Body adiposity in later life and the incidence of dementia: the health in men study.

Objective: To determine if adiposity in later life increases dementia hazard.
Methods: Cohort study of 12,047 men aged 65-84 years living in Perth, Australia. Adiposity exposures were baseline body mass index (BMI), waist circumference (WC) and waist-to-hip ratio (WHR). We used the Western Australian Data Linkage System (WADLS) to establish the presence of new cases of dementia between 1996 and 2009 according to the International Classification of Diseases (ICD). Crude and adjusted hazard ratio (HR, 95% confidence interval, 95%CI) of dementia for each adiposity marker was calculated using Cox regression models. Other measured factors included age, marital status, education, alcohol use, smoking, diet, physical activity, and prevalent hypertension, diabetes, dyslipidaemia and cardiovascular disease.
Results: Compared with men with BMI<25, participants with BMI between 25-30 had lower adjusted HR of dementia (HR = 0.82, 95% CI = 0.70-0.95). The HR of dementia for men with BMI ≥ 30 was comparable to men with BMI<25 (HR = 0.82, 95%CI = 0.67-1.01). Waist circumference showed no obvious association with dementia hazard. Men with WHR ≥ 0.9 had lower adjusted HR of dementia than men with WHR <0.9 (HR = 0.82, 95%CI  =  0.69-0.98). We found a "J" shape association between measures of obesity and the hazard of dementia, with the nadir of risk being in the overweight range of BMI and about 1 for WHR.
Conclusions: Higher adiposity is not associated with incident dementia in this Australian cohort of older men. Overweight men and those with WHR ≥ 0.9 have lower hazard of dementia than men with normal weight and with WHR<0.9.