Abstract
Epigenetic clocks have been widely used to predict disease risk in multiple tissues or cells. Their success as a measure of biological aging has prompted research on the connection between epigenetic pathways of aging and the socioeconomic gradient in health and mortality. However, studies examining social correlates of epigenetic aging have yielded inconsistent results. We conducted a comprehensive, comparative analysis of associations between various dimensions of socioeconomic status (SES) (education, income, wealth, occupation, neighborhood environment, and childhood SES) and eight epigenetic clocks in two large U.S. aging studies: The Multi-Ethnic Study of Atherosclerosis (MESA) (n=1,211) and the Health and Retirement Study (HRS) (n=4,018). In both studies, we found robust associations between SES measures in adulthood and the GrimAge and DunedinPoAm clocks (Bonferroni corrected p-value < 0.01). In the HRS, significant associations with the Levine and Yang clocks are also evident. These associations are only partially mediated by smoking, alcohol consumption, and obesity, which suggests that differences in health behaviors alone cannot explain the SES gradient in epigenetic aging. Further analyses revealed concurrent associations between polygenic risk for accelerated intrinsic epigenetic aging, SES, and the Levine clock, indicating that genetic predisposition and social disadvantage may contribute independently to faster epigenetic aging.
Introduction
The persistent link between socioeconomic status (SES) and health outcomes is one of the most robust findings in public health. However, the biological mechanisms through which social disadvantage translates into physical decline—often described as social adversity "getting under the skin"—remain a subject of intense investigation. Epigenetic clocks, which utilize DNA methylation (DNAm) patterns to estimate biological age, offer a promising window into this process. This study evaluates the relationship between various dimensions of SES and eight distinct epigenetic clocks across two major U.S. longitudinal studies, seeking to clarify which measures of biological aging are most sensitive to social inequality.
Methodology and Data
The researchers utilized data from two diverse, large-scale cohorts to ensure the replicability of their findings:
- Multi-Ethnic Study of Atherosclerosis (MESA): $n=1,211$ participants, with DNA methylation profiled specifically in monocytes.
- Health and Retirement Study (HRS): $n=4,018$ participants, with DNA methylation profiled in whole blood.
The study compared eight epigenetic clocks, categorized by their underlying training methods:
- First-Generation Clocks: Horvath (1 & 2), Hannum, and Lin (trained to predict chronological age).
- Mitotic Clock: epiTOC/Yang (measures cell division errors).
- Second-Generation Clocks: Levine (PhenoAge), GrimAge, and DunedinPoAm (trained on physiological biomarkers, mortality risk, or the longitudinal "pace of aging").
To capture the complexity of social standing, the authors developed a multi-dimensional SES Index encompassing education, household income, wealth, occupation, and neighborhood environment. They also utilized Polygenic Scores (PGS) to account for genetic influences on aging and performed mediation analyses to assess the role of lifestyle factors.
Results: The Sensitivity of Second-Generation Clocks
The analysis revealed a clear distinction in how different clocks respond to social factors. While first-generation clocks showed inconsistent associations with SES, the second-generation clocks—particularly GrimAge and DunedinPoAm—demonstrated robust, statistically significant associations with socioeconomic disadvantage in both the MESA and HRS cohorts ($p < 0.01$).
In the HRS sample, a one-point increase in the SES disadvantage index was associated with a 1.4-year increase in GrimAge. Similarly, for the DunedinPoAm clock, social disadvantage was linked to an increased pace of aging of approximately 0.24 months per calendar year. Among the individual components of SES, lower educational attainment and lower household income emerged as the most consistent predictors of accelerated biological aging across the different models.
Beyond Lifestyle: Mediation and Genetics
A critical question in social epidemiology is whether the health gap is simply a byproduct of lifestyle choices. This study performed mediation analyses to determine how much of the SES-aging link could be explained by smoking, body mass index (BMI), and alcohol consumption. The results showed that these behaviors only mediated between 37% and 50% of the association. This suggests that at least half of the biological impact of social disadvantage persists independently of traditional health behaviors.
Furthermore, the study examined the interplay between social factors and genetic predisposition. The findings indicate an additive effect: genetic risk for accelerated aging and social disadvantage act as independent risk factors. There was no evidence that a favorable genetic profile "buffers" the impact of low SES; instead, individuals facing both high genetic risk and high social adversity experience a cumulative acceleration in biological decline.
Discussion and Limitations
The findings underscore the importance of using "second-generation" epigenetic clocks in social science research, as they appear more proximal to the biological pathways triggered by social stressors. However, several limitations must be considered. The cross-sectional nature of the analysis prevents definitive causal claims regarding SES and epigenetic changes. Additionally, the study may be subject to "healthy survivor" bias, as the cohorts consist of older adults who may represent a more resilient subset of the population. Finally, differences in cell-type specificity (monocytes in MESA vs. whole blood in HRS) can influence the performance of certain clocks.
Conclusion
This research provides compelling evidence that socioeconomic disadvantage is etched into the epigenome. By demonstrating that social factors influence biological aging independently of lifestyle and genetics, the study highlights the need for public health interventions that address the root social causes of health inequality. Future research should focus on training epigenetic clocks directly on social outcomes to better isolate the specific biological pathways triggered by the social environment.
Summary updated