UK Statistical System UK Statistics Authority Code of Practice
Casework

Ben Connah to Ed Humpherson: Presentation of modelling in the Module 2 report of the UK Covid-19 Inquiry

Dear Ed,   

Thank you for the letter of 19 February on the presentation of modelling within the UK Covid-19 Inquiry’s Module 2ABC report. Your Office is clear that public bodies must use statistics, data and wider analysis transparently and with integrity, clarity and accuracy. You are concerned that the Inquiry conveyed too great a level of certainty when referring to the results of modelling.  

I note that you are not questioning the Inquiry’s conclusions, including those on the timing of lockdowns or the contents of the report itself.  

The focus of your letter is this extract on page 5 of the Executive Summary of the report: “Had a mandatory lockdown been imposed on or immediately after 16 March 2020, modelling has established that the number of deaths in England in the first wave up until 1 July 2020 would have been reduced by 48% – equating to approximately 23,000 fewer deaths”.  

I accept that in trying to condense and simplify the 800-page Module 2ABC report into a short, accessible Executive Summary, framing of the statistic through use of the word “established” was too definitive and did not reflect the uncertainty inherent in all modelling. The main body of the report uses more appropriate language in saying it is “estimated” that locking down one week earlier “could have” led to a reduction of 23,000 deaths.  

The Inquiry’s findings are grounded in the whole of the evidence received. Baroness Hallett concluded that the outputs of this model were useful in illustrating a broader finding about the timing of the first lockdown based on the totality of the evidence before her – both written and oral. She observed:  

“Indeed, the Inquiry accepts the consensus of the evidence before it that a mandatory lockdown should have been imposed one week earlier”1 and that doing so would have saved many thousands of lives during the first wave. Omitting, in its Executive Summary, the inherent uncertainty within one model does not change that overall finding of the Module 2ABC report.  

The academic paper (“Knock et al”) on which Professor Ferguson’s evidence was based – and from which the 23,000 figure was derived – did not make clear that it accounted for the impact of voluntary restrictions in the week before a mandatory lockdown. You say that you confirmed this with Prof Ferguson but this does not change the position. In the body of the Module 2ABC report, the Inquiry sets out the inherent uncertainty in modelling of this kind and the fact it is based on assumptions. The Inquiry accepts, however, that it did not sufficiently express this uncertainty in the Executive Summary.  

Having identified the issue about the framing of the modelling data in the Executive Summary to the Module 2ABC report, the Inquiry has introduced stronger quality assurance of the presentation of statistical data in the documents that accompany main reports such as summaries. The Inquiry will publish eight more reports in the coming months and in each we will seek to ensure the highest standards of statistical analysis and communication.  

The Inquiry will be publishing this letter on our website alongside the Module 2ABC report page so that future readers are aware.   

Yours sincerely,  

Ben Connah  

Secretary, UK Covid-19 Inquiry  

Related links: Ed Humpherson to Ben Connah: Presentation of modelling in the Module 2 report of the UK Covid-19 Inquiry

Ed Humpherson to Ben Connah: Presentation of modelling in the Module 2 report of the UK Covid-19 Inquiry

Dear Ben,

Thank you for taking the time to meet with the Office for Statistics Regulation (OSR), the regulatory arm of the UK Statistics Authority. We provide independent regulation of all official statistics produced in the UK, and investigate concerns raised with the Authority.

As you know, we were contacted about the presentation of modelling in the Executive Summary of the Module 2 report of the UK Covid-19 Inquiry, specifically the section which says that “Had a mandatory lockdown been imposed on or immediately after 16 March 2020, modelling has established that the number of deaths in England in the first wave up until 1 July 2020 would have been reduced by 48% – equating to approximately 23,000 fewer deaths”. Evidence given to the inquiry by Professor Neil Ferguson from Imperial College London and from an academic paper (Knock et al) provided the figure of a 48% reduction in deaths. The 23,000 figure was calculated by the Inquiry by subtracting the estimated number of deaths from the actual number of deaths.

While the inquiry report does not fall within our statutory remit, we considered the issue on an advisory basis, focusing on the communication of these figures in relation to the Standards for the Public Use of Statistics, Data and Wider Analysis, within the Code of Practice for Statistics.

