Biological Age Test: What Epigenetic Clocks Measure and What Your Results Actually Mean
What Biological Age Actually Means
Chronological age is simple: the number of years since you were born. Biological age is an estimate of how old your body appears at the molecular level based on measurable biomarkers. Two 50-year-olds can have very different biological ages depending on genetics, lifestyle, disease burden, and environmental exposures. One might have the molecular profile of a typical 42-year-old. The other might look closer to 58.
The concept is not new. Physicians have long recognized that some patients age faster than others. What changed in the last decade is the ability to quantify this difference with molecular precision. The most validated approach uses DNA methylation, a chemical modification where methyl groups attach to specific sites (called CpG sites) on your DNA. These modifications change predictably with age, and by measuring patterns across hundreds of these sites, algorithms can estimate biological age from a single blood sample. A TruAge Complete test measures these methylation patterns across the genome to generate your biological age estimate.
How Epigenetic Clocks Work
Epigenetic clocks are mathematical algorithms trained on DNA methylation data from large populations. Researchers collect blood or tissue samples from thousands of people of known ages, measure methylation levels at hundreds of thousands of CpG sites across the genome, then use machine learning to identify the subset of sites whose methylation patterns most accurately predict chronological age. The resulting model assigns a weight to each selected CpG site. When you submit a new sample, the algorithm reads methylation at those specific sites and calculates a weighted sum that outputs an age estimate in years.
The first widely validated epigenetic clock was published by Steve Horvath in 2013. Using 8,000 samples from 82 datasets spanning 51 tissue and cell types, Horvath identified 353 CpG sites whose methylation levels, combined through an elastic net regression model, predicted chronological age with a correlation of 0.96 and a median error of 3.6 years in independent test data (Horvath, Genome Biol, 2013; PMID 24138928). The same year, Gregory Hannum and colleagues published a blood-specific clock based on 71 CpG sites from whole blood samples of 656 individuals aged 19 to 101, building a model that could quantify individual differences in aging rate (Hannum et al., Mol Cell, 2013; PMID 23177740).
First Generation vs. Second Generation Clocks
The Horvath and Hannum clocks are considered first generation epigenetic clocks. They were trained to predict chronological age as accurately as possible. This makes them excellent timekeepers, but it also means they were optimized for the wrong target. Chronological age is just a proxy for what people actually care about: disease risk, functional decline, and mortality. A clock that perfectly predicts your birthday tells you less about your health than one that predicts your likelihood of dying or getting sick in the next decade.
Second generation clocks addressed this limitation by training on health outcomes rather than chronological age. In 2018, Morgan Levine and colleagues developed DNAm PhenoAge, which incorporated clinical measures of phenotypic age (including albumin, creatinine, glucose, C-reactive protein, lymphocyte percent, mean cell volume, red cell distribution width, alkaline phosphatase, and white blood cell count) alongside chronological age. PhenoAge strongly outperformed previous clocks in predicting all-cause mortality, cancer incidence, healthspan, physical functioning, and Alzheimer's disease (Levine et al., Aging, 2018; PMID 29676998).
In 2019, Ake Lu, Steve Horvath, and colleagues took this further with DNAm GrimAge. Rather than predicting age directly, GrimAge uses DNA methylation to estimate plasma levels of seven proteins associated with mortality (including plasminogen activator inhibitor 1 and growth differentiation factor 15) plus smoking pack-years, then combines these surrogates into a composite mortality predictor. In validation across thousands of individuals, GrimAge dramatically outperformed all previous clocks for predicting time to death (Cox regression p = 2.0 x 10^-75), time to coronary heart disease (p = 6.2 x 10^-24), and time to cancer (p = 1.3 x 10^-12). It also showed strong associations with fatty liver disease, visceral fat accumulation, and comorbidity count (Lu et al., Aging, 2019; PMID 30669119).
DunedinPACE: Measuring How Fast You Are Aging Right Now
Most epigenetic clocks answer the question: how old does your body look? DunedinPACE asks a different question: how fast is your body aging right now? Published by Daniel Belsky and colleagues in 2022, DunedinPACE (Pace of Aging Calculated from the Epigenome) was built using a unique dataset: the Dunedin Study, a birth cohort of 1,037 people born in New Zealand in 1972-1973 and followed with repeated biological measurements across two decades. Researchers tracked within-individual decline in 19 indicators of organ system integrity (cardiovascular, metabolic, renal, hepatic, immune, periodontal, and pulmonary) across four time points from age 26 to 45 (Belsky et al., eLife, 2022; PMID 35029144).
This longitudinal rate of decline was then distilled into a single blood-based DNA methylation score using machine learning. The result is a speedometer rather than an odometer. A DunedinPACE score of 1.0 means you are aging at the average pace (one biological year per calendar year). A score of 1.2 means you are aging 20% faster than average. A score of 0.8 means 20% slower. In validation datasets, DunedinPACE showed high test-retest reliability and predicted morbidity, disability, and mortality with effect sizes similar to GrimAge, while adding incremental prediction beyond GrimAge alone. It also detected faster aging in young adults who experienced childhood adversity.
