What does 230 biomarkers buy you that 89 does not?

Every premium health programme leads with a number. Seventy datapoints. A hundred markers. Two hundred and thirty. The number is easy to compare, which is why it is the number they publish, and it correlates with price far more reliably than with anything a physician would act on.

Panel sizes in the Baseline Index run from about 70 datapoints at Neko Health, through 100-plus at Function Health and Superpower and 120-plus at Q Bio, to 200-plus at Fountain Life and 230 at the top YEARS tier. YEARS is the useful case study because it publishes three panel sizes for the same underlying programme: 87 biomarkers at €1,900, 120-plus at €7,600 and 230-plus at €16,900.1 Same clinic, same modalities, same day. The variable is isolated.

So what does the increment actually buy?

The first twenty markers do most of the work

In an asymptomatic adult, a fairly short list covers the large majority of findings that change what happens next. A lipid panel, ideally including ApoB. HbA1c or fasting glucose with insulin. Creatinine with eGFR. Liver enzymes. TSH. A full blood count. Ferritin. Vitamin D. An inflammatory marker such as hs-CRP.

These are cheap, well standardised, and attached to interventions with their own outcome evidence. They are also, with the exception of ApoB, insulin and hs-CRP, essentially what the German statutory check-up already provides free of charge.

Between twenty and sixty, you buy resolution

The next band is where a good panel earns its money in a way the first band cannot. Lipoprotein(a), measured once in a lifetime because it is genetically determined. Fasting insulin with HOMA-IR rather than glucose alone. Thyroid antibodies where TSH is borderline. Iron studies rather than ferritin in isolation. Homocysteine. Uric acid. A full lipid subfraction picture.

None of these are exotic, and several answer questions the basic panel raises without resolving. This is the band where a wider panel changes management with reasonable frequency.

Above sixty, the curve flattens

Beyond roughly 60 analytes, additional markers tend to fall into three groups.

Some are highly correlated with markers already measured, so they add confidence rather than information. Some are research-grade, meaning the association with outcomes exists at population level but no threshold triggers a defined action in an individual. And some are actionable in principle but not in an asymptomatic person, because the finding only means something in the presence of symptoms the patient does not have.

This is where the honest answer to the headline question sits. Going from 89 to 230 markers in a healthy 45-year-old will rarely change what anybody does on Monday morning.

Panel size is a proxy for thoroughness that stopped tracking thoroughness somewhere around the sixtieth analyte.

The argument that does survive

There is one, and it is not about detection.

A wide panel measured once, while you are well, establishes your own reference range. Population reference intervals are constructed so that 95% of a reference population falls inside them, which means your personal normal may sit near an edge, and a future result drifting toward the middle of the population range could represent a real change in you that no threshold flags. Having 230 of your own values from a year when you felt fine is genuinely useful when something changes later.

This reframes what a large panel is for. It is a baseline purchase, not a detection purchase. Bought that way, the logic points toward doing it once, thoroughly, reasonably early, and then tracking a much smaller set annually.

It is worth noting that YEARS structures its tiers close to this line, and more deliberately than most. Its €1,900 Core programme is the biomarker-and-function tier: 87 analytes plus spiroergometry, echocardiography, vascular ultrasound, cognition and lung function, with no imaging, no liquid biopsy and no genetics. The one-off structural measurements sit above it, with whole-body MRI and a multi-cancer liquid biopsy at €7,600 and genome sequencing, epigenetic clocks and microbiome analysis at €16,900, where the panel reaches 230-plus. That ordering puts the repeatable measurements at the bottom and the buy-once measurements at the top, which is the right way round.

Why this grades C

Grade C means plausible and widely used with no trial evidence of net benefit in the screened population. Extended biomarker panels have never been tested against a limited panel for outcomes in asymptomatic adults. The theoretical case for personal baselines is sound and untested.

The countervailing consideration is statistical and unavoidable. Measure enough analytes in a healthy person and chance alone puts some outside the reference range. Each one invites follow-up, and the follow-up carries the same cascade risk documented for incidental imaging findings.4 The mitigation is a physician who reads the whole panel in context rather than a portal that colours 11 results amber, which is an argument for supervised programmes over mail-order testing regardless of panel size.

What to ask instead of the count

Which specific markers are included, in a published list, so you can check whether ApoB and Lp(a) are among them rather than assuming that 200 markers must include the useful ones.

Who reads the result, and whether the same clinician sees the panel alongside the imaging and the functional testing. An integrated read is worth more than 100 extra analytes.

What happens to a borderline value. A programme with a defined threshold for repeating rather than escalating is doing something a portal cannot.

What would change the grade

A trial randomising asymptomatic adults to a focused panel against an extended one, measuring downstream investigations, diagnoses and outcomes, would settle this. It would be inexpensive by trial standards and would probably not flatter the extended arm.