Precision exercise medicine: understanding exercise response variability

consensus statement; no new data

N/A — synthesis of human twin/family studies plus rodent selection experiments.

BJSM 53(18):1141-1153. Consensus statement from a 2017 Pennington Biomedical symposium. Graded C to match this database's treatment of expert consensus statements (cf. Webborn 2015). Included deliberately as the strongest mainstream statement AGAINST this topic's scepticism about trainability — we do not get to cite only the papers that agree with us.

The expert consensus disagrees with this database's reading. It concludes there is "strong evidence from both animal and human studies that exercise training doses lead to variable responses" and that "a genetic component contributes to exercise training response variability", citing HERITAGE heritability of roughly 50% of CRF response variance (30-60% depending on adjustment) and rodent selective-breeding lines that diverge fourfold. TWO THINGS BLUNT IT. First, its human heritability figure is the SAME HERITAGE family-design upper bound this database already downgrades — it is not independent confirmation. Second, the statement itself concedes that "figures depicting the range of the observed individual responses will tend to inflate the true interindividual response variability", and its own design recommendations are exactly the ones that produce null results when applied: it recommends "study designs that include a randomly assigned control group (or condition) and have multiple assessments of CRF before and after", because "designs without a control group cannot isolate changes due to treatment" and "designs without multiple assessments before and after cannot isolate random error from interindividual response variability." It also notes that no study had then examined heritability of CRF response to maximal or near-maximal training doses. It makes NO recommendation for clinical genetic testing.

The paper's pro-genetic claim rests on the HERITAGE two-generation family design (confounds genes with shared household) and on rodent selection lines (which establish that response CAN be selected on, not that human individual differences are cleanly heritable).

The design standard it sets was met by Renwick 2024 (control-arm meta) and Bonafiglia 2022 (IPD meta), both of which failed to find strong evidence for true VO2max response heterogeneity.