I want to emphasise that we are not questioning the inquiry’s conclusions, including those on the timing of advisory and mandated lockdowns, nor would it be appropriate for us to do so. Our focus is the presentation of statistical evidence, and the risk that the wording chosen for the Executive Summary is a misleading representation of the underlying analysis. We thought it would be helpful to share our findings with you because in future modules of the Inquiry report, you may wish to present modelling or other analytical evidence and it will be important that this is communicated clearly and accurately.

Firstly, while steps have been taken to communicate uncertainty in the main body of the report, we consider that the Inquiry’s Executive Summary of the modelling does not sufficiently communicate the level of uncertainty associated with the analysis. Uncertainty is inherent in statistical models, some of which is quantifiable as an interval around an estimate, but there is also broader uncertainty due to unavoidable assumptions made in the modelling process. The use of the words “established” and “would have been reduced” overstates the confidence in the 48% figure and has the potential to mislead the public about the certainty of the finding.

The second aspect concerns what the ‘23,000’ figure represents. The Inquiry’s Executive Summary for Module 2 says the ‘23,000’ is an estimate of the effect of bringing the mandatory lockdown forward a week. However, in investigating this case, we have confirmed with Professor Ferguson that the ‘counter-factual’ (the hypothetical scenario which the model aimed to describe) was that both the mandatory lockdown and the voluntary measures announced on 16 March were each brought a week earlier. While we acknowledge that the Knock et al paper did not explicitly set this out as the counterfactual, there is a risk that the Executive Summary is misleading about what the ‘23,000’ actually refers to.

Given the weight put on the ‘23,000’ figure in the Executive Summary and in the subsequent media coverage, we consider it is important to have clarified this point about the counterfactual.

We invite you to set out the steps you are taking in response to our findings, to support the future communication of statistics and analysis by the UK Covid-19 Inquiry.

Yours sincerely

Ed Humpherson
Director General for OSR

Related links: Ben Connah to Ed Humpherson: Presentation of modelling in the Module 2 report of the UK Covid-19 Inquiry

Rob Kent-Smith to Alex Lambert and Liz McKeown: Family Spending statistics

Dear Liz and Alex

Today, we have published a review of the latest Family Spending statistics from the Living Costs and Food survey (LCF) against key elements of the Code of Practice for Statistics. We previously set out an intention to undertake further regulatory work on response issues impacting household surveys used across the statistics landscape. This assessment was undertaken as part of this work.

Based on the findings of this review, the Family Spending in the UK statistics, which are sourced from the Living Costs and Food survey (LCF), can continue to be published as accredited official statistics. The review found that, despite recent challenges with achieved sample size and timeliness, the Living Costs and Food Survey (LCF) continues to produce statistics of sufficient value and quality to meet user needs.

We found that the LCF is a valued data source and informs key analyses. Many users we spoke to mentioned appreciation for the survey and noted the unique role of the food diary, as there are no other free, publicly available datasets to get a comprehensive view of household food consumption.

ONS has begun to make improvements to improve the value and quality of these statistics. In response to user demand, ONS increased the LCF sample in April 2024, and will continue investing in LCF improvements, subject to funding and the conclusion of current business planning processes. ONS has also made initial steps to increase the resources available to the data production and analysis team, which should mean that other quality issues, such as timeliness, should be improved. Innovations such as modernised processing infrastructure, a digital interviewer-facing diary and receipt-reading tools are being developed to improve data accuracy and reduce the processing burden within the LCF data collection environment.

As well as requiring that ONS follow through and report on LCF improvements in its Economics Statistics Plan and its Survey Improvement and Enhancement Plan, we have three detailed recommendations from this review to further meet users’ needs. These focus on more engagement with users to understand their needs, being clearer about how ONS can speed up the publication timetable and providing more information about survey representativeness and comparability.

We note that ONS is already progressing work to ensure full compliance with the updated Code of Practice for Statistics (Code 3.0), and we welcome the commitment to embed these standards as part of ongoing development of the statistics.

We expect ONS to report on its progress against our recommendations for improvement from this review by April 2026.

Yours sincerely

Rob Kent-Smith

Deputy Head of the Office for Statistics Regulation

 

Alex Lambert and Liz McKeown to Rob Kent-Smith: Family Spending

Dear Rob,

Thank you for your letter of 10 February regarding the publication of the report covering your recent review of Family Spending statistics from the Living Costs and Food (LCF) Survey. We welcome both the publication of the review and its conclusion that Family Spending in the UK can continue to be published as accredited official statistics.

