The association between type 2 diabetes disease trajectories and dementia incidence

Introduction Type 2 diabetes (T2D) prevalence, severity, duration, and control are associated with dementia incidence, but prior literature is focused on specific pharmacologic, dietary, and exercise interventions in isolation while controlling for other co-occurring factors. Accounting for comprehensive life course experiences of the timing of diabetes onset, severity, treatment, and progression over a period of decades would provide a more comprehensive description of how life course diabetes progression and control is associated with dementia. Trajectories of diabetes diagnosis, pharmacological management, and disease progression are heterogeneous, and classifying these trajectories presents a significant methodological challenge. Methods Using deidentified survey and electronic health record data from Kaiser Permanente Northern California (KPNC) from the Research Program on Genes, Environment, and Health (RPGEH), we defined annual "states" for each eligible participant with T2D diagnosed between ages 50 and 70 based on KPNC, diabetes diagnosis, glycated hemoglobin, antidiabetes prescription count, and kidney dysfunction. We then employed sequence and cluster analyses to group participants into clusters with similar trajectories of these states. Finally, we estimated hazard ratios for incidence of Alzheimer's disease and Alzheimer's disease related dementias (AD/ADRD) for each of these clusters as well as individuals with type 1 or other diabetes types, relative to participants without diabetes at age 70, using covariate-adjusted Cox proportional hazards models. Results Using the 18,688 participants with T2D included in the diabetes trajectory assessment, sequence and cluster analysis identified 9 clusters of T2D treatment and control histories between ages 50 and 70. Clusters differed markedly in timing of onset of T2D, glucose control, antidiabetes drug use and kidney function. Associations of these clusters with incident AD/ADRD after age 70 was heterogeneous and patterned by diabetes control and treatment history, particularly by diabetes duration and treatment regime. Conclusions In conclusion, in this real-world data context, we find increased diabetes severity, increased medication use, and faster progression to kidney disease is associated with increased risk of dementia. We find some patterns of diabetes severity and control are associated with greater dementia risk. This information may be useful in the context of targeted screening and allocation of preventative services for ADRD.