Making statistics available: enhancing understanding of topics of high public interest

What we said in October 2021:

Transparency is essential for building public trust in statistics and retaining public confidence in government decisions. To demonstrate trustworthiness, statistics producers must be able to use their unique ability to act independently from the political process.

The pandemic exposed gaps in available data. To ensure that statistics best serve the public good, these gaps must now be filled.

What we found during October 2021-October 2022:

One of the main ways in which statistics can serve the public good is for them to help people to understand the world around them, providing the information they need about the issues they care about. In order for this to happen, statistics need to be available at the right time and be shared in a transparent way. A lack of transparency and data gaps can both impact the ability of statistics to enhance public understanding on important topics.

Intelligent transparency

Last year, in response to recurring issues with transparency identified through our casework, we launched our campaign for intelligent transparency in statistics and data. Throughout this year (October 2021-October 2022) we have continued to intervene both publicly and privately through our casework where lack of transparency is an issue – for example, on the use of unpublished statistics about the cost of the UK’s Test and Trace programme, vaccination status of COVID-19 patients in critical care beds in Wales and COVID-19 hospital admissions for children in Scotland. It is important to note that in these cases, producers were quick to recognise and respond to the problem. We have found that there is a better understanding of the need for intelligent transparency and that producers are being champions for it within their organisations. Producers now have good processes in place for the quick release of analyses, often as a result of the demand for this during the pandemic. We have also had feedback that our regulatory guidance on transparency, published in February 2022, has supported producers in their efforts to achieve intelligent transparency. We recently published a blog and answers to Frequently Asked Questions to support others to advocate for intelligent transparency across government data, statistics and wider analysis.

There are many good examples of new health and social care data being published over the last year, supporting transparency on topics of high public interest. For example, there have been new publications on antiviral drugs for vulnerable people in Wales, bed availability in England and the ambulance service in Scotland. Our rapid review of the Scottish Ambulance Service (SAS) statistics published in June 2022 highlighted that “These statistics were first published in November 2021 in response to high public interest on the topic… By publishing these statistics SAS is supporting transparency, enabling individuals and organisations to reach informed decisions and answer important questions”.

Intelligent transparency is about supporting an open and accessible approach to communicating numbers. It is achieved by following three core principles:

  • equality of access (data and statistics quoted publicly are made available to all)
  • enhancing understanding (data and statistics enhance understanding of societal and economic issues)
  • analytical leadership (decisions about the publication of statistics and data are independent of political processes)

These principles are applicable to official statistics, as well as other types of analysis such as research, management information and evaluations.

The pandemic exemplified why intelligent transparency is so important. There was a huge public demand for data and statistics about COVID-19 – probably best seen in the over 1 million users per day experienced during peak times by the UK Health Security Agency’s COVID-19 dashboard (created in 2020 by Public Health England). The statistical system’s response to meet the increased demand for information was impressive – there were lots of examples of publishing data and analysis of high public interest in an open and insightful way, including the UK dashboard and similar dashboards published in all four nations of the UK. The UK dashboard was the 2022 winner of our Award for Statistical Excellence in Trustworthiness, Quality and Value, awarded in partnership with the Royal Statistical Society and Civil Service World.

The benefits of intelligent transparency are clear. Firstly, it builds trust in the organisations which produce data and statistics. While it focuses on the Office for National Statistics (ONS), rather than the whole statistical system, the 2021 study by NatCen provides an indication of public trust in producers of official statistics, with 79% of respondents agreeing that COVID-19 statistics are accurate and 87% of respondents trusting ONS statistics. The main reason people gave for trusting ONS was that ONS did not have a vested interest in the results – this independence is at the heart of analytical leadership, one of the three core principles of intelligent transparency.

Secondly, transparency ensures that public conversations focus on the important issues. During the pandemic we found that a lack of transparency often resulted in confusion about where numbers came from or accusations that governments cherry pick or manipulate data. We saw examples of these kind of issues in the media and on social media, as well as through our casework, for example on testing numbers and modelling during the Omicron wave (see the case study below).

Finally, making data available also allows independent analysis to be carried out by others, enhancing its value. For example, the UK COVID-19 dashboard allows users to easily download the data or to access it via an API (application programming interface) in order to carry out their own analysis. During the pandemic this resulted in excellent visualisations and additional insights being created from the data by people outside of the statistical system – we blogged about some of our favourite examples in August 2022. And there have been many other examples of the value of making data available for reuse, such as the Travelling Tabby Scotland Coronavirus Tracker. The site was created and run by John Frace, a student at Argyll College, and was commended by Professor Linda Bauld, Professor of Public Health at the University of Edinburgh, who said of the site’s creator: “With his website he has provided easily accessible analysis and visualisation of coronavirus statistics from Scotland and the UK, alongside local data and international comparisons”.

