CCU040: Investigating why some people with diabetes have a greater risk of becoming seriously unwell or dying with COVID-19

Project lead:
Adrian Heald, University of Manchester

Diabetes is a condition affecting approximately 4 million people in the UK. People with diabetes have high levels of blood glucose, which if unmanaged, can lead to serious damage to the heart, feet, eyes and kidneys. Previous research has shown that people with diabetes are more likely to become seriously unwell, or die, following a coronavirus infection. However, the exact reasons for that are still largely unknown.

We will report on how much more likely a person with diabetes is to become unwell or die following a COVID infection. We will also look at what factors (such as weight, BMI, blood glucose level, age, sex, ethnicity etc.) affect the likelihood of poor outcomes in patients with diabetes and a COVID infection. Analysing these factors in patients with and without diabetes, will help us to unpick to what extent the worse outcomes are related to the diabetes diagnosis, as opposed to the underlying factors such as BMI. Type 1 and type 2 diabetes patients will be analysed separately as these are different conditions and the underlying factors may not be the same.

We recently submitted findings from the city-wide Greater Manchester Care Record (GMCR) database that include identification of specific risk factors for people with type 1 and type 2 diabetes in relation to serious illness or death following a COVID-19 infection. We now wish to extend the analysis from the GMCR data to a much larger dataset. The findings of this study will inform policy decisions at a national level in relation to reducing the risk of people with diabetes of all types becoming seriously ill after contracting COVID-19, as we move out of the pandemic to a ‘status quo’ of long-term coexistence with the virus.

Outputs

Sars-Cov-2 infection in people with Type 1 diabetes and hospital admission: an analysis of risk factors for England

  • Diabetes Therapy publication 25/08/23 can be viewed here
  • Code and phenotypes used in this study are available here

The challenges of replication: a worked example of methods reproducibility using electronic health record data

  • Paper submitted to a journal (decision pending)
  • Preprint 07/08/24 can be viewed here
  • Code and phenotypes used in this study are available here

Replicating a COVID-19 study in a national England database to assess the generalisability of research with regional electronic health record data

  • Paper submitted to a journal (decision pending)
  • Preprint 08/08/24 can be viewed here
  • Code and phenotypes used in this study are available here

Projects