Response to Comment on Celis-Morales et al. Type 2 Diabetes, Glycemic Control, and Their Association With Dementia and Its Major Subtypes: Findings From the Swedish National Diabetes Register. Diabetes Care 2022;45:634–641

We appreciate the interest by Meng and Wang (1) in our article (2), which highlights the need to conduct a competing risk analysis between all-cause mortality, type 2 diabetes, and risk of Alzheimer disease, as the early average mortality experienced by patients with diabetes compared with controls could explain the inverse associations reported in our study. We completely agree; this is why our study already included all-cause mortality in competing risk analyses (2). As described in our methods and results section, the inverse association is indeed explained by the early mortality experienced in patients with diabetes compared with controls, which reduced the probability of developing Alzheimer disease (2). That noted, vascular dementia rates were meaningfully higher in people with diabetes, irrespective of competing risks, a novel observation that suggests targeting vascular risk factors in patients with diabetes should not only lessen cardiovascular outcomes but also slow cognitive decline. More trials in this area would be beneficial.

We appreciate the importance of comparing patients with diabetes with different levels of HbA_1c with well-matched control subjects. However, due to the nature of our study (2), where key covariates were available only for patients with diabetes but not their counterparts free of diabetes, it was impossible to perform these comparisons. Moreover, observational findings, unless profound, seldom prove causal links between risk factors and outcomes.