In June 2026, the UK Supreme Court handed down a judgment that changes what legally counts as a deprivation of liberty — and with it, how care homes need to document decisions about residents who lack, or may lack, capacity.
If your care plans still lean on the old "acid test" from Cheshire West, they're already behind. Here's what changed, and what we've built into Bool-care to keep up with it.
What the Judgment Changed
The previous test for deprivation of liberty was largely binary: continuous supervision and control, plus lack of freedom to leave, equalled a deprivation of liberty requiring authorisation — regardless of the person's own wishes.
The 2026 judgment replaces that with a broader, multifactorial assessment. Inspectors and courts must now weigh:
- The type, duration, effects, and manner of any restriction
- The resident's own expressed wishes and feelings — even where capacity is in question
- Whether an objection to the arrangement is present, which is now legally relevant where it previously wasn't
In practice, this means a blanket policy — "bedrails for everyone at risk of falling," "the door is always locked" — is no longer defensible on its own. Each decision needs its own reasoning, tied to the resident in front of you.
What This Means for Care Plans
The Mental Capacity Act 2005 hasn't changed, but this judgment sharpens exactly what CQC and the Court of Protection expect to see documented against its 5 principles:
- Presumption of capacity — every resident is assumed able to decide for themselves unless assessed otherwise.
- Support to decide — a resident must be given all practicable support before being treated as unable to decide.
- The right to make an unwise decision — declining care is not, on its own, evidence of lacking capacity.
- Best interests — anything done on a resident's behalf must be in their best interests.
- Least restrictive option — before acting for someone who lacks capacity, ask whether a less restrictive option achieves the same purpose.
A care plan that simply states "resident declines personal care" is now a gap, not a note. What CQC and the courts want to see is closer to: "The resident has the right to make the unwise decision to decline personal care; capacity for this decision was assessed as present, so no best-interests process was required." Where capacity is lacking, the best-interests reasoning and the less restrictive option actually considered both need to be on the record.
How Bool-care's Care Plans Reflect This
We built this directly into care plan generation rather than leaving it as an extra step:
- AI-generated care plans and risk assessments now document capacity, best-interests, and least-restrictive reasoning as a structured field on every relevant section — not just buried in prose — so it's reportable as real CQC evidence, not a hope that the wording happened to be right.
- A deterministic check flags any section that describes a decline, refusal, or restriction with no capacity reasoning recorded, so it surfaces for review before the plan is published rather than after an inspector asks.
- Existing DoLS & Mental Capacity Records on file for a resident are cited directly during generation instead of a fresh assessment being invented.
- Manually-written care plans get the same structured prompt via an "Insert MCA considerations" template, so the standard doesn't only apply to AI-generated documents.
You can read the full breakdown in our care plans documentation.
How We Keep This Current
Legislation and regulator guidance don't stand still, so we built Legislation Watch — an automated weekly check of gov.uk, CQC, and SCIE guidance on the Mental Capacity Act and DoLS. When it detects a genuine change, an AI assessment summarises what changed and flags it to our team for review.
To be precise about what this does and doesn't do: it's a weekly check, not continuous monitoring, and it never rewrites care plan guidance automatically. Every change goes through a human review before it reaches a real care plan — legislation interpretation affecting resident care isn't something we think should be left to an unsupervised process, however good the model.
It's one more layer making sure the standard we've built in doesn't quietly go stale as the law and CQC's expectations keep moving — alongside the broader changes to CQC's assessment framework we've also written about.
How Bool-care Helps
Getting mental capacity documentation right isn't just about avoiding enforcement action — it's about staff being confident that what they've written actually reflects a defensible decision, made properly, for the person in front of them.
Bool-care builds that reasoning into care plan generation itself, flags the gaps before they become inspection findings, and keeps the underlying guidance current as the law changes. If you're curious how AI-generated care plans handle a decision like this in practice, book a demo and we'll show you.
Summary
| Before | Now |
|---|---|
| "Acid test" — supervision + control = deprivation of liberty | Multifactorial assessment — type, duration, effects, manner, and the resident's own wishes all weighed |
| Resident objection often treated as legally irrelevant | Objection is now legally relevant to the assessment |
| Capacity reasoning often left implicit or undocumented | Structured, per-section capacity/best-interests/least-restrictive field required as real evidence |
| Blanket restriction policies | Decision-specific reasoning for each resident |
| Guidance updates tracked manually, if at all | Weekly automated check, human-reviewed before anything changes |
Sources and further reading: