2025/26 casework themes
As with our previous annual casework reports, we have broken down our key takeaways from the last year into separate themes that further explore observed trends. This year we have identified five themes, which are set out below.
Supporting the appropriate use of statistics, data and wider analysis
In November 2026, OSR launched the third iteration of its Code of Practice (Code 3.0). This version introduces the Standards for the Public Use of Statistics, Data and Wider Analysis (Standards for Public Use). While earlier versions of the Code primarily focused on the publication and use of official statistics, these new standards extend beyond that scope to cover the presentation of numbers by public bodies in all public-facing communications, including press releases, policy documents and social media content. These standards are based on the concept of intelligent transparency, which is itself grounded in the core principles of the Code of Practice – Trustworthiness, Quality and Value.
The Standards for the Public Use align with, and have indeed been shaped by, the types of casework issue that consistently occur:
Equality of access: Public bodies should make any statistics, data and wider analysis used in the public domain available to all in a timely manner. Casework in 2025/26 often highlighted issues around data accessibility – for example, where figures were quoted before the underlying data or statistics were published, or where management information was referenced despite not being publicly available.
Supporting understanding: Public bodies should use statistics, data and wider analysis with integrity, and communicate them with clarity and accuracy. Casework queries in this area typically focus on how data are presented, such as the use of misleading visualisations – for instance, graphs with truncated or broken y-axes that exaggerate differences between values.
Decision making and leadership: Public bodies should seek and use impartial, expert advice when using statistics, data and wider analysis in the public domain. Casework related to this standard usually involves the publication of statistics that haven’t been signed off or verified by an appropriate official, such as a departmental Head of Profession for Statistics.
We believe that the Standards for the Public Use in the Code will enable OSR to more effectively guide those producing and using statistics on how to communicate statistics appropriately. OSR is already engaging with members of other government professions outside of statistics, such as communication professionals, to promote these standards. It is our aim to support departments that communicate publicly to do so in a way that demonstrates trustworthiness more consistently.
In October 2025, The Standard published an article titled “Sadiq Khan claims ‘diverse’ London is reason for American tourism surge”. The article states that figures analysed by City Hall (the Greater London Authority, GLA) showed that the number of US visitors to London has increased by 8% in 2024 and by a fifth since 2022. It also claimed that the capital saw 5.56 million American tourists last year, according to the International Passenger Survey (IPS).
Although the IPS was referenced, there was no link provided to the ONS data. We examined the IPS data directly but were unable to replicate the figures cited in the article. In addition, we were unable to locate the underlying analysis that City Hall conducted, after searching the GLA website. This led us to conclude that the statistics used in the news article were not accessible to all users and therefore did not meet the ‘equality of access’ principle in the Standards for Public Use.
While the GLA is not a producer of official statistics, it has voluntarily committed to applying the Code of Practice for Statistics. We contacted the GLA regarding this issue and established that the figures reported in the article – apart from the IPS data, which the journalist is believed to have gathered themselves – were based on internal analysis using relevant data sources, provided on request to a journalist. Following our engagement, the GLA published its analysis of recent tourism trends, making the statistics accessible to all.
In November 2025, we were contacted by the BBC Verify team regarding the Conservative Party’s presentation of data from the Ministry of Housing, Communities and Local Government (MHCLG) on net additional dwellings in a post on the social media site X. The post received significant engagement, with over 3.3 million views and 350 reposts.
The post included a bar chart comparing the number of new dwellings delivered in 2023/24 and 2024/25. It presented 2023/24 as a period under the Conservative Party and 2024/25 as being under the Labour Party, and it visually showed the 2023/24 figure being around twice as high as the 2024/25 figure. However, when the underlying data were examined, it emerged that 221,409 dwellings were delivered in 2023/24 and 208,600 in 2024/25 – a difference of around 6%, with the latter figure being lower.
As the data were presented by financial year, the 2024/25 figure also included three months during which the Conservative Party was still in office. This meant that the comparison did not accurately reflect housing supply under each administration.
While the data themselves were accurate, the presentation was misleading. The use of a truncated axis exaggerated the difference between the two figures and undermined clear interpretation. This fell short of the Standards for the Public Use of Statistics, which emphasise communicating data with clarity, accuracy and integrity.
