Blood sugar levels can be classified in three categories: normal, elevated, and high. High blood sugar level is an indication of type 2 diabetes. While elevated blood sugar level is an indication of prediabetes, a condition which increases the risk of progression to type 2 diabetes. Elevated and high blood sugar are also risk factors for diseases involving the heart or the blood vessels of the heart or of the brain, and for early death.
In patients with prediabetes, 30 minutes per day of moderate exercise such as walking, combined with weight loss can reduce the blood sugar level. The general practitioners will encourage patients to increase their physical activities and to lose weight. However, in some patients, this might not be enough to reach a normal blood sugar level. In this case, metformin can be prescribed to help reducing the blood sugar level back to normal. However, in the studies that have proven the benefit of metformin, only a small number of patients were aged above 60 years old.
Thus, our aim is to show that, in prediabetic patients above 60 years-old, metformin in top of advice from the general practitioner for exercise and weight loss, can reduce the risk of progression to type 2 diabetes, increase the chance to get back to normal blood sugar level, decrease the risk of diseases involving the heart or the blood vessels of the heart or of the brain, and decrease the risk for early death.
This study will focus on patients above 60 years-old at high risk of diabetes due to elevated blood glucose, namely prediabetic patients. More especially, it will compare prediabetic patients that initiate metformin on top of lifestyle advice (MET+LSA), as opposed to prediabetic patients that have carried on lifestyle advice (LSA) only.
Both cohorts will be compared in terms of risk of type 2 diabetes mellitus, hospitalization for myocardial infarction, hospitalization for heart failure, hospitalization for stroke, chronic kidney disease, death, and rate of conversion to normo-glycaemia. Hospital Episode Statistics data will be used to determine the outcomes related to the hospitalisations, the Office for National Statistics for death status, and CPRD Aurum for the other outcomes.
The study design is based on a parent cohort of patients above 60 years-old, newly diagnosed with prediabetes encompassing: one cohort of patients initiating MET+LSA (new users) matched to one cohort of patients carrying on LSA only (prevalent). Starting from the prediabetes diagnosis, time-based exposure sets will be defined. The time based exposure will provide time points in the disease course at which confounder patient characteristics will be measured. To be eligible in a specific time risk set, patients will be required to have a medical visit. At each time-based exposure set patients a time-conditional propensity score will be assessed. It will compute the propensity of initiating MET+LSA, versus carrying on LSA only, as a function of the patient characteristics measured up to this specific exposure set.
Patients will be followed until the occurrence of an outcome of interest, diabetes, change in prediabetes therapy, death, transfer out of the dataset, or end of study period.
Comparison will be assessed using Cox proportional hazards regressions and Fine and Gray proportional subdistribution hazard regressions with competing events as the change in prediabetes therapy (discontinuation, switch, addition).
Type 2 diabetes mellitus; conversion to normo-glycemia; hospitalization for myocardial infarction; hospitalization for stroke; hospitalization for heart failure; chronic kidney disease; all-cause death.
Caroline Foch - Chief Investigator - Merck Healthcare KGaA (Merck Group)
Caroline Foch - Corresponding Applicant - Merck Healthcare KGaA (Merck Group)
Emmanuelle Boutmy - Collaborator - Merck Healthcare KGaA (Merck Group)
Michael Batech - Collaborator - Merck Healthcare KGaA (Merck Group)
Patrice Verpillat - Collaborator - Merck Healthcare KGaA (Merck Group)
Ulrike Gottwald-Hostalek - Collaborator - Merck Healthcare KGaA (Merck Group)
HES Admitted Patient Care;ONS Death Registration Data