The review recognises the value users attach to the LCF and its importance to them as a data source, including the unique role played by the food diary. We appreciate your finding that the LCF continues to provide statistics of sufficient value and quality despite recent challenges with achieved sample size and timeliness.

As you note, work is already underway to address these recent challenges and further improve the value and quality of these statistics. ONS will continue to take forward the improvements laid out in the Economic Statistics Plan and the Survey Improvement and Enhancement Plan for Economic Statistics, with further investment subject to the conclusion of current business planning processes.

We also welcome OSR’s recommendations to further strengthen how these statistics meet user needs including through engaging with our expert users to better understand their requirements, particularly around information on data changes over time, improving the timeliness of publications, and providing clearer information on survey bias, representativeness and comparability. We recognise your expectation that ONS will report on progress against these areas by April 2026. We will ensure you are kept updated as this work advances, and we remain committed to maintaining open, constructive and transparent engagement with OSR throughout this period.

Yours sincerely,

Alex Lambert, Director of Social Surveys

Liz McKeown, Director of Economic Statistics Production and Analysis

James Benford to Ed Humpherson: ONS Reports on Progress with Economic Statistics, December 2025: OSR response

Dear Ed,

Thank you for your letter setting out the Office for Statistics Regulation’s (OSR) assessment of our first quarterly progress reports on the Economic Statistics Plan and the Survey Improvement and Enhancement Plan. Your suggestion that the Office for National Statistics (ONS) publish these quarterly reports has been helpful in creating a clear and transparent way to set out progress, surface risks and engage users and stakeholders.

I welcome your recognition of the early progress we are making. This includes the resumption of producer price statistics, the restoration of the achieved sample side for the Labour Force Survey to close to the pre-Covid point and the methodological improvements that came with the Blue Book to increase coverage of research and development in the economy and more accurately account for the output of multinational pharmaceutical companies. These changes directly improve the quality of the UK’s core economic statistics.

You will also have seen that yesterday we announced our ‘go’ decision to incorporate grocery scanner data into the consumer price index, alongside other changes to the measurement of hotels and computer game prices to reduce their volatility. These changes, and the broader modernisation of the data pipelines and analytical processes, represent a step change in the measurement of UK consumer prices. When we implement the changes in March, the index will be more comprehensive, with grocery price data covering purchases across all goods and days of the month and including the effect of store card discounts, and less sampling volatility. The modernisation of our systems and processes also gives us a platform to onboard more big data sets to improve further how we measure consumer prices as they become available.

I am grateful also for your recommendations on how we can further improve our quarterly reporting. The ONS is currently undergoing a multi-year business planning process that will be critical to carefully sequencing our work to improve statistics so that the Economic Statistics Plan is realistic and prioritises what matters most to users. Given the scale of our commitments – including double-running the existing and transformed labour force survey, delivering a new statistical business register, preparing for the new international macroeconomic standards and progressing Census 2031 – we are taking deliberate decisions about sequencing. Where necessary, this means pausing or slowing some planned activity in order to safeguard the quality and delivery of our most critical existing workstreams. As we announced last year, we are also working with key users to review the scope of the statistics the office produces and looking for opportunities to narrow the scope of our outputs and will set out our conclusions in the coming weeks. These prioritisation decisions are taken to protect statistics that underpin the most critical economic and societal decisions.

Our next progress report, to be published in April, will report progress against the milestones we set ourselves in 2025/26 and set out clearly our priorities and where work has been deferred as part of the planning process. The subsequent update in July, will come alongside our decision, informed by user engagement, on whether to progress immediately to the transformed labour force survey and will set out the detailed milestones that follow.

I would like to recognise also the work the OSR is undertaking to follow up its review of His Majesty’s Revenue and Customs (HMRC) statistics against the Code of Practice. Errors in data supplied to the ONS by third parties have been a major source of errors in ONS statistics over the past year, including data supplied to us for public finance and trade statistics. Our quarterly report noted these errors as a major issue we are facing as we seek to improve the quality of economic statistics. As you will be aware we have already strengthened procedures around the checking of source data for public finance statistics, adding comparisons with independent feeds during collective assurance, between HM Treasury, the ONS, HMRC and the Office for Budgetary Responsibility, of taxes data. Our data governance team has also set out a new framework for managing third party data sources and I welcome your initiative to set up a joint event between the OSR, the ONS and government data suppliers in the coming months to discuss practices in supplying data to us. We look forward to the findings from your follow up review and stand ready to do whatever more we can do to improve the quality and reliability of important source data from HMRC.

