Here’s an interesting study I recently came across that was published back in 2015. Researchers in England looked at 162 pairs of middle-aged female twins (thus making a total of 324 women in the study). They administered a baseline cognitive test to them, and performed brain MRIs. They also did tests of their leg strength and power. A decade later they saw the women back and repeated the cognitive tests and brain MRIs. What emerged was that even after controlling for any obvious variables they could think of (smoking, alcohol use, exercise habits, obesity, diabetes, etc), the women with the strongest legs at the start of the study fared better than their twin in terms of brain aging. Or put another way: having weak legs was a risk for starting to show signs of dementia a decade later.
One strength of this study is that each woman was compared to her twin, a person who grew up in the exact same house at the exact same time, and who shares at least some genetic traits, thus eliminating some of the confounding factors that might be present when comparing completely unrelated people.
On the other hand, all studies have limitations and this one is no exception. The most obvious issue with this study is that it is observational in nature, meaning the researchers were simply observing what occurred to the women, not making an intervention and measuring the results. Such a study inherently risks mixing up the cause and effect (in other words: what if leg weakness were an early symptom of memory loss, not a cause of it?)
It would be helpful to have a randomized controlled trial in which scientists randomly assign a group of people to do a leg strengthening program, and then follow them along with a control group to see if there are differences in dementia rates as both groups age. But such a study would be a lot more difficult to pull off and require a lot more resources, financially and otherwise.
For now, we’ll have to live with some uncertainty. But given that many other studies have shown that good muscle health is protective against the infirmities of ageing (dementia, but also heart disease, diabetes, osteoporosis, and frailty, to name just a few), I would bet that if a randomized controlled trial like the one I propose above were ever done, it likely would find a protective benefit to leg exercises. And even if it didn’t, it would remain a fact that engaging in regular strength training is very beneficial for your overall health.
Given this, I remain unreserved in recommending that everybody should try to engage in strength training at least a few times per week. And definitely don’t skip leg day.