Best longevity blood tests and biomarkers to monitor.
Knowing which blood tests to get is half the job; knowing what a good number looks like, how often to recheck, and what to do when it moves is the other half. This guide gives a target and a retest schedule for each longevity biomarker, and explains when a change is real and when it is just noise.
The biomarkers worth monitoring for longevity are the ones that predict hard outcomes and respond to something you can change: ApoB (target under roughly 80 mg/dL for most adults, lower with higher risk), lipoprotein(a) once, HbA1c (under 5.7%) with fasting glucose and insulin, hs-CRP (under 1 mg/L, repeated), blood pressure (under 120/80 where tolerated), plus organ function, ferritin, thyroid and vitamin D annually. Retest the modifiable ones every six to twelve months or three months after a deliberate change. Read each as a trend across draws; a single result outside range is a prompt to repeat, not to act.
- Targets matter more than 'normal ranges': laboratory reference ranges describe the average population, which is not a healthy population. ApoB and HbA1c both have outcome-derived targets well inside the lab's normal band.
- Frequency should match how fast a marker can move: Lp(a) once ever; ApoB and HbA1c every six to twelve months; hs-CRP repeated before any decision, because it spikes with any infection.
- Three months is the shortest useful retest interval for most markers — HbA1c reflects a three-month average and lipid changes take that long to settle after a change in diet, training or medication.
- Blood pressure, VO₂max and grip strength are not blood tests, predict mortality at least as well, and belong on the same monitoring sheet.
- The marker is not the goal. A result is useful only if it changes a decision; a monitoring plan without a planned response to each result is a subscription.
Targets, frequency and what moves each marker
The monitoring sheet
| Biomarker | Evidence-based target | Retest | What moves it | Noise to expect |
|---|---|---|---|---|
| ApoB | < 80 mg/dL general; < 60–70 with higher risk or existing disease | 6–12 months; 3 months after a change | Saturated fat, weight, fibre, statins and other lipid drugs | ±5–10% between draws; fasting not required |
| Lipoprotein(a) | < 50 mg/dL (or < 125 nmol/L) is the risk threshold | Once in a lifetime | Almost nothing — genetic; lipid drugs in development | Stable; confirm a high result once |
| HbA1c | < 5.7%; lower within that band is better | 6–12 months | Weight, activity, carbohydrate quality, sleep, medication | ±0.2–0.3% between draws; anaemia and haemoglobin variants distort it |
| Fasting glucose / fasting insulin | Glucose < 100 mg/dL; insulin in the low-normal range (roughly < 10 µIU/mL) | 6–12 months, with HbA1c | Same as HbA1c; insulin responds earliest | Glucose ±5–10 mg/dL; insulin varies more — true fast, morning draw |
| hs-CRP | < 1 mg/L; 1–3 intermediate; > 3 high | Twice, a month apart, before acting; then annually | Infection, injury, poor sleep, visceral fat, smoking | Large — any illness spikes it for weeks |
| Blood pressure | < 120/80 where tolerated; < 130/80 for most | Home readings weekly; clinic annually | Salt, weight, activity, alcohol, sleep, medication | ±10 mmHg reading to reading — average several |
| Ferritin (with full blood count) | Roughly 30–200 ng/mL; low end is deficiency, high end warrants a look | Annually; sooner with fatigue | Iron intake, blood loss, inflammation (raises it), liver | Rises with any inflammation — pair with hs-CRP |
| eGFR / creatinine; ALT, AST, GGT | eGFR > 90; liver enzymes within range | Annually; before and after any new supplement or drug | Hydration, muscle mass, alcohol, fatty liver, many medicines and supplements | Creatinine rises after hard training; ALT rises with alcohol and weight |
| TSH | Within reference range; symptoms decide further testing | Annually, or with symptoms | Thyroid disease, some medicines, iodine | Moderate; time of day and illness shift it |
| Vitamin D (25-OH) | 30–50 ng/mL (75–125 nmol/L) is sufficient | Annually; 3 months after starting supplementation | Sun, supplementation, body fat | Seasonal swing of 10+ ng/mL is normal |
Reading a trend instead of a result
- Every marker has an expected noise band, listed above. A change smaller than the band is not a change. ApoB moving from 82 to 78 is the same number; HbA1c moving from 5.9 to 5.4 is a real improvement.
- Standardise the draw: same laboratory, same time of morning, same fasting state, no hard training the day before, not within two weeks of an illness. Half of apparent 'changes' come from varying these.
- Three points make a trend; two make a guess. Where a result matters, retest before acting on it, and plot successive results rather than comparing the last two.
- Retest three months after a deliberate change — new training, new diet, new medication — because that is how long lipids and HbA1c take to reflect it. Earlier retests measure the transition, not the result.