We contacted the Conservative Party about this issue, but the chart was not amended. However, following our engagement with the BBC, OSR’s views were included in a BBC Verify article, alongside a chart that accurately represents the number of additional dwellings completed without using a truncated y-axis.
In December 2025, we were contacted by Medr, the organisation responsible for funding and regulating the tertiary education and research sector in Wales, regarding a post on X by the Welsh Labour party. The post stated that 100,000 apprenticeships had been delivered during the current Senedd term, across a range of sectors from engineering and hairdressing to artificial intelligence. This figure reflected a Programme for Government target of delivering 125,000 apprenticeships by the end of the Senedd term, which was revised down to 100,000 in 2024.
Medr is responsible for publishing official statistics on apprenticeships in Wales. It raised concerns that the figure cited in the post appeared to be a misuse of statistics and was not based on any publicly available information. At the time, Medr published quarterly statistics on apprenticeship delivery. The August 2025 release reported that 77,385 apprenticeships had been delivered since Q4 of 2020/21, a significantly lower number than the 100,000 claimed.
Medr explained that, due to increased scrutiny of apprenticeship statistics, it had also been providing monthly management information to colleagues within the Welsh Government. However, we were unable to identify the source of the figure used by Welsh Labour. Medr had already planned to start publishing monthly management information updates, but we advised Medr to publish the most up-to-date management information immediately. This was subsequently released in December 2025 and confirmed that the target of 100,000 apprenticeships had not been met at the time Welsh Labour made the post.
During this case, Medr demonstrated behaviours in line with the Standards for Public Use of Statistics. Medr highlighted the post on X to OSR quickly after it was published, informing us that the statistics had been misused. Medr asked for our advice and whether it was appropriate for OSR to intervene. Though the Welsh Labour party is not a public body and is not required to follow the Code, consulting Medr ahead of posting would have better demonstrated adherence to the Standards for Public Use of Statistics, which state that public bodies should seek and apply impartial, expert advice when using statistics in the public domain. In turn, the publication of a misleading claim could have been avoided.
As a result of Medr’s decision making and leadership, and OSR’s intervention, the Welsh Labour party deleted the post on X. Medr’s publication of management information ensured that accurate information was available publicly.
An increase in the complexity of cases
Our analysis of casework during 2025/26 clearly shows that the time required to resolve a piece of casework can vary considerably – from a few days to several months – depending on the complexity of the issue and the level of intervention needed. Over the past year, we have observed an increase in more complex cases requiring deeper analysis than has been typical for casework. In some of these instances, the work has developed into a broader regulatory project, often extending into a compliance review.
This shift appears to be driven by a growing number of queries about the nature and quality of statistical outputs; the proportion of cases we have received related to quality, reliability or trustworthiness increased from 20% of our total caseload in 2024/25 to 33% in 2025/26. These types of cases are now our second most common type, with the potential misuse of statistics still being our most common casework issue.
These types of queries require more-detailed examination of the underlying data and methods, which can be time-intensive and, in some cases, lead to further regulatory engagement. An example of where this occurred was our review of Treatment of Seasonality in quarterly GDP statistics. This started as casework after concerns were raised by users and media reports, suggesting there may be residual seasonality in recent quarterly GDP statistics. Due to the complexity of the required investigation, OSR determined a full compliance review was required, with a focus on this specific issue. In total, five cases led to more in-depth regulatory work in 2025/26. Of these, two cases centred on economic statistics; the other three focused on abortion statistics, gambling statistics and crime statistics. This level of scrutiny speaks to the importance and public relevance of each of these topics.
Our role in the May 2026 elections
Though the period leading up to the local and devolved elections on 7 May 2026 falls outside of the 2025/26 period covered in this report, it is important to cover our work during the elections, given its proximity to this report’s release.
OSR undertook both proactive and reactive work in the run up to and during the election campaigns that took place across England, Scotland and Wales during March to May 2026. Our aim was to help ensure the appropriate and transparent use of statistics during the election campaigns, so that members of the public could understand and verify the statistics they heard and have confidence in them.
