Risk factors for overtreatment and undertreatment of glucose-control in patients with diabetes mellitus and association of overtreatment with mortality

Date of ISAC Approval: 
Lay Summary: 
In healthy subjects, blood sugar levels (also known as glycaemia, the amount of glucose present in the blood) are stable. People with diabetes have higher glucose levels compared to healthy subjects (a condition defined hyperglycaemia). Hyperglycaemia for long time may cause complications to organs, such as problems to the heart and blood vessels (i.e., myocardial infarction). Therefore, it is important to reduce the concentration of blood glucose with appropriate interventions (for example, with oral glucose-lowering drugs or insulin) to reduce the risk of complications. However, an over-controlled blood glucose is also dangerous as it could result in some side effects, the most common and important being hypoglycaemia (very low glucose levels). Hypoglycaemia is also associated with complication, including as loss of consciousness, abnormalities of the heart function, and coma, all of which can result in death. It is important then to understand which people may be at higher risk of hyperglycaemia (under controlled) or hypoglycaemia (over controlled) with diabetes drug treatments. Using information from general practices in England, we will identify the types of characteristics that patients with low or high blood glucose levels have. We will also assess the risk of death in patients taking too many diabetes medications.
Technical Summary: 
Using data collected in the CPRD Gold database with linkage to HES admitted patient care and ONS death registration, the primary aim of the study is to characterise patients with type 2 diabetes mellitus at risk of overtreatment and undertreatment. Secondly, the characteristics associated with the risk of future outcomes will be identified in overtreated patients. In particular, this study will include a retrospective cohort of patients with a diagnosis of type 2 diabetes mellitus aged over 30 years old and risk prediction models will be developed to estimate the probability for a patient to be overtreated with glucose-lowering medications (3 consecutive values of HbA1c <7% in patients on any insulin or any sulphonylurea and over 70 years old) and undertreated (3 consecutive values of HbA1c >8% in patients on any glucose-lowering drug and younger than 70 years old). In addition, in overtreated subjects the characteristics associated with cardiovascular mortality and all-cause mortality will be assessed.
Health Outcomes to be Measured: 
- Cardiovascular mortality - All-cause mortality

Dr Francesco Zaccardi - Chief Investigator - University Of Leicester
Claire Lawson - Collaborator - University Of Leicester
Dr Francesco Zaccardi - Corresponding Applicant - University Of Leicester
Professor Kamlesh Khunti - Collaborator - University Of Leicester
Professor Melanie Davies - Collaborator - University Of Leicester
Dr Suping Ling - Collaborator - University Of Leicester

HES Admitted;ONS;Patient Townsend