Metabolomic age (MileAge) predicts health and life span: A comparison of multiple machine learning algorithms
2024

Metabolomic Age Predicts Health and Lifespan

Sample size: 225212 publication 10 minutes Evidence: high

Author Information

Author(s): Julian Mutz, Raquel Iniesta, Cathryn M. Lewis

Primary Institution: King’s College London

Hypothesis

Nonlinear machine learning models may better capture the complex relationships between metabolites and age.

Conclusion

The study found that metabolomic age, derived from machine learning algorithms, is strongly associated with health outcomes and mortality risk.

Supporting Evidence

  • Individuals with an older metabolomic age were frailer and had shorter telomeres.
  • Older metabolomic age was associated with a higher risk of chronic illness.
  • Those with a higher metabolomic age rated their health worse.
  • Metabolomic age was predictive of all-cause mortality.

Takeaway

Scientists used blood samples to create a new way to measure how old our bodies really are, which can help predict health problems and how long we might live.

Methodology

The study compared 17 machine learning algorithms using metabolomic data from the UK Biobank to develop aging clocks.

Potential Biases

There was a systematic overestimation of age in younger individuals and underestimation in older individuals.

Limitations

The study did not validate the findings in independent cohorts and may not generalize to other populations.

Participant Demographics

The sample included middle-aged and older adults, with a mean age of 56.97 years.

Statistical Information

P-Value

P < 0.001

Confidence Interval

95% CI, 1.43 to 1.59

Statistical Significance

p<0.001

Digital Object Identifier (DOI)

10.1126/sciadv.adp3743

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication