Profile avatar
rohitpojha.bsky.social
Director & Associate Professor, JPS Health Network Center for Epidemiology & Healthcare Delivery Research | Causal inference • Prediction • Evidence synthesis
26 posts 949 followers 82 following
Regular Contributor
Conversation Starter
Posts 21 Comments 17

“The results suggest that large language models do not have the same sense-making capabilities as human beings.”

submitted 3 hours ago • 0 comments

Changes in the national research infrastructure and uncertainty about future opportunities will create a pool of talented population health researchers seeking new opportunities.   Healthcare organizations could benefit by embedding these researchers within the organization. tinyurl.com/yh4n3tu6

submitted 9 days ago • 0 comments

1/ When using observational data for #causalinference, emulating a target trial helps solve some problems... but not all problems. In a new paper, we explain why and when the #TargetTrial framework is helpful. www.acpjournals.org/doi/10.7326/... Joint work with my colleagues @causalab.bsky.social

submitted 10 days ago • 3 comments

1/ If you were taught to test for proportional hazards, talk to your teacher. The proportional hazards assumption is implausible in most #randomized and #observational studies because the hazard ratios aren't expected to be constant during the follow-up. So "testing" is futile. But there is more 👇

submitted 25 days ago • 4 comments

NEW PAPER Really glad to see this one in print: the harm due to class imbalance corrections in prediction models developed using ML/AI Excellently led by @alcarriero.bsky.social onlinelibrary.wiley.com/doi/epdf/10....

submitted 32 days ago • 0 comments

Our systematic review of EHR-based prediction models to identify patients who may benefit from HIV PrEP: • 7 models identified • 𝗔𝗹𝗹 𝗺𝗼𝗱𝗲𝗹𝘀 𝗵𝗮𝗱 𝗵𝗶𝗴𝗵 𝗿𝗶𝘀𝗸 𝗼𝗳 𝗯𝗶𝗮𝘀 • Most reports were missing critical information Pre-print available here: www.medrxiv.org/content/10.1... #EpiSky

submitted 34 days ago • 0 comments

Please stop telling me about risk factors. 🙏😖 (ICYMI) statsepi.substack.com/p/sorry-what...

submitted 39 days ago • 9 comments

What effect is being estimated? Clarifying this question can help decision-makers better use the evidence. Recent developments in defining estimands (i.e., the effect of interest) can help. Here’s a nice summary to help think about estimands: academic.oup.com/aje/article/... #EpiSky #CausalSky

submitted 57 days ago • 2 comments

LLMs may be good for certain tasks, but encoding causal knowledge is not one of them.

submitted 72 days ago • 0 comments

Many measures used in evaluating AI prediction models do not measure the intended parameter. AUC, calibration plot, and decision curve analysis are recommended for evaluation.

submitted 74 days ago • 0 comments

Study reports that “promotional language” in funding proposals is associated with being awarded funding for biomedical research. Translation: Funders reward hype and spin rather than more balanced perspectives. What gets rewarded needs to change. #EpiSky #MedSky jamanetwork.com/journals/jam...

submitted 79 days ago • 0 comments

Outdated: Describing an exposure-outcome study as a “retrospective cohort study” just because the data were already collected. Updated: Describing a study based on timing of exposure and outcome measurement. More informative about timing of events and potential biases.

submitted 83 days ago • 1 comment

It's that time of year to bring out our @bmj.com xmas paper On the 12th Day of Christmas, a Statistician Sent to Me... from 2022 (w/ @richarddriley.bsky.social) with our list of commonly seen statistical faux pas🎄 www.bmj.com/content/379/... #StatsSky #EpiSky

submitted 85 days ago • 0 comments

Nice call to action for how epidemiologic research can make more meaningful contributions to policy and society. “Neglected topics with policy implications may generate more impactful epidemiologic research than frequently studied topics in need of refinement.” academic.oup.com/aje/article/...

submitted 89 days ago • 0 comments

⚠️Urgent public service announcement for anyone using Microsoft Office: by default Office will scrape your documents, unless you opt out, as described below:

submitted 94 days ago • 36 comments

This is a fantastic tool if you’re looking for people with overlapping interests (as judged by networks) to follow: bsky-follow-finder.theo.io

submitted 102 days ago • 16 comments

Characterising #machinelearning studies in healthcare, the full set Weak design/methods tinyurl.com/yc4easr9 Poor reporting tinyurl.com/55ed3j9k High risk of bias tinyurl.com/yk6m9sx5 Full of Spin tinyurl.com/yckubrnp Not open science tinyurl.com/437bfz8f #StatsSky #MLSky #mustdobetter 😬

submitted 106 days ago • 8 comments

In April I predicted LLMs were reaching diminishing returns. 𝗔𝗹𝗹 𝘀𝗶𝗴𝗻𝘀 𝗻𝗼𝘄 𝘀𝘂𝗽𝗽𝗼𝗿𝘁 𝘁𝗵𝗮𝘁. 6 related predictions below for 2024 also were correct. That leaves 1 open: 𝗱𝗶𝗺𝗶𝗻𝗶𝘀𝗵𝗶𝗻𝗴 𝗿𝗲𝘁𝘂𝗿𝗻𝘀 𝗹𝗲𝗮𝗱 𝘁𝗼 𝗰𝗼𝗹𝗹𝗮𝗽𝘀𝗲 𝗼𝗳 𝗚𝗲𝗻𝗔𝗜 𝘃𝗮𝗹𝘂𝗮𝘁𝗶𝗼𝗻𝘀. Hasn’t happened yet but could—as the implications of slowdown set in.

submitted 108 days ago • 8 comments

We've put together an #Epidemiology / #EpiSky starter pack with some of our members & friends! Please share! TESTIMONIALS 'The original data scientists' - @miguelhernan.bsky.social 'The new rock stars' - @nytimes.com 'A science of high importance' - @natureportfolio.bsky.social go.bsky.app/K6DXCGi

submitted 108 days ago • 25 comments

In clinical research, you will often receive feedback on study design, stats, and/or data analysis from an editor or reviewer that is simply wrong. Here is a list of common "statistical myths" and references you can use to push back. discourse.datamethods.org/t/reference-...

submitted 108 days ago • 48 comments

NEW PAPER in the BMJ (with richarddriley.bsky.social) - 1st in a 3 part series on the ‘Evaluation of clinical prediction models’. Part 1 is ‘from development to external validation’. —> tinyurl.com/n8fy5xvj

submitted 417 days ago • 1 comment