Familial aggregation of VO₂max response to training
The HERITAGE Family Study paper establishing the genetic component of trainability
Participants
n=481 sedentary adult Caucasians from 98 two-generation families.
Protocol
20 weeks of standardised exercise training
VO2max tested on a cycle ergometer twice before and twice after the program
Familial aggregation tested by ANOVA (between- vs within-family variance) and by model-fitting heritability estimates adjusted for age and sex
Key Findings
1
Mean increase in VO2max approximately 400 ml/min
2
Considerable heterogeneity: some individuals had little or no gain, others gained more than 1.0 l/min
3
2.5 times more variance between families than within families in the VO2max response
4
Maximal heritability estimate for the VO2max response: 47% (adjusted for age and sex), with a maternal transmission of 28% in one of the models
5
Conclusion: trainability of VO2max is highly familial and includes a significant genetic component
6
The abstract gives no percentages of "non-responders" or "high responders" and no percentage gain
Mechanism
Inference, not tested: familial clustering of the response (2.5× more variance between than within families, heritability 47%) implies inherited factors in how much VO2max rises to the same training; the abstract does not locate them (cardiac vs peripheral) and names no genes.
Practical Application
In 481 previously sedentary adults from 98 families, 20 weeks of the same standardised training raised VO2max by about 400 ml/min on average, but some people gained nothing and others more than 1.0 l/min, and the response clustered in families — 2.5 times more variance between than within families, heritability of the response estimated at 47%.Nearly half of the spread in how much a person's VO2max responds to a given program is inherited; the average response in a textbook is not what any individual should expect.
For an athlete: a weak VO2max response to a standard program is a real, partly inherited trait, not a failure of effort — but this study only shows the spread; it does not test whether a different stimulus rescues low responders.
Not from this paper: the shares of low/non-responders and high responders used in the base (about 15% each) are Couzens' practitioner heuristic (REF-436), not a measured distribution — REF-431 Mann 2014 contrasts high and low responders without giving percentages; rescue of low responders by higher intensity is in REF-995 and by a higher dose in REF-996 (both in adults who were untrained or not described as trained); how to analyse individual response is in REF-430 Atkinson & Batterham 2015.
Not from this paper; the base's own practice: track two to three training cycles before concluding a stimulus "doesn't work" for an athlete, and change intensity or modality before declaring low responsiveness.
Limitations
Sedentary cohort — generalisability to trained athletes uncertain
All Caucasian — population-specific
Single fixed protocol, no control group (single-arm); heritability from model fitting, not from identified genes
20 weeks may be too short for slow responders to show full adaptation
VO2max only; heritability of other adaptations not assessed
How this entry was checked and changed
Why this level
regraded 2026-09-27 from A: large uncontrolled training cohort with familial-variance analysis, n=481
Content correction 2026-09-27
"Non-responders (<5% gain) ~10-15%; high responders (>25% gain) ~15%" -> deleted; not in the abstract (pointed to REF-431 as "Not from this paper").
"~17% improvement", "3 sessions/wk", "fat oxidation, hemodynamics measured" -> deleted; the abstract gives ~400 ml/min, 20 weeks, VO2max tested twice before and twice after.
"Cardiac trainability has a stronger genetic component", "Bouchard 2011 21-SNP predictor" -> deleted; not in this abstract. Added maternal transmission 28%.
Level A -> B+ (uncontrolled cohort); Tier 0 and Sport lines added.
Resync 2026-09-28: Attributed the ~15% responder shares to REF-431, which gives no percentages -> attributed to the Couzens heuristic (REF-436).
2026-10-01: cross-reference to SYNTH-13 no longer names an archetype — that model was retired as an unvalidated composite.