OBJECTIVE We aimed to investigate the differences in clinical characteristics and pharmacological treatment associated with the presence of diabetes in a large cohort of patients with dementia.

RESEARCH DESIGN AND METHODS A cross-sectional registry-based study was conducted using data from the Swedish Dementia Registry (SveDem). Data on dementia diagnosis, dementia type, and demographic determinants were extracted from SveDem. Data from the Swedish Patient Register and Prescribed Drug Register were combined for the diagnosis of diabetes. Data on antidiabetic, dementia, cardiovascular, and psychotropic medications were extracted from the Swedish Prescribed Drug Register. Logistic regression was used to determine whether the variables were associated with diabetes after adjustment for confounders. In total, 29,630 patients were included in the study, and 4,881 (16.5%) of them received a diagnosis of diabetes.

RESULTS In the fully adjusted model, diabetes was associated with lower age at dementia diagnosis (odds ratio [OR] 0.97 [99% CI 0.97-0.98]), male sex (1.41 [1.27-1.55]), vascular dementia (1.17 [1.01-1.36]), and mixed dementia (1.21 [1.06-1.39]). Dementia with Lewy bodies (0.64 [0.44-0.94]), Parkinson disease dementia (0.46 [0.28-0.75]), and treatment with antidepressants (0.85 [0.77-0.95]) were less common among patients with diabetes. Patients with diabetes who had Alzheimer disease obtained significantly less treatment with cholinesterase inhibitors (0.78 [0.63-0.95]) and memantine (0.68 [0.54-0.85]).

CONCLUSIONS Patients with diabetes were younger at dementia diagnosis and obtained less dementia medication for Alzheimer disease, suggesting less optimal dementia treatment. Future research should evaluate survival and differences in metabolic profile in patients with diabetes and different dementia disorders.