CQC inspections

Report, unpacked: how one GP practice reached Outstanding

CQC AI Auditor team · · 13 min read · Download PDF

In short: We read CQC's assessment report for Holmside Medical Group, a GP practice in Newcastle upon Tyne rated Outstanding in April 2026, statement by statement. The practice scored 4 on 13 of its 33 quality statements and 3 on the other 20. There was not a single 1 or 2 anywhere. Two key questions, Effective and Well-led, reached the outstanding range, which is the minimum CQC normally needs for an overall Outstanding. Most of the 4s came with numbers: a baseline, a change and a re-measure. The rest reflected exceptional leadership, workforce development and governance, backed by what staff told inspectors and what the records showed.

Report, unpacked is our monthly look at one real CQC report for a GP practice, read in full. This month it is the Holmside Medical Group report. We chose it because it is recent, it covers a mainstream GP practice rather than a specialist service, and CQC said in its press release that "other practices should look to this report to see what they can learn." Holmside Medical Group has no connection with CQC AI Auditor. Everything quoted below is CQC's own wording, from pages read on 20 October 2026. Where we add our own arithmetic or practical suggestions, we say so.

The practice at a glance

Holmside Medical Group serves around 10,000 patients from its main surgery in Newcastle upon Tyne and a branch surgery at Chapel House. CQC's report notes that deprivation in its population is in the most deprived decile nationally, that the population is transient, and that some patients' first language is not English. Its previous inspection was in 2016, when it was rated Good overall and for all five key questions.

The assessment ran from 20 November 2025 to 21 January 2026. CQC gathered information remotely, carried out remote clinical searches of patient records on 20 January, and visited on 21 January. The report was published on 16 April 2026.

The result: Effective and Well-led moved from good to outstanding, Safe, Caring and Responsive stayed good, and the overall rating moved from Good to Outstanding.

Every quality statement score

Under the single assessment framework, CQC scores each quality statement from 1 to 4. In CQC's words, a 3 means "Evidence shows a good standard" and a 4 means "Evidence shows an exceptional standard". The key question score is the total divided by the maximum possible, as a percentage. This is how the practice scored.

  • Safe: Good, 78 out of 100. Safe and effective staffing scored 4. Learning culture, Safe systems, pathways and transitions, Safeguarding, Involving people to manage risks, Safe environments, Infection prevention and control, and Medicines optimisation each scored 3. That is 25 out of a possible 32.
  • Effective: Outstanding, 88 out of 100. Assessing needs, Supporting people to live healthier lives, and Monitoring and improving outcomes scored 4. Delivering evidence-based care and treatment, How staff, teams and services work together, and Consent to care and treatment scored 3. That is 21 out of a possible 24.
  • Caring: Good, 85 out of 100. Treating people as individuals and Workforce wellbeing and enablement scored 4. Kindness, compassion and dignity, Independence, choice and control, and Responding to people's immediate needs scored 3. That is 17 out of a possible 20.
  • Responsive: Good, 82 out of 100. Person-centred care and Care provision, integration and continuity scored 4. Providing information, Listening to and involving people, Equity in access, Equity in experiences and outcomes, and Planning for the future scored 3. That is 23 out of a possible 28.
  • Well-led: Outstanding, 93 out of 100. Shared direction and culture, Capable, compassionate and inclusive leaders, Governance, management and sustainability, Partnerships and communities, and Learning, improvement and innovation scored 4. Freedom to speak up and Workforce equality, diversity and inclusion scored 3. That is 26 out of a possible 28.
Safe78 out of 100
Effective88 out of 100
Caring85 out of 100
Responsive82 out of 100
Well-led93 out of 100

CQC's thresholds at key question level are "63 to 87% = good" and "88% and above = outstanding". If you want the full method, including the two limiting rules, our post on how CQC scores a GP practice walks through it, and our guide to the 34 CQC quality statements lists every statement in CQC's wording.

How two outstanding key questions became an Outstanding practice

The overall rating is not an average of the five percentages. CQC aggregates the key question ratings using its rating principles, published on its How we reach a rating page: "At least 2 of the 5 key questions would normally need to be rated as outstanding and 3 key questions rated as good before an aggregated rating of outstanding can be awarded."

This practice met that exactly: two outstanding, three good, nothing lower. That has a practical consequence. A practice aiming for Outstanding does not need to be exceptional everywhere. It needs to be at least good in every key question, and clearly exceptional in two.

