Key Takeaways:
- Biological age can be read from DNA methylation. Epigenetic clocks combine age-related methylation patterns at many sites on DNA into one estimate, and some, such as DunedinPACE, measure the pace of aging rather than an age.
- A sustained diet change slowed the pace of aging in a randomized trial. In the two-year CALERIE trial, caloric restriction slowed a blood-based pace-of-aging measure by 2–3% in healthy adults without obesity, and lifestyle pilot studies showed early signals.
- Repeated testing can track change over time. Epigenetic tests analyze DNA methylation directly, and comparing results from tests spaced months apart shows how biological age and pace of aging move as habits change.
Related Product:
Index: an at-home biological age test that analyzes DNA methylation to report biological age, cumulative rate of aging, and age scores for nine body systems, along with lifestyle recommendations.
Chronological age counts years lived, while biological age estimates how much aging-related change has built up in the body’s cells and systems. The most studied method reads DNA methylation, chemical tags on DNA that shift in predictable patterns with age. Researchers have begun testing whether those patterns respond to everyday habits. Can lifestyle changes lower biological age?
The short answer: Some lifestyle changes can shift DNA methylation measures of aging, but the evidence is still early and the effects are modest. The strongest result comes from a two-year randomized trial in which caloric restriction slowed a blood-based measure of the pace of aging by 2–3% in healthy adults. Smaller pilot studies found early signals from an eight-week diet-and-lifestyle program in men ages 50 to 72 and from six months of endurance training in adults ages 35 to 65. The caloric-restriction trial found no change on two other epigenetic clocks. Index was designed to measure biological age and pace of aging over time, not to diagnose a condition or promise a specific result from any one change.
The more useful question is which changes measurably affect these clocks, and how a person can tell whether their own changes are working. Here’s what researchers have studied so far—and where measurement may fit.
Why DNA methylation matters for biological age
Two people born the same year can differ in biological age, which describes the state of their cells and organ systems (and is unrelated to the fertility “biological clock”). Our guide to what biological age is and why it matters explains the distinction.
DNA methylation (DNAm) is the attachment of a small chemical unit, a methyl group, to DNA at CpG sites, where cytosine sits next to guanine. It helps determine which genes are active in a cell, and with age it rises at some CpG sites and falls at others in consistent patterns.
An epigenetic clock is an algorithm that combines methylation at many CpG sites into one estimate. The Horvath clock was trained to predict chronological age; PhenoAge and GrimAge were trained against physiological measures and mortality risk; and DunedinPACE estimates pace, the years of biological change per calendar year, where 1 is the typical rate.
Because the clocks were built differently, they can respond differently to the same intervention. Our guide to epigenetics, DNA methylation, and biological clocks explores this biology in greater depth.
Smoking and other exposures are linked to DNA methylation differences
Smoking produces one of the clearest signals in methylation research. In a cross-sectional study of 1,793 adults ages 32 to 81 from a population-based German cohort, researchers compared the blood DNA of current, former, and never smokers. At one site in the AHRR (aryl hydrocarbon receptor repressor) gene, current smokers had methylation 24.40% lower than never smokers, the largest difference the study detected. In former smokers, most smoking-linked sites sat close to never-smoker levels, depending on time since quitting and lifetime exposure, but the AHRR site had not fully returned even among people who had quit decades earlier.
“The results of our study confirm the broad effect of tobacco smoking on the human organism, but also show that quitting tobacco smoking presumably allows regaining the DNA methylation state of never smokers.”
—Sonja Zeilinger and colleagues, PLOS ONE
Because the study was cross-sectional, it cannot establish that quitting caused methylation to recover; it compared different people at one time rather than following the same people after they quit. It does show that smoking leaves a large, consistent methylation signature, and that former smokers’ profiles sit closer to never smokers’.
Other observational studies point the same way. In Taiwanese adults, fine-particle air pollution has been associated with lower methylation at the same AHRR site.
