An analysis pooling 74 cohort studies and more than 4.2 million adults found that leisure-time physical activity, such as walking, cycling, swimming or sports, was associated with a 24% lower relative risk of dementia (relative risk 0.76, 95% CI 0.65 to 0.88), while physical activity performed as part of a job was associated with a 20% higher relative risk (relative risk 1.20, 95% CI 1.01 to 1.42). The study, led by researchers including Mika Kivimaki and colleagues, was published in The Lancet Public Health.
Public health guidance has for decades treated physical activity as a single category, urging people to move more without distinguishing where that movement happens. This analysis is among the first to separate activity by domain across a large pooled dataset, and the authors say the divergence between leisure and occupational activity may help explain inconsistent results in earlier dementia prevention research that lumped all movement together.
The researchers and outside commentators caution that occupational activity differs from leisure activity in ways beyond intensity: manual jobs often involve repetitive movement, less choice over pacing, lower job control, disrupted sleep, and are more common among people with fewer socioeconomic resources and less access to healthcare, all factors independently linked to dementia risk. The study is observational, meaning it can show an association but cannot establish that occupational activity itself causes higher dementia risk, or that leisure activity itself causes lower risk; unmeasured differences between the groups of workers and leisure-exercisers being compared could account for part or all of both findings.
The authors call for research that examines why the two domains diverge, including whether cognitive stimulation, autonomy, and recovery time during leisure activity, factors largely absent from repetitive manual labor, play a role.