Excited to share results from a 4-year first-ever RCT across US jails. We (Crystal Yang & I) find that death rates are much higher in jails than officially reported & health care accreditation improves staff coordination, quality & save lives. www.nber.org/papers/w33357

it's happening

EC 2025 will be held at Stanford from July 7-12. Itai Ashlagi and I are the chairs. The abstract deadline is February 3, and the paper deadline is February 10. The scope is inclusive of many topics across CS, economics, and operations research. Submit your best work!

Attn Tuxedo Cat People of Bluesky There is a new Lego that looks like all of our cats! 🤩 @sarahgollust.bsky.social @dmaanderson.bsky.social

Academics from poorer socio-economic backgrounds are more likely to - not publish - have outstanding publication records - introduce more novel scientific concepts - less likely to receive recognition, as measured by citations, Nobel Prize nominations, and awards. www.nber.org/papers/w33289

tl;dr Healthcare access disparities cascade through the entire ML pipeline. Check out our working paper here: arxiv.org/pdf/2412.07712

Join me virtually next month for the International Biometric Society Distinguished Lecture on January 30! “Ethical AI is More than Loss Functions" www.biometricsociety.org/education/dls

Our study, the first post-implementation evaluation of the removal of race adjustment from the eGFR equation, is discussed in this Nature feature www.nature.com/articles/d41... We found no change in nephrology referrals & visits Our paper, led by Marika Cusick proceedings.mlr.press/v248/cusick2...

Was a pre-med junior when a faculty member sent me a flyer about a new program called the Summer Institute for Training in Biostatistics Had never heard of 'biostatistics' as a field; sounded like exactly what I wanted to do Was accepted into the program, ultimately led to a PhD in biostatistics

I mean...it feels like this is true, but this is not correct 😆

Hello everyone! I teach a class for graduate students on methods for reproducible research. 🔧 🔢 I would interested to hear about the barriers you face in implementing reproducible practices in your research (e.g., version control, code review, reproducible documents).

cat in a cat bowl

Ok #FAccT folks - I'm still looking around for people. Who am I missing? go.bsky.app/EQF5Ne1

definitely misread this as "causal mediterranean"

Our Post-PhD Health Data Science Careers Panel is this Thursday at 11AM Pacific! Hear from folks at Google, Mathematica, and J&J in careers focused on population health and health policy. Virtual and open to all who register: healthpolicy.fsi.stanford.edu/events/post-...

I made one for stats papers

Couldn't find a machine learning for health starter pack so I made one.  DM/Reply if you want to be added! go.bsky.app/PJKJ8vK

I made a starter pack for health policy statistics! 📈🔢

quote a favorite film using a gif only

Excited to share that our paper "Self-Calibrating Conformal Prediction" with Ahmed Alaa is accepted at #NeurIPS2024! 🚀 We combine model calibration and prediction intervals by integrating Venn-Abers into conformal prediction. #conformal #calibration arxiv.org/pdf/2402.07307

👋 In Tokyo this academic year, on sabbatical at the Institute of Statistical Mathematics. In town and interested in causal ML? Would love to grab coffee and chat.

Academics being asked if they want to write a book

First post! I'm recruiting PhD students this PhD admission cycle who want to work on: a) impactful ML methods for healthcare 🤖, b) computational methods to improve health equity ⚖️, or c) AI for women's health or climate health 🤰🌎 Apply via UC Berkeley CPH or EECS (AI-H) 🌉. irenechen.net/join-lab/

@amstatnews.bsky.social HPSS conference #ICHPS2025 is two months away. I'm teaching a workshop, and so is @statstas.bsky.social! The keynote is @sherrirose.bsky.social who will discuss "Ethical Use of AI: Standards, Rigor, and Reality." Register so you don't miss these fantastic topics. #rstats

I didn't appreciate how much of being an academic is a bajillion microdecisions and fending off requests for hour-long chunks of my scarce time for "brain picking" or "brainstorming"

I have somehow accidentally(?) joined another academic committee

Making Halloween candy purchase decisions based on last year's candy data, as one does...

Posted elsewhere by @datavisfriendly.bsky.social

I volunteered to potentially host & organize an event. My 75-year old academic colleague, who is awesome: "I love it that you are down for this as the young people say."

