Analysis of three waves of the English Longitudinal Study of Ageing—covering 8,389 (2004/05), 8,549 (2012/13), and 6,090 (2023/24) community-dwelling adults aged 50+—reveals sharply divergent health trajectories over two decades. Fair or poor self-rated health climbed from 27% to 34%, any pain from 37% to 47%, psychiatric problems from 6% to 10%, and diabetes rose on both self-report and biomarker measures. Meanwhile, mobility impairments fell from 58% to 52%, biomarker-defined high cholesterol dropped from 78% to 54%, and memory scores improved. Critically, trends varied substantially by age group and education level, meaning population averages concealed deepening inequality.
This preprint, not yet peer-reviewed, challenges the comfortable assumption that population ageing automatically tracks toward better health. The cholesterol paradox here—self-reported prevalence doubling while biomarker levels fell—almost certainly reflects expanded statin prescribing and diagnostic awareness rather than worsening biology, a measurement artifact worth flagging in any policy discussion. The mental health deterioration aligns with post-pandemic wellbeing data across Western cohorts, but this study's longitudinal design strengthens the signal. The education-patterned divergence is arguably the most consequential finding: gains in functioning and cognition appear concentrated among more educated, wealthier adults, while disadvantaged groups accumulate chronic conditions and pain. For longevity researchers and policymakers alike, this demands multidomain surveillance frameworks rather than single headline metrics. The findings are observational and England-specific, limiting causal inference and global generalizability.