“If the little grey cells are not exercised, they grow the rust.” ― Fictional detective Hercule Poirot by The A.B.C. Murders.
One can just hear actor David Suchet as Poirot saying those iconic words on the British ITV series. To be fair, Poirot seemed to dislike physical exercise itself; he was talking about closing his eyes, sitting quietly, and thinking through a puzzle. Closer to home, the late Tom Magliozzi, of CarTalk, used to say (paraphrasing): “Exercise may or may not make you live longer, but it sure will make it seem like longer!”
Physical exercise has been one of the main interventions touted to help in preventing, or slowing down dementia.
The World Health Org’s 2019 “Risk Reduction of Cognitive Decline and Dementia” gave recommendations on exercise (physical activity) after massive literature reviews:
Rec 1: Physical activity should be recommended to adults with normal cognition to reduce the risk of cognitive decline. Quality of evidence: moderate Strength of the recommendation: strong ; Rec 2: Physical activity may be recommended to adults with MCI to reduce the risk of cognitive decline. Quality of evidence: low Strength of the recommendation: conditional.
The Lancet Commissions 2020 dementia review had this to say: “Studies of physical activity are complex.” They cite a number of studies, but suggest that more work needs to be done to pinpoint good effects. The US Agency for Healthcare Research and Quality did do an AD review in 2020, addressing the semi-effectiveness of current drugs like donepezil and memantine, but didn’t address physical activity. They did examine that in 2017, but it was basically inconclusive.
Now a paper (published Mar 31) could not find any positive effects of physical activity on cognition, and suggests the effects seen in other studies might just be “reverse causation“! Huh?! To quote the paper’s key points:
“Findings This cohort study of 1718 women at midlife found that, with adjustment for socioeconomic characteristics, menopause symptoms, hormone therapy use, and presence of diabetes and hypertension, self-reported physical activity was not associated with measured cognitive performance in the domains of processing speed, verbal memory, or working memory.
Meaning These findings suggest that the cognitive protection effect of physical activity observed in later life may be an artifact of reverse causation.”
AdGadfly will try to unpack this: it’s a carefully done 20+ year longitudinal study (from the Study of Women’s Heath Across the Nation); they performed serial cognitive battery testing and collected self-reports of physical activity, median followup ~12 years, with subjects from across the country enrolling in their mid-40s.
The results basically showed an overall decline on their cognitive tests in everybody, even before age 60 (the figure above shows results of one of their three cognitive tests, the one selected for processing speed). Not only did the authors find no association between self-reports of physical activity and cognitive decline, but they suggest that subjects having subtle cognitive issues might avoid physical activities. Those less impaired might be more active. Of the factors examined, the graph above shows that financial issues had a substantial impact on decline.
The authors don’t say this explicitly, but AdGadfly interprets their use of “reverse causation” to mean something akin to the idea of an inadvertent “self-fulfilling prophecy” in previous research reports. Since theirs is evidently a younger cohort than other studies, they suggest that the older cohorts in many observational studies have a self-selected element, making any difference more pronounced.
The authors seem to be saying that instead of physical activity helping cognition, it really might be that better cognition is helping subjects participate in physical activity in the first place.
AdGadfly admits to not having a breadth of knowledge of the literature to put this paper in perspective, but this was eye-opening. It also reminds one of general selection bias in research, or maybe stage-migration bias (Will Rogers Phenomenon) seen in cancer studies.
The authors do a nice job in the Discussion and Limitations sections. They cite European twin studies, other studies of positive correlations between activity and cognition, and caveats, but also cite the problems associated with the cross-sectional (rather than longitudinal) aspect of most studies. There is an accompanying editorial, citing some animal studies that support physical activity and cognition, but it also points out that human studies could be improved.
Time to go into Poirot-mode and put the little gray cells, and other colors of cells, to work on this puzzle!
