The “Physical Activity Paradox” in Dementia Risk Revealed in New Study

A new study has revealed a so-called “physical activity paradox” in dementia risk. In a nutshell, even though physical activity is generally considered to be protective, in some cases it can actually increase your risk. Just looking at “physical activity” as a single factor misses a lot of crucial detail.

The Different Types of Physical Activity  

The researchers identify four different types of physical activity: leisure-time, occupational, household and commuting activity. They’re basically pointing out that having a physically-demanding occupation is different from doing chores around the house or cycling to the office.

However, much of the evidence focuses on leisure time physical activity, which only represents 12% of all physical activity around the world.

The Study: Reviewing the Evidence on Different Types of Physical Activity

The researchers basically wanted to update a previous review they published in 2022, adding evidence from the intervening years and performing the analysis for different types of activity separately. The new paper [1] uses the 57 studies included in that review, and adds 17 new cohort and case-control studies published between the start of 2021 and May 2026. This led to a combined sample of over 4.2 million participants.

Studies were excluded if they had a follow-up of less than a year, if diagnoses were based on self-report or cause of death data, if people had dementia or mild cognitive impairment at baseline, or any research focusing specifically on people with a diagnosed disease.

What They Found: Leisure Time Activity Protects, Occupational Activity Doesn’t

The analysis found that leisure time physical activity was associated with a 24% lower chance of all-cause dementia, while occupational activity increased risk by around 20%. These were the two areas with the most research, but less reliable findings also suggest that household activity reduces risk by 15% and commuting activity increases it by 10%.

Limitations to Keep in Mind

As always, there are issues to keep in mind. Firstly, physical activity was based on self-reporting, which is open to bias, and secondly, the type of studies included make it hard to rule out reverse causation. Notably, 93% of studies came from high income countries, so the evidence base is not very representative of the world. The results broadly held for specific types of dementia, but vascular dementia wasn’t as well covered. 

However, the core findings generally seem robust.