Two Vocabularies, One Practice: Restrictive practice across older and younger adult care — and what the Supreme Court has just changed

Two people live eleven miles apart. Neither can leave the building without a member of staff. Both have a door they cannot open. Both are given medication when they become distressed. Both have possessions kept in an office.

One is twenty-six, autistic, and lives in a supported living service. Every restriction in his life is written down, dated, authorised, reviewed and counted. There is a behavioural formulation. There is a reduction plan. His provider reports on it.

The other is eighty-four, has vascular dementia, and lives in a nursing home. Almost none of her restrictions are recorded as restrictions at all. They are recorded as care.

This is the largest and least examined inequality in the way we regulate restrictive practice in the United Kingdom, and it has just become considerably more urgent.

The vocabulary problem

The sector has two entirely separate languages for the same set of actions, and which one gets used depends almost entirely on how old the person is.

The practice	In younger adult services it is called	In older adult services it is called
A locked external door	Environmental restraint. Logged, reviewed, justified per person.	Keeping her safe. A feature of the building.
Medication given for distress	Chemical restraint. Triggers a behavioural review.	Settling medication. A PRN entry.
Continuous staff presence	Enhanced observation. Recorded as restrictive.	Close supervision. Good care.
No access to the kitchen	A blanket restriction. Challenged on principle.	Health and safety.
Cannot go outside unaccompanied	A deprivation requiring justification and review.	Falls prevention.
Personal items held by staff	A restriction on possessions. Individually authorised.	Looking after her things for her.

This is not a matter of semantics. Language determines what gets counted, and counting determines what gets reduced. A restriction described as environmental restraint enters a register, acquires a review date and generates a reduction plan. The same restriction described as keeping her safe enters nothing at all. It is simply how the home works.

Two systems that grew up separately

There are reasons for the divergence, and they are structural rather than a matter of anyone’s bad faith.

Younger adult services were reformed by scandal. Winterbourne View, and everything that followed it, built an entire apparatus around restriction in learning disability, autism and complex mental health services: positive behaviour support, restraint reduction training standards, Building the Right Support in England, Coming Home and dynamic support registers in Scotland, independent care and treatment reviews, and a strong expectation that every restriction is individually formulated and actively reduced. The system is imperfect and its results are uneven, but it exists, and it treats restriction as a clinical and ethical problem to be solved.

Older adult services were reformed by paperwork. Restriction in care homes was addressed almost entirely through a legal authorisation route — Deprivation of Liberty Safeguards in England and Wales, guardianship and intervention orders under the Adults with Incapacity (Scotland) Act 2000 north of the border. The question asked was not “how do we reduce this?” but “is this lawfully authorised?” Those are different questions, and only one of them changes anyone’s day.

The consequence is that older adult care never developed the internal machinery that younger adult care was forced to build. It did not need to. The authorisation appeared to be doing the work.

An authorisation records that a restriction exists. It has never, on its own, made a restriction smaller.

What changed on 2 June 2026

On 2 June 2026, a seven-justice panel of the Supreme Court unanimously overruled its own 2014 decision in P v Cheshire West and Chester Council. The judgment came in a reference by the Attorney General for Northern Ireland, but its effect reaches every part of the United Kingdom, because it concerns the meaning of Article 5 of the European Convention on Human Rights rather than any single domestic statute.

In short: the acid test is gone. For twelve years, the question of whether someone was deprived of their liberty had a deceptively simple answer — continuous supervision and control, not free to leave, lacking capacity to consent. The Court has held that this bright-line approach was too crude, over-extended the concept of deprivation of liberty, and was never in fact the position of the Strasbourg court. In its place is a return to a multifactorial assessment considering the type, duration, effects and manner of implementation of the restrictions in a person’s particular circumstances.

The second limb is more consequential still. The Court held that valid consent under Article 5 is an autonomous Convention concept, not a matter of domestic capacity law. A person who lacks capacity under the Mental Capacity Act may nonetheless be capable of accepting their care arrangements in a way that is legally meaningful. Compliance, apparent contentment, and the relative normality of a placement are all now live factors.

There is no grace period. The Cheshire West approach must not be applied from the date of judgment, and the very large body of guidance that refers to it has to be read in that light. The Department of Health and Social Care published initial guidance on 15 June 2026 with further material to follow, and providers should expect the ground to keep moving for some time.

It is worth noting for Scottish readers that the Mental Welfare Commission for Scotland intervened in the case, in broad support of a wider concept of valid consent, and that the Court was influenced by its observations on the weight properly given to the views of people with impaired decision-making capacity. This was not a judgment handed down over Scotland’s head.

Why this lands hardest on older people

The deprivation of liberty architecture is, in practice, an older people’s architecture. Roughly nine in ten deprivation of liberty applications in England concern people over sixty-five, and around four in ten concern people over eighty-five. Whatever this judgment does, it does mostly to older people in care homes.

The concerns raised since June are serious and deserve to be stated fairly. Fewer people falling within Article 5 means fewer people with an independent representative, fewer automatic routes of challenge, and less access to advocacy and legal aid. The British Geriatrics Society has warned that older people with complex needs may lose protections precisely when they are least able to assert them. Inclusion Scotland has put the objection at its sharpest: absence of objection is not consent, and a system that treats compliance as agreement assumes that care settings are safe by default.

