Was is Kissinger who said that Academic politics are vicious because the stakes are so low
Posts by Michelle Kelly-Irving
Bon bah focus sur les chaussettes
…
#jyvoisrien #FRAvENG #crunch
Surely letting an AI use the ukda data would be a breach of user licence rules?
Yes! Was quite a day today in the 6 nations (Italy just beat England!). Next week will be interesting 🤨 if Scotland play as they did today they’ll beat Irl
Congratulations, Dr. Meena Kumari, PhD winner of the 2025 overall best paper award. lnkd.in/euUyrnag
It is our pleasure to announce Meena Kumari, PhD as the winner of the 2025 Overall Best Paper award.
Co-authors on the paper are Elise Whitley, Michaela Benzeval, and Michelle Kelly-Irving.
🧵1/ Global health is under attack, and not just by pathogens. Governments are walking away from international law, blocking humanitarian access, and even targeting health workers. We need a principled response, which @tiagocorreia.bsky.social & I set out in an editorial in Eur J Public Health
It helps, however, to be a Paris-based researcher, closer to the seats of power and influence... after all, data is power these days
Regularly, by the time we have obtained data for a research objective, 18 to 24 months have passed, and funding is running out for the early career researchers who were planning to work on it
The principle is that of suspicion. Despite both the researchers and the data being paid for by public funds, researchers are treated with suspicion in our attempts to access the data.
The right hand does not know what the left hand is doing
The value of publicly-funded population cohort studies is not in how many bone fide public researchers make use of the data to the benefit of science and the French population, instead data are 'safeguarded' by layers needless regulations & personal conflicts of interest
The dataset connecting French social & health data requires permissions from 3 regulatory bodies, 3 sets of producer approvals & the use of an online platform oxymoronically named ‘simplified procedures’ which includes a 37pg checklist embedded within another checklist
📢 TRUTH 🔽
"Our politics is dominated by people who have never lived on a low income and know little of what such a life involves. There are remarkably few ways for people to truly shape government decisions that will determine their own lives."
www.theguardian.com/society/2026...
I’m still not sure if enough Americans fully understand how
they have subjected the rest of the world, and themselves to this…
by voting for it
Twice
#HealthInequalities are exacerbated in the health system when women’s symptoms are considered to be “atypical” or explained-away as being not ‘true’ physical symptoms. When building research projects on health this issue needs to be addressed in the design phase
Be critical of equations, algorithms, tools or instruments that “adjust” automatically for sex or race/ethnicity ex. adjusting lung or kidney function readings/ estimates. These are common in medicine, often applied without knowing about them. Check for these & question why they would apply
Social positions (gender, social class, race/ethnicity etc) are embodied physiologically. There are social-driven differences in biology between men & women, across social class and racialised minorities, and at the intersections of these positions which affect cardiovascular risk factors & outcomes
Despite the call being driven by well-known inequities in cardiovascular health care, the idea that women’s symptoms are “atypical” is still in the vocabulary of practicing clinicians.
They are not atypical. so don’t bake that into your design or you’re perpetuating avoidable inequalities
A couple of recent observations regarding knowledge about #HealthInequalities among scientists as I work with diverse sets of colleagues preparing grants for the Horizon Europe sex/ gender & cardiovascular disease
In case you missed it, an article collection on the “Social uses of the body”
www.sciencedirect.com/special-issu...
As much as i loath Teams, this ‘homegrown’ stuff will be a shitshow for those of us in French public sector. They rolled-out a new accounting system to “simplify” bureaucracy in my institution in Nov 2024 & I still dont have updated budgets today
Heres how the “let’s roll back public health” #MAHA experiment is going in the US
“Hundreds of children across more than 20 schools in the Upstate portion of South Carolina were quarantined for the measles, health officials reported on Tuesday”
www.wbtv.com/2026/01/27/h...
Hello! I’m looking for rich ethnographies of the recent feminist movements in Latin America around abortion rights, femicides, and/or “buscadoras” and the disappeared. Can someone please help? Much obliged #anthrosky #anthro #socialscience #sociocultural #academicsky #academicchatter #acwri
Oh, sounds good! I recommend Three Bead Salad podcast @threebeansaladpod.bsky.social for light silliness containing "people who are not sure what they're doing and whether they should be doing it" !
Headline from Reuters with Meta logo picture "UK announces Meta-backed AI team to upgrade public services"
Headline from the Guardian with Meta logo "Meta allowed minors access to sex-talking chatbots despite staff concerns, lawsuit alleges"
On the one hand, "The British government said... it had recruited a team of AI specialists to build AI tools to improve transport, public safety & defence, using funding from Meta"
On the other hand Meta is "Meta allowed minors access to sex-talking chatbots despite staff concerns, lawsuit alleges"
Postdoc position in Paris: come help develop new generation human brain computer interfaces ⚡🧠💻
Interested? Contact me if you have experience with machine learning (e.g. simulation-based inference, RL, generative/diffusion models) or dynamical systems.
See below for + details and retweet 🙏
This is a P.S. that doesn't really fit anywhere else: Dr. Carlson, I know you are resigned to the fact that the use of AI is inevitable and that you cannot avoid getting submissions from students that have used AI. I can't speak for everyone, but I want to say that I don't think you should give up on the idea of an AI-free classroom. I'm not sure if you had already looked at assignments when you spoke about this, and I obviously do not know what other people submitted, but I know that among myself and my friends, using AI is very frowned upon. I'm a bit of an older student. I finished college before AI became a popular tool the way it is now, so maybe I'm biased and the youths actually will make AI adoption inevitable... But I don't think it has to be, and I think much of the push for the idea that it's not going anywhere comes from huge tech investors. My key takeaway for you here: stick to your AI policy that you told us on the first day of school. You don't have to expect AI submissions. Students do better when there are people expecting better from them, especially such a respected instructor. I am very pro failing students who use AI for assignments (outside of spell check and such) and I know many other people are too. Those of us who don't use AI don't want to be in the public health field in 20 years and see that no one is able to do anything without it. It's honestly a very weird and kind of scary time to be someone in academia, and the AI of it all is making it so much worse. I really respect your policy and strong stance that you outlined on day 1, and I think it's more than fair. So TL;DR: please don't expect us to use AI, punish those who do, and know that your strong anti-AI stance is making a positive impact in our educations.
I replaced my final assignment this semester at the last second because I felt like I couldn't LLM-proof it, and told my students as much. I got this back on my course evals. The kids are alright
Voting is a stronger social determinant of health than education #Vote #SDoH #EpiSky #SocialEpi #PublicHealth #HealthEquity