There is renewed controversy about a pesticide called chlorpyrifos and like most science issues it has come down along political lines; Democrats, who have almost all anti-science activists but ironically also a lot of science academics under their umbrella, want to ban it, while Republicans, who have a lot of industry and private sector scientists under their umbrella, say it should continue to be regulated but not banned.
Even in the White House the stances have been predictable. The Obama administration, which relied a lot on work by Natural Resources Defense Council and Pesticide Action Network during its time in office, decided to go ahead and ban it at their request while the Trump administration's EPA reversed course and said it needed more study. Both say they had science on their side.
It depends on what constitutes science. To activists, correlation is science, and they readily accept epidemiology papers that show a statistical link. That's how we figured out smoking kills people, after all, we didn't do long-term experimental trials on kids.
But correlation is not a science result, it is instead an exploratory one, and so to bolster confidence animal models are also invoked. Animal models can be good or bad. In the hands of someone like Gilles-Eric Seralini or Tyrone Hayes, for example, animal models are given less authority because no one can trust the secret sauce that resulted in their data. Plenty of other animal models are good. But animal models are also exploratory - a mouse model can only exclude something in humans, it can never show it, which is why there are no drugs that get US FDA approval based on cell studies animal models, or predictions on a computer.
Yet together, and in ideologically neutral hands, we can make informed decisions. EPA often does that. But what about epidemiology that is beset by undeclared conflicts of interest and that used a methodology that was as simple as laying two maps over each other? There is running joke in science circles about correlation, namely that you can show organic food causes autism if that us your goal, but that is the methodology that many activists are saying is why the government should ban chlorpyrifos.
To head off the Systemic Conspiracy Theory claims of activists, I have never gotten a penny from Dupont or Dow, nor has Science 2.0 (unless maybe 10 years ago one of them bought an ad through our ad reseller). It's a nonprofit, so it will happily cash a check from anyone should they choose to send it, but this is not a defense of chlorpyrifos, it is instead a call to skepticism by the public, because you won't get informed straight answers from mainstream science media.
One key paper activists liked and one the Obama administration cited when they decided to ban it in their waning days was published in 2014 and I noted at the time it had crippling flaws. I was not concerned about chlorpyrifos being targeted, I was concerned about autism being abused to push an agenda. I have a son with autism so I am sensitive to people manipulating the emotions related to that for their own gain. Instead of that paper being for financial gain or cultural fame, it was instead a senior author epidemiologist who was trapped in 'if your only tool is a hammer, every problem is a nail' thinking. (1)
To try and show pesticides cause increased in autism, the authors were willing to forgo actually measuring anything, use a very small sample size, and accept a proxy based on questionnaires.(2) Any of those would be crippling confounders in a science paper but this was for Environmental Health Perspectives, the official journal of chemophobia, so study methodology did not matter. Nor did lying about conflicts of interest. (3)
The senior author, I discovered, was on the board of both an anti-chemical advocacy group and on the Advisory Board of Autism Speaks, a controversial organization with few actual autistic people anywhere to be found who were in the bottom 10% of nonprofits for how they used their money effectively. Neither of which were declared despite those being obvious COIs.
That they did not measure any pesticides, there were weak odds ratios and the statistical sample size was too tiny to be meaningful were ignored by the mainstream press.(4) The Obama administration EPA latched onto it and Phil Landrigan, the organic industry's go-to denier for hire, even turned the weakness of the into a positive. He bizarrely claimed that because the sample was far too small and no measurements were taken the results were probably too conservative. He thought conventional pesticides probably caused even more autism? No surprise Freedom of Information Act emails have revealed he routinely collaborates with organic industry trade groups and corporations.(5)
I asked at the time what activists would have done if a study claiming pesticides boosted IQ happened to be done by someone on the advisory board of a pro-pesticide advocacy group? They'd have been apoplectic. But when it came to an agenda that profited them, they rushed to endorse it.
More importantly, what would have happened if government accepted this slipshod methodology about other products? Poor Andrew Wakefield was born a generation too soon. In the 2010s, if he had wanted to show vaccines cause autism, he wouldn't have even needed 12 hand-picked kids, he could just have asked parents how many kids had vaccines, and then how many kids had autism, and written an epidemiology paper in Environmental Health Perspectives and the Obama FDA would have taken MMR shots off the market. At least if they used the strange belief system EPA did.
NOTES:
(1) Using agenda-driven simplistic demography, a nexus for autism far worse than living near farms is being in wealthy Los Angeles neighborhoods where there are lots of pediatricians. Kids who moved there and had no diagnosis before suddenly caught it.
(2) Like epidemiology itself, proxies are not bad unless they are being used for malevolence, as anti-science activists do. In 2014 I ridiculed other bad proxy results such as "Zebras evolved stripes due to flies", "Famous paintings are proof of climate change" and "Bicep size correlates to political conservatism."
(3) If you think living closer to pesticide-using farms means less autism - as the odds ratios for their paper showed - even though pesticides cause autism, you believe in magic, endocrine disruption, the theory of Atlantis, or all of the above.
(4) Other papers have been similarly weak, for other reasons, although interesting. A neurologist examine a drawing of a spiral and suggest a child had hand tremors and then creating a statistical link to pesticides may not be the first time interpretative art was linked to trace levels of chemicals but it's the first time I knew of it.
(5) Landrigan even wrote a paper with an economist promoting the organic process who was 100 percent organic industry funded and did not disclose it.
