Adiposity in middle and old age and risk of death from dementia: 40-year follow-up of 19,000 men in the Whitehall study.

Aims and objectives: to examine the hypothesis that obesity is protective for dementia, we compared the associations of death from dementia with body weight and body mass index (BMI) in both middle and old age.

Design: height and weight were measured in a prospective study of 19,019 middle-aged men in the Whitehall study in 1967-70 and in 6,158 surviving participants at resurvey in 1997. Cox regression was used to examine the associations of death from dementia over a 40-year period with weight or BMI measured by health professionals in middle and old age adjusting for age, smoking habits, employment grade and marital status.

Setting: central government employees in London, UK.

Main outcomes measure: death due to dementia in 320 participants.

Results: body weight measured in middle age was weakly inversely associated with death from dementia (hazard ratio 0.98 [95%CI: 0.97-0.99] per kg), but neither height nor BMI were related to risk of dementia. In contrast, body weight in old age was more strongly inversely related to deaths from dementia (0.96; [0.95-0.98] per kg) as was BMI (0.92 [0.86-0.97] per kg/m2). Weight loss over the 30 years between baseline and resurvey was associated with a higher risk of death from dementia, with an adjusted HR per kg/30 years of 1.04 [95%CI: 1.02-1.08] and the association with loss of BMI was even stronger (adjusted HR of 1.10 [1.03-1.19]) per kg/m2 decrease.

Conclusions: the stronger inverse associations of deaths from dementia with BMI in old age, compared with middle age, together with strong positive associations of loss of BMI or body weight between middle and old age casts doubt on previous suggestions that obesity protects against death from dementia.