Using the power of linked data to understand factors preventing people from working

Crowds of people walking through a busy street

Nearly 9 million working-age people in the UK were not working or looking for work – that is, economically inactive – from May to July 2023. This figure includes more than 2.5 million people inactive due to long-term sickness, an increase of half-a-million people since before the COVID-19 pandemic. Emma Rourke explains how linked, population-level data can improve our understanding of the interplay between health and work, with the goal of improving the wellbeing of individuals and the economy.

The Office for National Statistics (ONS) is working with the Department of Health and Social Care (DHSC) and Department for Work and Pensions (DWP) Joint Work and Health Directorate to understand how the coronavirus (COVID-19) pandemic has had an impact on people’s health, their ability to work, and the social security benefits they received. Our work will assess if, and how, this has varied for people in different socio-demographic groups (for example, by age, sex, ethnicity, region, and level of deprivation), disabled people, and those with certain pre-existing health conditions (such as cardiovascular and respiratory diseases).

Survey data have shown that historically, there was a narrowing employment gap between disabled and non-disabled people, but this trend was initially reversed at the start of the COVID-19 pandemic. However, there is a limit to what data from sample surveys can tell us, and linked administrative data covering the majority of the working-age population of England has the potential to provide more detailed insights.

This new evidence may inform policies aimed at removing employment barriers for people with disabilities or health conditions who wish to stay in or return to work, reducing societal inequalities, and the possible expansion of relevant healthcare services. At an individual level, this may contribute to improved health, wellbeing and income among disadvantaged groups. At a macroeconomic level, it could contribute to our understanding of the socio-demographic and health drivers of increased economic inactivity in the labour market, ultimately supporting economic growth.

Secure data linkage and analysis for the public good 

In its report on data sharing and linkage for the public good, the Office for Statistical Regulation (the independent regulator of official statistics in the UK) highlighted that “within government, data can inform the delivery of vital public services, policy developments, evaluation, and answer valuable research questions”. Furthermore, “when data are shared and linked across government this potential is magnified, enriching insights into society, stimulating innovation and ultimately enabling data, and government, to better serve the public good.”

To undertake this work, ONS is combining NHS healthcare data and DWP benefits records with internally held Census 2011 data and death registrations. This process of data linkage will allow us to draw insights from the combined data that wouldn’t be possible from any of the individual datasets on their own.

We do all this with our commitment to data security and protecting the confidentiality of individuals at the forefront of our minds. All the datasets have been de-identified, meaning information that can be used to directly identify individuals such as names, addresses and NHS numbers have been removed before the datasets are combined and analysed. In line with the Code of Practice for Statistics, the de-identified linked data will only be used for statistical production and research; it will never be used for operational purposes, such as making decisions over individuals’ access to healthcare or benefits. This work has been assessed by the National Statistician’s Data Ethics Advisory Committee as being ethically sound and for the public good, and we will only use the data shared with us in ways agreed with supplier organisations.

Datasets are shared with ONS via secure means, and they are then stored, processed and analysed in a secure virtual environment only accessible by ONS staff who have undergone government security vetting and relevant data security training. Staff from outside ONS, including in other government departments and the private sector, do not have access to this secure environment. Within the environment, analysts never have access to personally identifiable information needed for data linkage (such as names and addresses) alongside detailed attribute information (such as health conditions and financial records), meaning analysts can never directly identify individuals in the linked data. Research outputs can only be removed from the secure environment once disclosure control procedures have been applied, mitigating against individuals being identifiable.

Taking the work forward 

We will publish our first experimental results on 6 December 2023, describing how trends in benefit recipients varied across socio-demographic characteristics and health conditions over the first year of the COVID-19 pandemic. This release is intended to be a “proof of concept” illustrating the potential power of linked health and labour market data to generate insights for the public good.

However, we recognise that much more can be done in this area. Moving into 2024, we’ll be incorporating other datasets into our work programme, including Census 2021, NHS hospital records, and Pay-As-You-Earn (PAYE) data from His Majesty’s Revenue and Customs (HMRC). This will enable us to piece together a more rounded view of individuals’ socio-demographic, health and labour market experiences, including exploring the relationship between health conditions and employment status (including economic inactivity) and the effect of Long Covid on the labour market.

We have also been commissioned by central government to use this linked data resource to evaluate the labour market outcomes of health interventions such as weight loss surgery, the NHS Diabetes Prevention Programme and NHS Talking Therapies, and of receiving an endometriosis diagnosis.

We are interested in your views on our work programme using linked data for analysing health status and the labour market. If you have any questions or comments on our proposals, please email us at quoting “Work and Health” in the subject line.

Emma Rourke is deputy national statistician

Emma Rourke is Deputy National Statistician