Type 2 diabetes is a medical condition that involves an inability to process sugar from the diet properly. Living with type 2 diabetes increases the chance that a person will develop dementia, a condition that can develop later in life that takes away a person’s abilities to think and to go about their daily lives on their own. It has been proposed that newer treatments for type 2 diabetes, including sodium-glucose cotransporter-2 (SGLT2) inhibitors and glucagon-like peptide-1 receptor agonists (GLP1-RAs), may offer new potential to slow the development of dementia. Although some studies have suggested that these types of drugs may have benefits to the brain, there remains too little evidence to conclude whether they might slow down dementia in people living with type 2 diabetes. Using information from family doctor visits and health records in the UK, we propose to determine whether people who have used these drugs remained free of dementia later into their lives. In our study, we will compare people who used these newer types of diabetes drugs to people who used a common older type of diabetes drug, a dipeptidyl peptidase-4 (DPP4) inhibitor, which is unlikely to slow down or speed up dementia progression if used. The study aims to suggest which types of drugs for diabetes might offer the largest brain benefits, and whether these types of drugs might be good choices for further testing in clinical trials to protect the brain as people age.
Type 2 diabetes increases the risk of dementia, but there is little evidence on how to reduce this risk. Using an active-comparator new-user cohort design, the primary aim of this study is to compare time to incident dementia diagnosis 1) between SGLT2 inhibitors and DPP4 inhibitors and 2) between GLP1-RAs and DPP4 inhibitors. The secondary aim is to compare time to dementia between GLP1-RAs and SGLT2 inhibitors, as the comparative effectiveness of these classes is of interest. The study will use an active-comparator new-user cohort design, and three separate cohort analyses of individuals older than 60 will be conducted. New users are defined as individuals without a prescription of either the treatment of interest of the comparator in the past year, and the exposure definition will follow an intention-to-treat approach. Incident all-cause dementia as the primary outcome and incident Alzheimer’s disease as an exploratory outcome will be determined based on the primary care records. Incident vascular dementia as the other exploratory outcome will be ascertained using both the primary care records and hospital admission records. Confounding by indication will be addressed using propensity scores. The confounders include demographics, practice-level index of multiple deprivation, laboratory measures, comorbidities, and medications. A 1-year lag period will be used to address disease latency and reverse causality. The adjusted hazard ratios and confidence intervals for time to dementia will be analyzed using a Cox proportional hazard model.
All-cause dementia; Alzheimer’s disease dementia; vascular dementia.
All-cause dementia outcome in the primary analysis will be defined using primary care records based on the date of Read code (CPRD GOLD) or the date of SNOMED-CT classification code (CPRD Aurum) for all-cause dementia diagnosis. Alzheimer’s disease dementia and vascular dementia will be defined using a previously proposed algorithm (PMID: 23455986) based on Read code (CPRD GOLD) or SNOMED-CT classification code (CPRD Aurum) in primary care records. The algorithm for vascular dementia involves the identification of stroke events, and therefore HES admitted care data will be used to identify stroke cases that not captured in primary care records.
Walter Swardfager - Chief Investigator - University of Toronto
Walter Swardfager - Corresponding Applicant - University of Toronto
Baiju Shah - Collaborator - University of Toronto
Che-Yuan Wu - Collaborator - University of Toronto
Colleen Maxwell - Collaborator - University Of Waterloo
Jodi Edwards - Collaborator - University Of Ottawa
Moira Kapral - Collaborator - University of Toronto
Wajd Alkabbani - Collaborator - University Of Waterloo
Che-Yuan Wu - Collaborator - Sunnybrook Research Institute
HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Practice Level Index of Multiple Deprivation