Biological age
What actually moves a pace-of-ageing measure
Cohort data points at the same short list every time, and none of it is available in a bottle.
If measured biological age responds to anything, it is worth knowing what. Longitudinal cohort work has converged on an answer that is consistent, well-replicated, and commercially useless.
The short version
- A multi-cohort study published in eBioMedicine in December 2025 found smoking, higher BMI, elevated glucose, and poor blood pressure profiles associated with accelerated ageing on measures including DunedinPACE.
- Physical activity and better diet quality were associated with slower measured ageing.
- Complementary work in BMC Medicine linked favourable cardiovascular health to lower epigenetic age acceleration, with some sex-specific differences.
- No supplement appears on this list, because no supplement has been shown to move these measures in adequately powered trials.
The consistent findings
The December 2025 multi-cohort analysis reported that people moving into faster ageing trajectories tended to show worsening cardiovascular profiles over time. The accelerating factors were smoking, higher body mass index, elevated glucose, and poor blood pressure control. The decelerating factors were physical activity and diet quality.
Work in BMC Medicine found the same relationship from the other direction, linking favourable cardiovascular health to lower epigenetic age acceleration. It also found sex-specific patterns: nicotine avoidance and glucose regulation carried more weight in men, while physical activity, glucose regulation and healthy BMI were most influential in women.
Why this is unsatisfying and probably correct
This list is the same list a cardiologist has been reciting for forty years. That is not a weakness of the research. It is convergent evidence arriving through a new instrument, which is roughly what you want from a new instrument.
It is commercially unsatisfying because none of it is a product. Nobody can package blood pressure control. The market for biological age testing has therefore grown much faster than the market for the things that measurably change biological age.
The causation caveat
These are observational associations. They are strong, replicated across cohorts, and biologically coherent, but they are not randomised trial results, and pace-of-ageing measures have not been used as a primary endpoint in a large intervention trial.
What can be said is that the behaviours associated with slower measured ageing are also the behaviours with independent randomised evidence for hard outcomes. The two lines of evidence agree, which is more than can be said for most claims in this field.
Common questions
Can a supplement slow DunedinPACE?
Nothing has demonstrated this in an adequately powered randomised trial. Small uncontrolled studies exist and should not be treated as evidence.
How long before behaviour change shows up?
Unclear, and complicated by the reliability problem. Cohort studies observe change over years, not months.
Sources
- Epigenetic clocks: advancing biological age measures towards clinical use — https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12905613/
- Biological versus technical reliability of epigenetic clocks — https://onlinelibrary.wiley.com/doi/10.1111/acel.70635