University of Edinburgh

DL_2022_035

Diabetes is a common condition that affects more than one in 20 people in the general population and about one in 5 people who are in hospital. Early in the Covid pandemic diabetes was found to be an important risk factor for being infected, being admitted to hospital and having severe illness (admission to an intensive care unit or dying in hospital). This project will investigate whether the increased risk has changed over time, taking other important factors (such as age) into account. Another aspect of this project is to check how well a diagnosis of diabetes is recorded in different health records by comparing information collected by GPs and hospitals. If we find that recording is reliable this will be extremely helpful to support quality improvement programmes in the health service and achieve better outcomes for people with diabetes.

Although it is clear that diabetes is a risk factor for Covid infection and for severe disease (including hospital admission and death) it is not clear whether the risk relative to people without diabetes has changed during the pandemic. The null hypothesis is that there has been no change and the alternative hypothesis is that the excess risk reduced over time. This project also provides the opportunity to assess the ability to reliably identify diagnoses of diabetes in routine data. This is not only relevant for assessing the limitations of the current project but also allows cross-validation of different components of data on diabetes diagnoses held by DataLoch and the potential to conduct a capture-recapture study if necessary. The overall proportion of people with a diagnosis of diabetes in the UK population of all ages is 6-7% but 17-25% of hospital-inpatients have diabetes and these proportions are increasing. The presence of diabetes therefore has implications for individuals and health services and better understanding of the association both in relation to Covid and to other co-morbidities will improve motivation and resource allocation for primary and secondary prevention of diabetes. The ability to use routinely collected data would be extremely helpful to improve the quality and representativeness of audit and research beyond what is already possible using SCI-diabetes data.

Public Health Research

27/10/2023

De-Personalised

TRE