How to write a CQC-defensible care plan (2026)

What “CQC-defensible” means in 2026

The Care Quality Commission now inspects under the Single Assessment Framework (SAF), which replaced the old Key Lines of Enquiry from late 2023. The five key questions are the same as they have always been — is the service Safe, Effective, Caring, Responsive and Well-led? — but they’re now assessed through a set of quality statements, written as “we statements,” each backed by defined evidence categories. One important caveat for 2026: the framework is still being reformed, with the final version expected around summer 2026. Treat what follows as current and verify the detail against CQC’s latest published guidance.

Two shifts matter most for care plans. First, assessment is now continuous: CQC gathers evidence across the year rather than turning up once for a set-piece inspection, so your plans have to hold up all the time, not just on an announced day. Second, inspectors triangulate. The plan is never read in isolation — it’s checked against direct observation of care, the person’s own lived experience, what families say, and how staff describe their practice. When the plan and the reality disagree, the reality wins.

Underneath all of this sits a legal duty. Person-centred care is required by Regulation 9 of the Health and Social Care Act 2008 (Regulated Activities) Regulations 2014. It underpins all five key questions and shows up most directly under Responsive (care organised around the individual, people listened to and involved) and Caring (people treated as individuals, with dignity). A care plan is the main written evidence that this duty is being met.

The plan written for the inspector vs the plan written for the person

Here is the thing most guidance misses, and the thing working carers feel every shift. Open a lot of care plans and you find the same flat boilerplate: “I like to be washed in the morning. I should be encouraged to eat. I must take my medication. I like a cup of tea.” It reads as thorough. It is, in fact, the problem.

A plan written like that — a single stated routine, in rigid “I must” language — quietly makes carers tense. It reads as the one correct way to do things, so a carer becomes afraid to deviate, worried that trying something different would be “wrong” and might be held against them. The result is rote, anxious, defensive care: the carer follows the letter of the plan and stops actually noticing the person in front of them. That is the opposite of person-centred, even though every box is ticked.

And people are different. One person genuinely welcomes a carer who suggests sitting in the garden because it’s a nice day; another finds any change to the routine unsettling and wants things kept exactly as agreed. Neither is wrong — but a carer walking in for the first time has no way of knowing which person this is unless the plan tells them. A genuinely good plan says it plainly: this is how much latitude this person wants. That one piece of information is what lets a carer relax and respond, instead of freezing on the script.

Here is what that difference looks like on paper. Both plans describe the same woman and the same morning.

Nothing in the second plan is extra work. It is the same assessment, written down as if the carer reading it matters.

This is also why the “written for the inspector” plan ironically fails inspection. Because CQC triangulates, a plan that’s been polished to look person-centred but doesn’t match observed practice or the daily notes stands out as exactly the gap inspectors are trained to find — the well-documented gap between a provider’s intention to be person-centred and the generic reality of its plans and practice. The plan written for the person — specific, honest, in their words, clear about latitude — is the one that survives triangulation, because it describes the care that’s actually happening.

What a good, defensible plan includes

A practical checklist — written from the point of view of a carer reading the plan before a shift. Each item notes the CQC key question it most helps you evidence.

• 1

  The person in their own words

  Capture how the person describes themselves, their day and what they want — quoted, not paraphrased into clinical language. “I don't really wake up until I've had my first cup of tea” tells a carer more than “requires morning beverage.” Their voice is the single clearest sign a plan is about them and not a template.

  EvidencesCaringResponsive

• 2

  What matters to them — not just the tasks

  Tasks (personal care, medication, meals) are the easy half. The defensible half is what the person is trying to live: seeing the grandchildren on Sundays, getting to the allotment, never missing the racing on the telly. Write the outcomes the tasks are in service of, so care has a point the person recognises.

  EvidencesResponsive

• 3

  Preferences AND the latitude they want

  Record not only how the person likes things done, but how much room they want a carer to use judgement. Some people welcome a carer suggesting a walk because it's sunny; others find any change unsettling and want the routine kept exactly. Say which this person is. This single line is what lets a carer relax into responsive care instead of freezing on the script.

