Most of them move, and a few genuinely do not. Triglycerides, VO2 Max, waist-to-height ratio, strength and the functional tests all respond substantially to training and diet. HbA1c, ApoB, blood pressure and resting heart rate respond partly, with a genetic floor underneath. Lp(a) is largely inherited and does not respond meaningfully to anything you do.
The three buckets
| Bucket | Markers | Realistic movement |
|---|---|---|
| Responds well | Triglycerides, VO2 Max, waist-to-height, body fat, strength, sit-to-stand, grip, walk test | Large, within months |
| Responds partly | HbA1c, ApoB, blood pressure, resting heart rate, hs-CRP, liver enzymes | Real, against a genetic floor |
| Largely fixed | Lp(a), cystatin C, TSH and free T3 | Little to none from habits |
What moves, and how much
The first bucket is where effort converts most directly into a different number.
Triglycerides move fastest of any lipid, with consensus guidance putting the effect of aerobic exercise alone at roughly 10 to 30 percent, before touching alcohol or added sugar. VO2 Max moves furthest: an untrained person has enormous room, and it carries more weight in the scoring than anything else.
Body composition and the functional tests belong here too, and the functional ones deserve emphasis because people assume otherwise. Sit-to-stand and grip improve with training at any age, including in the eighties.
One caveat that applies across this bucket: trainability itself is partly inherited. The HERITAGE family study found the VO2 Max response to an identical training program varied several-fold between individuals. Everyone improves; not everyone improves at the same rate, and a slower responder is not doing it wrong.
What moves against resistance
The middle bucket is the honest one, and the one most often misrepresented in both directions.
ApoB responds to diet and exercise, and it also has a substantial genetic component through receptor activity and cholesterol synthesis. Two people making identical changes will land on different numbers. That is not a failure of effort on the part of the one who lands higher.
The same applies to HbA1c, blood pressure and resting heart rate. All three move meaningfully with training, weight loss and dietary change, and all three have a floor below which a given person will not go without medication.
The right conclusion is not resignation. It is that the target should be your own trajectory rather than someone else's absolute number.
What does not move
Lp(a) is the clearest case in the entire model. Concentration is largely set by variation at the LPA gene, and the app treats it as largely genetic, with heritability estimates of roughly 70 to 90 percent. That specific figure is our own working estimate rather than one taken from a named study, and the benchmark sources page records it as such.
Diet does not shift it. Exercise does not shift it. Statins, which lower LDL substantially, do not lower Lp(a). This is why the app excludes it from projections entirely and says so on screen rather than drawing a hopeful line for a marker that will not follow one.
Cystatin C and the thyroid markers sit here for a different reason. They can change, but they track kidney and thyroid function, so they move when those change rather than in response to habits, and not on any timescale a projection could honestly represent.
Why the distinction changes your plan
Sorting markers this way answers a practical question: where should the next six months go?
Into the first bucket, almost always. It contains the heaviest-weighted domain in the model and the markers with the most room to move. The middle bucket rewards the same work and rewards it more slowly. The third rewards nothing you do, and its value is entirely as context.
What counts as elevated
The page would be no use telling you to measure Lp(a) without saying what the number means. The 2022 European Atherosclerosis Society consensus statement sets it out.
Below 30 mg/dL (75 nmol/L) rules Lp(a)-attributable risk out. Above 50 mg/dL (125 nmol/L) rules it in. Between the two is an explicit grey zone where the answer depends on your other risk factors.
Those cut-offs are a clinical convention rather than a biological cliff. The relationship between concentration and cardiovascular risk is continuous, with no true threshold effect.
Roughly 20% of people are above 50 mg/dL, and about 0.3% above 175 mg/dL (438 nmol/L). Testing rates run at around 1 to 2%. Updated ESC/EAS guidance recommends measuring it at least once in every adult's lifetime, which is the strongest argument for doing it that exists.
The tier boundaries the app uses below 75 nmol/L are our own interpolation rather than guideline values, and the benchmark sources page says so.
That context is worth having, though. An elevated Lp(a) does not respond to effort, and precisely because it raises your baseline, it makes the effort you put into the first two buckets worth more than it would otherwise be.
That has been measured, not just argued. In EPIC-Norfolk, among people with Lp(a) above 50 mg/dL, those with few traditional risk factors had a one-third to two-thirds lower risk of an atherosclerotic cardiovascular event over 11.5 years of follow-up than those with an unhealthy lifestyle. Same unchangeable genetic burden, substantially different outcomes.
One note on the boundary of this third bucket. It is fixed with respect to habits, and that is what the page means by fixed. Several Lp(a)-lowering therapies are in late-stage clinical trials, so the category may not stay closed indefinitely.
Sources
Key references for the claims on this page. Where a figure is attributed to a specific study or body, it is named here.
- Kronenberg F, Mora S, Stroes ESG, et al. Lipoprotein(a) in atherosclerotic cardiovascular disease and aortic stenosis: a European Atherosclerosis Society consensus statement. European Heart Journal, 2022;43(39):3925. Source of the thresholds, the grey zone, the prevalence and testing figures, and the EPIC-Norfolk finding. PMID 36036785
- Bouchard C, An P, Rice T, et al. Familial aggregation of VO2max response to exercise training: results from the HERITAGE Family Study. Journal of Applied Physiology, 1999;87(3):1003–1008. Trainability itself varies substantially between individuals. PMID 10484570
Frequently asked
If Lp(a) cannot be changed, why measure it?
Because it changes what your other numbers are worth. Cardiovascular risk is multiplicative, so an elevated Lp(a) raises the value of controlling ApoB, blood pressure and smoking status rather than lowering it. It is also heritable, which makes it information your family may want.
Does a genetic floor mean effort is wasted?
No. It means the achievable range is narrower than the population range, not that the range is zero. Someone with a genetically higher ApoB can still move it substantially with diet, exercise and, if warranted, medication. They may simply not reach the same absolute number as someone who started lower.
Which single marker rewards effort most?
For most people, VO2 Max. It carries the heaviest weight in the scoring, the untrained have the most room to move, and the improvement is achievable with time rather than talent. Triglycerides move fastest, but VO2 Max moves furthest.
Are the functional tests trainable?
Very. Sit-to-stand, grip and the walk test all improve with resistance training and practice, and they improve at any age. They sit firmly in the responsive bucket.