Five things this report teaches

1. Outstanding was built on 3s, with no weak spots

Twenty of the 33 scores were 3. Safe, the key question many practices worry about most, scored 3 on seven of its eight statements and was rated good. What mattered just as much was what the report did not contain: no 1s and no 2s.

That matters because of CQC's limiting rule: "If the key question score is within the outstanding range, but one or more of the quality statement scores is 1 or 2, the rating is limited to good." A single 2 in Effective or Well-led would have held that key question at good, however strong the rest. A solid, consistent good standard across every statement is the foundation that the 4s sit on.

2. Effective was outstanding by the narrowest margin

Effective scored 21 out of 24, which CQC published as 88 out of 100, the first point of the outstanding range. On our arithmetic, one of its three 4s scored as a 3 would have made it 20 out of 24, about 83%, and good. The same works the other way. Caring scored 17 out of 20. One more 4 would have made it 18 out of 20, 90%, in the outstanding range.

The lesson for any practice is that the difference between good and outstanding at key question level can be a single quality statement. It is worth knowing which statements in your own practice are closest to exceptional, and what evidence would show it.

3. Most of the 4s came with numbers, the rest with leadership

In Effective, Caring and Responsive, most of the scores of 4 were backed by figures: a starting point, what the practice changed and what happened next. A few examples from the report:

  • Long-term condition reviews. The practice used the year of care approach and could show reviews rising from 1,656 in 2023 to 1,839 in 2024 and 2,145 in 2025, which CQC describes as an increase of "almost 30%". It also introduced a housebound review service in 2025, designed by its lead nurse, with 143 housebound reviews that year. Assessing needs scored 4.
  • Finding people who would otherwise be missed. A link worker phoned patients who would not usually come in for a health check. In a year, 368 checks were carried out; 70 patients were identified with high blood pressure or cholesterol, and 55 were referred for hypertension management. Supporting people to live healthier lives scored 4.
  • Two-cycle audits that changed outcomes. Pneumococcal vaccine uptake among patients with learning disabilities rose "from 4.9% to over 55%" after targeted work. Annual blood glucose testing for patients with a history of gestational diabetes rose from 36% to 68% on re-audit. A missed PSA test, logged as a significant event, led to an audit and a change in policy. Monitoring and improving outcomes scored 4.
  • Learning disability health checks. 89% of patients on the learning disability register had their annual health check in 2024/25. CQC notes that the Quality and Outcomes Framework would expect between 49 and 80%. Treating people as individuals scored 4.
  • Mental health on the same day. Mental health practitioners were embedded in the clinical team so patients could be seen the same day after triage, which the report says "freed up 766 extra GP appointments in a year". Person-centred care scored 4.

Not every 4 rested on figures. Safe and effective staffing, Workforce wellbeing and enablement, and most of the Well-led 4s were awarded for exceptional leadership, culture, workforce development and governance. For Workforce wellbeing, CQC wrote: "We received consistent and overwhelmingly positive feedback from staff about working at the practice." For Governance, management and sustainability: "Governance was well-embedded into the running of the service." In each case the report backs the judgement with what staff told inspectors and what inspectors saw in audits and records.

Compare that with the 3s. The report's narrative for a 3 is often a short confirmation that systems are in place and working: policies known to staff, training up to date, audits completed. The 4s go further, either with measured improvement or with leadership and culture that staff and records show to be exceptional.

4. The practice's context was part of the case

This was not a practice with easy numbers. Its National GP Patient Survey result for overall experience was 70.4%, against a national average of 75.4%. Screening and immunisation targets were, in the practice's own words to CQC, challenging, because of deprivation, a transient population and language barriers.

What the report credits is the response to that context. The practice made targeted calls to patients, some in specific languages with interpreters, to improve cervical screening uptake, and sent appointment reminders in specific languages. CQC records the practice's own figures for cervical screening and child immunisation as "unverified", and still describes the improvement work. The leaders' understanding of their population runs through Shared direction and culture, which scored 4: "an exceptional understanding of the challenges and the needs of people and their communities."

For a practice in a deprived area, the takeaway is that context is not an excuse CQC accepts for weak care, but it is the frame in which your improvement is judged. Showing that you know your population, and what you did about it, is evidence.

