The NHANES analysis · April 2026

What 50 years of national blood-lead data actually say.

The CDC has measured blood lead in a representative slice of the United States population every few years since 1976. Nobody had ever drawn the curves end-to-end. So we did. Six charts, three waves of preventable death, one open question.

NHANES 10 rounds · 99,961 records · 1976–2023 · CC BY 4.0

The curve nobody had drawn before.

Three NHANES eras stacked on the same age axis: 1976–80 (red, peak leaded gas), 1988–94 (orange, phasedown), 2017–18 (blue, post-ban). Same toddler peak, same adult plateau, same late-life bone-release uptick. What changed is the absolute dose, not the shape.

Read the lifetime arc →

Dementia is the new heart attack.

Two waves of lead-attributable death, plotted on the same axis. Heart attack peaked in the 1970s and has been receding for forty years. Dementia is rising and overtakes CVD around 2033, at roughly 95,000 deaths each. Same generation, two different organs, two different latencies.

Read the crossover

The other five

02 · Cardiovascular wave

~21 million Americans died early because of lead.

Lanphear 2018 hazard ratios applied to NHANES blood-lead distributions, 1960 to 2023. The first wave of preventable death from leaded gasoline. Peak was the 1970s and 1980s. It is receding. It is not over.

"Cumulative 1960 to 2023: ~21.1 million lead-attributable cardiovascular deaths."

03 · Dementia wave

The dementia wave from lead is just starting.

The 1945–1965 cohort is now entering the dementia incidence window. The same kids who breathed peak leaded exhaust are turning 75. Projected lead-attributable dementia deaths rise through 2035.

"Approximately 1.1 billion IQ points lost across United States birth cohorts since 1930."

04 · Skeletal remobilization

Bones don't forget.

Even decades after the gasoline-lead ban, the lead stored in peak-era American skeletons keeps re-entering the bloodstream as bones age and resorb. A third wave, mechanistically distinct, peaking around 2042.

"The lead from 1965 is still being released into the bloodstreams of the people who carried it as kids."

05 · The open question

Which exposure window did the damage?

Childhood, adult-life, or current bone release. Three hypotheses, three different dementia death curves. They diverge by 20 years and a factor of five. The 1945–1975 birth cohort is the natural experiment that will resolve it.

"Whichever curve matches the observed dementia rate in the next 20 years will tell us which exposure window did the damage."

06 · By birth year

Every NHANES round, on the same axis.

Ten rounds spanning 1976 to 2023, plotted by birth year, not by survey year. NHANES II towers above the rest. The continuous rounds since 1999 zoom into a fraction of the original signal.

"Older Americans don't have more lead because they're older. They have more lead because they grew up with it."

Download · Source · Reproduce

↓ The full 6-page PDF ↓ Lifetime arc PNG ✉ Request the CSV + Python

Source · detectlead.com analysis of CDC NHANES II, NHANES III, and Continuous NHANES rounds 1976–2023. Methodology · Integrated Lead Burden Model (ILBM). License · CC BY 4.0. Charts and writeup current April 2026. Contact · eric@detectlead.com