Diabetes is a disease that occurs when a person’s sugar level in the blood is too high. People who have diabetes often need to take medications to lower the sugar in their blood. This study will use medical records from many patients to compare two medications for diabetes. This study will verify which of the two drugs is better at reducing the sugar levels and body weight. The results will be compared to the results of another study that looked at the same question but used fewer patients. This project is part of a bigger project that compares the results of many studies using medical records data with results from smaller studies. The goal is to better understand when and how questions on drug effects can be answered using only data from medical records. Findings from this study can be used to help patients and doctors better understand and apply results from these larger studies that use patients that are followed by general doctors in a typical clinic.
This project aims to emulate the SUSTAIN 2 trial as part of the RCT DUPLICATE initiative. The RCT DUPLICATE initiative was created to assess when and how certain drug effect questions can be answered using observational healthcare data. The SUSTAIN 2 trial was a phase III, multicentre, open-label, randomized trial that assessed the efficacy and safety of semaglutide, a new glucagon-like peptide-1 antidiabetic treatment administered subcutaneously as an add-on to metformin compared to sitagliptin among patients with type 2 diabetes. This trial will be emulated in the CPRD database using methods reported for the RCT for designing this real-world-data (RWD) study. We will evaluate change in HbA1c and weight outcomes at 30 and 56 weeks after cohort entry. Change in HbA1c will be calculated as the difference between the last baseline HbA1c measure and the closest follow-up measure at 30 weeks (+/-60 days) and 56 weeks (+/- 90 days). When an HbA1c measure does not occur in the predefined time window, they will be considered missing data. Change in weight will be calculated as the difference between the baseline weight (closest measure) and the follow-up weight measure at 30 and 56 weeks (closest measure). A missing weight measure will be considered no change in weight, but will require a visit during the period leading up to the 30- and 56-month timepoints (assuming the GP had the opportunity to measure the weight but chose to measure weight only if a change in weight was expected or observed). Patients will be censored at drug discontinuation, switching or augmenting to the other study drug, end of coverage, end of study period (03/31/2022), death, a maximum follow-up of 56 weeks, whichever occurs first.
Emulating SUSTAIN2 as part of RCT DUPLICATE will help patients and physicians better understand and apply results from RWD studies.
Change in hemoglobin A1c at 30 and 56 weeks
Change in weight at 30 and 56 weeks
Samy Suissa - Chief Investigator - Sir Mortimer B Davis Jewish General Hospital
Samy Suissa - Corresponding Applicant - Sir Mortimer B Davis Jewish General Hospital
Elisabetta Patorno - Collaborator - Brigham & Women's Hospital
Gregory Brill - Collaborator - Brigham & Women's Hospital
Karine Suissa - Collaborator - Brigham & Women's Hospital
Laurent Azoulay - Collaborator - McGill University
Sebastian Schneeweiss - Collaborator - Brigham & Women's Hospital
Shirley Wang - Collaborator - Harvard University
HES Admitted Patient Care;HES Outpatient;Practice Level Index of Multiple Deprivation