The multifactorial test in AGNI and under 16s

On 2 June 2026, the Supreme Court handed down its decision in The Attorney General for Northern Ireland’s Reference (Deprivation of Liberty) [2026] UKSC 16 (“AGNI”). This decision significantly changed the meaning of deprivation of liberty in domestic law, replacing the two-stage “acid test” in P v Cheshire West and Chester Council [2014] UKSC 19 (“Cheshire West”), with a new multifactorial test.

Whilst the decision in AGNI expressly applies to adults aged 16 and over, it also has significant implications for under 16s. Re Mustafa (A Child) [2026] EWHC 2159 (Fam), is one of the first reported cases to apply the new multifactorial test to the circumstances of a 15-year-old looked after child, living in a specialist residential school.

Factual and Procedural Background

Mustafa was a 15-year-old boy, who had a diagnosis of Prader-Willi Syndrome, and presented with developmental delay and additional learning needs. Mustafa was subject to a full care order, and lived full-time in a specialist residential school. Mustafa was subject to a number of significant restrictions at school, including:

(a) Locked doors,

(b) 24-hour 1:1 support, increasing to 2:1 support when necessary, and

Pursuant to the old legal test in Cheshire West, the arrangements for Mustafa’s care and support at school had constituted a deprivation of liberty, which deprivation of liberty had been authorised by order of the High Court. On 4 June 2026 (two days after the decision in AGNI was handed down), the local authority applied to extend this authorisation. The local authority subsequently applied to withdraw its application, on the basis that Mustafa was not deprived of his liberty under the new legal test. The Guardian opposed the application to withdraw.

It was agreed as between the parties that Mustafa was not Gillick competent[1]. Nonetheless, Mustafa was able to express his views, was positive about his placement at school, and was generally accepting of the requirement for physical restraint. The local authority submitted that, on this basis, Mustafa was validly consenting to the arrangements for his care and support at school. The Guardian disagreed, raising concerns that Mustafa was restrained relatively frequently (approximately every 9 – 10 days), and that Mustafa’s acceptance of the need for restraint may be interpreted as compliance as opposed to positive consent.

The multifactorial test in AGNI

Article 5 of the European Convention on Human Rights (“ECHR”) provides that no person should be deprived of their liberty, save:

(a) in limited circumstances (including the detention of a child for the purposes of educational supervision, and the detention of persons of unsound mind); and,

(b) where authorised by a legal process.

Article 5 ECHR is given effect in domestic law by the Human Rights Act 1998 (“HRA”). The HRA requires domestic courts to take into account decisions of the European Court of Human Rights when determining any question about a Convention right. The European case law[2] has established that a deprivation of liberty requires three components (often referred to as the Storck limbs):

(a) The objective element of a confinement in a particular restricted space for a not negligible length of time;

(b) The subjective element of a lack of “valid consent” to the confinement in question;

In Cheshire West, the Supreme Court condensed the objective element of the Storck test into an “acid test”, consisting of:

(a) Continuous supervision and control; and,

(b) Lack of freedom to leave.

Further, the Supreme Court accepted in relation to the subjective element of the Storck test that “valid consent” could only be given by a person who had capacity to do so (within the meaning of the Mental Capacity Act 2005). Similarly, subsequent case law confirmed that children could only give “valid consent” if Gillick competent to do so[3].

AGNI overruled Cheshire West, and enacted a fundamentally different interpretation of the Storck limbs. Pursuant to AGNI:

(a) The objective limb of the test requires a “multifactorial” assessment, taking account of “a whole range of factors including the type, duration, effects and manner of implementation of the measure in question”. A person’s lack of objection to the arrangements for his or her care and support is highly likely to be relevant to the objective limb: if a person is accepting of such arrangements, they are less likely to be subjectively experienced as restrictive. Further, the “relative normality” of the placement is a relevant factor.

(b) The subjective limb of “valid consent” is not synonymous with capacity under the Mental Capacity Act 2005. A person may lack capacity to make decisions about their residence and care, but have sufficient understanding of their circumstances to express their view and to provide “valid consent”.

The application of the multifactorial test in Mustafa’s case

The High Court held that Mustafa was not deprived of his liberty by the arrangements for his care and support at school.

The High Court confirmed that, following AGNI, a child who is not Gillick competent may nonetheless have sufficient understanding of their circumstances to provide “valid consent”. The High Court determined that Mustafa was validly consenting, and that the subjective element of the Storck test was therefore not made out.

Responding to concerns that Mustafa may “seek to withdraw any consent he has given, if he does not like the particular restraints placed on him”, the High Court noted that schools have a statutory power to physically restrain pupils in the circumstances set out in section 93 of the Education and Inspections Act 2006 (“EIA”), “regardless of the child’s wishes”. As such, “[w]here the school uses physical restraint, it either has Mustafa’s consent or it is entitled to use restraint under its common law powers. Either way, there is no deprivation of liberty such as to give rise to a violation of Article 5(1).”