Among middle-aged participants in a U.S. cohort, higher habitual alcohol intake has been associated with faster biological aging in a dose-dependent pattern, though not in adults ages 24 to 44. In the U.S. Health and Retirement Study, short sleep and insomnia have been associated with GrimAge acceleration, a clock estimate older than chronological age.
These studies do not mean that any single habit sets a person’s biological age, or that changing one will move a clock by a predictable amount. They do show that smoking, air quality, alcohol, and sleep are each linked to measurable methylation differences.
Biological age, genetics, and the pace of aging
Part of the variation in aging is inherited, but how much is contested: a 2026 analysis in Science argued that genetics explains more of lifespan variation than earlier twin-study estimates, and other researchers have published critiques of its modeling.
The questions here do not depend on that dispute, because clocks can be measured repeatedly and trials have begun to test whether they respond to deliberate change. Sample type matters too: the trials below used blood or saliva, and a result in one should not be assumed to match the other.
“Epigenetic clocks are emerging as promising biomarkers of biological aging, yet their sensitivity to short-term interventions remains unclear.”
—Menno Van Damme and colleagues, GeroScience
The statement describes the measurement science, not evidence that any habit works, and it is a reason to read any single result cautiously.
The rationale behind lifestyle change here is to slow aging-related change, not to drive a score ever lower. Removing exposures linked to faster methylation change and adding supportive habits aim to bring the pace of aging to, or below, the typical rate. That fits the data: the clearest trial result so far is a modest shift in a pace measure.
Can lifestyle changes lower biological age?
Caloric restriction slowed a blood-based measure of the pace of aging
The strongest evidence comes from the CALERIE trial, a randomized controlled trial at three U.S. centers in which 220 healthy adults without obesity (men ages 21 to 50 and premenopausal women ages 21 to 47) were assigned either to cut calorie intake by 25% or to continue eating as usual for two years. The restriction group achieved about 12% on average. In a post-hoc analysis of blood samples from 197 participants, caloric restriction reduced DunedinPACE by the 12-month visit and maintained that reduction at 24 months, corresponding to a 2–3% slower pace of aging. PhenoAge and GrimAge did not differ between the groups at either time point.
The analysis was post hoc, the effect sizes were small, and healthy volunteers do not represent the general population. Follow-up ended with the intervention, so longer-term effects are unknown. Still, the trial matters: a sustained diet change shifted an epigenetic measure of aging in a randomized test, and the disagreement among clocks shows that the choice of clock shapes the answer.
A diet-and-lifestyle program produced an early signal in saliva
In an eight-week pilot randomized trial, 43 healthy men ages 50 to 72 were assigned either to a combined program or to no intervention. The program combined a plant-centered diet rich in folate and betaine; at least 30 minutes of exercise on five or more days a week at 60–80% of maximum perceived exertion; at least seven hours of sleep; twice-daily breathing exercises; and two daily supplements, a fruit-and-vegetable powder and a probiotic. Among the 38 men analyzed, DNA methylation age measured in saliva with the Horvath clock was 3.23 years lower in the program group than in controls after eight weeks. The 1.96-year decline within the program group did not reach statistical significance, and most blood markers, including homocysteine, glucose, and cholesterol, did not change compared with controls.
“Larger-scale and longer duration clinical trials are needed to confirm these findings, as well as investigation in other human populations.”
—Kara N. Fitzgerald and colleagues, Aging
The study offers a promising signal, not a final answer. It was small, included only men, used a first-generation clock in saliva, and bundled every component, so none can be credited with the result. It was also industry-funded.
Endurance training produced an early signal that tracked fitness gains
In a six-month pilot study, 42 adults ages 35 to 65 enrolled in a cycling-based endurance training program; 38 completed it and 33 met the adherence threshold. VO₂max, the maximum rate at which the body can use oxygen, rose by 20%. GrimAge fell by an average of 7.44 months relative to the trajectory expected from chronological age, and the size of that change tracked improvements in VO₂max but not changes in body composition.
