In Bell & Anor v The Tavistock and Portman NHS Foundation Trust  EWCA Civ 1363, the Court of Appeal upheld the appeal of the Tavistock and Portman NHS Foundation Trust against the declaration made by the Divisional Court as to the relevant information that a child under the age of 16 would have to understand, retain and weigh up in order to have competence to consent to the administration of puberty blocking drugs. As is common practice now, the Court of Appeal has provided a helpful summary of its judgment; the summary below sets out the key points and passages.
By way of context, the judicial review leading to the decision of the Divisional Court been brought by two claimants, one of whom (Keira Bell) was a former patient of the Tavistock who was treated with puberty blockers as a 16-year old, progressed to cross-sex hormones and began surgical intervention as an adult to transition from female to male. She terminated her treatment having changed her mind and regretted having embarked upon the treatment pathway. The second claimant (Mrs A) was the mother of a child who suffered from gender dysphoria and had been referred to Tavistock, but had not yet had an appointment. The purpose of the judicial review had been to require, as a matter of law, the involvement of the court before anyone under the age of 18 was prescribed puberty blockers, thus denying the opportunity of consent to such treatment either individually or with the support of their parents or legal guardians. The argument was that those under 18 were not capable in law of giving valid consent to the treatment. The Divisional Court did not accept this proposition, but, rather than dismissing the judicial review, it (1) made the declaration above; (2) gave extensive guidance as to practice and procedure, in particular as to when the involvement of the court would be appropriate.
The question before the court was whether the Divisional Court was right to do both of these things. As the Court of Appeal noted, the arguments that it had heard about the Divisional Court’s approach to the evidence provided the background to this question.
The Court of Appeal held that the Divisional Court had erred in its approach to the evidence. Having noted that evidence adduced by the claimants appeared to have informed the Divisional Court’s conclusion that the treatment was experimental, and in relation to the conclusion that it was highly unlikely that a child under 14 could give valid consent to puberty blockers and improbable that a child aged 14 or 15 could do so, the Court of Appeal identified that:
38. The claimants made no application for permission to rely upon the expert evidence they produced. Although some expert evidence was served with the claim the majority was served shortly before skeleton arguments were due to be lodged. None of it complied with the rules regarding expert evidence and a good deal of it is argumentative and adversarial. Tavistock sought to exclude the expert evidence on the grounds that it was inadmissible because it was not necessary to resolve the legal issue before the court; and also because it comprehensively failed to comply with the rules regarding expert evidence in any event. The issue was not resolved. Much of it was adduced to contradict the evidence given by Tavistock and the Trusts. Such evidence is rarely admitted but a particular difficulty here was that there was no way of resolving evidential disputes. The court supported the guidance it gave “in the light of the evidence as it has emerged”: see para . It would have been preferable for the status of the claimants’ expert evidence to be resolved. It was controversial and would not, as we have said, ordinarily be preferred over that of a defendant in judicial review proceedings.
At paragraphs 62ff, the Court of Appeal returned to this theme:
62. The correct approach was not in dispute. It was not for the court hearing a judicial review to decide disputed issues of fact or expert evidence (see paras ,  and ). That principle is only subject to exceptions that are not relevant to this case. The question is whether, notwithstanding its acceptance of the principle, the Divisional Court placed reliance on the contested and untested expert evidence of the claimants as Tavistock and the Trusts contend. The claimants submit that the salient facts decided by the court were taken from Tavistock’s own evidence so that they were effectively common ground.
63. This dispute applies most significantly to the two findings to the effect that treatment of gender dysphoria with puberty blockers was “experimental” (see paras , , , and ), and that the vast majority of patients taking puberty blockers go on to cross-sex hormones and are on a pathway to much greater medical interventions (see paras  and ). The Divisional Court recorded at para  that Professor Butler had “explained that it is very common for paediatric medicines to be used off-label and that this factor does not render the treatment in any sense experimental.” It nonetheless concluded at para  that the treatment was experimental in the sense it explained in that paragraph (real uncertainty over the short and long-term consequences of the treatment with very limited evidence as to its efficacy). The argument may, in one sense, be semantic, but, respectfully, we think that it would have been better to avoid controversial factual findings.
