ONS Health Index beta webinar - report

On 11 February 2021 the RSS Official Statistics Section (OSS) organised a webinar presenting the ONS Health Index beta. In December 2020, the ONS published a beta version of the Health Index, which is a new statistic measuring a broad definition of health, and how health differs both over time and between geographical areas. The Health Index aims to fulfil the previous Chief Medical Officer, Dame Sally Davies’, wish for a composite index to track health and healthiness over time, and reflect the multi-faceted determinants of the population’s health.

The Health Index presents a single headline measure of health overall, but can be disaggregated into separate domains, allowing trends in particular aspects of health to be observed and compared. The ONS has launched a public consultation alongside the publication of this beta version of the Health Index, to ensure this measure can suit the needs of users from a variety of backgrounds.

The meeting was chaired by the RSS OSS Chair Alison Macfarlane. The webinar was split into two segments, the first represented by our two ONS speakers and the second half moved on to the user perspective and Q&As.

Download slides for the webinar (PPTX)


  • Myer Glickman - ONS (Office for National Statistics)
  • Greg Ceely - ONS (Office for National Statistics)
The following is a general overview of the presentations by both speakers. They explain how the Health Index was developed; what it adds to the health data landscape; the provisional findings from this first version; and some of the possible ways in which the Index could be used and developed further in future.
Introduction to the Health Index and its purposes (Myer Glickman):
  • The background to this index comes from Dame Sally Davies (former Chief Medical Officer) mentioning in her 2018 annual report that we need a new health index that will include not only the measurement of health outcomes, but also factors that are known to contribute to health outcomes.
  • Approach 1: Example of a source of health data is PHE’s Fingertips tool, and these types of tools are very valuable and provide lots of information and data, and have this at local level too. So, this index will not be replacing these valuable tools.
  • Approach 2: Then there are existing indices like GDP are what the public sees and government and decision makers use and they are expressed in a simple way, i.e either will go up or down and it does not matter that the public does not understand how to calculate GDP, as long as they know that when it is up it is good and when down it is bad and so can assess what it means for the country.
  • What the ONS Health Index is going to do is take the best of these two approaches, so to present and see headline measures as indicators of change and inequality. This will be something simple for the media and public to understand. It will enable policymakers to clearly identify areas for improvement and measure health impacts consistently. For the users of the data like analysts, academics, think tanks it will help them with their interest of working with health data and seeing how it improves overtime
  • Developing the Health Index: there were many people involved in the development as Dame Sally Davies had advised, these were Expert Advisory Group featuring central and local government officials, academics and health-related organisations
  • The beta version was created to test the index and highlight any improvement before the Health Index goes live
  • The structure of the Health Index is a concept of wider determinants of health, and so not just measuring health but measuring healthiness in a broad sense and feeding into a structure and interpretation of the index. The below diagram explains the different levels and structure of the index.

Methodology and data visualisation (Greg Ceely):
  • Data selection: there was a comprehensive review of the data landscape across concepts within definition of health e.g. WHO and Dahigren & Whitehead Model. Having a time series (beta version looking at 2015-2018) was important to compare over the years. It was important to make clear a movement in one direction or another was indicating a health improvement or not for health as a whole. The index is available at sub-national level (beta version is at UTLA level), making it easier for people to compare and understand inequalities in the country, feedback received from users has indicated they would like lower geographies and so this is being explored at the moment. The index will be measuring health rather than health services. The data included in the Health Index is publicly available data for now due to time constraints, but in the full version there is exploration into including other type of data. There are 58 indicators total in the beta version.
  • Methods: the index needed to have a national figure tracked over time, scores for lower geographies which can be compared to the national figure and to other geographies (relative and absolute terms) and allow relative and absolute changes in those lower geographies’ scores to be trackable over time. The example of the IMD index was given here as it does not allow for this. Imputation was used for missing values in UTLA’s. Data was scaled so that a value increasing meant that health was improving, and so all the indicators had a consistent direction. Skewness and kurtosis in the indicators were addressed by functional transformations (e.g. log transformation). Time series standardisation was used by calculating the mean and SD for each indicator for a base year (for beta version 2015) and then apply those mean and SD for the entire time series. These were aggregated to England values using population size for each UTLA.
  • Domains split the index into three broad areas:
    • Healthy People – focusing on health outcomes such as life expectancy, health conditions and personal well- being.
    • Healthy Lives – health-related behaviours and personal circumstances such as smoking, alcohol misuse, unemployment and working conditions
    • Healthy Places – wider determinants of health, environmental factors including access to green space, air pollution and access to amenities (such as housing, GP surgeries)
  • To explore the Health Index beta further please refer to https://healthindex.lcp.uk.com/
Next steps, and questions for you:
  • The speakers would like to get some key insights from you:
  1. What do you think of the concept of a health index?
  2. How likely are you to use the Health Index for your own analysis?
  3. Which elements of its structure, data and content, or methodology could improve for you to be more likely to use it?
  4. How would you want others to use the Health Index?
  5. Would you be interested in a Health Index for lower geographies, or more frequently than annual, at the expense of some indicators?

The session then moved onto a Q&A and discussion which can be read in the following document (DOCX). 

Overall, the webinar was very informative and the contributions from each speaker were well received, especially since this is a very busy period for them and therefore their time was much appreciated.

Hira Naveed is the RSS Official Statistics Section (OSS) meeting secretary.

Load more