A systematic review published in npj Exercise Medicine and Health pooled 61 Mendelian Randomization studies to ask whether being genetically predisposed to more physical activity, less sedentary time, or better fitness causally affects brain health. The short answer is that the evidence for general physical activity is thin and inconsistent. Thirty of the 61 studies reported only null or directionally opposite results for brain health outcomes.
The review was led by researchers at Amsterdam University Medical Center and collaborators at the University of Manchester. Mendelian Randomization uses inherited genetic variants as stand-ins for lifelong differences in a trait, which can reduce the bias from confounding and reverse causation that plagues observational studies. That design is only as good as its genetic instruments, however. The authors judged 52 percent of the studies (32 of 61) to be of low methodological quality, and only 11 percent had sufficiently strong genetic instruments. Only 26 studies used independent summary statistics for a given exposure-outcome pair, so some apparently consistent findings may come from reanalyzing the same data.
The clearest signals came from fitness rather than activity volume. Greater handgrip strength and greater appendicular lean mass were associated with lower depression risk and better cognitive performance in several studies. The authors note these measures may also reflect frailty and underlying health, which makes them hard to read as independent protective factors. Aerobic exercise such as swimming and cycling was linked to lower depression risk across all three studies that examined it, though the authors stress this does not directly prove an exercise program works. Prolonged television viewing was repeatedly associated with higher depression risk, although several estimates weakened after adjusting for body mass index.
The authors conclude that genetic evidence cannot establish that routine physical activity uniformly prevents psychiatric disorders. They argue the answer will come from combining genetically informed designs with randomized trials and family-based cohorts, and they point out that a lack of genetic association does not rule out benefits that genetic studies may simply fail to capture.