64. The same points apply to the finding that the vast majority of patients taking puberty blockers go on to cross-sex hormones and are on a pathway to much greater medical interventions. The evidence filed by Tavistock indicated that more than half of those who embark upon a course of puberty blockers go on to cross-sex hormones. For the Divisional Court to have reached with confidence the conclusion set out at  that the “vast majority of patients taking [puberty blockers] go on to [cross-sex hormones] and therefore that s/he is on a pathway to much greater medical interventions”, it would, we think, have been necessary not only to look at the limited data provided by Dr de Vries and Dr Carmichael, but also to evaluate evidence as to how patients were chosen for puberty blockers, the progression of the treatment, and multiple issues affecting progression between treatment pathways, including the consent processes for subsequent treatment stages. Tavistock and the Trusts argue that the Divisional Court failed to appreciate the difference between a causal connection and an association, whatever the proportion of those who move from one treatment to another. The correlation may be the result of effective selection of those for puberty blockers and information sharing at the consent stage. The point, however, is that these judicial review proceedings did not provide a forum for the resolution of contested issues of fact, causation and clinical judgement.
65. As will appear from what we say in the next section of this judgment, we have concluded that the declaration implied factual findings that the Divisional Court was not equipped to make.
Turning to the question of whether the Divisional Court was right to make the declaration, the Court of Appeal identified that no example of a declaration being granted in judicial review proceeding in which a clear legal challenge had failed had been drawn to their attention. It then noted that:
70. The declaration [made by the Divisional Court] is in terms which not only states the law but also identifies an exhaustive list of the factual circumstances that must be evaluated in seeking consent from a child and specifies some matters as conclusive facts. It comes close to providing a checklist or script that clinicians are required to adopt for the indefinite future in language which is not capable of clear and uniform interpretation and in respect of which there were evidential conflicts. Some of the factors identified in the declaration are simple statements of fact. Others beg questions to which different clinicians would give different answers.
The Court of Appeal was particularly struck by the fact that:
71. […] The declaration would require the clinicians to suspend or at least to temper their clinical judgement and defer to what amounts to the clinical judgement of the court on which key features should inform an assessment of Gillickcompetence, influenced by the views of other clinicians who take a different view and in circumstances where Mr Hyam accepts that the service specification, which sets out criteria for referring a child for puberty blockers, is not unlawful.
72. The ratio decidendi of Gillick was that it was for doctors and not judges to decide on the capacity of a person under 16 to consent to medical treatment. Nothing about the nature or implications of the treatment with puberty blockers allows for a real distinction to be made between the consideration of contraception in Gillick and of puberty blockers in this case bearing in mind that, when Gillick was decided 35 years ago, the issues it raised in respect of contraception for the under 16s were highly controversial in a way that is now hard to imagine. A similar conclusion was reached by Silber J in connection with abortion in R (Axon) v. Secretary of State for Health  QB 539 at para .
The Court of Appeal identified that:
78. The legal issue before the Divisional Court was not a general inquiry into the content of information and understanding needed to secure the informed consent of a child, although we have great sympathy with the Divisional Court given the large volumes of materials which informed that clinical issue. The declaration which the Divisional Court made does not sit happily with the observations of Lord Phillips [in Burke, as to the dangers of a court being used as a “general advice centre,” and also declarations which did not resolve issues between the parties but “appeared intended to lay down propositions of law binding upon the world”]
81. A formal declaration states the law. In so far as it specifies facts as part of the law (itself a difficult concept) they remain the law. There is a great deal of difference between the declaration originally sought in these proceedings (“no prescription of puberty blockers without court approval”) or in Gillick (“no contraceptives without parental consent”) and the declaration made here. It turns expressions of judicial opinion into a statement of law itself. In addition, it states facts as law which are both controversial and capable of change. Both Lords Fraser and Scarman in Gillick expressed views about the matters which a clinician would have to explore with a patient, without being prescriptive and recognising that it was for the clinicians to satisfy themselves, in their own way. No declaration was contemplated to capture the essence of that thinking. It would have been inconsistent with the ratioof the case that clinicians must be trusted to make the decisions for the court effectively to give them a manual about how to do so. It is instructive to consider the language of Lord Scarman on the main issue in Gillickat pages 188H to 189A:
“I would hold that as a matter of law the parental right to determine whether or not their minor child below the age of 16 will have medical treatment terminates if and when the child achieves a sufficient understanding and intelligence to enable him or her to understand fully what is proposed. It will be a question of fact whether a child seeking advice has sufficient understanding of what is involved to give consent valid in law.”
