Dr Ellie Murray, ScD
@epiellie.bsky.social
Epidemiologist and science communicator | newsletter: epiellie.substack.com | cohost @casualinfer podcast | Causal inference for public health #epitwitter | Canadian in US 🇨🇦 | she/her/Dr
You’re late to the party. @causalkathy.bsky.social has been doing it for years.
October 31, 2025 at 12:54 PM
You’re late to the party. @causalkathy.bsky.social has been doing it for years.
Yes, we show the DAG required for diff-in-diff here: pubmed.ncbi.nlm.nih.gov/33791189/
Difference-in-Difference in the Time of Cholera: a Gentle Introduction for Epidemiologists - PubMed
We use the famous example of John Snow's investigation of the cause of cholera mortality in London to illustrate the difference-in-difference approach and corresponding assumptions. We conclude by arguing that this method deserves a second-look from epidemiologists interested in asking causal questi …
pubmed.ncbi.nlm.nih.gov
October 28, 2025 at 12:53 PM
Yes, we show the DAG required for diff-in-diff here: pubmed.ncbi.nlm.nih.gov/33791189/
The equation should be:
1 > P(T=t | Z=z) > 0
Because it’s not sufficient for the prob of treatment to simply be greater than 0. The probability of *every* treatment value has to be greater than zero, which means no one can have prob treatment equal to 1 either.
1 > P(T=t | Z=z) > 0
Because it’s not sufficient for the prob of treatment to simply be greater than 0. The probability of *every* treatment value has to be greater than zero, which means no one can have prob treatment equal to 1 either.
October 27, 2025 at 2:34 PM
The equation should be:
1 > P(T=t | Z=z) > 0
Because it’s not sufficient for the prob of treatment to simply be greater than 0. The probability of *every* treatment value has to be greater than zero, which means no one can have prob treatment equal to 1 either.
1 > P(T=t | Z=z) > 0
Because it’s not sufficient for the prob of treatment to simply be greater than 0. The probability of *every* treatment value has to be greater than zero, which means no one can have prob treatment equal to 1 either.
So sorry for your loss
October 26, 2025 at 10:19 PM
So sorry for your loss
This is a 100% true story, by the way.
It's from case notes taken during George's stay at the Great Ormond Street Hospital, which the wonderful people at the Historic Hospital Admission Records Project have digitized.
It's from case notes taken during George's stay at the Great Ormond Street Hospital, which the wonderful people at the Historic Hospital Admission Records Project have digitized.
October 26, 2025 at 2:49 PM
This is a 100% true story, by the way.
It's from case notes taken during George's stay at the Great Ormond Street Hospital, which the wonderful people at the Historic Hospital Admission Records Project have digitized.
It's from case notes taken during George's stay at the Great Ormond Street Hospital, which the wonderful people at the Historic Hospital Admission Records Project have digitized.
You'll appreciate today's post I think. It is the story of 8-year old George Williams who got sick in the summer of 1852. It perfectly encapsulates what life was like before vaccines for so many children.
bsky.app/profile/epie...
bsky.app/profile/epie...
What was life really like before vaccines? The story of what happened to 8-year old George Williams in August 1852 gives us a glimpse.
open.substack.com/pub/epiellie...
open.substack.com/pub/epiellie...
Here's what life before vaccines was really like
The morning of August 2nd 1852 promised a lovely summer’s day in London. But not for 8 year old George Williams.
open.substack.com
October 26, 2025 at 2:45 PM
You'll appreciate today's post I think. It is the story of 8-year old George Williams who got sick in the summer of 1852. It perfectly encapsulates what life was like before vaccines for so many children.
bsky.app/profile/epie...
bsky.app/profile/epie...
Diff-in-diff explained for health researchers: pubmed.ncbi.nlm.nih.gov/33791189/
Difference-in-Difference in the Time of Cholera: a Gentle Introduction for Epidemiologists - PubMed
We use the famous example of John Snow's investigation of the cause of cholera mortality in London to illustrate the difference-in-difference approach and corresponding assumptions. We conclude by arguing that this method deserves a second-look from epidemiologists interested in asking causal questi …
pubmed.ncbi.nlm.nih.gov
October 26, 2025 at 1:09 PM
Diff-in-diff explained for health researchers: pubmed.ncbi.nlm.nih.gov/33791189/
Exactly! Like, I’m happy for Rob that he didn’t need to know childrens hospitals existed when he was a child. Not every child is so lucky.
October 25, 2025 at 6:15 PM
Exactly! Like, I’m happy for Rob that he didn’t need to know childrens hospitals existed when he was a child. Not every child is so lucky.
today is like if my neck were this tree 😭
October 22, 2025 at 3:43 PM
today is like if my neck were this tree 😭