Applying Mendelian Randomization to the Correlation Between Fitness and Health
It is a struggle to derive evidence for causation from human data. It is well established that physical fitness correlates with a lower risk of age-related disease and mortality in humans, and well established that greater physical fitness causes a lower risk of age-related disease and mortality in animal studies. But as a practical matter one can't run the sort of study that would be needed to obtain direct proof of causation in humans. So researchers turn to approaches such as Mendelian randomization, in which one adds an additional set of genetic correlations in order to try to infer at least some support for causation. There are indeed genetic correlations with a tendency to greater physical fitness, and those do correlate in turn with risk of age-related disease and mortality.
We investigated potentially causal associations between genetically predicted aerobic fitness and multiple health phenotypes using a two-stage phenome-wide Mendelian randomization (MR) study. We screened 712 health-related phenotypes as outcomes using publicly available European-ancestry genome-wide association study (GWAS) summary statistics from OpenGWAS (Discovery GWAS n > 5,000), prioritizing non-UK Biobank/non-FinnGen datasets for Discovery when available and selecting an independent GWAS for Validation.
We identified 108 Discovery associations, of which 34 remained valid and statistically significant after Validation. Higher genetically determined aerobic fitness was associated with lower lacunar stroke risk, lower arterial stiffness, higher heart rate variability, lower diastolic blood pressure, more favorable anthropometric measures, lower use of antidiabetic drugs, lower asthma risk, lower C-reactive protein, higher bone mineral density, favorable liver function biomarkers, favorable platelet-related traits, multiple blood-count-derived hematological cell indices and counts, as well as higher years of schooling. Adverse associations were confined to atrial fibrillation, valvular heart disease, and systolic blood pressure.
Genetically determined aerobic fitness is linked to a broad pattern of favorable cardiometabolic, inflammatory, musculoskeletal, respiratory, hepatic, and hematological phenotypes, alongside a narrow set of potential cardiovascular hazards.