The Court’s own reasoning runs the other way, and should also be stated fairly. It held that Article 5 was never meant to carry the entire safeguarding burden, and that Articles 2, 3 and 8, the Mental Capacity Act and its Code, mandatory annual review of restrictions, and local authority safeguarding duties remain fully in place. On that reading, the judgment removes a procedural over-reach rather than a protection.

Both positions can be held at once. What is not in dispute is the practical consequence for providers: an external check that many services had come to rely upon has narrowed, and it has narrowed for the age group whose restrictions were least likely to be examined internally in the first place.

The point most commentary has missed

The safeguard was never the form.

If a legal authorisation was the only thing standing between an older person and an unexamined restriction, then that restriction was never actually being examined. It was being processed. A provider that maintained a live account of every restriction in its service, reviewed each one on a named date, and required someone to argue for keeping it rather than for lifting it, is entirely unaffected by what the Supreme Court did in June. A provider whose entire assurance rested on a folder of authorisations now has very little.

That is the honest test of the last few weeks, and it is a governance question rather than a legal one.

What older adult care should borrow

The answer is not to invent something. It is to take the infrastructure that younger adult services were compelled to build after Winterbourne View and apply it to a population that has never had it.

—  A restriction register for every service: every restriction in operation, who it applies to by name, who authorised it, what risk it addresses, what less restrictive alternative was tried, and when it will next be reviewed.

—  A review date on every restriction — a date, not “ongoing” — with the burden of proof placed on retention rather than removal.

—  Restraint, PRN and observation data that can be disaggregated by age, ethnicity, sex, diagnosis and communication need. A provider who cannot disaggregate cannot claim proportionality; it can only say it has not looked.

—  Formulation rather than management: understanding what distress is communicating before deciding what to do about it. This is standard in learning disability services and remains rare in dementia care.

—  Debriefing after every episode, including with the person themselves wherever that is possible, which in dementia care is far more often than teams assume.

None of this is novel. All of it is ordinary practice in a well-run supported living service for a twenty-six-year-old. Almost none of it is ordinary practice in a nursing home for an eighty-four-year-old.

FIVE QUESTIONS FOR ANY BOARD OR REGISTERED MANAGER

—  How many restrictions are currently operating in each of our services, and can we produce that list today?

—  Which of them apply to everybody, and who decided that?

—  Can we break our restraint and PRN data down by age, ethnicity, sex and diagnosis?

—  If an authorisation lapsed tomorrow, what in our own governance would notice?

—  What have we stopped doing in the last twelve months?

Age is a social location

There is a deeper reason for the asymmetry, and it is uncomfortable.

Younger adult services ask what a good life would look like for this person, and then treat every restriction as an obstacle to it. Older adult services ask whether she is safe. The second question is not a smaller version of the first. It is a different question, and it produces a different service.

Behind it sits an assumption that older people, and particularly older people with dementia, are no longer building a life — that there is a future being protected in one case and only a present being managed in the other. That assumption is rarely stated, would be rejected by almost everyone who acts on it, and shapes an enormous amount of practice.

It compounds, too. An older woman who is Black, or who does not speak English as a first language, or who is a lesbian, or who has a learning disability alongside her dementia, is not experiencing several risk factors in sequence. She occupies a position that the service has almost certainly never designed for, and the restrictions placed on her are correspondingly less likely to be questioned by anyone.

Would I accept this in my own home? And would my answer change if the person were forty years younger?

Where this leaves providers

The regulatory direction of travel has not altered. The Care Inspectorate’s position on restrictive practice applies across every registered service and every age group, and its updated self-evaluation tool, published in May 2026 with the Scottish Physical Restraint Action Group, is explicitly not age-bounded. The Care Quality Commission continues to treat restrictive practice as a failure of person-centred care planning and a marker of closed culture. The Mental Health Act 2025 and the forthcoming Liberty Protection Safeguards consultation both point towards less coercive practice, not more.

What has altered is how much of the thinking a provider can outsource. For the last twelve years, a great deal of the reasoning about restriction in older people’s services was done elsewhere, by someone completing an assessment. Less of it will be from now on.

Services that had already built the habit will find this a quiet summer. Services that had not should treat the judgment as the prompt they were unlikely to get any other way.

Sources and further reading

A Reference by the Attorney General for Northern Ireland [2026] UKSC 16, judgment of 2 June 2026, overruling P v Cheshire West and Chester Council [2014] UKSC 19.

Department of Health and Social Care, initial guidance on the 2026 Supreme Court judgment on deprivation of liberty, published 15 June 2026; further guidance awaited.

British Geriatrics Society, commentary on the 2026 Supreme Court decision to redefine deprivation of liberty, June 2026.

Inclusion Scotland, “Rolling back on rights is not a waiting list solution: absence of objection is not consent”, June 2026.

Care Inspectorate, policy position on the use of restrictive practices, and updated restrictive practices self-evaluation tool, May 2026.

Care Quality Commission, cross-sector policy position on reducing restrictive practice, August 2023.

Mental Welfare Commission for Scotland, Rights, Risks and Limits to Freedom, 2021.

This paper is intended as sector commentary and professional comment. It is not legal advice, and the law in this area is developing rapidly. Providers should take their own advice on individual cases.