82. His conclusion on the law is found in the first sentence but the second recognises that the question whether valid consent is given in any case is a question of fact. That depends upon the individual circumstances of any child and the surrounding circumstances of the clinical issues. Both he and Lord Fraser identified at a high level what they could expect a clinician to take into account in making a clinical decision. Turning their observations into formal declarations (all the more so if they included immutable facts) would have been inappropriate. It is a matter of clinical judgement, tailored to the patient in question, how to explain matters to ensure that the giving or refusal of consent is properly informed. As Lord Fraser observed at page 174F, medical professionals who do not discharge their responsibilities properly would be liable to disciplinary sanction. The law of informed consent culminating in Montgomery also exposes the vulnerability of clinicians to civil action from someone they have treated who shows that they did so without first obtaining informed consent.
The Court of Appeal was therefore clear that that the Divisional Court was wrong to make the declaration. It was equally clear that it was wrong to have given guidance, although it recognised that it “stemmed from the understandable concern of the Divisional Court for the welfare of children suffering from gender dysphoria who, it is common ground, are deeply distressed and highly vulnerable.” Critically, however, the Court of Appeal found that the Divisional Court was “was not in a position to generalise about the capability of persons of different ages to understand what is necessary for them to be competent to consent to the administration of puberty blockers” (paragraph 85). The Court of Appeal also noted that:
86. […] the effect of the guidance was to require applications to the court in circumstances where the Divisional Court itself had recognised that there was no legal obligation to do so. It placed patients, parents and clinicians in a very difficult position. In practice the guidance would have the effect of denying treatment in many circumstances for want of resources to make such an application coupled with inevitable delay through court involvement. Furthermore, the guidance that there should be an application to the court in circumstances where child, parents and clinicians all consider the treatment to be in the best interests of the child would be inconsistent with the conclusion of the Supreme Court in An NHS Trust (discussed at  above).
89. We conclude that it was inappropriate for the Divisional Court to give the guidance concerning when a court application will be appropriate and to reach general age-related conclusions about the likelihood or probability of different cohorts of children being capable of giving consent. That is not to say that such an application will never be appropriate. There may be circumstances where there are disputes between one or more of clinicians, patients and parents where an application will be necessary, even if they are difficult to envisage under the service specification and SOP with which this case is concerned.
The conclusions of the Court of Appeal merit reproduction in full:
92. We should not finish this judgment without recognising the difficulties and complexities associated with the question of whether children are competent to consent to the prescription of puberty blockers and cross-sex hormones. They raise all the deep issues identified in Gillick, and more.Clinicians will inevitably take great care before recommending treatment to a child and be astute to ensure that the consent obtained from both child and parents is properly informed by the advantages and disadvantages of the proposed course of treatment and in the light of evolving research and understanding of the implications and long-term consequences of such treatment. Great care is needed to ensure that the necessary consents are properly obtained. As Gillick itself made clear, clinicians will be alive to the possibility of regulatory or civil action where, in individual cases, the issue can be tested.
93. The service specification and SOP provide much guidance to the multi-disciplinary teams of clinicians. Those clinicians must satisfy themselves that the child and parents appreciate the short and long-term implications of the treatment upon which the child is embarking. So much is uncontroversial. But it is for the clinicians to exercise their judgement knowing how important it is that consent is properly obtained according to the particular individual circumstances, as envisaged by Gillickitself, and by reference to developing understanding in this difficult and controversial area. The clinicians are subject to professional regulation and oversight. The parties showed us an example of a Care Quality Commission report in January 2021 critical of GIDS, including in relation to aspects of obtaining consent before referral by Tavistock, which illustrate that. The fact that the report concluded that Tavistock had, in certain respects, fallen short of the standard expected in its application of the service specification does not affect the lawfulness of that specification; and it would not entitle a court to take on the task of the clinician in determining whether a child is or is not Gillick competent to be referred on to the Trusts or prescribed puberty blockers by the Trusts.
94. Once it was conceded by the claimants that the Divisional Court had made no findings of illegality, the focus of this appeal was squarely on Gillick and whether, by making the declaration accompanied by guidance requiring (probably frequent) court intervention, the Divisional Court had placed an improper restriction on the Gillick test of competence. In our judgment, whilst driven by the very best of intentions, the Divisional Court imposed such a restriction through the terms of the declaration itself, by the utilisation of age criteria and by the requirement to make applications to the court. As we have said, applications to the court may well be appropriate in specific difficult cases, but it was not appropriate to give guidance as to when such circumstances might arise.
The Court of Appeal were at interesting pains to make clear what they were not considering:
- The situation where a court is asked to approve life-sustaining treatment for under-18s to which they or their parents are unable or unwilling to consent (recently considered by Sir James Munby in Re X (A child)(No 2)  4 WLR 11) (paragraph 82).