  EvidencesResponsiveCaring

• 4

  Strengths — what the person can do

  A plan that lists only deficits invites carers to do everything for the person, which erodes independence and dignity. Name what they can and want to do for themselves, so support is enabling rather than taking over.

  EvidencesEffectiveCaring

• 5

  Risk written as enabling-with-safeguards

  Positive risk-taking is person-centred care. Frame risk as “how we help them keep doing this safely,” not a blanket ban. A walk to the corner shop with an agreed check-in beats “do not allow to leave the property.” Show the thinking, the safeguards, and the person's own view of the risk.

  EvidencesSafeResponsive

• 6

  Where the boundaries ARE fixed — and why

  Carers need to know exactly where judgement is welcome and where it is not. Make the few hard lines explicit and give the clinical or safety reason: a thickened-fluids instruction for a swallowing risk, a specific moving-and-handling method, a medication that must not be skipped. Naming the fixed boundaries is what makes the latitude elsewhere safe to use.

  EvidencesSafeEffective

• 7

  How the person and their family were involved

  Evidence that the plan was made with the person, not for them — who was consulted, what they said, where their wishes shaped the plan, and how consent and capacity were considered. This is also what inspectors look for when they ask whether people are genuinely listened to.

  EvidencesResponsiveWell-led

• 8

  Review dates and triggers

  State when the plan is reviewed and what events trigger an earlier review — a fall, a hospital stay, a change in how the person is coping. A plan with a clear review rhythm shows the record is alive, not filed and forgotten.

  EvidencesWell-ledEffective

• 9

  Clear, accessible language

  Write so the person, a new carer on their first shift, and a family member could all read it and recognise the same person. Plain language, no unexplained jargon, no copy-paste phrasing that could describe anyone.

  EvidencesCaringResponsive

Common mistakes that fail inspection

• Generic, copy-pasted content

  The same phrasing across multiple people's plans is the fastest way to signal the care isn't actually individualised. It's the gap between intention and implementation that CQC reports flag most often.

• The plan contradicts the daily notes

  If the plan says one thing and the visit notes show another, inspectors believe the notes. A polished plan that doesn't match what's recorded day to day is a red flag, not a strength.

• Out of date and unreviewed

  A plan describing needs the person no longer has — or missing ones they've developed — fails under continuous assessment, because the mismatch is visible all year, not just on an inspection day.

• Restrictive language with no rationale

  “Do not allow,” “must not,” and blanket prohibitions with no clinical reason read as risk-averse, non-person-centred care — and as possible unlawful restriction. If a boundary is real, give its reason.

• No trace of the person's voice

  A plan written entirely in the third person and the passive voice, with nothing the person actually said, is hard to defend as person-centred however thorough it is.

• Tasks without outcomes

  A checklist of duties with no sense of what they're for tells a carer what to do but not what good care looks like for this person — and gives an inspector nothing to triangulate against lived experience.

Keeping it defensible, day to day

Because assessment is continuous, defensibility isn’t a document you finish — it’s a state you maintain. Three habits keep a plan strong between reviews. Keep the plan matched to practice: when how you care for someone changes, change the plan, so the two never drift apart. Make sure the daily notes corroborate the plan: the visit record should read like the plan being lived, including the moments a carer used the latitude the plan allowed. And make sure staff can speak to it: a carer who can explain, in their own words, what matters to the person and where their judgement is welcome is the strongest evidence of all — far stronger than a tidy folder.

A note on scope

This guide is written for domiciliary and home care — care delivered in people’s own homes. Residential and nursing care are CQC-regulated too, and the same person-centred principles and key questions apply, but the evidence you’ll gather and the operational realities differ (24-hour environments, shared spaces, different observation opportunities). Take the principles here as transferable and the specifics as home-care-shaped.