What the Dash Review Published — and Did Not Publish — About the National Guardian’s Office

In July 2025, Dr Penny Dash’s review of patient safety across England’s health and care system made a direct finding about the National Guardian’s Office.

Finding 8 said that the National Guardian’s Office duplicated work carried out by providers.

The review went on to recommend that its responsibilities should be incorporated into providers, with related national staff-voice functions aligned elsewhere in the NHS and the Care Quality Commission checking whether effective Freedom to Speak Up arrangements were in place.

The National Guardian’s Office closed on 30 June 2026. From 1 July, NHS England took on some of its activities while providers and commissioners took greater responsibility for local arrangements.

That makes the evidence behind Finding 8 more than a historical question. The recommendation helped reshape a national speaking-up system.

TWIS therefore asked a narrow question: does the published Dash record show, function by function, what the National Guardian’s Office duplicated and what evidence supported replacing or redistributing each function?

The answer from the published record is: not at that level of specificity.

What Finding 8 actually says

The review describes the National Guardian’s Office as leading, training and supporting the Freedom to Speak Up guardian network. It notes the growth of that network and the number of concerns raised through guardians.

It then questions whether an independent national oversight body is still required. The report says the office had at times found it difficult to engage NHS leadership and, as a result, could lack the impact required. It also says hosting the role within CQC made it “too distant” from the people it needed to support and influence.

Its proposed answer was to move responsibility closer to commissioners and providers. The report says this should increase the profile of speaking-up functions locally and allow a more rapid response.

Recommendation 6 develops that position. It says responsibilities of the National Guardian and the office should be incorporated into providers, while related staff-voice activity should be aligned with NHS England and CQC should assess whether effective arrangements exist.

These are clear conclusions. The harder question is how the publication demonstrates them.

The report records functions that are not purely local

Elsewhere, the Dash review records a broader National Guardian remit.

Appendix 8 says the office was established as an independent function within CQC. It records a requirement for an independent annual report to be laid before Parliament, including the purpose of holding government and the wider system to account and advocating for change.

It also describes national data and learning activity, support for guardians, Speak Up reviews and the use of those reviews to identify thematic learning.

That creates a testable question for a duplication finding: for each material national function, which provider, commissioner, NHS England or CQC function duplicates it, and on what evidence is equivalence or replacement based?

Appendix 11 is a landscape map, not that chain of evidence

The review does contain a functions table.

Appendix 11 calls it a “visual illustration of the approximate mapping” of 16 functions to 18 organisations or types of organisation across the patient-safety landscape. It marks whether organisations largely undertake broad functions such as engagement, governance, training, monitoring, enforcement and regulation.

That table is relevant. It shows that functions overlap across the system and includes the National Guardian’s Office among the organisations mapped.

But it does not provide the narrower chain needed to test Finding 8 function by function. It does not take each material NGO responsibility, identify the particular provider or national function said to duplicate or replace it, then set out comparative evidence showing that the alternative performs the same role effectively.

Nor does the published report present an NGO-specific comparison of response times, outcomes or effectiveness to substantiate the proposition that redistribution would produce a more rapid response. TWIS also found no NGO-specific cost-benefit or value-for-money analysis in the published review supporting closure or redistribution.

That does not mean such analysis never existed. It means it is not demonstrated in the material the review published.

The NGO submitted evidence to the review

There is another reason not to turn a publication gap into a claim that no evidence existed.

An NHS England Board update records that the National Guardian met Dr Dash four times and that the office supplied five returns to the review.

Those submissions are therefore part of the evidential history even though they are not reproduced with the published report.

DHSC says it does not hold the requested underpinning records

On 18 August 2026, TWIS asked the Department of Health and Social Care for recorded information directly underpinning Finding 8 and Recommendation 6.

The request was deliberately specific. It asked for any function mapping, options appraisal, evidence table or working paper comparing NGO functions with supposedly duplicative functions elsewhere; evidence supporting the “too distant”, impact and “more rapid response” propositions; comparative performance material; cost or value analysis; analysis of the effect on independence; and the five NGO returns if held.

DHSC replied on 16 September 2026:

“DHSC does not hold the information you have requested.”

The department said the Dash review was commissioned as an independent review and pointed to the stakeholders engaged and references consulted in the published report.

That response matters, but its limits matter too. A Freedom of Information response saying DHSC does not hold the requested material is not proof that the records never existed, were never used, or are not held by another public authority.

CQC holds some information and is still considering disclosure

TWIS separately asked CQC for the five NGO returns, covering correspondence and records of the National Guardian’s four meetings with Dr Dash where the relevant issues were discussed.

On 16 September, CQC confirmed that it does hold some recorded information in relation to the request.

CQC said disclosure of some material may engage section 31 of the Freedom of Information Act because of possible prejudice to its regulatory activities. It is carrying out the required public-interest test and has extended the response period. It says it will respond, or provide a further update, no later than 14 October 2026.

That prevents a stronger conclusion today.

Until the CQC process is complete, TWIS cannot responsibly say whether its records contain a more detailed function mapping, comparative analysis or other evidence that fills the gap in the published report. Equally, CQC’s confirmation that it holds some information does not establish that the material supports the Dash conclusions.

It is evidence still to be examined.

What can be concluded now

The distinction is straightforward.

The published Dash review gives readers the conclusion that the National Guardian’s Office duplicated provider functions, alongside broad reasoning about distance, impact, localisation and response. It also gives a system-wide approximate functions map.

What it does not publish is a function-by-function evidential chain showing how each material NGO responsibility duplicates or is equivalently replaced by another body, supported by NGO-specific comparative performance, effectiveness or cost analysis.

DHSC’s subsequent FOI response does not supply that missing chain: it says the department does not hold the requested records.

CQC’s response means the wider record is not closed. Some information responsive to the related request is held and remains under consideration for disclosure.

So the defensible conclusion is narrow: Finding 8 is stated more strongly than its function-by-function basis is demonstrated in the published record.

That is not the same as saying Finding 8 was wrong.

The next evidence can change this article

If CQC discloses the NGO returns or meeting records, TWIS will test them against the same questions used here: which NGO function, what claimed duplicate or replacement, what evidence of equivalence or improved effectiveness, and what consequence for independence and accountability?

If those records supply the missing evidential chain, this article should be updated accordingly.

The point of an evidence audit is not to defend its first conclusion. It is to make the conclusion change when the evidence changes.