Understanding the association between diabetes and severe COVID-19 infection

Date of ISAC Approval: 
Lay Summary: 
It appears that people with diabetes are more likely to become severely unwell with COVID-19. The reasons for this are unclear, although early research has shown that this is not entirely explained by the fact that people with diabetes are more likely to be older and have other health conditions. This means it is likely that something about having diabetes itself is making people more at risk. We will study a large number of people with diabetes, to gather evidence on the characteristics of people with diabetes who get very unwell with COVID-19. We will look in detail if the type of diabetes a person has, and how severe the diabetes is, is linked to the risk of developing severe COVID-19. We will explore the influence of, in particular blood sugar levels and routine blood tests, diabetes duration, medication use, and diabetes complications. Finally, we will describe the longer-term health of people with diabetes who recover from COVID-19, up to one year after infection. For the study we will make use of routinely collected data from GP practice to provide very detailed information on people with diabetes, in combination with hospital care records and national death records. Findings will be shared rapidly to help guide care for people with diabetes.
Technical Summary: 
Diabetes is one of the most common comorbidities seen in people with COVID-19. As the pandemic has progressed, worse prognoses for people with diabetes who contract COVID-19 have been reported in population-based studies, with an increased risk of intensive care admission and higher mortality in the majority of hospital cohorts. This study will use Clinical Practice Research Datalink (CPRD) primary care records, linked to Second Generation Surveillance (SGSS) data for COVID-19 testing, and COVID-19 hospital records (Hospital Episode Statistics (HES), COVID-19 Hospitalisation in England Surveillance System (CHESS), Intensive Care National Audit and Research Centre (ICNARC)). In people with diabetes we will identify those infected with COVID-19, and those who develop severe COVID-19. We will examine diabetes specific clinical prognostic factors for development of severe COVID-19 (a composite endpoint of mortality, admission to hospital critical care, ventilation, renal dialysis/filtration, or high-low oxygen), in particular blood glucose control and routine blood tests, diabetes duration, diabetes complications and medication. Advanced methods to control for measured and unmeasured confounding, including the Prior Event Rate Ratio method, will be used to evaluate specific prognostic factors, in particular medication use. We will also examine longer-term outcomes, up to one year, in people with diabetes who recover from COVID-19 infection, compared to a matched control population of COVID-19 unexposed people with diabetes. The study will provide robust data on whether potentially modifiable risk factors such as hyperglycaemia and medication alter risk of developing severe COVID-19, and establish if there are subgroups of people with diabetes that differ to a clinically relevant degree in COVID-19 outcome. We will work with Diabetes UK, the study funder, to rapidly disseminate findings.
Health Outcomes to be Measured: 
Primary outcome: Severe Covid-19 infection identified, in patients with confirmed COVID-19 infection, using a composite of: All-cause death, or admission to a critical care unit, or mechanical ventilation, or non-invasive ventilation, new commencement of renal dialysis/filtration, or requirement for high flow oxygen, within 30 days of COVID-19 infection. Secondary outcomes: • Admission to hospital with COVID-19, defined by either a new hospital admission over the study period, and either a confirmed COVID-19 infection or ICD-10 code for confirmed or suspected COVID-19. • Admission to critical care with COVID-19, defined by either a new hospital admission over the study period, and either a confirmed COVID-19 infection or ICD-10 code for confirmed or suspected COVID-19. • Confirmed or suspected death from COVID 19, as recorded on death certificate or death occurring in a patient with confirmed COVID-19 infection. • Long-term health outcomes in people with diabetes who recover from COVID-19 infection.

John Dennis - Chief Investigator - University of Exeter
Andrew McGovern - Collaborator - University of Exeter
Andrew Hattersley - Collaborator - University of Exeter
Bilal Mateen - Collaborator - Wellcome Trust
Beverly Shields - Collaborator - University of Exeter
Jack Bowden - Collaborator - University of Exeter
John Dennis - Corresponding Applicant - University of Exeter
Kashyap Patel - Collaborator - University of Exeter
Katherine Young - Collaborator - University of Exeter
Nicholas John Thomas - Collaborator - University of Exeter
Spiros Denaxas - Collaborator - University College London ( UCL )
Sebastian Vollmer - Collaborator - University of Warwick
William Henley - Collaborator - University of Exeter