MoreYears reader guide · biomarkers
A measurement can move without health improving.
Biomarkers can help describe biology, risk, or response. Their practical meaning depends on whether they are measured reliably, validated for the population and purpose, and connected to outcomes people actually experience.
From number to meaning
Six questions before “better” means better.
A useful biomarker claim must survive each layer. Passing one layer does not automatically answer the next.
Would the measurement be similar if repeated?
Collection, storage, timing, device variation, laboratory methods, and ordinary biological fluctuation can all change a value.
Does the marker track with an outcome?
An association can make a marker useful for description or prediction, but correlation does not show that changing the marker will change the outcome.
Has it been tested for this population and purpose?
Validation should match how the marker will be used. Performance in one reference population, setting, or sample type may not transfer to another.
Can change in the marker stand in for a real outcome?
A surrogate endpoint needs evidence that intervention-driven change reliably predicts a meaningful outcome. Correlation alone is not enough.
Does the change map to how people feel or function?
A laboratory shift, improved physical function, fewer disabling events, and longer healthy life are distinct outcomes and should be named precisely.
Is there evidence about length of life?
A biomarker can be associated with lifespan without establishing that deliberately moving it extends life. That stronger claim requires stronger outcome evidence.
Common failure modes
Where biomarker stories outrun the data.
Prediction becomes causation
A marker can forecast risk because it reflects underlying biology without being the cause or a useful treatment target.
A lower value becomes “younger”
A model output or composite score needs validated interpretation. A directional change does not by itself establish rejuvenation.
A group average becomes personal certainty
Population-level performance leaves uncertainty around an individual result, especially outside the reference population.
Before you trust a biomarker claim, ask:
- How repeatable is the measurement?
- What collection and laboratory conditions matter?
- Is the marker associated with or causal for the outcome?
- Which population and intended use were validated?
- Is it an accepted surrogate for a meaningful outcome?
- Did the study measure function, healthspan, or lifespan directly?
- How large is the uncertainty around an individual result?
- Does changing the marker improve the outcome, or only the number?