Analysis

The decision in Re Mustafa (A Child) provides helpful confirmation that, following AGNI, a non-Gillick competent child may provide valid consent to arrangements which would otherwise constitute a deprivation of liberty.

Beyond this, the decision highlights the importance of ensuring that there is separate consideration of the legally distinct concepts which are at play in deprivation of liberty cases, and which may be at risk of being elided. In particular:

(a) The Strasbourg concept of “valid consent”, as distinct from the domestic law concept of “Gillick competence” to consent; and,

(b) The existence of a lawful basis for physical restraint, as distinct from consideration of whether the arrangements as a whole (including, inter alia, physical restraint) constitute a deprivation of liberty.

Valid consent vs Gillick competence

A child is “Gillick competent” to consent to care or treatment, if he or she is of sufficient maturity and comprehension to make a decision about the care or treatment in question. In general terms, where a Gillick competent child consents to care or treatment, his or her consent provides lawful authority for such care or treatment. Where a child is not Gillick competent, another form of consent or authority is required (whether via parental consent (if available / appropriate), statutory authority, or court order).

As such, Gillick competence is relevant to determining whether a professional has authority to provide a particular act of care or treatment (whether physical restraint or otherwise) to a child.

On the other hand, AGNI has reaffirmed that “valid consent” for the purposes of the subjective element of the Storck test is a different, and legally autonomous, concept to that of capacity / competence. “Valid consent” in this context appears to require practitioners and the courts to zoom out, and to consider whether the arrangements as a whole are being imposed against, or in accordance with, the person’s will. Importantly, the fact that a person is “validly consenting” to the arrangements for the purposes of Article 5 ECHR, does not of itself provide lawful authority (in the form of capacitous / competent consent) for any particular act of care or treatment.

Authority for physical restraint vs deprivation of liberty

In Re Mustafa (A Child), the High Court noted that, regardless of whether Mustafa was Gillick competent to consent, or in fact provided such consent, to physical restraint, his school was empowered to physically restrain him in the circumstances set out in section 93 of the EIA. This section provides that:

(1) A person to whom this section applies may use such force as is reasonable in the circumstances for the purpose of preventing a pupil from doing (or continuing to do) any of the following, namely-

(a) committing any offence,

(b) causing personal injury to, or damage to the property of, any person (including the pupil himself), or

(c) prejudicing the maintenance of good order and discipline at the school or among any pupils receiving education at the school, whether during a teaching session or otherwise.

This statutory authority for physical restraint is comparable to sections 5 – 6 of the Mental Capacity Act 2005 (“MCA”), which set out the circumstances in which a person may provide care and treatment to, and may physically restrain, an incapacitous adult, without incurring liability (for example, for the tort of battery or of false imprisonment).

In Re Mustafa (A Child), the High Court appears to suggest that the existence of a statutory power for schools to physically restrain pupils may make it less likely that a pupil will be considered to be deprived of his liberty by the arrangements for his care and support (which include such physical restraint). However, it is not entirely clear from the judgment whether legal argument was heard on this point, and pending further case law clarification, practitioners may wish to treat this aspect of the judgment with caution.

An alternative way of considering the distinction between the authority for an act of physical restraint, and the existence of a deprivation of liberty, is to think about layers of safeguards for children and vulnerable adults.

The first layer is that, for each particular act of care or treatment which may otherwise constitute a battery or false imprisonment, there must be a lawful authority to carry out that particular act (whether via the consent of a capacitous adult / competent child, a statutory authority, a court order, or via parental consent (if available / appropriate)).

The second layer involves zooming out to consider the arrangements for the person’s care and treatment as a whole, and whether they may constitute a deprivation of liberty. If they do, then Article 5(1) ECHR requires a further, separate, authorisation of the arrangements (“in accordance with a procedure prescribed by law”).

Conclusion

Re Mustafa (A Child) is an early and important example of how AGNI applies to children under the age of 16. It confirms that non‑Gillick competent children may “validly consent” to restrictive arrangements which would otherwise constitute a deprivation of liberty. However, it also highlights the importance of maintaining a separate analysis in respect of the domestic law concept of capacitous consent, and the Strasbourg concept of “valid consent”.

The implications of AGNI across children’s and adults social care are enormous, and will no doubt continue to be developed and refined by case law in the coming months and years.

Holly Littlewood is a barrister at Spire Barristers, specialising in Court of Protection and education law. She can be contacted on 0113 200 2400 or by email at This email address is being protected from spambots. You need JavaScript enabled to view it..

This article is written for general information purposes. It does not constitute legal advice, and should not be relied on as such.

^[1] As defined in Gillick v West Norfolk and Wisbech AHA [1986] AC 112

^[2] Storck v Germany (2005) 43 EHRR 6 ("Storck")

^[3] See e.g. Re QX (Parental Consent for Deprivation of Liberty: Children under 16) [2025] EWHC 745 (Fam)