Thank you for your continued engagement on our plans and your focus on improving the quality, trustworthiness and value of ONS statistics.

Best wishes,

James Benford

Director General, Surveys and Economic Statistics Group

Ed Humpherson to Chris Mullin: Supporting appropriate use of gambling-related harms evidence review

Dear Chris,  

Supporting appropriate use of gambling-related harms evidence review

I wrote to colleagues in the Department for Health and Social Care in September 2024 regarding a case raised with the Office for Statistics Regulation (OSR) about the 2023 report ‘The economic and social cost of harms associated with gambling in England’ by the Office for Health Improvement and Disparities (OHID).  

We were grateful for the chance to meet with OHID to discuss the concerns raised and identify opportunities to improve the clarity and transparency of the information. My letter set out several recommendations that OHID committed to take in response to our findings. 

Since then, the OHID report has continued to be a focus of cases raised with OSR. While these cases concern the communication of statistics by different individuals and organisations, they are united by references made about the OHID report that incorrectly refer to harm caused by problem gambling, rather than being associated with gambling.  

We appreciate that the report does state that the links between problem gambling behaviours and harms are associative and do not demonstrate causality. However, the examples shared with us suggest that the distinction between association and causation is not being fully understood by users. This undermines the analysis and creates the potential for people to be misled.  

We have tried to engage with OHID on several occasions to strongly encourage it to consider actions that could be taken to support the appropriate use of these figures. For example, including a prominent disclaimer or banner on the publication itself to clarify this point. However, we have lacked any recent engagement from OHID to engage with the importance of this issue. 

I am therefore writing to urge you to put measures in place to prevent the potential misuse of this analysis. I would be grateful for your commitment and continued support in ensuring statistics and data are communicated in a way that supports the public good and strengthens public debate.  

Yours sincerely  

Ed Humpherson
Director General for OSR 

Ed Humpherson to Tony Dent: ONS Economic Statistics Plan, Consumer Prices and UKSA Board and Panels

Dear Tony,

ONS Economic Statistics Plan, Consumer Prices and UKSA Board and Panels

Thank you for your letter dated 4 September 2025 and your follow-up enquiries on 5 September.

You expressed concerns with the status of the Office for National Statistics (ONS)’s CPI and CPIH statistics as accredited official statistics. As mentioned in my letter in March 2025, we are intending to re-assess CPI and CPIH once ONS has incorporated grocery scanner data into its inflation statistics. We recognise that there are competing views about different elements of the calculation of the consumer price indices, and we are continuing to encourage ONS to improve transparency around its approach. Our thinking remains open on this issue, and we plan to discuss this with a range of users, including you and your colleagues at Better Statistics.

You referenced the need for a way for users to communicate their concerns directly to the appropriate advisory committee, particularly the two UKSA consumer prices advisory panels (the Technical and Stakeholder Panels). These advisory panels can consider issues raised by users; if this has not been your experience, please let me know and my team can follow up. Separately, you mentioned that you would like to see the prompt release of minutes for UKSA Board meetings. We have referred this request to the Board Secretariat to see what can be done.

Thank you for your comments on the ONS Plan for Economic Statistics and the additional ONS Survey Improvement and Enhancement Plan. Following the publication of OSR’s 2025 State of the Statistical System report in July, I wrote to the new ONS Director General for the Economic, Social and Environmental Group, James Benford.

While in this letter I reaffirmed that the Plan for Economic Statistics met “OSR’s immediate requirements as a plan of activities to restore confidence, ensure strategic transparency and enhance focus on the quality of data inputs”, I also agreed with your view that this represents a very ambitious programme of work, and that OSR expects further actions from ONS over the longer term to build confidence. OSR has also requested quarterly reporting of progress from ONS against the milestones. So our judgement, as set out in our public statements, remains one of wanting to see how well ONS is able to deliver on these plans.

In your follow-up email, you asked for more information about the users we spoke to for our compliance review on the transformation of CPI and CPIH. These included users from the Bank of England, HM Treasury, the Royal Statistical Society, the Trades Union Congress, the Resolution Foundation, the Northern Ireland Statistics and Research Agency and the University of Southampton.

