Bill Hanage @BillHanage.bsky.social

Don't forget - vaccines don't help on their own. We need *vaccination*.

Recently my daughter told me that she used to think that, at the climactic moment in ‘Let it go!’ instead of “I’m free!” Elsa was actually singing “Highbury!” Proud #arsenal dad :)

There is a lot of talk about claims about vaccinations increasing autism in the news lately, with a focus on MMR vaccines. I hope anyone discussing these issues is aware of the existing scientific literature investigating such potential links that have been done since the (fraudulent, retracted) work of Wakefield raised concerns about this question many years ago. For example, this excellent Danish cohort study by Anders Hviid @anders_hviid and collaborators (he does great studies, BTW) published in Annals of Internal Medicine in 2019, investigating >650k children born in Denmark between 1999 and 2010, with follow up through August 2013. They compared MMR vaccinated children with MMR unvaccinated children, using time to autism as the response and adjusting for potential confounders age, sex, birth year, other childhood vaccines, sibling history of autism, and autism risk factors. They found absolutely no evidence for increased risk of autism after MMR vaccine in the Danish children, with hazard ratio of 0.93 (95% confidence interval 0.85 to 1.02) [note, HR=1 means no difference, HR<1 means vaccinated lower risk of autism, HR>1 means vaccinated higher risk of autism]. The fact that the confidence interval contained 1 means effect was not statistically significantly less than zero, so the study cannot conclude that MMR vaccines reduce the risk of autism, but there is certainly no evidence at all or even of a suggestion they increase risk. Personally, I think it is great if scientific researchers want to do additional studies looking claims like this with rigorous, well designed studies, but they should probably look at this paper and the others in the existing literature that have been done in the past 20 years to investigate this question to identify gaps in the literature that their studies can address to best use research and financial resources to address unanswered questions. I hope people don't get the impression from any rhetoric online or in the media that this question has not been extensively studied already, because it clearly has, even if the studies did not give the results some were hoping to see. proceedings.mlr.press/v238/katta24a/…

@baym I can think of two easily…

This is subtweeting at least three different recent things

You know an industry is in a bubble ready to burst when real technical critique is derided as being too pessimistic and not having vision. People who are more interested in building something real than hype welcome and learn from criticism

I wish that American schools had reopened in the fall of 2020, and been smarter about doing so (making more use of pooled or rapid testing, investing much more in indoor air quality and outdoor classrooms to limit spread). (1/x)

Making pumpkin pie from scratch. Happy thanksgiving to all celebrating it today!

Big if true sciencedirect.com/science/articl…

Just a thought. academic.oup.com/aje/article/19…

Finally a modest suggestion for how we might investigate transmission and get more information on the range of severity academic.oup.com/aje/article/19…

And can’t really say anything about possible transmission. Would be interesting to know about symptoms among household contacts. Although these could easily have been caused by any of the many other viruses out there…

Important study in @CDCMMWR on evidence of prior infection with H5N1 among dairy workers in MI and CO. Results suggest many more such infections have been happening, but that they’ve been mild cdc.gov/mmwr/volumes/7…

For all that I’m a big fan of Albert Camus, I think we should all be able to agree that Sisyphus is entitled to ask “what is it with rolling this bloody rock AGAIN?”

This November, we're spotlighting @CCDD_HSPH! Led by @mlipsitch & @BillHanage, CCDD works to improve mathematical modeling & statistical methods for infectious disease data, while engaging policymakers & educating the public. Follow us to learn about their work & initiatives!

Looks like @RoryStewartUK is using a Lenovo thinkpad while @campbellclaret and everyone else on the coverage are apple users… The divide is real youtube.com/watch?v=sAmC9F…

How to show you don’t understand casual inference without using the words ‘causal’ or ‘inference’

Which was, for a few months in 1990, my favorite REM song…

@dfallik As in before your reply :)

@dfallik I actually just watched it!

Just thinking, not for the first time, that “I’ll be your mirror” by the Velvets is the most moving love song I’ve ever heard. youtu.be/KGZWb1SIiR4?fe…