In preparation for the elections, a dedicated webpage was developed on the OSR website to bring together our new and existing guidance to help users navigate statistical claims surrounding an election and to support those working with statistics in upholding the Code of Practice for Statistics. The webpage includes our updated series of election explainer statements (15 in total) on topics including health, the economy and migration. In addition to the election explainers, Penny Young, the Interim Chair of the UK Statistics Authority, wrote to the leaders of political parties in advance of the elections asking for the appropriate and transparent use of statistics during election campaigns. By 7 May, our election pages had collectively been viewed more than 1,800 times.
From 26 March 2026 to 7 May 2026 (this period covers the pre-election periods for each election), OSR opened 38 pieces of casework. Of these, 22 were related to the local or devolved elections. Concerns with election leaflets were common during the pre-election period, with 11 of the cases we received taking issue with statistics found in leaflets. Other common sources of casework during the pre-election period included claims made by politicians whilst campaigning and concerns with statistics presented in political party manifestos.
OSR can continue to intervene publicly during formal pre-election periods, under our role of promoting and safeguarding the production and publication of official statistics. However, as set out in our interventions policy, our threshold for intervening publicly does increase. We are likely to limit public interventions to occasions where a statement is materially wrong or misleading to the point it could significantly impact on the political debates that are presented to voters.
During the pre-election period, we published a statement on understanding NHS planned care waiting times statistics across the UK to address widespread misuse of these statistics. The statement was picked up and drawn on by several media outlets and secured a correction to a statistic relating to greenhouse gas emissions in the Scottish National Party manifesto. We also wrote to several parties privately advising them about their use of statistics in leaflets and or public discourse.
We were unable to formally intervene on all cases brought to us during the pre-election period. This is because, as previously stated, it is outside of OSR’s remit to formally intervene on issues related to statistics that are not official statistics. Of the 22 cases relating to local or devolved elections we received, 9 were deemed to be outside of OSR’s statutory remit for intervention, though we did still offer advice on how statistics could be used appropriately.
NHS waiting times were an issue of high public interest in the 2026 elections, with high-profile claims by politicians and in the media comparing waiting times statistics between the countries of the UK.
In advance of the elections, OSR published guidance advising against making comparisons between waiting lists across different UK nations and areas of care. This reflected the devolved nature of the UK health system, which operates under differing legislative frameworks, policies and priorities, commissioning processes, and patient data systems.
In Scotland specifically, there is no single, comprehensive dataset covering all patients waiting for treatment. Instead, data are held across multiple individual waiting lists, which cannot be combined to produce an overall total or a consistent measure of waiting times. This means there is no definitive figure for the total number of people waiting, or the length of time they have been waiting, in Scotland.
Despite these warnings, in April 2026, during the pre-election period, the UK Secretary of State for Health, Wes Streeting, published a post on X stating that “the number of patients waiting longer than two years for treatment is almost 30 times higher in Scotland than in Wales”.
We were unable to identify a data source that corroborated the claim made in the post. Given the interest in this statement, and other similar statements that were being made, OSR published a statement on understanding NHS planned care waiting times statistics across the UK. This statement reinforced our previous guidance and more clearly set out why direct comparisons between NHS waiting lists across UK nations should be avoided. Our statement was quoted across several media outlets, helping to counter the potentially misleading statements made based on the statistics.
Influence beyond official statistics
Though our statutory remit covers official statistics, we take a broad view of what we will consider through our casework function and what falls within our scope. For many users of statistics, the distinction between official statistics and other data is not always clear, and members of the public tend to have a broad interpretation of what counts as ‘statistics’. We have observed this in our own research into how individuals use statistics to shape personal decisions.
As in recent years, there have been many cases raised with us in 2025/26 that sit outside of our statutory remit of official statistics. In such instances, we consider whether an OSR intervention could improve the use or interpretation of the figures, how our involvement might be perceived, and the extent to which it could have a positive impact on the statistical system or public debate if we were to intervene. As a result, we may engage with issues that resemble official statistics, regardless of whether they are formally classified as such, on an advisory basis. Broadly speaking, if the concern raised relates to how statements represent numerical conclusions based on analysis in the public domain, we may comment from an advisory perspective, based on the Standards for Public Use. If the concern raised relates to the technical processes underpinning the analysis, then we will not comment. For example, if a numerical statement does not quote a source, we would consider that an issue of use. But if the concern is around the methodology used to generate the numbers (e.g. the assumptions underpinning a macroeconomic model) then we would not comment.
