Research shows DNA methylation as a robust predictor of later life outcomes

Reviewed by Emily Henderson, B.Sc.Feb 22 2023

A new study utilizing data from the Health and Retirement Study at the University of Michigan’s Institute for Social Research is offering a new look at the effects of aging and what they mean for population health.

Jessica Faul, associate research professor at ISR’s Survey Research Center, and Eileen Crimmins, professor of gerontology at the University of Southern California, found that examining epigenetic changes to the human genome, such as DNA methylation, can offer some ability to predict future health outcomes.

“Our main aim was to use data from a nationally representative and diverse sample to see whether biological age acceleration as measured from DNA methylation was predictive of health outcomes and mortality. And it turned out that is largely the case,” said Faul, co-director of the Biospecimen Lab at ISR and also affiliated with the Michigan Center on Demography and Aging.

“You can use certain DNA methylation markers to predict age and age-related outcomes quite well.”

Methylation is a process by which genes are regulated over time. The new research by Faul and Crimmins suggests that tracking methylation can help us measure biological age acceleration. Previous studies have looked at summary indicators for rates of aging, but many have been applied to smaller samples. This project examines a large, representative sample of the U.S. population, strengthening the conclusions drawn from the research.

For years, we’ve been using things like hypertension and cholesterol and weight as health indicators, things that are downstream than this. The idea is we’re trying to get (at) more basic processes of aging in this sample population.

I think the bottom line from this paper is yes, epigenetic aging and DNA methylation adds something to what we could do without this measure. In other words, we’re explaining a little bit more of the health outcomes than we did before, but we are not explaining much of the population difference in the rate of aging.”

Eileen Crimmins, Professor of Gerontology, University of Southern California

The researchers analyzed data from 3,581 adult participants with a median age of 68 years, evaluating their models’ predictive ability for age-related health outcomes. And while epigenetic aging processes such as DNA methylation do show some promise, both Faul and Crimmins are cautious about labeling it a cure-all for analyzing this kind of data.

“There are probably social as well as biological mechanisms we still need to measure better,” Crimmins said. “But this isn’t a be-all and end-all. It doesn’t get you a lot further.”

However, it could present a promising option for addressing certain health issues within the U.S. population. DNA methylation occurs in line with other social factors, such as a respondent’s socioeconomic status or demographic background, in addition to being affected by health behaviors, such as smoking or heavy alcohol consumption. Adding a new data point to the aging picture could help set policy that helps improve population health.

“We used to be working at the clinical level of biomarkers, and we were putting together many, many of the clinical indicators to indicate biological age,” Crimmins said. “Now we’re at a different level, down at the molecular and cellular level. If we can do a good job of actually capturing the long run outcome in some of these earlier life measures, you could potentially treat the processes that lead to diseases and other negative health outcomes.”

But regardless of where it fits, Faul and Crimmins believe they have added an important piece to the puzzle of aging.

“DNA methylation is a robust predictor of later life outcomes, but does not take away the importance of social and demographic and health behaviors in predicting those outcomes,” Faul said. “It’s a piece of the puzzle, but only a piece of it.”

Three grants from the National Institute on Aging supported this research, and an additional NIA grant supports the Health and Retirement Study.