Dr Ellie Murray, ScD
@epiellie.bsky.social
36K followers 400 following 1K posts
Epidemiologist and science communicator | newsletter: epiellie.substack.com | cohost @casualinfer podcast | Causal inference for public health #epitwitter | Canadian in US 🇨🇦 | she/her/Dr
Posts Media Videos Starter Packs
Pinned
epiellie.bsky.social
If you’ve been following the RFK Jr autism news, then you’ve probably heard that there’s a systematic review “proving” Tylenol causes autism.

Here’s my review of that paper👇🏼

open.substack.com/pub/epiellie...
The best evidence Tylenol causes autism isn't great
On Monday, RFK Jr announced Tylenol ‘causes’ autism referencing three studies as evidence. Let's dive in.
open.substack.com
epiellie.bsky.social
It’s easy to miss the errors in papers like this because the problems are hidden in the things they don’t say. That’s how papers like this get through peer-review so often too.
epiellie.bsky.social
Basically, they used some data that had already been collected for a different reason, and then failed to do any sort of sanity checks on whether it was appropriate to use or not.
epiellie.bsky.social
That would certainly have been the first question I asked if this were my data. They don’t mention it.
epiellie.bsky.social
Yes exactly. I dont wanna sound like a hater, but I gotta point out the flaws!
epiellie.bsky.social
epiellie.bsky.social
My review of the third autism-Tylenol study RFK Jr mentioned is out now.

If you weighed in on "what proportion of kids have neurodevelopmental disorders" yesterday, this paper is why I asked the question. Most of you will be very surprised by their answer!

open.substack.com/pub/epiellie...
RFK Jr's third Tylenol-autism study is just run-of-the-mill bad research
I’m reading through RFK Jr’s supposed evidence that Tylenol causes autism and it’s not looking good for his conclusions.
open.substack.com
epiellie.bsky.social
I struggled to write this, because there are so many things wrong with this paper. They are all things that you see frequently in low-quality research, and it was hard to decide what to focus on.
epiellie.bsky.social
My review of the third autism-Tylenol study RFK Jr mentioned is out now.

If you weighed in on "what proportion of kids have neurodevelopmental disorders" yesterday, this paper is why I asked the question. Most of you will be very surprised by their answer!

open.substack.com/pub/epiellie...
RFK Jr's third Tylenol-autism study is just run-of-the-mill bad research
I’m reading through RFK Jr’s supposed evidence that Tylenol causes autism and it’s not looking good for his conclusions.
open.substack.com
epiellie.bsky.social
Even real trials have an ideal target trial
epiellie.bsky.social
The thing that’s really concerning about this paper is that the author team *looks* like they should know better. And yet.
epiellie.bsky.social
also, “alcohol use” on any medical record ever
epiellie.bsky.social
birth to 21 years 🤷🏼‍♀️this wont help you
epiellie.bsky.social
and yet, in this study i’m reading, it counts 🤦🏼‍♀️
epiellie.bsky.social
That’s what I’d think too
epiellie.bsky.social
Anyway, the real question is .. if I told you a sample of kids was less than one-third "neurotypical" would you think I had randomly sampled them?
epiellie.bsky.social
Formally diagnosed, but choose your preferred set of diagnostic categories
epiellie.bsky.social
Personally, I wouldn't base either definition on "alcohol use, unspecified" but maybe that's why RFK Jr isn't hyping my papers 🤷‍♀️
epiellie.bsky.social
Follow-up question:

if you were attempting to identify children and young adults with "neurotypical development" vs with "any neurodevelopmental disorder" would you count "alcohol use" as neurotypical or a disorder?
epiellie.bsky.social
Quick question: if you had to guess, what would you expect the prevalence of “any neurodevelopmental disorder” to be in a random sample of young people?
epiellie.bsky.social
Yeah, I’m def gonna continue in my paper but it’s a bit of a bummer to see someone else had a v similar idea.