5. In Well-led, the 4s reached beyond the practice

Well-led scored 93, the practice's highest key question. Much of the evidence behind its five 4s, and across the report, was about learning that travelled outside the building:

  • The practice manager had been recruited by the NHS transformation team to teach other practice managers in the locality.
  • The practice planned to share its housebound review model with the integrated care board and the local GP federation, "with a view to this approach being adopted by other practices."
  • It provided minor surgery, coil fitting and removal, contraceptive implants and pessaries for pelvic floor prolapse to other practices in the area.
  • Staff development was visible and specific: a receptionist who trained as a healthcare assistant and then a practice nurse, a nurse and a pharmacist who completed independent prescribing qualifications, and apprenticeships at several levels.

The two Well-led statements that scored 3 are useful too. Under Freedom to speak up, CQC fed back that two members of staff had said communication could be improved, and "The practice immediately put in place a communication strategy to address this." A 3 does not mean nothing could be better. It means a good standard, with concerns heard and acted on.

One more detail worth noticing. The report scored seven Well-led statements, and environmental responsibility still appeared in the narrative, under Shared direction and culture and Learning, improvement and innovation, with examples such as "using CFC free inhalers" and safe recycling of batteries and waste.

Using this report in your own practice

These are our practical suggestions, drawn from the report rather than from CQC guidance.

  • Check for 1s and 2s first. Before working towards 4s, make sure no quality statement in your practice would score below a good standard. One weak area limits the whole key question.
  • Pick two key questions to be exceptional in. The overall rating rule rewards depth in two areas more than effort spread thinly across five. Choose the two where your practice already does something unusual.
  • Write your improvement work as before, change, after. For each project, have the baseline figure, what you changed, the re-measured figure and the date of each. Two-cycle audits are the obvious format.
  • Show who else has benefited. Learning shared with your primary care network, your integrated care board or other practices is evidence for Well-led, and it is easy to leave out because it feels like extra work rather than core work.
  • Describe your population. Keep a short summary of your practice's context, the barriers it creates and what you have done about each one. It frames everything else.

A note on what is changing at CQC

The scores in this post follow CQC's current method. CQC has announced changes in its initial response to the Better regulation, better care consultation (page last updated 25 March 2026): "We will remove scoring from our assessment approach", and "We will no longer make formal judgements below the level of our 5 key questions." Quality statements are to be replaced by supporting key lines of enquiry. CQC says it will publish a full response during 2026.

What does not change is what sits under every score in this report: evidence of how care is planned, delivered, measured and improved. A practice that can show its improvement work clearly today will be able to show it under any framework.

How CQC AI Auditor helps

The evidence behind this practice's 4s, audits, improvement projects, minutes, training records and policies, is the kind of evidence CQC AI Auditor works with. The AI assisted audit reads the documents you choose to give it, identifies where your evidence is strong and where it is thin, maps it to each quality statement and scores it, with the reasoning written out. Each area includes a Path to Outstanding, describing what your practice would need to show to move beyond good. You can see the format on our sample report page, and how it works explains the two levels. It is our assessment of your evidence, not a prediction of your CQC rating. You can run your first self check free.

Frequently asked questions

What score does a GP practice need for an Outstanding CQC rating?

At key question level, CQC's threshold is "88% and above = outstanding", with no quality statement in that key question scoring 1 or 2. For the overall rating, CQC says at least 2 of the 5 key questions would normally need to be outstanding and the other 3 good.

Does every quality statement need to score 4 for Outstanding?

No. The practice in this report scored 4 on 13 of its 33 statements and 3 on the rest. Its two outstanding key questions still included 3s. What it did not have was any score of 1 or 2.

Can a GP practice be rated Outstanding with a key question rated requires improvement?

Not under CQC's normal rating principles. CQC says outstanding overall would normally need at least two key questions rated outstanding and three rated good, so a requires improvement in any key question would normally rule it out.

What evidence did CQC accept for a score of 4?

In this report, most 4s were backed by measured improvement: review numbers rising year on year, two-cycle audits with before and after figures, and appointments freed up. Others, mainly in Well-led, reflected exceptional leadership, workforce development and governance, backed by staff feedback and what inspectors saw in audits and records. The statements scored 3 were mostly described as good systems working well.

Ask us

We do not run comments on Insights. If you have a question about this report or about reaching Outstanding, ask on our LinkedIn page or reply to the newsletter.

Sources

Contains public sector information licensed under the Open Government Licence v3.0.