This is one encouraging result rather than a settled effect. The published abstract describes no control group, and GrimAge changes moved closely with shifts in white blood cell proportions, so part of the response may reflect immune-cell composition. Whether training produces lasting changes in epigenetic aging requires larger, controlled studies.
Results vary by clock, population, and eating pattern
In a 12-week randomized trial of 88 adults with obesity, with an average age of 43, participants followed an eight-hour self-selected eating window, a 15% calorie reduction, or their usual diet. Compared with usual eating, neither approach produced a statistically significant difference in weight (−1.4 kilograms with time-restricted eating and −2.5 kilograms with calorie reduction), and body composition and glycemic measures were similar across groups. A shorter eating window was associated with lower calorie intake and less visceral fat. The trial did not measure biological age, and its findings cannot be generalized to people without obesity.
Taken together, the positive and null findings show why the clock, population, and duration matter. One randomized trial has slowed a blood-based pace measure and smaller studies have produced early signals, but the same caloric restriction left two clocks unchanged, and meal timing has not been shown to add benefit beyond eating less.
Human trials at a glance
| Study | Participants | Intervention | What changed | What did not |
|---|---|---|---|---|
| CALERIE caloric restriction trial | 220 healthy adults without obesity, ages 21–50 | 25% calorie restriction prescribed (about 12% achieved) for two years | DunedinPACE slowed 2–3% | PhenoAge and GrimAge |
| Diet-and-lifestyle pilot trial | 43 healthy men ages 50–72 | Diet, exercise, sleep, breathing practice, and two supplements for eight weeks | Saliva DNA methylation age 3.23 years lower than controls | Within-group change not significant; most blood markers |
| Endurance training pilot | 42 adults ages 35–65 | Cycling-based endurance training for six months | VO₂max up 20%; GrimAge 7.44 months below expected | Body composition did not track GrimAge change |
| Time-restricted eating trial | 88 adults with obesity, average age 43 | Eight-hour eating window, 15% calorie reduction, or usual diet for 12 weeks | Shorter window associated with lower calorie intake | Weight, body composition, and glycemic measures; biological age was not tested |
What the evidence means for adults tracking biological age
Taken together, the human research supports three conclusions:
Everyday exposures are linked to measurable methylation differences. Current smokers had 24.40% lower methylation at one AHRR site than never smokers in a large German cohort.
A sustained diet change can slow a blood-based measure of the pace of aging. In the CALERIE trial, caloric restriction slowed DunedinPACE by 2–3% over two years in healthy adults without obesity.
Shorter programs have produced early signals, but not consistently across clocks. An eight-week diet-and-lifestyle program in men ages 50 to 72 and six months of endurance training in adults ages 35 to 65 were each followed by lower clock estimates, while caloric restriction did not change PhenoAge or GrimAge.
Age, baseline habits, health status, the clock, the sample type, and duration may all influence how much a measure moves. At present, the evidence does not justify assuming that someone with more room for improvement will see a larger change in a test result—or that someone with already healthy habits has nothing left to gain.
Ways to support healthy aging and track change
Smoking, alcohol, and air quality
Start with what to stop. If you smoke, quit: smoking leaves one of the largest methylation signatures studied. With alcohol, the association with faster biological aging rises with intake, so less is better. Where you can, spend less time in heavily polluted air.
Exercise
Aerobic and resistance training build fitness, strength, and muscle mass in ways no supplement can replace, and higher cardiorespiratory fitness has been associated with lower all-cause mortality in a large UK cohort.
The diet-and-lifestyle trial offers a concrete benchmark: at least 30 minutes of exercise on five or more days a week at moderate-to-vigorous effort. Regular strength work belongs alongside it. Supplements can complement this work but cannot replace it.
