People are making Rapture jokes like there's no tomorrow

Demystifying Clone‐Censor‐Weighting to Studying Treatment Initiation Windows: An Example Using Publicly Available Synthetic Medicare Claims Data - Webster‐Clark - 2025 - Pharmacoepidemiology and Drug Safety - Wiley Online Library

A sentence I just read in a published study: "By identifying and adjusting for biases such as confounding, selection bias, and information bias, epidemiologists accurately estimate the genuine relationship between exposures and outcomes" ...the genuine relationship...

Currently attending a conference & our field is quite a bit into fancy modeling, so it’s time to repost this blog post. Don’t try to squeeze your research question into whatever model is fashionable right now; try to build the right model for your research question. www.the100.ci/2024/08/27/l...

Oligarchs are working overtime: To destabilize any ability you and I have to govern ourselves. To foster unrest. To attack democracy. To destroy checks and balances. To remove privacy and private communications. To poison the information ecosystem. Nothing to do with "right" vs "left".

Question to academics and PhD students: Does your department have a protocol defence or proposal evaluation for early year PhD students? And if so, what do they look like?

Question to academics and PhD students: Does your department have a protocol defence or proposal evaluation for early year PhD students? And if so, what do they look like?

It feels like a lot of people in the world these days are using Pascal’s wager but to get to hell

Brilliant piece on how to engage in conversation with strangers about vaccines and public health @statnews.com @unbiasedscipod.bsky.social www.statnews.com/2025/09/06/c...

Science is messy. Data are messy. Results are messy. Conclusions are messy. The one thing that you have complete control over and therefore should not be messy is your research question.

There's something deeply intellectually satisfying about taking something that is highly technical and breaking it down to the point that even non-technical people can follow along and derive some own intellectual satisfaction from it.

When you're not an expert on something, it's hard to know which experts to trust. But these days there are many "experts" saying stuff that is so easily disproved with a simple search, that there's no excuse for anyone to believe anything they say.

Universal basic income is UNIVERSAL. That's a key detail. You can do a pilot that provides unconditional cash to a thousand people, but unless they all live in the same small town of 1,000 people, it's not truly testing UBI. That's where we see the big impacts, like boosted local GDP and less crime.

Applications are now open! We are recruiting 20 Assistant Professors in a wide range of subject areas. We're looking for early-career researchers with strong scientific merits and future potential. 🔗 All positions: ki.se/en/about-ki/...

I think about this quote a lot...

This was the infant mortality rate in the Netherlands 50 YEARS AGO

“The Mississippi health department declared a public health emergency Thursday over rising infant mortality rates in the state. There were 9.7 deaths for every 1,000 births in Mississippi in 2024, the highest rate in more than a decade, according to a news release from the state health department.”

What the heck is 'the longitudinal association'? I see the term used in many lifecourse epidemiology studies. I have no idea what it means. Cc @jeremylabrecque.bsky.social

The biggest trick the devil ever pulled is to call LLMs “artificial intelligence.”

Lots of different ways to do observational causal inference—IV, proximal inference, etc. What if you could compare those strategies more directly? New preprint w/ @melodyyhuang.bsky.social tries to do just that. Here's one cool figure—we're able to visualize bias of 3 estimators on the same plot

I'm astonished by the similarities between Lysenko (en.wikipedia.org/wiki/Trofim_...) and the charlatans we have working in the federal US health institutions now. Overconfidence from minor successes, an astonishing, dangerous attachment to stupid viewpoints, and chance support from politicians

I increasingly believe that having that much money distorts one's human interactions such that billionaires are in a constant state of progressive moral decay. The option to avoid this fate by philanthropically disposing of your gross excess assets is permanently available yet rarely exercised.

This is outrageous. Cubans and Venezuelans who came to the U.S. *legally*—fleeing socialism & communism—are being fast-tracked for deportation by the Trump administration without so much as a hearing. I wrote about some of them. Anyone who claims to hate those oppressive regimes can't defend this:

The CEO of a a multi-trillion dollar company went to the White House, openly handed the US president a gift made of gold, and his company got a special exemption to tariffs the president has illegally imposed. Also, the Founders took bribery so seriously the Constitution bans it 3 separate times.

