The study derived a neighborhood exposome at birth using 22 census tract–level variables mapped to four domains: area deprivation, social fragmentation, population density, and environmental quality. Researchers applied self-organizing maps to group census tracts into profiles that reflect combined social and built-environment characteristics rather than single-variable exposures. The resulting neighborhood profiles served as the primary exposure in subsequent analyses of epigenetic aging trajectories.
This work used data from Project Viva, a pre-birth cohort that enrolled pregnant women in eastern Massachusetts between 1999 and 2002. The analytic sample comprised 570 children with available peripheral blood DNA methylation data, complete neighborhood characteristics at birth, and full covariate information. DNA methylation–based epigenetic age acceleration (EAA) was calculated at three time points: age 3 years (n=84), age 8 years (n=301), and age 13 years (n=434). The investigators computed EAA using three established epigenetic clocks: the Horvath clock, the Skin & Blood clock, and the Wu clock.
To describe longitudinal patterns of EAA across childhood and early adolescence, latent class linear mixed models were used to identify distinct trajectory groups for each epigenetic clock. These models accounted for repeated measures of EAA within individuals and allowed classification into trajectory classes such as stable, increasing, and decreasing (for Horvath and Skin & Blood clocks) or a low-baseline pattern (for the Wu clock). Associations between neighborhood profiles at birth and membership in EAA trajectory groups were then estimated using multinomial mixed models, producing adjusted odds ratios and 95% confidence intervals for the comparisons.
Three neighborhood exposome profiles were identified from the self-organizing map analysis:
These profiles reflect combined social and physical environmental patterns rather than isolated census measures, enabling assessment of complex neighborhood contexts present at birth.
Across the three clocks, three distinct EAA trajectory groups were observed. When comparing neighborhood profiles, children born into Profile 2 (urban, residential instability, single-person households, highway proximity) had greater odds of belonging to an increasing EAA trajectory for the Wu clock relative to Profile 1. The reported adjusted odds ratio was 1.67 with a 95% confidence interval of 1.03 to 2.69. Profile 3 (urban, high socioeconomic deprivation) did not show statistically significant associations with EAA trajectories in the reported analyses. The Horvath and Skin & Blood clocks likewise demonstrated three trajectory groups (stable, increasing, decreasing), but the primary reported neighborhood association reaching statistical significance was with the Wu clock increasing trajectory and Profile 2.
The authors interpret these findings as evidence that early-life neighborhood conditions combining social instability and adverse physical features (notably highway proximity) are associated with longitudinal increases in epigenetic age acceleration measured by the Wu clock. Because EAA has been associated with greater disease risk in adults, the study suggests that neighborhood environments at birth may set developmental trajectories that could influence later health. Consequently, neighborhoods with high residential turnover, many single-person households, and close proximity to major roadways may represent targets for public health or policy interventions aimed at mitigating long-term disease risk.
Strengths include the longitudinal design with repeated epigenetic measures spanning early childhood to early adolescence, use of multiple epigenetic clocks, and a multivariable neighborhood exposome approach capturing social and environmental dimensions. Limitations reported in the source include that Project Viva data are not publicly available due to participant privacy and governance restrictions. The source notes that statistical analysis code can be requested from the corresponding author. The source did not report further sensitivity analyses, effect modification tests, or mechanistic biomarker data beyond the methylation-based clocks in the abstract and provided sections.
The study received ethical approval from the Harvard Pilgrim Health Care Institutional Review Board. Authors reported no competing interests. Funders declared in the source included the National Institute of Mental Health, the National Institute of Environmental Health Sciences, and internal Emory University support among others. The preprint is posted on medRxiv under a CC-BY-NC-ND 4.0 license. For data access, interested investigators should contact the Project Viva study team due to governance constraints.