With the inclusion of the new Standards for Public Use in the Code, we are on firmer ground to advise on a broader range of government communications, such as press releases. We have made several interventions based on the standards since their introduction, including on announcements made by organisations that do not produce official statistics. We hope this will strengthen the trustworthiness of all government communications of statistics, not just those published in statistical bulletins.
Similarly to last year, we continue to maintain mutually supportive relationships with organisations such as Full Fact and BBC Verify. These organisations often come across, or are requested to investigate, the same issues which are raised with OSR, although their remits are much broader and not restricted to official statistics. On occasion, this can result in us passing cases on to Full Fact or BBC Verify, particularly for cases which sit outside our formal remit. We also collaborate to address cases together when appropriate.
The below case study illustrates a high-profile example of our intervention on a case that fell outside of our scope of official statistics. It was also one of the first examples of OSR publicly offering advice based on the newly established Standards for the Public Use of Statistics, Data and Wider Analysis.
On 29 November 2025, OSR received a complaint regarding a claim made in the Executive Summary of Module 2 of the UK Covid-19 Inquiry. This report was published by the UK Covid-19 Inquiry team, which is an independent public inquiry set up to examine the UK’s response to, and impact of, the COVID-19 pandemic. The claim in question was 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.”
As both the modelling data found in the inquiry report and the report itself were not official statistics and therefore outside of OSR’s statutory remit, we approached this case from an advisory basis, focusing on the communication of the statistics in relation to the Standards for the Public Use of Statistics, Data and Wider analysis.
We summarised our findings in a public letter to the Inquiry Secretary in February 2026. There were two points of concern:
- That there was a lack of communication of the inherent uncertainty found in statistical modelling.
- That it could have been clearer that the 23,000 figure represented a hypothetical scenario in which both the mandatory lockdown and the voluntary measures announced on 16 March were brought in a week earlier, rather than just the mandatory lockdown.
Shortly after we published our letter, we received a response from the Covid-19 Inquiry stating that it accepted our advice and that it had introduced stronger quality assurance of the presentation of statistical data in the documents that accompany main reports, such as summaries. The Inquiry also uploaded both letters to the inquiry website alongside the Module 2 report.
This highlights the effectiveness of the new Code standards. It also shows how influencing and fostering constructive communication, and advice sharing beyond our remit, can result in positive outcomes for both producers and users.
Improving our casework service and the coming year
The volume of casework we received in 2025/26, while lower than in 2024/25, demonstrates the value of OSR’s casework function as one of OSR’s main regulatory tools.
Looking ahead, the complexity of cases that we have received this year has demonstrated a need for us to further build our casework function so that it can continue to act in the interest of the public good and respond to concerns in a timely manner. We have recently recruited several new regulatory staff, which will increase our capacity and ability to deal with cases faster and allocate more resource to complex cases. We are continuing to roll out regular casework training to both new and experienced regulators, to ensure that OSR’s casework function continues to effectively, consistently and fairly tackle the concerns that we receive.
OSR will continue to use the Standards for Public Use of Statistics, Data and Wider Analysis to support our casework interventions and intends to look in more detail at how we apply these standards to cases that fall outside of our statutory remit. As OSR continues to offer advice based on the new standards, we will be monitoring the types of cases we intervene on, to ensure we are consistent. In terms of consistency of advice in these cases, we believe the standards act as an effective guardrail to ensure we continue to act in cases where we can have a positive impact for the wider public.
OSR also intends to conduct a review of the work we did ahead of and during the 2026 elections this summer, with an aim to further improve our readiness and delivery during future elections, our explainers, and our ability to respond quickly to the misuse of statistics in political debate.
We will continue to seek feedback on our casework process via our casework feedback survey, which will also inform our delivery plans going forward.
Casework will continue to be at the centre of OSR’s regulatory activity in the future: it supports each theme of OSR’s new 3-year strategy. Our ability to intervene and to aid members of the public, and others, to resolve concerns with statistics continues to be our most visible form of regulatory work and allows us to spread the message of the Code of Practice for Statistics to a wider audience.
Back to top