Abstract
BACKGROUND AND OBJECTIVES: It remains unknown whether the association between diabetes mellitus (DM) and cognitive function differs in Eastern and Western populations. This study aimed to elucidate whether DM is associated with worse cognitive performance in both populations. METHODS: The Shanghai Aging Study (SAS) and the Mayo Clinic Study of Aging (MCSA) are two population-based studies with similar design and methodology in Shanghai, China and Rochester, MN, USA. Non-demented participants underwent cognitive testing, and DM was assessed from the medical record. Separate analyses were performed in SAS and MCSA regarding the association between DM and cognitive performance. RESULTS: A total of 3,348 Chinese participants in the SAS and 3,734 American subjects in the MCSA were included. Compared with MCSA subjects, SAS participants were younger, less educated, and had lower frequency of vascular disease, APOE ɛ4 carriers and obesity. Participants with DM (compared to non-DM participants) performed significantly worse on all the cognitive domains in both the SAS and MCSA. After adjustment for age, gender, education, and vascular covariates, DM was associated with worse performance in executive function (β=-0.15, p = 0.001 for SAS, and β=-0.10, p = 0.008 for MCSA) in the total sample and in the cognitively normal sub-sample. Furthermore, DM was associated with poor performance in visuospatial skills, language, and memory in the SAS, but not in the MCSA. CONCLUSIONS: Diabetes is associated with cognitive dysfunction and, in particular, exerts a negative impact on executive function regardless of race, age, and prevalence of vascular risk factors.