Can You Actually Change Your Biological Age?
This is the question everyone wants answered, and the honest answer is: early evidence is encouraging but preliminary. The most cited intervention study is a 2021 pilot randomized controlled trial by Kara Fitzgerald and colleagues. They enrolled 43 healthy men aged 50-72 in an 8-week program that included a plant-centered diet rich in methylation-supporting nutrients (folate, betaine), exercise, sleep optimization, relaxation practices, and supplemental probiotics and phytonutrients. The treatment group showed a 3.23-year decrease in Horvath DNAmAge compared to controls (p = 0.018). Within the treatment group alone, DNAmAge decreased by an average of 1.96 years, with a trend toward significance (p = 0.066) (Fitzgerald et al., Aging, 2021; PMID 33844651).
These results are promising but come with important caveats. The sample was small (43 participants), all male, all healthy, and the study lasted only 8 weeks. The measurement used (first generation Horvath clock on saliva samples) has known limitations in sensitivity to short-term changes. No follow-up data showed whether the changes persisted after the intervention ended. Larger-scale, longer-duration trials using second generation clocks and DunedinPACE are underway, but the field does not yet have the kind of robust, replicated evidence that would let anyone claim with confidence that a specific protocol reliably reverses biological aging by a specific amount.
What the broader literature does support is that lifestyle factors influence epigenetic age acceleration. Smoking, obesity, poor diet, sedentary behavior, and chronic stress are consistently associated with accelerated epigenetic aging across multiple clocks and cohorts. Conversely, physical activity, healthy diet patterns, normal BMI, and not smoking are associated with slower aging. The direction of effect is consistent even if the precise magnitude of reversibility remains an open question.
What a Biological Age Test Can and Cannot Tell You
A biological age test can tell you whether your epigenetic markers suggest you are aging faster or slower than the average person your age. It can provide a baseline measurement that you can track over time to see whether lifestyle changes, interventions, or treatments are moving the needle. Second generation clocks and DunedinPACE scores correlate meaningfully with future disease risk and mortality at the population level. For someone who is serious about proactive health optimization, this information adds a dimension that standard blood work does not capture.
A biological age test cannot diagnose any specific disease. It does not tell you which organ system is aging fastest (though some newer panels attempt tissue-specific estimates). Results can vary by several years between tests taken weeks apart due to biological variability, batch effects in laboratory processing, and differences between tissue types (blood vs. saliva vs. buccal cells). The field has not yet established universally agreed-upon clinical thresholds for when biological age acceleration is "concerning" versus normal variation. And while the population-level statistics are strong, translating a group-level hazard ratio into an individual prediction is inherently imprecise.
Reproducibility remains an active area of research. Different clocks can give different biological age estimates from the same sample because they measure different aspects of the aging process. Horvath's clock measures something closer to a developmental and maintenance program. GrimAge captures mortality-related physiological decline. DunedinPACE tracks the current pace of multi-organ deterioration. These are complementary perspectives, not competing answers to the same question.
Who Should Consider Getting a Biological Age Test
Biological age testing is most valuable for people who are already investing in their health and want a deeper, molecular-level readout of whether their efforts are working. If you have made significant lifestyle changes (diet, exercise, sleep, stress management) and want to know whether those changes are reflected at the epigenetic level, a baseline test followed by a retest 6-12 months later provides data you cannot get from standard lab panels. The same applies to anyone working with a longevity-focused physician and considering interventions that target the biology of aging itself.
People with a family history of premature aging-related diseases (cardiovascular disease, cancer, neurodegenerative conditions) may also find value in understanding their baseline biological age. And for anyone simply curious about where they stand, it is a reasonable starting point that often motivates meaningful behavior change. The test is less useful if you are not prepared to act on the results or if you are looking for a definitive medical diagnosis.
How to Interpret Your Biological Age Results
If your biological age is lower than your chronological age, your epigenetic markers are aging slower than average. If it is higher, they are aging faster. A gap of 1-3 years in either direction is within the normal range of measurement error for most clocks (recall that Horvath's clock has a median error of 3.6 years in test data). Differences of 5 or more years are more likely to reflect genuine biological differences in aging rate.
For DunedinPACE, the interpretation is different because it measures pace rather than cumulative age. A score near 1.0 is average. Scores below 0.95 suggest meaningfully slower aging. Scores above 1.05-1.10 suggest meaningfully faster aging. Because DunedinPACE was designed to detect changes over time, it is particularly well suited for tracking the effect of interventions in serial measurements.
The most actionable approach is to treat your first test as a baseline, not a verdict. One measurement is a snapshot. Two or more measurements, spaced 6-12 months apart, reveal a trajectory. Combine your biological age results with standard clinical markers (metabolic panels, inflammatory markers, lipid testing) for a more complete picture. No single test captures the full complexity of aging, but together they paint a much richer picture than any one metric alone.