The CEO-to-worker pay ratio at S&P 500 companies last year was 285-to-1. The median employee would have had to start working in the year 1740 to earn what the average CEO received in 2024. Is it any wonder why so many think the system is rigged?

I wonder how people would react if they were told something like: "You know that drug you're taking? Its final FDA approval was given by a team who just recently learned to spell strawberry and are working really hard on spelling blueberry."

Oh so it does have phd level intelligence

It really does feel like a lot of formerly reasonable people have been infected by some sort of terrifying brain parasite when it comes to unthinking and total acceptance of using AI tools

I don’t know why this incenses me as much as it does, but it does: we very likely have a thousand plus Mozarts *right now*. But they make non-Western art in economically destitute conditions so they have to do other work to pay bills; if you cared, UBI’d be more important to you than going to space

This take is so damn HOT

"this product is good because it's like having a pocket full of PhDs" just suggests to me you've never actually been in a room with a bunch of experts on a topic. You think you're going to come away with a quick, simple, easy-to-digest summary of their expertise? Think again.

If you've never considered reading Wikipedia to relax/unplug before bed, may I recommend my beloved: the list of lists of lists. I pick a topic that sounds cool/ informative, but not too close to my research or current events to stress me out, and away I go... en.wikipedia.org/wiki/List_of...

this is probably in the "being an academic 101" syllabus but I've found that even reviewers who are completely wrong or missed the point can provide an opportunity to improve wording so that other readers are less likely to make the same error, even if you don't do what they ask

"Reducing our dependence on whales as an energy source will send us sliding down the slippery, spermaceti-lubricated slope to socialism. If we give up whaling, what will those lackwits come for next? Will I have to surrender my peg leg for a titanium prosthetic?"

"[T]he scientific enterprise is now witness to widespread, organized defection from the scientific public goods game. Large swaths of players, among them many scientists, reviewers, editors and publishers, are choosing to no longer make genuine contributions to the pot."

I always wonder if people are aware that people in the future can check their predictions

Frankly, releasing LLMs which can give what seems like expert advice (on any subject) without having actual subject matter experts rigorously test it should be considered gross negligence.

The main problem here is the asymmetry. Nobody ever talks about drivers as entitled, for example, but if you step back and think about it, it is pretty wild that we built the single biggest structure in human history and that only people who spend thousands of dollars on private property can use it.

Should statistical software that estimates causal effects also tell you the causal assumptions under which that estimate can be interpreted as causal? I don't know but my PhD student Maurice Korf has some thoughts (and software) to get the conversation going: academic.oup.com/ije/article/...

This is funny. But doesn’t the dad know that ‘not significant’ doesn’t imply the null is true?

100% this, null results are valuable and I don’t trust the people upthread saying that they aren’t!

SAVE THE DATE: The 2026 IEA European Congress of Epidemiology and 70th @socsocmed.bsky.social Annual Conference will take place in London, UK on 8th-11th September 2026! #EpiSky #EuroEpi2026

I do this with papers I really love. Recently it led to writing a paper together with someone I contacted in this way!

It’s not like there aren’t enough important and genuinely complex issues we face, but it feels like the current trend is to take non-issues with obvious solutions and do the opposite.

A good peer reviewer - who engages thoughtfully and constructively, teaches, notes what's good about a paper and what can be improved - is such a positive force. As a mentor (and as an editor and an author), I send out deep thanks to everyone who works hard to write kind and useful reviews.

What is the "true propensity score"? To me, the true PS (in a non-experimental setting) can only take the values 0 and 1 and it's equal to the observed treated status (which is kind of useless). But it's often used to mean some kind of true probability underlying the P(treatment)?

I wish the same skepticism about evidence for alien intelligence or consciousness was applied to the evidence for artificial intelligence or consciousness.

Sometimes I hear "those researchers are doing their best, don't be hard on them." I hate criticizing people's work, especially when they are doing their best. But, I also think about the people who gave time and risked their privacy to be study participants. We owe it to them to get things right.