That physical activity reduces the risk of dementia is a well-established idea in scientific literature for years. However, a meta-analysis published in The Lancet Public Health and led by David Raichlen (USC Dornsife College of Letters, Arts and Sciences) adds a nuance: not all physical activity protects equally. The context in which it is performed - work or leisure - seems to determine if the movement benefits the brain or not.
The research team analyzed data from 74 cohort and case-control studies, involving more than 4.2 million people aged 45 to 93 in 30 countries, with a median follow-up of 10 years. Individuals with higher levels of leisure physical activity - such as walking, swimming, running, cycling, or playing sports - had a 24% lower risk of dementia than less active individuals. Domestic physical activity also appeared beneficial, although evidence comes from a single study.
Regarding occupational physical activity, it was associated with a 20% higher risk of dementia. Active commuting to work showed a modest increase in risk, although this latter finding is based on only two studies.
"I prescribe you a dose of squats": the benefits of doctors also 'prescribing' exercise
Exercising daily: the forgotten 'pill' to mitigate the effects of aging
"The results challenge the long-assumed belief that 'all movement counts' for brain health," stated Natan Feter, first author of the study, postdoctoral researcher at Raichlen's Evolutionary Biology of Physical Activity Laboratory at USC, and researcher at the Graduate Program in Epidemiology at the Universidade Federal do Rio Grande do Sul (Brazil).
A well-known paradox in cardiology
The authors frame the finding within the so-called physical activity paradox, well documented in cardiovascular research: physically demanding jobs do not provide the same health benefits as voluntary exercise and could even increase the risk of disease. This meta-analysis is, according to its authors, the first to systematically examine if this paradox extends to dementia as well.
Lucrecia Moreno, professor of Pharmacology and director of the DeCo Chair for the Study of Cognitive Impairment at CEU Cardenal Herrera University, in statements to SMC Spain, places the finding within the existing evidence: "There is a lot of evidence on the so-called physical activity paradox and what is concluded in the meta-analysis: more total exercise leads to more dementia prevention and more physical occupation during working life leads to less prevention," she explains. And she points out a possible explanation: "Perhaps due to less intellectual dedication; it's many years of working life."
Not categorizing all physical jobs "under the same umbrella"
For the authors of the meta-analysis, the harms associated with occupational physical activity probably do not stem from the movement itself, but from the social and environmental conditions that often accompany physically demanding work: psychosocial stress, less task control, exposure to occupational risks and pollution, and fewer opportunities for recreational exercise due to fatigue or lack of time.
At this point, Moreno highlights a limitation in the way occupational activity is classified: "We need to be more precise and not categorize everyone under the same umbrella. The population classification on work activity divides between physical (or manual) and intellectual jobs. I believe that here lies the bias of only considering physical or sedentary activity and not taking into account cognitive reserve or the purposeful aspect of life."
According to the pharmacologist, "it is very important to study how all aspects interrelate and not look at each factor in isolation."
The meta-analysis is not without limitations, and the authors themselves clearly point them out. Evidence on domestic activity and active commuting is still scarce. Most of the available studies come from high-income countries, although it is expected that low- and middle-income countries will experience the greatest increase in dementia cases in the coming decades. And almost all are based on self-reported physical activity, which can introduce memory bias.
Moreno agrees with this assessment: "The authors are quite clear about the limitations: most studies use self-reported physical activity data. There are few high-quality studies, scarce long-term follow-ups, etc. But, above all, there is a lack of randomized prospective studies that allow establishing causality."
Due to being an analysis based on observational studies, the researchers emphasize that the results do not imply that physically active jobs should be avoided, nor can a causal relationship between occupational activity and dementia be established.
Raichlen places these limitations within the study's underlying message: the context of physical activity is as important as the movement itself. "Improving access to safe parks, walking routes, and recreational facilities, ensuring that people have enough free time for exercise, and reducing harmful occupational exposures could become as important as promoting physical activity itself," he stated.
Managing work stress and caring for brain health
Beyond the epidemiological evidence, the study raises a reflection on how to translate this knowledge into clinical practice and the general population. For Moreno, the message left by the meta-analysis is twofold: "The study reminds us of the importance of disseminating knowledge so that people are aware of the importance of cognitive reserve if you only dedicate your work activity to the physical and manual part. And vice versa, the importance of physical exercise in sedentary occupations."
She adds a conclusion that she says is the most important of all: "Learning to manage work stress in both cases and caring for brain health."