Prolonged smoking does not protect against Alzheimer's disease, say researchers from the Radcliffe infirmary, Oxford. The study, led by Professor Sir Richard Doll, contradicts previous reports which have suggested an inverse relation between dementia and smoking.
The prospective study (British Medical Journal 2000;320:1097) looked at a cohort of 34,439 British male doctors, followed since 1951, with their smoking habits reviewed every six to 12 years. Of the 24,133 deaths occurring up until 1999, dementia was mentioned on the death certificate of 483 cases.
Ten cases were excluded from the main analyses either because dementia was attributed to head injury or brain tumour or because death occurred less than 10 years after smoking habits were first recorded.
Overall, the study showed no significant relation between any type of dementia and smoking (relative risk 0.96, 95 per cent confidence interval 0.78 to 1.18). Dementia involving Alzheimer's disease was also neither directly nor inversely related to continued smoking, the proportion of long-term smokers being 40 per cent in those who developed the disease and in the matched controls (relative risk 0.99, 0.78 to 1.25).
The authors also discuss the limitations of previously published retrospective and short-term prospective studies and conclude that the protective effects of smoking reported in some of these studies were largely artefactual.