Matching an Epigenetic Clock to Physical Function

Physical function is itself a measurement of biological aging.


Exercise clockExercise clock

A new, three-year study published in The Journals of Gerontology: Series A examined the relationship between epigenetic clocks and physical performance in older women.

Epigenetic clocks and biological age 

Epigenetic clocks attempt to estimate age based on measurable DNA modifications. However, not every individual ages at the same rate; people may have biological ages that are higher or lower than their chronological ages. Unlike chronological age, the best method to quantify biological age is not clear.

Such a method would be an invaluable tool to longevity researchers to estimate mortality risk, study the basic biology of aging, and measure the effect of anti-aging interventions on a shorter time scale. It is hypothesized that the difference between the age predicted by epigenetic clocks and actual age could be a potential measure of biological aging.

Indeed, multiple epigenetic clocks have used this difference to estimate people’s risks of suffering from a variety of age-related diseases [1]. However, most of these studies are cross-sectional, and few longitudinal experiments exist. Further, the number of epigenetic clocks is increasing, and comparisons are needed between these clocks to determine which are best for different applications [1].


Eterna is a clothing company with a focus on longevity.

Epigenetic clocks and physical function

Physical function is a good proxy for biological aging. It incorporates multiple organ systems (nervous, cardiovascular, pulmonary, and musculoskeletal), making it closer to a whole-body assessment than disease-specific measures. Additionally, it follows a similar pattern to canonical aging, first declining slowly around age 30 and then rapidly towards the end of life.

While many studies have been done on the predictive ability of epigenetic clocks for mortality and disease risk, limited information is available for functional outcomes such as physical decline. Physical function is incredibly important for healthspan, quality of life, living independence, and health concerns such as falls and fall-related injuries.

In a longitudinal study, researchers at the University of Jyväskylä in Finland have related physical decline to multiple epigenetic clocks, providing insight into their ability to measure biological aging broadly and predict physical decline specifically [2].

Study design

413 Finnish women between the ages of 63 and 76 from The Finnish Twin Study on Aging were included in this research. The epigenetic age of blood samples was measured using four different epigenetic clocks: Hannum, Horvath, PhenoAge, and GrimAge. Physical function was measured via the Timed Up and Go test, the 10-meter walk test, the six-minute walk test, and the isometric muscle strength tests of grip strength, ankle plantar flexion strength, and knee extension strength. Epigenetic age and age acceleration (the difference between epigenetic and chronological age) as measured by each clock were related to each physical function at baseline as well as the change in physical function after three years.

GrimAgeAccel best predicts physical function and decline

At baseline, a higher GrimAgeAccel (the difference between epigenetic age and chronological age using the GrimAge clock) was significantly associated with worse performance on the Timed Up and Go and 6-minute walking tests, but it was not associated with the 10-meter walking test or any of the strength measurements. Similar comparisons with the Horvath, Hannum, and PhenoAge clocks did not yield any statistically significant associations.


An advertisement banner for PartiQular supplements.

Looking at physical function after three years, a higher GrimAgeAccel was significantly related to declined performances in participants’ Timed Up and Go tests, 10-meter walking tests, 6-minute walking tests, ankle plantar flexion strength, and knee extension strength, but not grip strength relative to baseline. Other clocks did not show significant relationships with declines in physical function, and these findings were all maintained when adjusting for smoking, alcohol usage, and chronic diseases.

The present study employed a longitudinal design to investigate associations between markers of biological aging, i.e., epigenetic clocks and validated physical functioning phenotypes. Both the “first-generation” clocks and novel “second-generation” clocks were utilized in the analysis. We found that DNAm GrimAge, which was developed to predict lifespan and healthspan, associated with a decline in physical functioning, while other clocks showed no associations with physical functioning. More specifically, GrimAgeAccel was associated with lower performance in the TUG test and six-minute walk test at baseline and with declining performance in the TUG test, six-minute and 10-meter walking tests, and ankle plantar flexion and knee extension strength test during the three-year follow-up. However, chronological age provided very similar estimates in cross-sectional and longitudinal analyses. In sum, our results suggest that DNAm GrimAge outperformed other epigenetic clocks in predicting age-related decline in physical functioning. Current epigenetics clocks, however, do not provide special benefits in in predicting later decline in physical functioning, at least during a rather short follow-up period and narrow age range.


The most striking finding from this study was the performance of GrimAge compared to other clocks. These results provide strong, although not fully conclusive, evidence that GrimAge is a better predictor of both physical function and physical decline than the Horvath, Hannum, and PhenoAge clocks. This is valuable information for those researching mechanisms and developing treatments for age-related physical decline. However, it should be noted that the other clocks may outperform GrimAge for other applications beyond physical function.

These results also do not preclude future studies from discovering contradicting findings. In particular, the specific details of this experiment may limit its generalization to broader contexts. For example, these results were with Finnish, female participants in a very narrow age range and over a somewhat short (three-year) follow-up. The participants were also relatively healthy for their age. It included few smokers, a healthy average BMI, and only participants who were mobile and independent enough to participate in the study. It is possible that studies done in other populations may make different conclusions.

Perhaps most importantly, it is not clear from this study whether GrimAgeAccel outperformed chronological age. Ideally, measures of biological age should better predict mortality, disease onset, and functional decline than chronological age. However, this was not addressed in detail, other than to establish that GrimAgeAccel and chronological age show similar trends in this study. Ultimately, future research will be needed to answer these questions and continue to push forward the field of epigenetic clocks.

To do this, we need your support. Your charitable contribution tranforms into rejuvenation research, news, shows, and more. Will you help?


[1] Jylhävä, J., et al. Biological age predictors. EBioMedicine (2017). https://doi.org/10.1016/j.ebiom.2017.03.046


[2] Föhr, T., et al. The association between epigenetic clocks and physical functioning in older women: a three-year follow-up. The Journals of Gerontology: Series A (2021). https://doi.org/10.1093/gerona/glab270

About the author

Greg Gillispie

Greg is a recent graduate from the Wake Forest Institute for Regenerative Medicine. He strongly believes that age-related diseases have common underlying mechanisms at play and that an ounce of prevention is worth a pound of cure. In addition to writing for LEAF, Greg continues to conduct laboratory research in stem cell regeneration and cellular senescence. He is also an avid runner, curious reader, proud dog owner, and a board game enthusiast.