Heavy metal exposure and conditional survival time in U.S. adults: a censored quantile regression cohort study
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Chronic low-level exposure to lead, cadmium, mercury, and arsenic remains a determinant of premature mortality in the U.S. general population, but previous hazard-ratio analyses do not characterize how exposure shifts the lower tail of the survival distribution, where premature mortality is concentrated.
Objectives
We estimated the association of whole-blood lead, whole-blood total mercury, urinary cadmium, and the sum of urinary inorganic and methylated arsenic species with the 10th, 25th, and 50th conditional quantiles of follow-up time to all-cause mortality among U.S. adults aged 40 years and older.
Methods
NHANES Continuous 1999–2018 was linked to the National Death Index through December 31, 2019 (n = 29,652). Censored quantile regression was fit per metal on the log 2 scale at quantiles ∈ {0.10, 0.25, 0.50}. A restricted-cubic-spline (RCS) censored-quantile-regression was fit for blood lead and urinary cadmium to investigate threshold effect.
Results
Over a median follow-up of 9.1 years, 7,215 deaths were ascertained. A doubling of urinary cadmium was associated with -1.57 years of follow-up (95% CI: -2.08, -1.07) at the 10th conditional quantile, -1.50 (−2.04, −0.96) at the 25th, and -1.49 (-1.93, -1.04) at the median (Benjamini–Hochberg q < 0.001 throughout). A doubling of whole-blood lead was associated with -0.70 years (95% CI: -0.99, -0.40) at the 10th conditional quantile, -0.62 (-0.92,-0.31) at the 25th, and -0.61 years (-0.89, -0.34) at the median; the absolute loss was largest at τ = 0.10 for both metals. Urinary arsenic-metabolite sum was not associated with conditional follow-up at the estimable quantiles. Despite adjustment for dark and fatty-fish intake or DHA/EPA, whole-blood total mercury was associated with longer follow-up (i.e., negatively associated with mortality risk) — possibly due to residual confounding by broader dietary or socioeconomic factors, rather than a true protective effect. The cadmium association was additionally robust to mutual adjustment of lead.
Discussion
Low-to-moderate urinary cadmium and whole-blood lead were associated with fewer years of follow-up survival at the lower-tail and median conditional quantiles of survival, with the largest absolute losses at the lower tail of the conditional survival distribution, where premature mortality is concentrated. These findings support continued reductions in U.S. cadmium exposure and lead with particular benefit for adults most vulnerable to premature death.