Another incredible data jamboree (feat. @malcolmbarrett.malco.io) for the Health Policy Data Science Lab! We had sessions on efficient coding patterns, projects as packages, lessons from code review, and "what is a model?" as well as our journal club and SQL book club. What a fun nerdy time. 🤓

Join us virtually for a Post-PhD Health Data Science Careers Panel focused on population health and health policy! Panelists: Irina Degtiar, @mathematicanow.bsky.social Jack Ching, Google Ben Ackerman, J&J November 21 @ 11AM Pacific Registration: healthpolicy.fsi.stanford.edu/events/post-...

New policy brief on our work studying the removal of race adjustment from the eGFR equation: hai.stanford.edu/policy-brief... Algorithmic changes are not enough to address social & structural health inequities. Paper: proceedings.mlr.press/v248/cusick2...

Our new paper in AJE, led by PhD student Oana Enache, highlights problems in demographic tables and proposes graph-based reporting for understanding intersectional identities. Clinical Research Reporting Paradigms May Incompletely Describe Participant Identities: academic.oup.com/aje/advance-...

Radicalize researchers to look at each other's code ✊

my emeritus research assistant turned 15 years old and still likes to debate base R vs. tidyverse

bsky is growing, hi! Some areas I work in: Causal ML drsherrirose.org/targeted-learning-book Generalizability onlinelibrary.wiley.com/doi/10.1111/... Ethical ML www.annualreviews.org/doi/abs/10.1... Plan payment www.journals.uchicago.edu/doi/10.1086/... CKD proceedings.mlr.press/v248/cusick24a.html

COS is hiring a new Data Scientist position. This is a critical role for evaluating the effectiveness of our solutions, and offering insights for improvement to advance our mission. If you are passionate about open science, don't wait a second to apply! jobs.lever.co/cos/4ad3c942...

New paper! arxiv.org/abs/2405.08675 tldr: automatic differentiation can be used to derive efficient influence functions and construct efficient estimators.

me, an elder millennial, teaching quarto: who has heard of Sweave?

We've developed our Lab Manual over the past year in the Stanford Health Policy Data Science Lab: stanfordhpds.github.io/lab_manual/ Includes major contributions from @malcolmbarrett.malco.io leading us on improved processes for code review and reproducible reports! 📈 📄

then later that day, you'll be asked to review 1 of these papers that was started 3 months ago & is now under review in a clinical journal. you'll spend 4 hrs writing up 5 fundamental flaws that make the paper unsalvageable. the journal will publish it as is anyway because it has AI/ML in the title.

Welcome (albeit overdue) JAMA manifesto on causal inference jamanetwork.com/journals/jam...

Our work evaluating the removal of race adjustment from the eGFR equation at Stanford Health Care found no change in outcomes after 2 years arxiv.org/abs/2404.12812 (accepted @ CHIL 2024) Modifying the equation is likely insufficient for health equity as many social & structural inequities remain

We need less low-quality AI/ML research! New TRIPOD+AI guidelines published in BMJ: www.bmj.com/content/385/... Glad to be part of the team led by @gscollins.bsky.social See the thread below where Gary highlights a focus on transparency, ethics, open science, patient/public engagement & more

🔥 NEW PAPER out today in BMJ 🔥 The long awaited TRIPOD+AI: reporting recommendations for studies developing or validating prediction models for use in healthcare based onML/AI or traditional regression models (hence the "+") t.co/ed9oiB1MVG

NEW PAPER out today in the BMJ TRIPOD+AI: reporting recommendations for studies developing or validating prediction models for use in healthcare that use #machinelearning methods bmj.com/content/385/... #ArtificialIntelligence #AIstandards #OpenAccess

I became this much sooner in my academic career than anticipated

“Do you want to hop on a quick call about this?” N 　 O 　　　 O 　　　　 o 　　　　　o 　　　　　 o 　　　　　o 　　　　 。 　　　 。 　　　. 　　　. 　　　 .

Our paper on modeling population level task functional connectivity using Fourier series is published! We propose an estimator of population level task FC and discuss this contribution in the context of existing approaches for modeling FC. academic.oup.com/jrsssc/advan...