Although there is no cure for Alzheimer's disease or other dementing illnesses, some evidence has suggested that regular physical exercise may prevent or delay the onset of dementia in older persons. Increased cerebral blood flow resulting from physical exercise seems to reduce age-related brain tissue atrophy. Also, epidemiologic studies have shown an association between physical activity and lower dementia risk. Larson and colleagues conducted a prospective cohort study to examine the relationship between regular exercise and the onset of dementia in adults 65 years and older.

The 1,740 participants were a subset of patients enrolled in the Seattle-area Adult Changes in Thought study who scored above the 25th percentile on a screening test for cognitive function and did not have a previous dementia diagnosis. At baseline, participants reported how often in the previous year they had participated in physical activities including walking, hiking, bicycling, aerobics, swimming, and weight training. Regular exercise was classified as performing any of these activities at least three times per week for at least 15 minutes at a time. Participants underwent testing for or provided information on potential confounding variables such as physical function; depression; health conditions; tobacco, alcohol, and dietary supplement use; years of education; and apolipoprotein E genotype (which is associated with increased risk of Alzheimer's disease). Cognitive screening was repeated every two years for a mean follow-up period of 6.2 years. Participants whose cognitive scores fell below a predetermined threshold underwent a comprehensive clinical and neuropsychological evaluation, and criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th ed., were used to diagnose incident dementia.

At the study's conclusion, 158 participants had developed dementia, 397 had died or withdrawn from the study, and 1,185 were alive and had not developed dementia. At baseline, patients in the dementia-free group were statistically more likely to have higher cognitive and physical function scores and less likely to have depression than those in the dementia group. Health conditions such as coronary heart disease, cerebrovascular disease, and hypertension were less common in the dementia-free group. After accounting for these factors, age, and sex, participants who exercised regularly were significantly less likely to develop dementia than those who did not, with an adjusted risk ratio of 0.62 (95% confidence interval, 0.44 to 0.86). The difference in incidence of dementia between the two groups also was striking: 13.0 per 1,000 in the regular exercise group and 19.7 per 1,000 in the group who exercised less often. Regular exercise produced the greatest dementia risk reduction in participants with the poorest physical function scores at baseline.

The authors conclude that physical exercise for 15 minutes or more at least three times per week appears to protect against incident dementia. By limiting this study to persons with high levels of baseline cognitive functioning, they minimized the possibility that some participants exercised less because they already had early (but as yet undiagnosed) dementia. Noting that participants with low levels of physical functioning benefited the most from regular exercise, the authors suggest that the preservation of physical function may be responsible for exercise's protective effect against dementia.