TallyAge Review: What the Epigenetic Age Test Can and Cannot Tell You

Short answer: TallyAge is a real DNA-methylation aging test built on a peer-reviewed buccal-cell epigenetic clock. The underlying CheekAge model has published associations with lifestyle, disease-related signals and mortality. But a TallyAge number is still a biomarker estimate, not a diagnosis, a direct measurement of “how old your body really is,” or proof that lowering the score will make you live longer.

The science behind the test has improved substantially since the original version of this review was published in 2023, so this article has been updated to separate what has now been validated from what remains a commercial or biological-age interpretation.

What is the TallyAge test?

TallyAge is an at-home cheek-swab test from Tally Health. The laboratory measures DNA methylation, chemical marks on DNA that vary with age, cell type, exposures and biological state.

The result is reported as an epigenetic-age estimate. Tally Health currently sells a standalone test for $249. Its membership currently costs $129 per month and includes repeat testing every six months plus action plans and supplements. Prices and membership contents can change, so the company’s current site is the appropriate source for purchasing details.

Current TallyAge test page.

What is CheekAge?

The scientific model behind TallyAge is called CheekAge, a next-generation epigenetic clock trained using DNA methylation from more than 8,000 adult cheek-cell samples.

Unlike early epigenetic clocks that were optimized mainly to predict chronological age, CheekAge was developed to capture methylation patterns associated with health and lifestyle as well as age.

A 2024 peer-reviewed paper reported associations between CheekAge and variables including BMI, smoking, alcohol use, exercise, sleep and other health-related factors.

CheekAge: a next-generation buccal epigenetic aging clock associated with lifestyle and health.

Does CheekAge predict mortality?

A 2024 follow-up study applied the CheekAge model to DNA methylation data from the Lothian Birth Cohorts and found that the score was significantly associated with mortality.

This is a stronger result than merely showing that the clock can guess chronological age. A biomarker associated with future mortality may be capturing biologically relevant variation.

But there is an important caveat: the validation was performed in blood methylation datasets, despite CheekAge being trained on cheek cells, and nearly half of the model’s CpG inputs were unavailable in those data. The authors used the remaining information and still found a mortality association.

That makes the result scientifically interesting, but it does not mean an individual TallyAge score gives a precise personal life-expectancy forecast.

CheekAge is predictive of mortality in human blood.

Does a younger TallyAge mean you are healthier?

On average, the clock is associated with a range of health and lifestyle signals. A 2025 study reported associations between CheekAge and multiple diseases and conditions across publicly available methylation datasets.

That supports the idea that the score contains information beyond calendar age.

But an association at population level does not make the number a clinical diagnosis for an individual. Two people with the same TallyAge can still have very different cardiovascular risk, cancer risk, fitness, metabolic health and life expectancy.

CheekAge associations with diseases and conditions.

If your TallyAge drops, have you reversed aging?

Not necessarily.

This is the key interpretation problem with all aging clocks.

A score can change because the methylation pattern measured by the algorithm changed. To claim that aging itself has been reversed, we would need evidence that an intervention-induced change in the clock reliably predicts better hard outcomes such as lower disease incidence, preserved function or longer survival.

That causal bridge has not been established for TallyAge.

Tally Health reports that a majority of members lower their epigenetic age over time. This is company member data, not a randomized controlled trial. Members who keep retesting may also differ from non-members in health behavior, motivation and other factors.

How repeatable is the test?

DNA methylation measurement itself has technical variability, and biological samples also vary with cell composition and other factors. Tally Health describes the test as intended for repeated measurements and trend tracking.

The company has published methodological work on methylation platforms and epigenetic-clock reproducibility. That is useful, but it is worth noting that much of the CheekAge validation literature is authored by Tally Health scientists.

This does not make the research invalid. It does mean that independent replication is especially valuable before interpreting a commercial biomarker as a clinical endpoint.

Is TallyAge a medical diagnostic test?

No. It should not replace standard medical assessment.

Aging clocks are research and wellness biomarkers. They do not replace blood pressure, lipids, glucose testing, cancer screening, exercise capacity, body composition, kidney function or other validated measurements that answer specific clinical questions.

If your TallyAge result moves in an unfavorable direction while your validated health measures improve, the test should not automatically overrule those real clinical outcomes.

What can TallyAge be useful for?

I think the most defensible use is as an experimental longitudinal biomarker.

If you enjoy quantified-self experiments, repeating the same test under reasonably similar conditions can add another data series alongside established measurements such as weight, waist circumference, blood pressure, fitness and laboratory markers.

The interesting question is not whether one test says you are “five years younger.” It is whether a sustained change in the score tracks with meaningful improvements in other health variables over time.

What are the biggest limitations?

No proven lifespan endpoint. There is no trial showing that people who lower TallyAge live longer because they lowered the score.

Commercial validation. Much of the peer-reviewed CheekAge literature comes from researchers employed by Tally Health.

Model dependence. Different epigenetic clocks are trained on different outcomes and can give different age estimates from the same person.

Individual interpretation. A population-level association does not translate directly into a precise personal prognosis.

Regression and measurement variability. Repeat values can move partly because of biological or technical variation, especially when the first score is unusually high or low.

Bottom line

TallyAge is more scientifically interesting in 2026 than it was when I first reviewed it in 2023. CheekAge now has peer-reviewed evidence linking it with lifestyle, disease-related signals and mortality.

But the correct interpretation remains narrower than the marketing phrase “how old are you really?”

The test estimates a methylation pattern associated with aging-related variables. It does not directly measure total biological age, and lowering the result has not been shown to extend life.

For a quantified-self experiment, that can still be useful. I would treat it as one experimental biomarker among many, not as a verdict on whether a lifestyle intervention is making you younger.

References

1. Shokhirev MN et al. CheekAge: a next-generation buccal epigenetic aging clock associated with lifestyle and health. Geroscience. 2024.

2. CheekAge, a next-generation epigenetic buccal clock, is predictive of mortality in human blood. 2024.

3. Various diseases and conditions are strongly associated with the next-generation epigenetic aging clock CheekAge. 2025.

4. Contextualizing aging clocks and properly describing biological age. Aging Cell. 2024.