Case study: Supporting public understanding of the number of Omicron infections

On 13 December 2021, the Secretary of State for Health and Social Care delivered a statement to parliament saying that “The UK Health Security Agency estimates that the current number of daily infections are around 200,000”. At this time, the UK COVID-19 dashboard showed that the number of cases was around 45,000. The public were familiar with these figures, which were regularly covered by the media and used in press briefings, and so were confused when they heard the 200,000 figure – had cases suddenly jumped four-fold? Where did this new figure come from and what did it mean? The answer was that the two figures measured different things. The 200,000 figure was a modelled estimate of the number of Omicron infections in the population. This included people who had just been infected but didn’t know that yet. The number presented in the dashboard was the number of positive cases reported through the testing programme. But, based solely on the statement, it was hard for people to know this.

In response to this confusion, UKHSA worked quickly to publish an explainer about the 200,000 figure on 16 December. This included detail about underlying assumptions, data sources and limitations of the calculation. UKHSA also quickly published a daily overview of Omicron cases, hospitalisations and deaths, starting on 14 December. On 17 December we wrote to UKHSA, a relatively new organisation established in October 2021, welcoming the publication of data to support the statement made by the Secretary of State. We acknowledged the commitment UKHSA had shown to share data to support public understanding in such a fast-moving environment.

UKHSA reflected afterwards that issues like this can distract from the key public health message and issues, and can risk delaying buy-in from the public if trust is undermined. UKHSA also identified that ad-hoc releases can be a useful way to provide short explanations to support figures used by ministers. UKHSA now has processes in place for rapidly producing ad-hoc releases, both for proactive and reactive releases.

Data gaps

The ability to inform the public on issues which matter to them can be hampered by the absence of statistics and data, as well as issues with transparency. Our 2021 report highlighted critical data gaps exposed by the pandemic in social care, mental health and on protected characteristics, particularly ethnicity. Users continue to raise these data gaps with us, as well as gaps in other areas such as independent healthcare and healthcare spending. We have also used our voice over the last year to raise the ongoing issue of data gaps – for example, discussing gaps in social care, primary care and ethnicity data in an evidence session with the Scottish Parliament’s COVID-19 Recovery Committee in May 2022.

We are encouraged to see ongoing efforts to fill many of these gaps by producers across the UK. For example, on social care, individual level data is now being collected in England from early adopters in the adult social care sector. NHS Digital is working towards mandating this dataset as of April 2023 and aims to use this as the source of official statistics in future. In Wales, a new data collection is planned in which each social care provider will be required to submit an annual return. This requirement is set out in legislation under the Regulation and Inspection of Social Care (Wales) Act 2016, with both the Act and The Regulated Services (Annual Returns) (Wales) Regulations 2017 specifying the information which must be collected. While the pandemic delayed the rollout of returns, these will commence from April 2023 and will provide insight into staffing, the number of people supported, charges and accommodation facilities. The Act also requires the publication of annual returns and these data will also be made publicly available at an individual service level. Finally, the Scottish Government is working with Public Health Scotland and other partners to carry out a review of nationally collected care home data during 2022/23. The aims of the review are to ensure a coherent suite of data collections, reduce the burden on data providers and meet the existing and emerging needs of data users. A key element of the project will be identifying data issues and gaps, including in relation to equalities data, to inform how best to collect this information. The care home data review forms one strand of a wider review of the entire social care data landscape in Scotland, in light of the Scottish Government’s policies on health and social care integration and the establishment of a National Care Service.

On mental health, NHS Digital has started to publish an interactive dashboard which presents time series data from the Mental Health Services Monthly Statistics publication. This dashboard was developed using a user centred design approach meaning that views were gathered from a range of sources including submitters of data, charities, media and journalists, as well as policy leads. Alongside the dashboard development, NHS Digital has developed its demographic reporting for mental health statistics, including the reporting of some characteristics for the first time, such as gender, age, ethnicity and deprivation. Whilst other newly published breakdowns, such as accommodation status, employment status, disability and sexual orientation, do not have high data quality at present, NHS Digital hopes to work with providers to improve data quality in future. In Northern Ireland, following our review of mental health statistics, the Mental Health Outcomes Framework Group has made progress in identifying the data needed to support Outcomes Based Accountability reporting on mental health services. The Framework Group will make recommendations on how data should be gathered at source, particularly for the community sector, and is likely to recommend that the changes needed to collect robust and consistent activity data should be taken forward alongside the roll-out of a new administrative system in Northern Ireland. In the shorter-term, the Department of Health in Northern Ireland is focussing on Child and Adolescent Mental Health data, with the aim of publishing statistics about this service.

On ethnicity, during the pandemic NHS Digital created an ethnic category information asset to provide a near population level view of ethnicity in England for COVID-19 planning and research. This approach combines data from general practice and hospital episode statistics to provide 94.8% coverage for the most recent data collected. NHS Digital also includes breakdowns by ethnic group in its publications on maternity services and activity, adults receiving long-term social care support and on the use of Psychological Therapies for treating people with depression or anxiety. Similar breakdowns by ethnic group are available in the excess mortality reports published by the Office for Health Improvement and Disparities. As noted in the section on horizon scanning, once new operational data streams are underway, new data should become available to enable producers to publish statistics which continue to close existing data gaps.

Back to top
Download PDF version (313.64 KB)