Function Health review: 114 biomarkers, and about six false alarms
A full membership year, two panels, and an uncomfortable amount of arithmetic. The testing is real, the value is genuinely good, and the results screen does something no laboratory would be allowed to do.
Function Health is the best value in consumer blood testing: 114 biomarkers across two panels a year for $365, about $1.60 per result. The assays are standard and clinician review is included. What's missing is any warning that a healthy person should expect roughly six out-of-range flags by chance.
Excellent value for a baseline. Take the results to a clinician, and don't subscribe forever.
Function Health covers 114 biomarkers across two panels a year for $365 — roughly $1.60 per result, the best data-per-dollar in consumer blood testing. The assays are standard and run by reference laboratories. What it doesn't do is prepare you for the near-certainty that a healthy person will get several out-of-range flags from a panel this size.
Not medical advice. Nothing here diagnoses or treats a condition, and none of it replaces a clinician who knows your history. Results and programmes are a starting point for a conversation, not a substitute for one.
The score, axis by axis
Composite = mean of six axes × 10, rounded. Axis weights are equal by design; see methodology. Current price: $365/year · 114 biomarkers · two panels.
What we liked
- 114 biomarkers, run twice a year, for $365 — about $1.60 per result
- Clinician-reviewed notes included as standard
- Blood-based biological age at no extra cost
- The included mid-year retest is what makes a flagged result interpretable
- Price cut from $499 to $365 during 2026, unusually in the consumer's favour
What we didn't
- No warning that ~6 out-of-range results are statistically expected
- Broad screening in asymptomatic adults isn't recommended by preventive-health guidelines
- Twice-yearly testing doubles annual exposure to false positives
- Per-marker guidance is thinner than InsideTracker's prioritised plan
- Cascade testing costs are yours, and can dwarf the membership
The value case, which is strong
Function lowered its annual price to $365 in 2026, down from $499, and includes a mid-year retest. That's 228 marker-results a year across 114 distinct biomarkers — about $1.60 per result, or $3.20 per distinct marker.
Set against InsideTracker's 48-marker Ultimate panel at roughly $489 plus a $149 membership — about $13.29 per result — this is a rout on data-per-dollar. Against buying the same analytes self-pay from Quest or Labcorp it's less dramatic, but you get interpretation and a dashboard those don't provide.
The arithmetic nobody shows you
A laboratory reference range is defined as the central 95% of results in a healthy reference population. That means a perfectly healthy person has a 5% chance of falling outside the range on any single marker — not because anything is wrong, but because that is what a reference range means.
Across 114 markers, the probability of at least one out-of-range result is 1 − 0.95114 ≈ 99.7%, with roughly six flags expected. Real biomarkers correlate within panels, so the true figure is somewhat lower — but the direction and rough magnitude hold, and running two panels a year doubles your annual exposure.
None of this appears on the results screen. It is one paragraph of copy that would make the product meaningfully safer, and its absence is the single biggest reason this scores 72 rather than the low eighties its value would justify.
You tested 114 markers. About six will read out-of-range by chance alone. Here are the two that matter. That paragraph does not exist, and it should.
What the flags cost you
Not the test — the follow-up. A flagged marker leads to a repeat test, then perhaps imaging, then a specialist referral. Each step carries cost, time and its own false-positive rate, and none of it is included in the $365.
The mitigating feature, and it's a real one: the included second panel. A single out-of-range value on a large panel is usually noise; the same value twice is a signal. Building the confirmatory test into the membership is genuinely good design and it's the strongest reason to prefer Function over Superpower, which includes one panel.
Clinician review, and its limits
Function includes clinician-reviewed notes with each draw, which is more than most consumer testing offers and the closest thing in the category to the interpretation layer these products need. It is not the same as a clinician who knows your history, your family, your medications and what you were worried about last year.
Worth stating plainly: broad biomarker screening in asymptomatic adults is not currently recommended by major preventive-health guidelines. That doesn't make the product useless — a considered baseline in a year when you have a reason is defensible — but it does mean this is not a screening programme, whatever the marketing atmosphere suggests.
How we'd actually use it
Once. Buy a single deliberate snapshot in a year when you have a reason to look, take the results to a clinician who knows you, and let them tell you which of your six flags is signal. At $365 for a comprehensive baseline that is genuinely good value against private panel pricing.
What we would not do is run it twice a year indefinitely. Repeated broad screening on an asymptomatic person is how people end up with a scan they didn't need and six months of worry they can't unlearn.
Who should use it
- Use it if you want the broadest baseline available for the money, once.
- Use it if you'll take the results to a clinician who knows your history.
- Consider Quest or Labcorp self-pay if you only want the numbers and have your own interpreter.
- Skip if you're anxious about health, if you have symptoms that need a clinician now, or if you won't act on the results either way.
The bottom line
Function Health earns 72/100. The value is the best in consumer biomarker testing and the assays are standard and sound. The score is held down by an interpretation layer that hands you six red numbers without telling you that six red numbers were the statistically expected outcome.
Sources
- Function Health — $365 annual membership, 114 biomarkers, two panels, clinician-reviewed results. Accessed 1 Sep 2026
- Applandica calculation: probability of at least one out-of-range result = 1 − 0.95n at a 95% reference interval; n = 114 → 99.7%, expected flags ≈ 5.7. Accessed 1 Sep 2026
- Clinical and Laboratory Standards Institute convention for reference intervals: the central 95% of a healthy reference population. Accessed 1 Sep 2026
Corrections & updates. None yet. We re-test every app every 12 months or every two major releases, whichever comes first, and publish any score change here, dated and signed.