Objectives: The negative effects of overweight on cardiometabolic health is well-known. An increasing body of evidence extends the negative effects of midlife overweight to dementia. However, a different picture emerge when overweight is assessed in late life. The time-varying effect of weight status on dementia was evaluated in two prospective Nordic population-based studies.

Methods: The participants included in the Swedish Twin Registry self-reported their height and weight in 1963 (mean age 52.5 years). About 25 years later these twins were either included in the SATSA study (50 years and older) or the OCTO-Twin study (80 years and older). Dementia was consequently screened for and diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria presently used at the time of diagnoses. The participants weight and height was assessed at baseline of the Finnish Lieto Study (mean age 70.8) and dementia was screened for and diagnosed according to the DSM-IV criteria eight years later.

Results: Logistic regression analyses indicated that midlife overweight was associated with a greater risk of all cause dementia, odds ratio 1.55 (95% confidence interval (CI) = 1.18-2.04), when demographic and cardiometabolic risk factors and diseases were controlled for. However, Cox regression analyses indicated that for each unit increase in BMI score in late life, the risk of dementia decreased 8% (hazard ratio = 0.92, 95% CI = 0.87–0.97), when demographic and cardiometabolic risk factors and diseases were controlled for. The association remained significant when individuals who developed dementia during the first four years of follow-up were excluded from the analyses.

Conclusions: Our results indicate there might be a time-varying effect of weight status on dementia. Preclinical dementia might blur the association between weight status and dementia in late life. This needs to be further analysed in studies following the same sample over the life course.