Physical activity and brain health: a systematic review of Mendelian Randomization studies

ObjectiveObservational studies consistently link physical activity (PA) to better brain health; psychiatric, neurocognitive, and mental well-being outcomes. While PA intervention studies generally find positive mental health effects, it is unclear whether day-to-day PA, not manipulated through an intervention, is beneficial. Design/DataSystematic review of Mendelian Randomization (MR) studies using genetic instrumental variables to assess causal effects of PA on brain health. In Embase and medRxiv, 557 articles were identified, of which 35 met inclusion criteria. Study quality was determined based on a MR-specific scoring system. Eligibility criteriaMR studies exploring any PA exposure (self-reported PA, sedentary behaviours, strength-related traits) on mental health, cognition, or cognition-based neurodegenerative outcomes (e.g., Alzheimers). ResultsOf 35 studies, 43% received a low quality score and 57% a moderate score. There was no consistent evidence for causal effects of day-to-day PA. For sedentary behaviours (e.g., TV-watching) there was consistent evidence for increasing effects on depression. For strength-related traits (e.g., grip-strength) there was consistent evidence for protective effects, particularly on cognitive outcomes. For depression specifically, device-measured PA more often showed protective effects on depression than self-reported PA. ConclusionMR studies on PA and brain health are generally lacking in quality, due to low sample sizes and/or poorly measured variables. Results are mixed with the most consistent evidence indicating that better physical strength and less sedentary behaviour are beneficial for brain health. Since core MR assumptions are difficult to fulfill for highly complex PA traits, triangulation with other methods and improved phenotyping is needed in future work. SummaryO_ST_ABSWhat is already known?C_ST_ABSO_LIThe prevalence of the most common mental health disorders (i.e., depression, anxiety) has been increasing in the last 20 years in the United States and Europe. C_LIO_LIbservational studies report a consistent association between physical activity and brain health outcomes, but the high risk of confounding precludes a causal interpretation. C_LIO_LIPhysical activity intervention studies find positive mental health effects, but do not show whether day-to-day physical activity, not manipulated through an intervention, is beneficial. C_LI What are the new findings?O_LIverall, the Mendelian randomization literature offers limited support for a broadly protective effect of physical activity. C_LIO_LISelf-reported physical activity showed inconsistent associations with depression, schizophrenia, and bipolar disorder, possibly due to misclassification and confounding. C_LIO_LIbehaviours were more consistently associated with negative mental health outcomes, particularly depression. C_LIO_LIStrength-related traits, such as handgrip strength and appendicular lean mass, showed the most consistent and strong, protective effects, especially on cognitive outcomes. C_LIO_LIRandomization relies on core assumptions that are difficult to fulfill for highly complex traits, like physical activity. Triangulation with other methods and improved phenotyping is needed in future work. C_LI