Biological age is an estimate of how old your body is functioning, as opposed to how many birthdays you have had. Two people born the same year can have biological ages twenty years apart. One has the arteries, blood sugar, and muscle of someone in their 40s. The other has the numbers of someone in their 60s. Chronological age is a fact. Biological age is a measurement, and like every measurement it depends on the instrument. A biological age test collects traces of that damage (blood markers, DNA chemistry, fitness capacity, body composition), compares them against a large reference population, and reports the chronological age at which the average person has your numbers.
That last part matters, because four different tests will give you four different body ages, and the marketing rarely tells you which one you are buying.
"Body age" is the consumer term for biological age. Your gym's scale, a smartwatch, a fitness assessment, and a $500 blood test all use it. They are not measuring the same thing.
Every version starts from the same idea. Aging is not one process but the slow accumulation of damage across many systems, and that damage leaves traces you can measure. Each test picks some of those traces and reads an age off them.
So a body age of 52 in a 45-year-old means: your measured markers look like the average 52-year-old's. It does not mean you have seven fewer years to live. It means the instrument you used is reading older than your birthday, on the markers that instrument happens to measure.
| Method | What it measures | Typical cost | What it is good for |
|---|---|---|---|
| Fitness-based (VO2 max, grip strength, body composition) | Functional capacity now | Free to $150 | Tracking training. VO2 max carries the most mortality data of the group. |
| Phenotypic (blood biomarker) age | Nine routine blood values (glucose, albumin, creatinine, CRP, white cell count, and others) run through a published formula | Free if you already have a CBC, metabolic panel, and CRP from the last year; otherwise the cost of those three tests | Cheap, repeatable every few months, and responds to lifestyle change |
| Epigenetic clocks (Horvath, GrimAge, PhenoAge clocks) | DNA methylation patterns at hundreds of sites | $300 to $500 | Research-grade estimate of accumulated age. Moves slowly. |
| Pace of aging (DunedinPACE) | How fast you are aging right now, in biological years per calendar year | $300 to $500, often bundled with a clock | The one built to measure rate of change rather than accumulated age |
This is what the gym scale and the smartwatch report. It takes your VO2 max (measured or estimated from heart rate), sometimes grip strength, body fat, and resting heart rate, and compares them with age norms. A 2018 study in JAMA Network Open of more than 122,000 patients found that cardiorespiratory fitness was as strong a predictor of death as smoking, diabetes, or heart disease, and in the least fit group stronger. So a fitness age is not trivial. But it only sees fitness. A marathoner with a fasting insulin of 25 will get a flattering number.
Morgan Levine's group at Yale published PhenoAge in 2018 in the journal Aging. It uses nine values from a standard blood panel plus chronological age, and it predicted mortality better than chronological age alone in a national sample. The formula is public. If you have had a comprehensive metabolic panel, a CBC with differential, and a C-reactive protein in the last year, your phenotypic age can be calculated today for free. This is the version I use most in practice because it moves within months when someone fixes their metabolic health, and it costs nothing extra.
In 2013 Steve Horvath at UCLA published a clock built on DNA methylation, chemical marks on DNA that change predictably with age. Later clocks, GrimAge and the epigenetic PhenoAge, were trained to predict mortality rather than chronological age and do it better. Two cautions. First, the number moves slowly, so a retest in three months mostly measures the test's noise. A 2022 analysis in Nature Aging found that technical noise alone produced deviations of up to nine years between replicates of the same sample on six common clocks, and that a reliability-corrected version of each clock brought most replicates within 1.5 years. Ask your lab which version it runs. Second, different companies use different clocks, and the same blood sample can come back with body ages several years apart depending on which clock is run.
DunedinPACE, published in 2022 in eLife by Daniel Belsky's team, is built differently. Instead of asking "how old do you look," it asks "how fast are you aging," calibrated against two decades of repeated measurements in the Dunedin birth cohort. A score of 1.0 means one biological year per calendar year. A 0.8 means you are aging 20% slower than average. This is the test we run in the longevity program because it is built to measure the rate of aging rather than the total, which is the question an intervention asks.
Trust the direction more than the number.
A single biological age result is a noisy snapshot. Two results from the same test a year apart, with the same lab, tell you something. My order of usefulness for a healthy adult who wants to act on the result:
What I would skip: any body age that comes from a bathroom scale, and any test that will not tell you which clock or formula it uses.
If you want the phenotypic age run on a real panel, see how Rebel Health Alliance does it.
The same things that move every marker on the list above. In our own members, 35 with labs through March 2026, fasting insulin fell 38% in the group that started above 15, from 26.7 to 16.7, over the period of a practice audit with no control group. Individual results vary. Insulin resistance drives CRP and glucose, two inputs to phenotypic age. The full numbers and the caveats are on our Results page.
The levers, in order of evidence: cardiorespiratory fitness (see what counts as cardio), muscle mass, insulin sensitivity (see the best exercise for insulin resistance), sleep, not smoking, and body fat, visceral fat in particular. Supplements on their own, marketed to "reverse" epigenetic age, have not shown that in a controlled trial I would cite.
Rebel Health Alliance members start with an initial panel of more than 100 biomarkers, drawn from more than 3,000 available, enough to calculate phenotypic age on day one, and the longevity program adds DunedinPACE for pace of aging. Your physician reads it with you, and you can message them any time, not on a schedule. Rebel Peak adds a dietitian and strength coach to build the plan. See how it works, book a 20-minute call, or read the FAQ. If you want the bigger picture on when decline starts, read how your body ages.
What is a good biological age?
Lower than your chronological age. Five or more years younger on a phenotypic age test is a gap larger than that test's noise, so it likely means something. On an epigenetic clock a single result can carry that much noise on its own, so treat the first one as a baseline, not a verdict. A DunedinPACE below 1.0 means you are aging slower than average.
Can biological age go down?
Measured biological age can go down. In a 2021 pilot trial published in Aging, eight weeks of diet, sleep, exercise, and relaxation changes, plus two supplements, lowered epigenetic age by about three years compared with a no-intervention control group, in a small group of men aged 50 to 72. Metabolic and fitness markers move reliably with lifestyle change; whether that adds years of life is the part still being proven.
Is a body age from a smart scale accurate?
No. It estimates body composition from electrical impedance and maps it to age norms. It is a rough fitness proxy, not a biological age test.
How often should I test biological age?
Blood-based phenotypic age every three to six months if you are actively changing something. Epigenetic clocks and DunedinPACE once a year at most.
Dr. Alec Weir is the Chief Medical Officer at Rebel Health Alliance (https://rebelhealthalliance.io), a 100% virtual, physician-led longevity practice in all 50 states. This article is for education only and is not medical advice.