Cash Payments Changed an Epigenetic Aging Measure in Children. What Does That Mean?

Can additional income during early childhood leave a measurable mark on biology?

A randomized trial involving families with low incomes suggests that it might. Four-year-old children whose mothers received a larger monthly cash payment had a more favourable result on one DNA methylation measure called DunedinPACE than children in the lower-payment group.

The study is important because the payments were randomly assigned. That gives researchers a stronger basis for causal inference than a conventional comparison between richer and poorer families.

It is still a biomarker result. The study did not show that the children will live longer, avoid disease, or experience better adult health.

How the cash-transfer trial worked

The findings come from Baby’s First Years, a long-running United States trial that enrolled 1,000 mothers shortly after childbirth. All reported household income below the federal poverty threshold when they joined.

Participants were randomly assigned to receive either $333 each month or $20 each month on a debit card. The larger payment added about $4,000 a year, an estimated 18 percent increase in family income. Payments began after the child’s birth and continued through early childhood.

Randomized social interventions of this kind are unusual. Income is normally entangled with education, neighbourhood, family health, discrimination, employment, and many other influences. Randomization helps separate the effect of additional cash from those pre-existing differences.

At age four, researchers collected saliva for DNA methylation analysis. Quality-controlled data were available from 735 children, including 312 in the higher-payment group and 423 in the lower-payment group. Samples were also available from 777 mothers.

What DNA methylation clocks measure

DNA methylation refers to chemical tags attached to DNA. These tags help regulate gene activity and can change with development, environment, illness, and age.

Researchers combine methylation measurements from many sites into statistical models. Some models estimate age, while others are designed to predict mortality risk, health, cognition, or the pace of physiological change.

DunedinPACE belongs to the last category. It was trained using repeated measurements from a New Zealand birth cohort and is intended to estimate how quickly multiple age-related physiological systems are changing.

Calling it a clock can be misleading. It does not directly measure years of life gained or lost, and its interpretation in four-year-old children is less established than its use in adults.

The result was real but narrow

The larger cash payment was associated with a 0.17 standard-deviation reduction in children’s DunedinPACE relative to the smaller payment. The 95 percent confidence interval ranged from 0.01 to 0.32 standard deviations lower, and the reported P value was 0.037.

That is a modest effect, with the confidence interval close to no difference. DunedinPACE was a preregistered secondary outcome, which increases confidence that the analysis was not selected only after researchers saw the data.

Other measures did not tell the same story.

The researchers found no evidence that the larger payment improved Epigenetic-g, a methylation measure related to cognition. Supplemental measures based on PhenoAge and GrimAge were expected to show no effect in children and did not provide broad confirmation of slower aging.

The study also found no effect of the cash payment on the mothers’ epigenetic measures.

Most importantly, children’s DunedinPACE was not associated with concurrent brain activity, executive function, receptive vocabulary, body mass index, or parent-reported overall health in this analysis.

Why disagreement among clocks matters

Different epigenetic measures are built from different populations and outcomes. They can respond differently because they capture overlapping but non-identical aspects of biology.

One positive clock alongside several null results does not automatically invalidate the finding. DunedinPACE may be more sensitive to environmental conditions during early development.

But the disagreement limits the claim. If additional income had broadly slowed a unified biological-aging process, researchers might expect several independent measures or health outcomes to move in a consistent direction.

For now, the safest interpretation is that the intervention changed one molecular pattern associated with the pace of aging in adults. Whether that pattern has the same meaning in young children remains uncertain.

How income could influence biology

Additional money could affect a child’s environment through many routes. Families may be able to buy more or better food, improve housing stability, obtain childcare, reduce exposure to hazards, or handle unexpected expenses with less disruption.

Income can also affect parental stress and the emotional climate of a household. Stress hormones, inflammation, sleep, nutrition, and environmental exposures can all influence biological development and DNA methylation.

The trial does not establish which pathway produced the DunedinPACE difference. Other Baby’s First Years analyses have found a mixed pattern of effects across family processes and developmental outcomes. The biomarker should not be treated as a simple meter translating dollars into slower aging.

What longer follow-up must show

The decisive question is whether the molecular difference persists.

Researchers will need repeated measurements as the children grow, ideally across different tissues and with several validated biomarker systems. They will also need to test whether the result predicts meaningful differences in physical health, cognition, mental health, educational development, or later disease risk.

Replication in other populations and policy settings would help show whether the effect is robust. A larger biomarker-specific study could also produce a more precise estimate.

The trial offers evidence that a social intervention can reach biology. It does not show that money literally slowed the children’s aging or changed their eventual lifespan.

That distinction does not make the finding trivial. It makes it a starting point for a much longer test of how childhood conditions become biologically embedded.

Primary study: https://doi.org/10.1038/s41562-026-02568-4