Diabetes is a long-term condition characterized by high blood sugar levels which can lead to health issues. We will use information from routine blood sugar level tests (HbA1c) over time and study how it is related to the risk of developing vascular complications (i.e., heart attack, stroke) and death in people with type 2 diabetes. We want to see how the full profile of glucose control over many years, and not just a single measure of glucose control, is associated with subsequent health issues and death. More fundamentally, we also want to find out what the risks of having these health issues is in people with diabetes, and how having had a complication, such as a stroke, changes the risk of having further problems. This study may help clarify the long-debated relationship between glucose control and vascular complications (particularly of large vessels) and help doctors tailor glucose-lowering medications.
The current study will investigate the longitudinal relationship between HbA1c and complications and mortality, map out the natural history of macrovascular complications, and assess the impact of prior or repeated events on the risk of future events in patients with type 2 diabetes (T2DM). Using CPRD data linked to HES APC and ONS, we will investigate the association between repeated HbA1c measurements and diabetes-related complications (myocardial infarction, stroke, peripheral vascular disease, heart failure, chronic kidney disease) and mortality (all-cause and cause-specific mortality) in patients with T2DM using a joint modelling of longitudinal (i.e., HbA1c) and survival (i.e., time to complications) data in Stata and R {1, 2}. Key confounders of the associations, such as age, sex, ethnicity, glucose-lowering medications, and comorbidities at or before the diagnosis of T2DM will be accounted for while determining the relative and absolute risk of complications related to the longitudinal values of HbA1c. Furthermore, we will use repeated events analysis (recurrent events) and multistate models to describe the total burden of diabetes complications and the natural history of transitions across diabetes complications, respectively. Multi-state models permit estimation of state-specific hazards of such event to help predict risk of macrovascular complications in patients with different past histories of complications.
Primary (cardiovascular) outcomes: hospitalisation for heart failure and 3-point MACE (major adverse cardiovascular events) defined as nonfatal stroke, nonfatal myocardial infarction and cardiovascular death.
Secondary (non-cardiovascular) outcomes: non-cardiovascular death, all-cause death.
David Kloecker - Chief Investigator - Leicester Diabetes Centre
David Kloecker - Corresponding Applicant - Leicester Diabetes Centre
Atanu Bhattacharjee - Collaborator - University of Leicester
Francesco Zaccardi - Collaborator - University of Leicester
Kamlesh Khunti - Collaborator - University of Leicester
Melanie Davies - Collaborator - University of Leicester
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation