Biological age, honestly.

A clear, careful guide to what biological age actually is, the four ways scientists measure it, what the peer-reviewed literature says about moving it, and how Heaven Health tracks yours over time. No hype, no promises — just the picture as it stands in 2026.

What biological age is

Chronological age is the number of birthdays you have had. It moves forward at the same speed for everyone. Biological age is something else — an estimate of how old your body behaves at a cellular and physiological level. Two people born on the same day can have biological ages five or ten years apart, depending on how their genes, environment and lifestyle have interacted over decades.

Researchers introduced biological age because chronological age alone is a clumsy predictor of who will get sick when. A more useful question for medicine and for individuals is: how worn is this body, right now, compared to other bodies its calendar age? Biological age is the field's attempt at a number that answers that question.

It is still an active research area. The methods are improving every year. The Heaven Research Labs pipeline reads new biological-age papers from PubMed daily — over the last year we have indexed several hundred — and the picture in 2026 is: more reliable than it was in 2018, less settled than it will be in 2030.

The four ways scientists measure it

There are roughly four families of biological-age estimate in the literature today. Each is good at something different.

1. Epigenetic clocks (DNA methylation)

These look at patterns of methylation — small chemical tags — on your DNA at thousands of sites across the genome. The pattern shifts predictably with age. The most cited clocks include:

• Horvath clock (2013) — the original, trained on 8,000 samples. Accurate to within a few years for most tissues.
• Hannum clock (2013) — blood-specific, similar accuracy.
• PhenoAge / DNAm PhenoAge (2018) — refined to predict mortality and morbidity, not just calendar age.
• GrimAge (2019) — currently the most predictive of mortality in the literature.
• DunedinPACE (2022) — measures rate of aging rather than age. Built from the Dunedin longitudinal cohort.

Epigenetic clocks require a blood or saliva sample sent to a regulated lab. They are the most validated method we have. Heaven Health does not perform epigenetic testing — but if you do an epigenetic test, you can log the result inside the app and Luna will track it over time alongside your other markers.

2. Phenotypic age (composite blood biomarker formulas)

A different approach: take nine common blood markers — albumin, creatinine, glucose, C-reactive protein, lymphocyte percentage, mean corpuscular volume, red cell distribution width, alkaline phosphatase and white blood cell count — and feed them into a regression formula calibrated against mortality. The output is "Phenotypic Age" (Levine et al., 2018).

The appeal of PhenoAge is that the inputs are markers most clinicians already test. If you have had a standard blood panel in the last year, Heaven Health can compute a PhenoAge-style estimate for you. The estimate is not a substitute for an epigenetic test, but it correlates meaningfully with one and it is far more practical to repeat.

3. Telomere length

Telomeres are the protective caps at the ends of your chromosomes. They shorten with cell division, and shorter telomeres are associated with older biological age and various chronic conditions. Telomere length can be measured from a blood sample.

The current research view: telomere length correlates with age but the year-to-year noise within an individual is high, which limits its usefulness for tracking change. It is more useful for population studies than personal tracking. Heaven Health does not lean on telomere measurement for this reason.

4. Biomarker-derived models

The most pragmatic family: combine measurements that are easy to collect at scale — blood pressure, resting heart rate, HRV, sleep quality, body composition, grip strength, VO₂ max, basic blood markers — into a composite "physiological age" estimate. The output is not as precise as an epigenetic clock but it tracks change well over months and is built from data you can collect from a phone, a watch and an annual blood panel.

This is the family Heaven Health works in. We are honest about its limits and we tell you what each input contributes.

How Heaven Health tracks your biological age

Inside the Heaven Health app, your biological-age estimate is computed from data you already provide. The more you provide, the more accurate it becomes.

• Bloodwork. The largest single input. Any panel you photograph and upload — your annual physical, a longevity-focused panel from Function Health or InsideTracker, a private order — feeds a PhenoAge-style composite. We extract the markers automatically.
• Sleep. Duration, regularity, and (if your wearable measures it) deep and REM proportions. Sleep is the single most controllable input on the list.
• HRV and resting heart rate. Long-term proxies for autonomic balance and cardiovascular fitness. Drawn from Apple Health or Google Fit.
• VO₂ max estimate. If your watch reports it, Luna reads it. Higher VO₂ max is associated with lower biological age in multiple cohorts.
• Heaven Score trend. The four-week trend in your Heaven Score contributes a small modifier — sustained high scores are associated with healthier biology.
• Self-reported variables. Smoking status, alcohol, medication list. Honesty here matters.

Luna shows you the estimate and the curve. She does not show you a precise decimal value — biological-age estimates are not that precise and false confidence is worse than honest uncertainty. The interesting number is the trend over months.

What the literature says actually moves the number

The peer-reviewed evidence base for "interventions that slow or reverse biological age" is small but growing. The Heaven Research Labs engine has indexed the major studies. Here is the careful picture, with sources you can read for yourself.

Caloric restriction (moderate, sustained)

The CALERIE-2 trial (Belsky et al., 2023, Nature Aging) found that two years of ~12% caloric restriction in healthy adults was associated with a slower rate of aging measured by DunedinPACE. The effect was modest but statistically meaningful. This is among the strongest pieces of human evidence we have.

Exercise — both cardio and resistance

Multiple cohort studies link higher VO₂ max and regular resistance training to lower biological-age estimates. The mechanism is not a single pathway — it is mitochondrial function, glucose regulation, inflammation, and more. Zone-2 cardio three times a week and two strength sessions a week appears in most longevity protocols for this reason.

Sleep regularity

Sleep regularity — the consistency of when you sleep, not just how long — has been associated with better metabolic and cardiovascular markers in several recent studies. This is one of the easiest levers and one of the cheapest. Luna nudges you toward it.

Diet patterns: Mediterranean, plant-forward, sufficient protein

The pattern with the strongest longevity literature is Mediterranean-style: olive oil, fish, vegetables, legumes, modest dairy, low ultra-processed food. The PREDIMED trial and many cohorts support it. Protein adequacy — at least 1.2g per kg body weight per day for most adults, more for those over 60 — has its own evidence base for sarcopenia prevention.

Stress and connection

Chronic stress accelerates several biological-age clocks. Social connection — measured crudely as "do you have people you can talk to" — predicts mortality independently of other variables across many large cohorts. The data is more robust than the wellness world tends to admit.

What does not have strong evidence (yet)

This is the part the supplement industry does not want said. As of 2026, the human evidence for the following lowering biological age in a measurable, replicable way is weak or absent:

• NMN supplements — early-stage human data, no robust biological-age outcomes yet
• Resveratrol — mouse data did not replicate well in humans
• Most "anti-aging" cream-based interventions
• IV vitamin therapies marketed for longevity

This may change. The Heaven Research Labs engine reads the new studies daily and Luna will tell you when the evidence shifts. Until then, Heaven Health does not promote these.

How often to test

For a clinically validated epigenetic test: once a year is usually enough. Twice a year if you are actively running a longevity protocol and want a tighter feedback loop. More often than that is generally over-testing — the year-to-year change is small in healthy adults.

For the Heaven Health estimate: it updates continuously as your bloodwork, sleep and activity data come in. Most users see a meaningful trend stabilise around month three.

What to ask your doctor

If you are interested in biological age and you want your clinician's input, the questions worth asking on your next visit:

• Can my annual panel include ApoB, fasting insulin, hs-CRP, HbA1c, ferritin and homocysteine?
• Given my numbers, what one to two markers would most benefit from active management?
• Is a VO₂ max test available locally and would you support it?
• What is your view on epigenetic age testing for someone with my history?

Most clinicians are interested. The longevity literature is no longer fringe — major academic medical centers (Harvard, Stanford, UCSF) now have geroscience clinics. Your GP may not, but they will usually engage thoughtfully with the questions.

Primary sources cited above

For readers who want to read the studies, not just the summary:

• Horvath, S. (2013) "DNA methylation age of human tissues and cell types." Genome Biology. PMID: 24138928
• Levine, M. E. et al. (2018) "An epigenetic biomarker of aging for lifespan and healthspan." Aging. PMID: 29676998
• Lu, A. T. et al. (2019) "DNA methylation GrimAge strongly predicts lifespan and healthspan." Aging. PMID: 30669119
• Belsky, D. W. et al. (2022) "Quantification of the pace of biological aging in humans through a blood test." eLife. PMID: 35029144 (DunedinPACE)
• Waziry, R., Ryan, C. P., Corcoran, D. L. et al. (2023) "Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial." Nature Aging. doi: 10.1038/s43587-022-00357-y
• Estruch, R. et al. (2018) "Primary Prevention of Cardiovascular Disease with a Mediterranean Diet Supplemented with Extra-Virgin Olive Oil or Nuts." NEJM. PMID: 29897866 (PREDIMED)