A new paper accompanied by a scary-looking map claims "people who lived in cities with lead-contaminated water as children had worse baseline cognitive functioning at age 72" and may make every new parent worry their child's grades can be blamed on the water supply, but it leaves out important scientific context.

That lead level is also in 80 percent of actual people in US cities of the last 150 years. And nearly 100 percent of ancient Romans and a lot of European cities for the last 2,000. Nearly all the people who put humans on the moon? High lead levels. Because they went to college and then lived in or near cities. Invented the internet? High lead levels in all those ARPANet participants. 

Through the 1970s, nearly everyone in every city drank water with higher lead levels than those in Flint, Michigan that caused so much controversy a few years ago. Yet the paper is claiming that an alarming number of people in two generations are going to have cognitive impairment.

No one is advocating for more lead in waters, of course, but journalists hyping lead or any compound often misunderstand the language of science, and epidemiologists exploit it for attention and, let's be honest, maybe a new grant. "No safe level" does not mean any trace amount is harmful, it means there is no known level that can be harmful. Or safe. Like with salt. Any government organization that says X is the amount of salt you should have to be safe or harmed is looking at a population level average, which no doctor in the world would do in a clinical setting. Scientists know using correlation that way is wrong but some don't want to cast harm on their allies - like Katie Couric when things Ruth Bader Ginsburg said in an interview ended up on the cutting room floor because Couric wanted to 'protect RBG from herself' - while some let it go because they are in the field, like social psychologists or astrologers peer-reviewing each other. Others want it to be true for their own reasons.

Knowing all that, there are things to keep in mind. First, and most importantly, cognitive decline is now diagnosed a lot more than at any time in history. That's due to better health care and a good thing, but like using modern thermometers alongside the temperature some farmer claimed he had at his house in 1963, you can't Frankenstein together those two things and call it science. The hockey stick of cognitive decline is going to go up not because it is happening more, but because it is diagnosed more.

If you just look at two curves going the same way, the cause of cognitive decline could be organic food. Or the price of steel. Or the existence of Science 2.0. All of them can be linked to cognitive decline because they have all gone up. So of course levels of lead that were (and often are, New York City isn't replacing all its pipes) in 80% of older people can be linked to cognitive decline, but so can organic food. Or anything. In the modern world, we can literally detect anything in anything. Which means we can link anything to anything. You just have to ignore dose, and biology, chemistry, and toxicology. Basically, all of science.

That is the tactic of that third group who want papers to be true for their own reasons. They may want a spot on an IARC Working Group so they can get fat contracts with trial lawyers as an expert witness so they want to show they are True Believers. They may already be on an IARC working group and be consultants for an environmental group who happens to be targeting a chemical. 

They not only get science wrong, sometimes they get their own agenda wrong. We saw it a few years ago, when a study claimed that French men were becoming less fertile, and tried to blame a 'chemical cocktail' of endocrine disrupting chemicals. Except the activists creating the paper and promoting it did just the opposite. If French male motility was higher before the advent of the organic process and numerous bans on science, then clearly the organic substitutes advocated today are the fertility problem - because conventional pesticide use had declined since 1985. They hilariously ended up stuck with the claim that fewer pesticides cause infertility

If you are reading here, you are already literate enough to know that correlation is scientifically meaningless. Every epidemiology paper needs to have EXPLORATORY watermarked on every page so activists can stop fooling journalists into "suggesting" their correlation is valid risk, or causal. The science community should leave such woo to WebMD or Cleveland Clinic.

No one is saying the authors of this paper had anything but positive intent, but if we're ever going to get back to a world where the public trusts science without political filters, scientists will need to do less of this stuff. Oddly, the last time both political parties had trust in science was 60 years ago, when nearly everyone was drinking lead in water.