Finally, you requested any documentation about how the ONS applies hedonic regression in its calculations of inflation measures. This information is not currently located in one place. For details on its use in private rents statistics, for example, see Section 6 of the Price Index of Private Rents Quality & Methodology Information (QMI) document. For the calculation of consumer prices inflation, in our view the best resource is the Consumer Prices Indices Technical Manual.

In our regulatory work, we have set out requirements (for example Requirement 2 in our PIPR Assessment and the recommendations in our recent CPI-CPIH Compliance Review) asking ONS to improve the way it disseminates its information about its data and methods to a range of audiences. This includes setting out end-to-end documentation outlining the complete production process.

Our requirements remain outstanding, and the example of hedonic regression which you have highlighted illustrates that ONS has more to do to address this need. Please let us know of any specific thoughts you have about what ONS could do to better communicate its approach to hedonic regression.

Yours sincerely

Ed Humpherson
Director General for OSR

Related

Tony Dent to Ed Humpherson: ONS Economic Statistics Plan, Consumer Prices and UKSA Board and Panels

Ed Humpherson to Dr James Noyes: SMF report on gambling tax reform

Dear James,

SMF report on gambling tax reform

Thank you for taking the time to meet and for allowing us the opportunity to provide more information about the work of the Office for Statistics Regulation and our approach to investigating concerns raised with us.

As you are aware, we were asked to look at the Social Market Foundation’s (SMF) report on The Duty to Differentiate, specifically the language used when referring to analysis taken from the Gambling-related harms: evidence review published by the Office for Health Improvement and Disparities (OHID).

We are pleased that prior to our meeting, you published a correction to the report that explains the SMF report mistakenly referred to “due to” rather than “associated with” when discussing gambling related harm and that the drafting was not intended to imply a causal relationship. This correction is clear and accessible in line with best practice, and we consider the matter closed.

We welcome your continued commitment to the clear and transparent communication of statistics and data.

Yours sincerely

Ed Humpherson
Director General for OSR

Tony Dent to Ed Humpherson: ONS Economic Statistics Plan, Consumer Prices and UKSA Board and Panels

Dear Ed,

Thank you and Rob for the recent discussion, it was gratifying to hear that the OSR has valued some of our commentary.  However, without wishing to carp in any way, I continue to believe our contribution would have been greater had there been a clearer understanding of the structure and purpose of the various elements of UKSA.  We have frequently commented on how our ‘outsider’ perspective differs from your ‘insider’ view and we believe the difference results from the failure of UKSA to communicate its activities effectively.

There are many examples of the difficulties created by poor communications but, probably, none are more significant than the fact that Better Statistics CIC do not understand how the CPI and the CPIH retain their status as Accredited Official Statistics, given the many errors we and others have exposed.  Moreover we fail to understand how they can be considered to be “statistics for the public good”.

Whilst we recognise that the OSR’s role as the assessor of the efficacy of these measures must be influenced by the two Advisory Panels on Consumer Price Statistics (APCP), we have no explanation as to how the various parties actually inter-relate to achieve a co-ordinated opinion.  In our view this lack of clarity contributes to the confusion evidenced by the public’s inability to recognise UKSA’s opinion of the value and purpose of these measures.  A confusion recently highlighted by the BBC announcer who said that “most households” would be experiencing a higher inflation rate than the July rate published by the ONS.  This dissonance, between personal experience and the ONS measures, lies at the core of the mistrust which many people have with our statistics.  A mistrust that long preceded the controversies of the past year or two, with consequences that UKSA continue to ignore.

Turning to the wider issue of our economic statistics, we note that the web-site announcement of the OSR’s State of the UK Statistical System 2025, includes the assurance that “In our view, the Office for National Statistics (ONS)’s recently published plan for economic statistics provides the right foundations to restore confidence in ONS”,  however we fail to see how that view is supported by the publication.  Have we missed something?  To us, the plan lacks explanation as to how the ambitious series of deadlines for improvements across every area will be achieved, even considering the additional ONS Survey Improvement and Enhancement Plan for Economic Statistics.  Both plans appear to lack the details required to explain how resources will be deployed, with the main plan lacking explanation as to the objectives of, or reasons for, many of the activities.