- Pair markers that confound each other: ferritin with hs-CRP (inflammation inflates ferritin), HbA1c with a full blood count (anaemia distorts it), creatinine with a note of recent training.
What each result should change
The planned response
| If this is out of target | First response | Escalate to a clinician when |
|---|---|---|
| ApoB | Saturated fat down, fibre up, weight toward target; retest at 3 months | Still above target, or Lp(a) is also high, or existing cardiovascular risk — lipid-lowering medication has the strongest outcome evidence on this page |
| HbA1c / fasting insulin | Weight, daily activity, resistance training, carbohydrate quality, sleep; retest at 3–6 months | HbA1c ≥ 6.5%, or rising despite change — this is where metformin and GLP-1 medicines have outcome evidence |
| hs-CRP (confirmed on repeat) | Visceral fat, sleep, activity, smoking; look for a hidden source — dental, gut | Persistently > 3 mg/L with no explanation |
| Blood pressure | Salt, alcohol, weight, activity; home monitoring for two weeks | Average ≥ 130/80 despite change — treatment has decades of outcome evidence |
| Ferritin low / high | Low: dietary iron, find the source of loss. High: alcohol, weight, check hs-CRP | Low without an obvious cause (needs investigation); high with normal CRP (possible haemochromatosis) |
| Liver enzymes or eGFR | Review every supplement and medicine; alcohol; weight | Any abnormal result — do not self-manage; several supplements on this site can cause it |
| Vitamin D low | Supplement 1,000–2,000 IU daily; retest at 3 months | Still low on supplementation, or very low with bone symptoms |
Frequently asked questions
What ApoB level should I aim for?
Under roughly 80 mg/dL for most adults, and under 60–70 mg/dL with existing cardiovascular disease, diabetes or a high lipoprotein(a). Laboratory 'normal' ranges run well above this because they describe the average population, not a low-risk one. Retest six to twelve months after any change.
How often should I test longevity biomarkers?
Match the interval to how fast the marker moves: lipoprotein(a) once in a lifetime; ApoB, HbA1c and fasting insulin every six to twelve months, or three months after a deliberate change; hs-CRP twice a month apart before acting on it; organ function, ferritin, thyroid and vitamin D annually or around any new supplement or medicine.
What HbA1c is optimal for longevity?
Under 5.7% is the non-diabetic threshold, and lower within that band tracks with lower risk in cohort studies. Note that anaemia and haemoglobin variants distort HbA1c; pairing it with fasting glucose and insulin gives a more reliable picture, and insulin rises earliest as insulin resistance develops.
How much change between two blood tests is real?
It depends on the marker's noise band: roughly 5–10% for ApoB, 0.2–0.3 points for HbA1c, 10 mmHg for a single blood-pressure reading, and a great deal for hs-CRP and ferritin, which move with any inflammation. A change smaller than the band is not a change; a change larger than it should still be confirmed on a repeat draw before treatment changes.
Do I need to fast for longevity blood tests?
For fasting glucose and insulin, yes — a true overnight fast, morning draw. ApoB and HbA1c do not require fasting. Standardising the conditions of each draw — same lab, same time, no hard training the day before, not within two weeks of illness — matters more for interpreting a trend than fasting does.
Which biomarkers are not worth monitoring?
Telomere length, NAD⁺ levels, routine hormone 'optimisation' panels without symptoms and most proteomic 'organ age' scores — none has a validated target, a known noise band or a planned response. Epigenetic-age tests are real science with individual precision too poor to act on from a single result.
Keep reading
- Which blood tests are best for tracking longevity?
Which markers to run, and why — the companion to this monitoring guide.
- Biological age tests: what they measure and what they miss
Why epigenetic clocks are hard to monitor from a single draw.
- VO₂ max testing for longevity
The strongest non-blood biomarker on the sheet.
- Early disease detection
Where monitoring ends and screening begins.
- Free stack check
Several supplements move liver and kidney markers — screen before you test.
More in Biomarkers & testing
- What lab tests monitor efforts to extend lifespan?
Lab tests ranked by whether they meaningfully monitor lifespan-relevant progress: blood pressure and lipid panels, HbA1c and fasting glucose, kidney and liver function, inflammatory markers, and popular 'biological age' tests that do not change any actionable decision.
- Which blood tests are best for tracking longevity?
ApoB, Lp(a), HbA1c, hs-CRP and a handful of others predict lifespan in large studies and cost little. Here is the core panel, what each marker means, and what to skip.
- VO2 max testing for longevity: what it measures, and what the evidence actually shows
VO2 max is one of the strongest predictors of all-cause mortality in the research. What the test actually measures, how accurate estimates are, and where to get one.
- Biological age tests: what they measure and what they miss
Epigenetic clocks are real science with real limits. What each test measures, how much your result can move between draws, and when it changes a decision.