A Graham Digital Special Report
Care-Planning Problems in Care-Home Inspection Reports
National census analysis of England’s registered care homes
Prepared 10 August 2026 · Full population pulled (13,713 inspected homes) · 3,570 reports classified, region-weighted
Executive summary
Just under a third (31.5%) of care home inspection reports that assess care planning flag it as a problem. The issue is dominated by plans that are too thin and out of date. The most common finding is insufficient detail, especially on risk, present in 62% of flagged reports.
Flagging rates have risen sharply over time, from under 15% in 2018 to around 49% in 2025. Nursing homes flag problems more often (40%) than residential homes (29%), and the flag rate climbs steeply as overall ratings worsen, including 79% among homes rated Requires Improvement.
Method
The population is based on the CQC directory from 5 August 2026: 13,713 inspected homes. Of those, 8,597 reports assessed care planning, after excluding 3,361 COVID-era infection-control inspections. Graham Digital classified 3,570 reports against a seven-type taxonomy and post-stratified the sample by region to produce a national estimate.
Reports were coded with programmatic integrity checks and independent full-text re-review of borderline cases, confirming high accuracy. “Any negative finding” is a deliberately broad threshold.
1. How many reports flag care planning?
Graham Digital’s national estimate is 31.5%: just under a third of reports that assess care planning contain at least one negative finding on it.
31.5%
National flag rate
3,570
Reports classified
13,713
Inspected homes in frame
2. What kinds of problem?
Each flagged report often raises more than one issue. The chart below shows both how often a problem type appears in a flagged report and how often it is the main, primary problem.
Insufficient detail (esp. risk)
Not up to date / not reviewed
Missing plan for a known need
Inaccurate / contradictory
Lack of personalisation
Person or family not involved
Records disorganised / inaccessible
Weighted to the national population. ‘Appears’ columns can overlap; ‘main problem’ sums to 100%.
| Care planning problem type | Appears in flagged report | Is the main problem |
|---|---|---|
| Insufficient detail (esp. risk) | 62% | 39% |
| Not up to date / not reviewed | 37% | 18% |
| Missing plan for a known need | 37% | 15% |
| Inaccurate / contradictory | 28% | 12% |
| Lack of personalisation | 20% | 9% |
| Person or family not involved | 11% | 3% |
| Records disorganised / inaccessible | 10% | 4% |
3. Change over time
There has been a steep rise in flagged care planning, roughly tripling between 2018 and 2024–25. This likely reflects the move to a Single Assessment Framework and closer scrutiny of records, rather than a simple decline in actual care quality. The time trend should be read as directional.
By year of each home’s latest inspection report.
4. By region
Flag rates range from 24% in London to 40% in the East of England, around Graham Digital’s 31.5% national estimate.
East
Yorkshire & Humberside
North West
West Midlands
South West
North East
East Midlands
South East
London
National 31.5%
Dashed line marks the national 31.5% average.
5. Home type and validation
Nursing homes flag care planning more often than residential homes, consistent with their higher-acuity residents. The flag rate also climbs steeply with worse overall ratings, which is the expected pattern if the coding captures real signal.
By overall rating
Outstanding
Good
Requires improvement
Nursing vs residential
Nursing
Residential
Limitations
The estimate rests on 3,570 classified reports weighted to the national population (effective sample ~3,000, margin about ±1.7 points). “Latest report” from the API is typically a home’s most recent full inspection, so newer Single Assessment Framework assessments are somewhat under-represented and the time trend should be read as directional. Classification uses a defined rubric applied by a language model with automated integrity checks and a pilot verification pass; “any negative finding” is a deliberately broad threshold. Single-year figures have smaller cell sizes and wider uncertainty than the national and regional numbers.
Appendix. Taxonomy
- Lack of personalisation
- Missing plan for a known need
- Not up to date / not reviewed
- Inaccurate / contradictory
- Insufficient detail (esp. risk)
- Person or family not involved
- Records disorganised / inaccessible
Accompanying data: the classified dataset of all 3,570 reports (rating, region, home type, year, flag, all problem types, primary type, evidence quote, and weight), plus the full frame of 13,713 inspected homes.
Data source and attribution
Source. All figures in this report are derived from Care Quality Commission (CQC) care-home inspection reports and ratings. Registration and directory data are taken from the CQC care directory (published 5 August 2026); inspection report text and current ratings were retrieved via the CQC Syndication API (api.service.cqc.org.uk/public/v1).
Licence. Contains public sector information provided by the Care Quality Commission and licensed under the Open Government Licence v3.0. CQC data is © Care Quality Commission.
Disclaimer. This is an independent analysis. It is not produced, reviewed, or endorsed by the Care Quality Commission, and any errors of interpretation are the authors’. Classifications of whether a report “flags care planning” are analytical judgements applied to CQC report text, not CQC assessments.
References. Care Quality Commission, Using CQC data: https://www.cqc.org.uk/about-us/transparency/using-cqc-data. CQC Syndication API developer portal: https://api-portal.service.cqc.org.uk. Individual inspection reports are available at https://www.cqc.org.uk/location/<location-id>. Accessed August 2026.
What this means for care homes
Incomplete, out-of-date, and poorly personalised care plans remain one of the most common inspection findings. Graham Digital builds AI care-planning systems that sit inside existing workflows, check for gaps and contradictions, and keep records inspection-ready, with clinicians in control at every step.