Diabetes is a serious illness that occurs when the body cannot regulate its blood sugar levels. Whilst it is commonly thought that a family history of diabetes is the most important risk factor, being exposed to viruses can also trigger this disease. This is because viruses trigger immune responses in the body, and in some cases, these immune responses can attack healthy cells in the body. In the case of diabetes, immune cells can attack the pancreas, the organ which manages the amount of sugar in the blood. This results in high blood sugar levels, which over time can lead to damage to the eyes, kidneys, nerves and blood vessels.
The recent SARS-CoV-2 (COVID-19) pandemic raised concerns about the possible development of post-COVID complications that may be triggered by the body’s immune response. One such concern is the development of new onset diabetes. Therefore, the aims of this study are to:
1. Evaluate the association between SARS-CoV-2 and new onset diabetes
2. Determine whether associations between SARS-CoV-2 and new onset diabetes are moderated by age and/or sex
Finding associations between SARS-CoV-2 and new onset diabetes will increase our understanding of both diseases, and alert clinicians to monitor patients at high risk of disease.
Two studies will be carried out using different methodologies to address these aims. The first is a cohort study where the primary exposures are a diagnosis of COVID-19 or a positive SARS-CoV-2 test result, and the comparator is the absence of a diagnosis of COVID-19 or absence of a positive SARS-CoV-2 test result. The outcomes are new onset T1D or new onset T2D.
The study population will be all people in CPRD GOLD with >1 year data availability without diagnosis/medication of/for T1D/2 any time before index date and without previous SARS-CoV-2 infection (identified by positive PCR test). The observation period will start at 01/01/2021 and end at end of data availability. The exposure and comparator cohorts will be matched (1:1) with months as “enrolment” windows during 2021, without replacement.
Large scale propensity score matching methods will include age, sex, obesity, and number of healthcare visits forced into the model. The statistical analyses will encompass cohort diagnostics for the outcomes and exposure/comparator cohorts; propensity score distribution and covariate balance after matching; summary descriptive statistics for baseline characteristics for matched cohorts at index date; Cox-proportional hazard regression models for both outcomes.
The second is a Self-Controlled Case Series (SCCS) where the primary exposures are a diagnosis of COVID-19 or a positive SARS-CoV-2 test result (compared to patient time pre-COVID). The outcomes are new onset T1D or new onset T2D. The study population will be all people in CPRD GOLD with >1 year data availability and without diagnosis/medication of/for T1D/2 any time before study start (01/01/2021) and without previous SARS-CoV-2 infection before study start (01/01/2021). The statistical analyses will encompass cohort diagnostics for the outcomes and exposure/comparator cohorts; summary descriptive statistics of the study population; incidence rates and incidence rate ratios (IRR) calculated for pre-COVID and post-COVID patient time.
A new onset diagnosis of T1D or T2D.
Martí Català Sabaté - Chief Investigator - University of Oxford
Pablo Spivakovsky Gonzalez - Corresponding Applicant - University of Oxford
Annika Jodicke - Collaborator - University of Oxford
Antonella Delmestri - Collaborator - University of Oxford
Nicola Barclay - Collaborator - University of Oxford