See for example: “A2.3 Sustainable operation:

We are planning to extend the coverage of our Large Case Unit and our Account Management Units, and so extend their proven impact on data quality and simplify engagement with more businesses. We plan to expand their combined coverage from 150 today to:

  • around 850 business units this year (2025 to 2026)
  • increasing to in the region of 1,800 business units in 2026 to 2027
  • rising to a combined total of around 4,000 business units by the end of 2027 to 2028.

BSC would prefer to see some explanation for these numbers – specifically what benefits are expected from the increased sample sizes?   We would also prefer to see consideration of alternative approaches, such as the use of cluster sampling to reduce Household survey costs and a trial of mix-mode data collection to include alternative methods such as time use surveys.  We consider that the latter would more readily capture the details of those persons seeking to hold down two or three jobs, or others with complex lifestyles.  However, although the OSR is charged with preparing the annual reports on the state of the UK Statistics System, we understand that comments  such as those above are not within your purview.

Meanwhile we welcome the ambition of the ONS for a “Refreshed Citizen Relationship – to improve trust, brand recognition and understanding of the importance of our surveys” and we hope that this will encourage more urgent consideration of the inflation measures as mentioned above.  Also, we hope that the ONS now accept that not all citizens listen to radio 4!  The frequent mentions of ONS activities on the Today Programme had been evidenced by the ONS as sufficient publicity, when BSC had suggested advertising the Covid-19 Infection Survey to improve the response rate from the lamentable 11% of households sampled.

Anyway, to return to our recent meeting.  I am sure you will recall our opinion  that the provision of prompt and informative minutes of the Advisory and Main Board meetings would greatly improve UKSA’s communications.  It being expected that such meetings should be  more robust in considering statistical details, because a statistic is a relatively ephemeral, point in time, estimation that, normally, will be different on another occasion. It is NOT immutable and one would expect the relative advisory boards to be more open in their discussions as to the level of reliability required and the methods used.

The current suppression of any disagreement is actually anathema to a healthy statistical operation, which should be more concerned with discussion of the policy risks associated with unreliable data. Public discussion of the important issues underlying such matters can only be for the public good.  One clear recommendation we would make to improve public trust would be to provide a conduit for users to communicate their concerns directly to the appropriate advisory committee, particularly the two committees advising on inflation measurement.  The SUN platform could be used for this purpose, were it to be more concerned to encourage public participation, instead of controlling it.

I hope you will give these remarks further thought and, possibly, discuss them with your colleagues.  In particular, BSC would appreciate a direct answer as to the accreditation of the CPI and CPIH and, accordingly, we propose to publish this letter in due course.

Finally, we have given further thought to  the proposed seminar “Growing the UK Economy – how can business contribute?“.  That seminar was partly inspired by the fact that the business community was not well represented at the Assembly earlier this year, nor is business much engaged with the Government Statistical System in general.

We had hoped to make a contribution to a ‘refreshed citizen relationship’ with this seminar and we’d like to discuss that with you at your convenience.

With all good wishes,

Tony

On behalf of Better Statistics CIC

Related

Ed Humpherson to Tony Dent: ONS Economic Statistics Plan, Consumer Prices and UKSA Board and Panels

Ed Humpherson to Stephanie Howarth: Response to National Survey for Wales request to temporarily suspend Accredited Official Statistics status 

Dear Stephanie,   

Thank you for your letter requesting a temporary suspension for the accredited official statistics status for outputs based on the 2024-2025 National Survey for Wales data. This request is due to a smaller than expected achieved sample size for the survey in 2024-25 which will impact on the quality of estimates at lower levels.  

Based on your investigations, we agree with your request to suspend the accreditation of the National Survey for Wales statistics. We understand that data derived from the National Survey for Wales are used to inform national well-being indicators in Wales. Therefore, please do consider any potential quality impacts on other statistical outputs that rely on this data, such as the annual Wellbeing of Wales statistics. 

We welcome your plans to publish the release of 2024-2025 results as official statistics in development and your commitment to be transparent in your guidance for users on how the data should be used and interpreted. We also support your future ambition to increase the achieved sample size by moving from a telephone-first to an online-first approach. 

My team will be happy to engage with you about the most appropriate timing for a review of these statistics. This will be with a view to reinstating accreditation once you feel assured that they fully comply with the Code of Practice for Statistics.    

Yours sincerely,  

Ed Humpherson