Toxic Lead Levels

Posted by Valerie on February 19, 2009

Lead poisoning can cause serious harm and has hurt many people throughout history. At high levels, lead can cause disability, even death–but there’s an interesting side to the recent history of lead “toxicity.”

In 1960, the toxic lead level was 60 micrograms/deciliter (mcg/dl)
In 1970, 40 mcg/dl
In 1975, 35 mcg/dl
In 1985, 25 mcg/dl
In 1991, 10 mcg/dl

Yes, lead levels that were considered safe and healthy just 20 or 30 years ago are now considered toxic and, yes, the “toxic” lead level did drop by more than half in just the six years between 1985 and 1990.

There’s a push right now to drop it to 5 mcg/dl–or lower.

In 2000 (the last year for which I could find complete data), 0.001% of American children had a blood lead level that would have been considered toxic in 1985, and all estimates point toward a further decrease since then. You can imagine that there wouldn’t be many lead programs left to fund if we didn’t keep changing the definition of “lead poisoning.”

The effects of lead on the human body haven’t changed and lead levels under 40-50 mcg/dl still generally call for no treatment  other than discovering and avoiding lead and repeating the testing. What are we told time and time again about modern lead poisoning? “Lead poisoning is invisible. There are no symptoms, and it can only be detected by testing.” (There are now and always have been symptoms over 60-70 mcg/dl.)

The progressive decrease in the lead toxicity scale has been based almost exclusively on large studies detecting very small and weak correlations between blood lead levels and IQ. Two points in one study, four points in another, eight points in a third–and from those we, supposedly, somehow know that minute quantities of lead actually cause deficiences in mental development.

Medical researchers would very much like to be exempted from the rules of logic, but even for them, the principle applies: correlation does not imply causation. Or, post hoc ergo propter hoc, which is a fancy Latin way of suppposing that because the rooster’s crow precedes the sunrise, all that noise wakes the sun.

Maybe minimal amounts of lead cause a lower IQ, but maybe children with slightly lower IQ’s have a slightly greater tendency to chew on paint chips.

A population with “toxic” lead levels and IQ’s 2-4 points below average shares many other characteristics. A child with a blood lead level over 10 mcg/dl tends to be the child of lower income unemployed or minimally employed  immigrants who speak English as a second language, lives in an urban area in run-down rental housing, tends to be less well-nourished than the general population, tends to go to an inferior public school, and tends to be dosed with folk and herbal remedies. Besides these kinds of characteristics, which are repeatedly stated in the literature directly or indirectly, there are likely dozens if not hundreds more correlations, probably including such things as sleep habits, food preferences, and entertainment choices.

Within a group like this there may be millions or more possible relationships among combinations of lifestyle data points. This being the case, it’s impossible to assign one particular characteristic as being the cause of another particular characteristic. For all that we know, within this group, the difference in IQ has nothing whatsoever to do with (even) relative intelligence and everything to do with differences in culture, language, and life experience. (In other words, if a large number of our target group speaks English as a second language could that account for tiny differences in testing outcomes? In the real world, one would think so.)

At these very low levels, which were formerly considered entirely benign, lead may still be a harmless variant. 

Why is it that there is a tiny difference in IQ between very low and low lead levels–and an almost equally tiny difference in IQ between very low and moderate lead levels? Why does a much greater dose of lead not make a more severe impact on intelligence, if it is really lead that is making the difference at low levels?

There’s actually a fascinating set of correlations among 1) the dramatic disappearance of symptomatic lead poisoning in the United States, 2) a decrease in federal funds directed against lead poisoning, 3) progressive revisions of toxicity standards, and 4) vigorous appeals for more federal money. Not that correlation implies causation here either, but what’s good enough for the gander’s certainty is plenty good enough for this goose’s tentative hypothesis.

I tend to think that if less medical research were funded by the government there would be far fewer efforts to use weak correlations and remote, relative risks to create useless headlines and a great deal more interest in the far more naturally lucrative business of relieving people of painful and debilitating illnesses.

Without government funding we most likely wouldn’t have the enormous and tremendously expensive Nurses’ Health Study, which recently released the latest “relative risk” gem among many. They’ve now discovered that women who drink four or more cups of coffee per day have a 20% lower risk of stroke. In fact, this correlation proves nothing  about either strokes or coffee. This data notwithstanding, for all we know about causation, four or more cups of coffee per day might not be such a good choice for some stroke-prone women.

A lead level over 10 mcg/dl (or 20 mcg/dl) might decrease IQ. It’s certainly possible, but we have no evidence that it’s so.