Hawkins 2023)). They noted that refusal ‘is not always simply the obverse of consenting to a particular treatment’ (Gilmore Reference Gilmore and Herring2011). They made a clear distinction between two types of refusal: refusal of a specific treatment while remaining open to the idea of an alternative; and refusal of all treatment. Self-evidently, refusal of all treatment holds graver consequences and ‘requires the patient to direct his or her mind in different directions, and thus to answer different questions’. They justified the separation of acceptance and refusal in three ways. First, based on the sliding scale described above. Second, that some decisions demand a ‘rich’ form of autonomy and some do not. Third, that decisions taken by minors have impacts on family relationships, and those impacts are greater depending on the gravity of the decision. Their model demands a richer form of autonomous decision-making capacity for refusal of all treatment in grave situations. This fits with the idea of a test of competence that demands decision-making capacity plus an appraisal of maturity.

A practical exercise in examination of the derived position

The use of two fictitious vignettes allows an exploration of the derived position described in this article (Box 9). Both Amelia and Brody are operating with Coggon's best desire autonomy (as elucidated in part 1 (Hawkins 2023)). But Brody's situation is graver and demands a more sophisticated engagement with the clinical advice. He will have to demonstrate more maturity and more competence to have his autonomous choice respected. This supports the sliding scale of competence described in case law (Re T [1992]).

Both Amelia and Brody must show greater reflection than a person accepting all treatment. But the refusal of all treatment by Brody requires him to move his mind in a different direction again to Amelia. Not just to move his mind, but to engage in different thought processes and be seen to be doing so. Thus, differential treatment of acceptance, refusal of treatment with minimal consequences and refusal of treatment with grave consequences appears logically justified.

Amelia's decision would be likely, in everyday clinical practice, to be fully respected. Brody's decision would be listened to, but it would be likely that medical professionals, with parents alongside, and perhaps the court, would seek to secure acquiescence by any legal device that preserved life. This might be to raise the bar of competence, to use concurrent parental consent, to use the Mental Health Act or use the authority of the court. His voice would be heard, but his autonomy would not be fully respected.

Thus, in everyday clinical practice, autonomy can be fully respected. In grave situations, a minor's voice and opinions might be heard, but the preservation of life would dominate the clinical decision. In such a situation, any legal and available means would be used to limit the exercise of autonomy. This position is compatible with the rights agenda and is, in our opinion, ethically justifiable.