Helping Others May Slow Biological Aging, but Too Much Informal Help Could Backfire

The research team, led by Seoyoun Kim of the University of Michigan and including collaborators from Queen’s University Belfast, the University of Southern California, and Boston College, drew on two of the most rigorous aging studies in the world: the Health and Retirement Study (HRS), a nationally representative survey of roughly 20,000 Americans aged 51 and older, and the Northern Ireland Cohort for the Longitudinal Study of Ageing (NICOLA), which follows more than 8,000 adults using a parallel design. Because both studies collected DNA methylation data in their 2016 waves, the investigators could align the timing of biological measurements across the two countries, reducing the risk that differences in survey timing would distort the comparison.

At the heart of the study are epigenetic clocks, statistical tools that read chemical tags called methyl groups attached to DNA and convert those patterns into estimates of biological age. The researchers used five clocks spanning three generations of development. First-generation clocks, Horvath and Hannum, were trained simply to predict chronological age, using 353 and 71 methylation sites respectively. Second-generation clocks, PhenoAge and GrimAge, were trained on phenotypic markers of immune functioning and on mortality predictors, making them more sensitive to health-relevant decline. The third-generation measure, DunedinPACE, estimates the pace of aging itself, derived from changes in 18 biomarkers of organ system integrity tracked over 12 years in a New Zealand birth cohort. The team used principal component versions of these clocks, a computational refinement shown to improve reliability across populations, and adjusted for blood cell composition, since methylation patterns differ across leukocyte subtypes and could otherwise confound the results.

The analytical design also attempted to address a stubborn problem in volunteering research: healthy people are more likely to volunteer in the first place. To mitigate this self-selection, the researchers applied inverse probability of treatment weighting using data from the 2014 HRS wave, balancing volunteers and non-volunteers on age, gender, race, education, health, depressive symptoms, marital status, wealth, income, work status, and informal helping. This method, which mimics some features of a randomized controlled trial, was combined with HRS sample weights to produce a weighted analysis of 3,919 Americans with valid DNA methylation data. The NICOLA analysis, constrained to cross-sectional data, included 1,830 participants aged 51 and older.

The descriptive picture differed markedly between the two countries. In the American sample, about 35 percent of participants reported some volunteering in the past year, with 6.3 percent contributing 200 or more hours annually. In Northern Ireland, where helping hours were reported per week, roughly a quarter volunteered at least occasionally, but only 0.7 percent reached 20 or more hours weekly. Informal helping, defined as unpaid assistance to friends, neighbors, or relatives outside the household, was common in both countries: nearly half of American participants reported some informal help, while about 31 percent of Northern Irish participants did so.

The results from the United States were striking. Low-level engagement, between 1 and 49 hours per year, was consistently associated with slower epigenetic aging across all five clocks, for both volunteering and informal helping. The associations were strongest and most consistent for the second- and third-generation clocks, which are more closely tied to morbidity, mortality, and physiological functioning. On GrimAge, the mortality-linked measure, volunteering at 1 to 49 hours was associated with decelerated aging, and remarkably, so was volunteering at 200 or more hours per year. Informal helping told a different story at the high end: 200 or more hours of helping per year was associated with accelerated epigenetic age on the first-generation clocks and on GrimAge, with the GrimAge coefficient of 0.49 indicating meaningful acceleration even after extensive covariate adjustment.

The Northern Irish findings were directionally similar but statistically weaker. Lower levels of informal helping, 1 to 9 hours per week, were associated with less epigenetic age acceleration across all clocks, reaching significance only for GrimAge. Higher levels, 10 or more hours per week, trended toward accelerated aging on most clocks, though the estimates did not reach statistical significance. Formal volunteering in NICOLA showed only one significant association, between 1 to 4 hours of weekly volunteering and decelerated GrimAge. The authors attribute the muted results partly to the smaller sample size, which was less than half that of the HRS, and partly to differences in how the questions were worded.

That measurement difference matters. The HRS asks broadly about unpaid help to friends, neighbors, or relatives who do not live with the respondent, capturing everything from errands to home repairs. The NICOLA question asks whether respondents look after others because of long-term physical or mental ill-health, disability, or problems related to old age, a formulation that leans toward caregiving-like support. The authors also note a striking scale mismatch: 200 hours per year in the American data, the level at which informal helping turned harmful, is roughly equivalent to the upper end of the lowest weekly category in the Northern Irish data. The two countries’ highest-intensity helpers are therefore not directly comparable, a limitation the researchers flag for future work as additional NICOLA waves become available.

The divergent pattern at high intensity fits two classic theoretical frameworks. Self-determination theory holds that helping is most health-supportive when it is autonomous and freely chosen, conditions that structured volunteering, where people select causes aligned with their values, tend to satisfy. Role strain theory, by contrast, predicts that helping becomes burdensome when it is frequent, obligatory, or emotionally demanding, as intensive informal helping of family and close friends often is. At 200 or more hours per year, informal helping may begin to resemble caregiving, with its documented physical strain, time pressure, and emotional burden, offsetting the social and psychological benefits that lighter helping provides. Consistent with this, the authors point to prior evidence that high-intensity caregiving is associated with worse mental and physical health outcomes.

The study’s broader lesson may be about how we measure aging itself. The associations between helping behaviors and epigenetic aging were most evident for GrimAge, the mortality-linked clock, and weaker or absent for the chronological-age clocks, underscoring that DNA methylation measures are not interchangeable indicators of a single biological aging construct but distinct windows onto different aging processes. The authors are careful to note that the study is largely associational: inverse probability weighting improves balance on observed covariates but cannot eliminate unmeasured confounding, and the NICOLA analysis lacked a longitudinal weighted design. Still, the findings, supported by funding from the US-Ireland R&D Partnership Programme and the National Institute on Aging, suggest that prosocial engagement may modestly mitigate biological aging across cultures, with the strongest molecular rewards accruing to those who help in structured, chosen, and moderate doses, and a molecular warning for those whose kindness tips into obligation.

Subject of Research: Associations between formal volunteering, informal helping, and epigenetic age acceleration in older adults in the United States and Northern Ireland

Article Title: Association between Helping Behaviors and Epigenetic Aging: Findings from the United States and Northern Ireland

Article References: Kim, S., Potter, C., Halvorsen, C., Hill, C., Smyth, L., Scott, A., McKay, G., Kee, F., McGuinness, B., McKnight, A.-J., Crimmins, E., & Faul, J. (2026). Association between Helping Behaviors and Epigenetic Aging: Findings from the United States and Northern Ireland. SSM – Population Health, Article 101977. https://doi.org/10.1016/j.ssmph.2026.101977

Image Credits: AI Generated

DOI: 10.1016/j.ssmph.2026.101977

Keywords: epigenetic clocks, DNA methylation, volunteering, informal helping, biological aging, Health and Retirement Study, NICOLA, GrimAge, DunedinPACE, prosocial behavior, older adults, caregiving