Abstract
Abstract
Purpose.
In small, selected samples, biological-age clocks have been associated with low back pain and disability, raising the prospect of an aging-based risk marker. It remains untested whether a clinical clock measurable at the bedside carries the same signal as the epigenetic clocks that cannot reach it, and whether either tracks low back pain in an unselected national population. We compared routinely measurable clinical aging clocks with epigenetic clocks for their association with low back pain and disability in a nationally representative sample.
Methods.
Cross-sectional analysis of NHANES 1999–2002 adults, reported following STROBE. Clinical clocks (phenotypic age [PhenoAge], Klemera–Doubal biological age) were derived from routine blood chemistry; epigenetic clocks (first-generation, second-generation, and pace-of-aging) from the DNA-methylation subsample. Outcomes were self-reported low back pain in the past three months and physical-function disability. We used survey-weighted regression with hierarchical false discovery rate control; the same-name pair, clinical PhenoAge versus its methylation counterpart (DNAm-PhenoAge), was the primary contrast.
Results.
In the DNA-methylation subsample (n = 2,532 with complete clock and covariate data; design-effective n ≈ 1,268), no aging clock was associated with low back pain in the past three months. Adjusted odds ratios per SD ranged from 0.64 to 0.99, and every 95% confidence interval crossed 1 (clinical PhenoAge 0.64, 0.25–1.65; DNAm-PhenoAge 0.84, 0.49–1.42). The same-name pair showed no difference between clinical PhenoAge and DNAm-PhenoAge (ratio of odds ratios 0.80, 0.15–4.25; p = 0.79). No clock was associated with physical-function disability after false discovery rate control. The only association reaching nominal significance was between clinical PhenoAge and the deficit-accumulation frailty index (p = 0.012), an expected correlation given their shared physiological inputs.
Conclusion.
Neither routinely measurable clinical aging clocks nor epigenetic clocks were associated with low back pain or disability, and the two phenotypic-age measures showed no significant difference. The aging-clock signals reported in small, selected cohorts did not generalise to the population. Biological-age acceleration is not a bedside risk marker for low back pain, and epigenetic clocks add no value over clinical clocks for this outcome.