Diet, sleep, and recovery
Leafy greens, cruciferous vegetables, beets, and legumes supply folate and betaine, and fish and eggs supply vitamin B12; the body uses these nutrients in the pathways that add methyl groups to DNA. Caloric restriction has the strongest trial evidence but does not suit everyone, so discuss a large, sustained cut with a clinician first.
For sleep, keep consistent sleep and wake times, get morning light, and keep the bedroom cool and dark; resting heart rate and heart rate variability can show recovery trends. Time with other people matters too: social isolation and loneliness have each been associated with higher mortality across 70 prospective studies, though that research did not measure biological age.
Supplements and testing
None of the studies above tested a supplement on its own against an epigenetic clock, so the clearest roles for products are supporting specific biology and measuring change.
Epigenetic tests analyze DNA methylation directly; proxy measures such as blood panels, VO₂max, and grip strength are cheaper but less specific. Every result carries measurement noise, so judge change across tests spaced months apart.
Index is Elysium’s at-home biological age test. Developed with Yale researcher Morgan Levine, Ph.D., it analyzes DNA methylation to report a biological age, a cumulative rate of aging, age scores for nine body systems, and lifestyle recommendations. In technical replicates, Index results agreed within 0–1.5 years; our guide to the technology powering Index explains how that precision was measured.
Precision describes measurement consistency, not diagnostic power: Index is informational, not intended to screen for or diagnose any condition, and biological age should not be used to determine or alter medical treatment unless advised by a healthcare professional.
Matter, Elysium’s supplement for brain aging, combines a tri-vitamin complex of vitamin B6, folic acid, and vitamin B12, developed with the University of Oxford, with omega-3 fatty acids, and is made with ingredients backed by human clinical trials. Its B vitamins are involved in homocysteine metabolism, part of the pathway that supplies methyl groups to DNA.
The research behind Matter studied brain aging and did not measure epigenetic clocks. The two products do different jobs: Index measures biological age and pace of aging over time, while Matter supplies B vitamins involved in methylation-related metabolism. Neither replaces the changes above.
Supplements for biological age versus creatine for strength
None of the trials above isolated a supplement’s effect on biological age; in the eight-week trial, the supplements were bundled with lifestyle changes. Some supplement categories do have an established effect on a specific outcome.
Creatine, by contrast, has a long research record supporting strength, power, and adaptation to resistance training. Creatine+ is Elysium’s advanced creatine system for longevity, strength, and cognition. It supports the training itself and should not be presented as lowering biological age.
Creatine and the lifestyle changes above work at different levels and can be combined: creatine supports the training, and the habits themselves remain the main way to influence biological age.
How to evaluate a biological age test
If you want a test to track lifestyle changes, look for:
- Direct measurement. Does the test analyze DNA methylation itself, or estimate biological age from proxies such as standard bloodwork?
- Published reliability. Has the company shown how closely repeated measurements of the same sample agree? That matters when you compare results taken months apart.
- Results you can interpret. Does it report a pace or rate of aging alongside an age estimate, and explain what each measures?
- Clear limits. Does it state that results are informational and not a diagnosis or a basis for changing medical treatment?
Where lifestyle change fits in slowing biological aging
Epigenetic clocks turn age-related DNA methylation patterns into estimates of biological age and pace of aging. The clearest intervention result is modest: in the CALERIE trial, caloric restriction slowed DunedinPACE by 2–3% over two years in healthy adults without obesity.
Pilot studies of a diet-and-lifestyle program in men ages 50 to 72 and endurance training in adults ages 35 to 65 add early signals, but neither is established. Caloric restriction also left two other clocks unchanged, and the eating-window findings in adults with obesity cannot be generalized to people without obesity.
Any one change, or any one test result, is best understood as one component of a long-term approach to healthy aging. Quitting smoking, drinking less, training consistently, eating well, and protecting sleep do the work, and a test such as Index can show how those measures move over time—alongside those habits, not in place of them.
