The research, published in the European Journal of Ageing, examined how area-level deprivation relates to multiple measures of ageing simultaneously, from DNA methylation patterns to self-reported physical functioning, and found consistent social gradients across the entire adult age span.
Roughly five months after waves 2 and 3 of the main survey, participants aged 16 and older received a Health Assessment in which trained nurses collected biomedical measures and blood samples. In total, 13,286 participants provided blood, and because of financial constraints, DNA methylation data were collected from a subsample of 3,654 people restricted to those of White/European ethnicity. Respondents from Northern Ireland were not included in the Health Assessment, so the analysis covers Great Britain only.
After excluding individuals with missing data, the analytic sample comprised 13,174 participants from the Health Assessment and 3,636 for the DNA methylation measures. The researchers measured area deprivation using three different indices: the Index of Multiple Deprivation, the official small-area measure in England that combines seven domains including income, employment, education, health, crime, housing barriers and living environment; the Townsend index, based on homeownership, unemployment, car ownership and household overcrowding; and the Carstairs index, built from four census-derived measures including low social class and unemployed males. Each index was divided into quintiles, with the fifth quintile representing the most affluent areas. The three indices correlated strongly with one another, with correlation coefficients ranging from 0.830 to 0.906, but the authors note they are not interchangeable.
To capture biological ageing, the team used the DunedinPACE DNA methylation algorithm, which estimates the pace of biological ageing in years of accelerated ageing per chronological year, and the Biological Health Score, a composite measure constructed from 16 biomarkers across six physiological systems that reflects multi-system physiological dysregulation. Cognitive function was assessed by combining episodic and working memory tests into a memory domain score, and fluid reasoning, category fluency and numeracy indicators into a non-memory domain score. Subjective functioning was measured using the physical component score of the Short Form 12-item Survey and self-rated general health on a five-point scale. Linear regression models with standard errors clustered at the level of lower-layer super output areas were used for cross-sectional analyses, and growth curve models tracked subjective health trajectories over 13 years of follow-up.
The cross-sectional results showed a clear and consistent pattern. Participants living in the least deprived areas, defined as the most affluent quintile of the Index of Multiple Deprivation, had significantly lower DunedinPACE values than those in the most deprived quintile, with a coefficient of −0.045 (95% confidence interval −0.061 to −0.030, p < 0.001). Similar associations emerged for the Townsend index (coefficient −0.047) and the Carstairs index (coefficient −0.059). The Biological Health Score showed the same graded pattern, with higher biological risk scores in more deprived areas and consistent reductions moving from deprived to affluent quintiles across all three indices. Because higher values on both measures indicate biologically older profiles, the findings suggest that area deprivation is associated with faster molecular ageing and greater multi-system physiological wear and tear.
Cognitive performance followed the same gradient. Compared with residents of the most deprived quintile, those in the most affluent areas scored better in the memory domain (coefficient 0.039, 95% CI 0.029 to 0.048) and in the non-memory domain (coefficient 0.036, 95% CI 0.029 to 0.044), with associations of similar magnitude when deprivation was measured by the Townsend and Carstairs indices. Subjective health measures also exhibited deprivation gradients: relative to the most deprived quintile, the physical component score was 3.536 points higher in the least deprived quintile of the Index of Multiple Deprivation (95% CI 2.849 to 4.222), with comparable results for the Carstairs and Townsend indices. Self-rated health showed the same pattern. The authors describe these trends as graded and monotonic, consistent with a dose–response pattern, and note that the results were robust to multiple sensitivity checks.
Longitudinal analyses using repeated measures from waves 3 to 15 of the main survey, spanning 2012 to 2024 and covering 118,862 observation points, revealed a more nuanced picture. Baseline area deprivation was associated with both lower initial levels and, in some cases, divergent trajectories of subjective functioning over time. For physical functioning, significant deprivation-by-time interactions emerged for the Townsend and Carstairs indices, indicating widening inequalities over follow-up, with participants in more affluent areas experiencing slower declines. The Index of Multiple Deprivation, by contrast, was associated primarily with persistent baseline inequalities rather than significant divergence. For self-rated general health, deprivation gradients remained relatively stable over follow-up, suggesting persistent rather than substantially widening gaps.
Subgroup analyses, which stratified the sample by sex, age group, education, occupation, income, urbanity, partnership status and ethnicity, found broadly consistent patterns across most population groups, with most interaction p values exceeding 0.05. No strong effect modification was observed for the molecular and organ-system ageing measures, although a somewhat weak association for DunedinPACE appeared in the youngest (16–39 years) and oldest (66 and over) age groups. Effect modification by education and occupation that survived Bonferroni correction was detected for the association between the Townsend and Carstairs indices and cognitive function, and some age variation and education and occupation effects emerged for subjective functioning. The authors suggest that individual-level socioeconomic assets may act as a buffer against neighbourhood adversity for cognitive and subjective outcomes, whereas molecular ageing biomarkers appear less responsive to these individual differences.
The researchers are careful to stress that their findings should not be interpreted as evidence that neighbourhood deprivation itself causes accelerated ageing. Residential environments reflect both contextual exposures and selective residential mobility, and individuals do not distribute randomly across neighbourhoods. When the cross-sectional models were additionally adjusted for baseline health behaviours and long-standing illness or disability, the area-deprivation coefficients were noticeably attenuated, though most gradients maintained statistical significance, indicating that individual composition and pre-existing health status account for a meaningful portion of the area-level variation. The study also has limitations: relatively few participants were aged 80 and over, the DNA methylation subsample was restricted to White/European participants, selective retention likely produced a healthier analytic sample, and the deprivation indices capture structural disadvantage at a single time point without reflecting lifetime exposure. The authors conclude that the findings should be read as associations rather than causal effects, while underscoring neighbourhood-level disadvantage as a structural determinant of ageing that may warrant policy attention aimed at improving housing, environmental conditions and local infrastructure.
Subject of Research: Associations between area deprivation and biological, cognitive and subjective measures of ageing in Great Britain
Article Title: Area deprivation and multi-domain measures of ageing: biological ageing, cognitive, and subjective health across the entire adult age span
Article References: Wang, W., Haworth, S., Bao, Y., & Kumari, M. (2026). Area deprivation and multi-domain measures of ageing: biological ageing, cognitive, and subjective health across the entire adult age span. European Journal of Ageing, 23(1), Article 42. https://doi.org/10.1007/s10433-026-00937-6
Image Credits: AI Generated
DOI: 10.1007/s10433-026-00937-6
Keywords: area deprivation, biological ageing, DNA methylation, DunedinPACE, cognitive function, self-rated health, UK Household Longitudinal Study, health inequalities, Index of Multiple Deprivation, neighbourhood effects, observational study, Great Britain
News Source: Beatrice Stafford. (November 5, 2026). Deprived neighbourhoods linked to faster biological ageing across adult life. Scienmag.
