Providing a better understanding of suicides among higher education students

People talking in support group

Today, the ONS has published updated statistics on suicides among higher education students using improved methods to provide more accurate information. In this blog David Mais summarises these changes and the steps we have taken to assure their quality.

A recent national review of higher education student suicide deaths warned of a rising number of students taking their own life and provided recommendations for tackling this serious issue. Reducing suicides in university students is also highlighted in the Suicide Prevention Strategy for England: 2023 to 2028. ONS statistics on higher education student suicides are critical to monitoring progress against these commitments and guide decision making amongst those working to improve student mental health.

Getting these statistics right is important but comes with challenges, as no information on higher education students is provided through the death registration process in England and Wales. As such, to produce these figures we have developed unique methods to link HESA (Higher Education Statistics Agency) data and death registration data.

A new approach

We have previously published figures on higher education suicides in 2018 and 2022. In these previous releases, we worked to link death registration data and HESA data. Names (first names and surnames, and middle names where available), date of birth, sex, postcodes and academic year/year of death were the variables used to link the two datasets. Based on these variables, a series of rules were developed, each consisting of a set of criteria that must be met to make a link . By cross-referencing these data sources, we were able to identify that there were approximately 100 cases a year of a higher education student taking their life.

However, our current linkage has been improved by using the Demographic Index (DI). The DI is part of the Reference Data Management Framework (RDMF) which is a set of reference data that enables the ONS to connect information with data from other departments to produce analysis. The DI integrates education, health, and tax and benefit administrative data to provide a composite data source of the population interacting with administrative data sources. It draws together records of individuals, and uniquely identifies them via an Entry ID. By individually joining HESA data and death registration data to the DI using HESA student ID and NHS number respectively , we were able to directly link these two sources together, as a unique entry ID from the DI was then common across the two sources.

Testing techniques

To ensure the quality of our linked results, we also manually checked all links for false positives – cases where our approach had identified a suicide that was not a higher education student. This review found that only 1% of records were likely to be false positives, which were removed from our analysis. After this cross-examination, we were able to identify that there were approximately 160 cases a year of a higher education suicide.

The figure of around 160 cases per year is higher than the 100 cases per year identified by the previous analysis. To confirm that this reflects our improved methods being able to identify additional cases, rather than a substantial increase in the number of higher education student suicides, we re-examined data from a previous analysis using our new method. When we compared the results, we discovered that our new techniques had allowed us to identify many additional cases across the whole period covered by the previous analysis (1 August 2016 to 31 July 2020) with nearly all the records that were linked previously also linked using our new method. This comparison has given us high confidence that we have linked more records than were linked in the previous analyses due to better linkage methods, rather than reflecting a trend in higher education student suicides. It is therefore important not to make comparisons between our updated analysis and previous releases.

Looking ahead

We will continue to explore further improvements to linking death registration data and HESA data going forward to ensure we can continue to provide robust statistics on higher education student suicides. We worked collaboratively with the Department for Education and Office for Students to deliver this analysis, and we will continue to work with them and other users to provide the support needed to understand these statistics. We are also continuing to work with policymakers, stakeholder groups, and other users to ensure they are useful statistics that can help people working in higher education suicide prevention.

David Mais

David Mais is Head of Mortality Statistics at the ONS