- The situation where a child (by which the Court of Appeal must mean a child of 16 or 17) lacks capacity to make the decision to consent to puberty blockers/cross-sex hormones applying the MCA 2005, although the Court of Appeal observed – somewhat cryptically – that “[w]e do not think that a comparison between the exercise of assessing Gillick competence and the process envisaged under the Mental Capacity Act 2005 […] assists in this case.” The Court of Appeal referred to the judgment of Sir James Munby in Re X (A child)(No 2)  4 WLR 11 in relation to this point (the reference to paragraph 72 of that judgment must be a typographical error), in which Sir James observed that he considered that:
the tests of capacity and of Gillick competence have nothing very obvious in common, not least because they are rooted in different areas of scientific knowledge and understanding. Capacity, or, more precisely, lack of capacity, derives from what Butler-Sloss LJ referred to in Re MB as “some impairment or disturbance of mental functioning”, what in section 2(1) of the 2005 Act is referred to as “impairment of, or a disturbance in the functioning of, the mind or brain.” Gillick competence, in contrast, is tied to the normal development over time of the typical child and teenager. In the first, one is therefore in the realm of psychiatry. Indeed, it is notorious that Thorpe J’s analysis in In re C, from which everything since has flowed, was modelled on the analysis provided in the expert evidence of a psychiatrist, Dr Eastman. In the other, one is not in the realm of psychiatry, rather that of child and adolescent psychology.
- Children covered by s.8 Family Law Reform Act 1969, which provides that the consent of a minor over 16 to 16 “to any surgical [or] medical treatment … shall be as effective as it would be if he were of full age.”
As (in a different context) a previous Court of Appeal had been in relation to the case of Leslie Burke, this Court of Appeal was equally robust in identifying that – in essence – the lower court had been lured out of its role in determining a dispute before it into seeking to resolve ethical dilemmas.
The Court of Appeal’s identification that the responsibility for determining whether a young person has the Gillick competence to consent to treatment (of any kind) lies with the clinician proposing that treatment is robust and clear. However, to the extent that the judgment could be read as saying that the courts simply cannot consider the question at all, it is more problematic. It would certainly come as a surprise to judges of the Family Division who have considered for themselves over the years whether they are satisfied as to the Gillick competence of a child to accept or refuse medical treatment (see, for instance, the judgment of Baker J (as he then was) in An NHS Trust & Anor v A & Ors  EWHC 1135 (Fam), in which the judge took into account both clinical evidence and “informal oral evidence” on the part of the young man in question when considering his competence). The concept of Gillick competence has also escaped the gravitational pull of medical decision-making, and has been considered by judges, for instance, in relation to the ability of a young person to consent to the adoption of her child, and, again, it would likely come as a surprise to such judges to learn that they have no ability to resolve a dispute as to whether valid consent has been given.
In the circumstances, therefore, it is suggested that the proper approach (in line with the position in Re Y, expressly referred to by the Court of Appeal) is that where there is consensus, there is no need to approach the courts, but, there is dispute or disagreement, or the issue of competence, is finely balanced that (1) the court can be approached to reach a resolution of the question; and (2) the courts can determine – with the benefit of appropriate evidence – what is (as the House of Lords identified in Gillick) ultimately a question of fact.
It should be noted in any event that even if the Court of Appeal did, in fact, intend to say that the court can never take on the role of determining a dispute as to whether a child is or is not Gillick competent in respect of the decision in question, this cannot apply to the question of whether a person of 16 or over has or lacks the mental capacity to make a decision. Section 15(1)(a) MCA 2005 expressly empowers the Court of Protection to make exactly such a decision.
Finally, the Court of Appeal in this case was clearly troubled by the fact that the Divisional Court had embarked upon an abstract exercise in relation to the question of identification of what information a young person should be able to process to be able to consent to the administration of puberty blockers. It is in this regard striking that the appellate courts are entirely comfortable with the idea of setting down the types of information that a person should be able to understand, retain, use and weigh to make a decision for purposes of the MCA 2005. Perhaps the difference is that they do not seek to reduce these to declarations; or perhaps the difference is that they do not enter into medical debates. But it is a difference which will no doubt fall for further consideration in due course.
 There was, of course, no such debate in this case in relation to whether a specific child patient was actually Gillick competent to give consent (and in AB v CD, decided subsequently, and endorsed by the Court of Appeal in this case (see paragraph 48) the court did not have to confront the issue head-on: see paragraph 51).