
phd student biomedical informatics @stanford | immunologist➡️biostatistician | clinical trials | @columbiamsph @uchicago | swammer
Joined May 2021
- Tweets364
- Following415
- Followers45
- Likes629
Bryan Bunning retweeted
The @US_FDA now says it will now allow *pharmacists* to prescribe @pfizer's Paxlovid pills to Americans who test positive for COVID-19, as long as they have
"sufficient information to determine their eligibility"
fda.gov/news-events/press-an…
Bryan Bunning retweeted
The biopharmaceutical industry, and the patients who rely on it, have a stake in the outcome of challenges to FDA preemption that will be triggered by the Supreme Court's decision to overturn Roe v Wade. ($) biocentury.com/article/64419…
Bryan Bunning retweeted
Fearless grad student @veronical1411 led this team science effort to answers the question, why don't you feel hungry after a hard workout? @nature @SnyderShot @StevenMBanik @proferikrichter @DrYongXu1 @Stanford_ChEMH @StanfordPath @WuTsaiAlliance nature.com/articles/s41586-0…
Bryan Bunning retweeted
Generalized Additive Models (GAMs) are fully interpretable ML models, unlike DNNs, but it's hard to make them efficient. We're sharing research on scaling GAMs to real world tasks w/o sacrificing accuracy or interpretability. arxiv.org/abs/2205.14120 arxiv.org/abs/2205.14108 [1/4]
Bryan Bunning retweeted
The brain's energy consumption is similar to that of a very bright LED bulb -- or 1/5th of that of a standard incandescent bulb. It's not exactly a data center.
Bryan Bunning retweeted
1. FTC has voted to order the six largest pharmacy benefit managers (PBMs) to produce documents and data as part of an inquiry into PBMs and their impact on patients, pharmacies, payers, and doctors.
ftc.gov/news-events/news/pre…
Bryan Bunning retweeted
You must to read this JACC paper CI folks!! Observational studies is a false solution.
pubmed.ncbi.nlm.nih.gov/3273…
@_MiguelHernan @yudapearl @EpiEllie @statsepi @stephensenn @f2harrell
Rarely are you so excited by a KM curve that looks like this!
Half the chemo, same outcomes!
The MRD revolution has started. In this trial (n=455), stage II CRC pts were randomized to standard management or ctDNA-guided (no adj chemo if ctDNA negative). ctDNA guidance halved the use of chemotherapy, preserving outcomes.
We need the same in #bcsm!
nejm.org/doi/full/10.1056/NE…
Bryan Bunning retweeted
"Men who lose access to Medicaid eligibility are 14% more likely to be incarcerated in the subsequent two years relative to a matched comparison group. The effects are entirely driven by men with mental health histories."
by @elisajacome
elisajacome.github.io/Jacome…
Bryan Bunning retweeted
Ultimately, our goal is to show how hospitals are not all that far from implementing AI-QI. Maybe it’s time to bring together clinicians, statisticians, ML experts, IT specialists, and others to create AI-QI teams in the hospital. 4/4
Bryan Bunning retweeted
Check out our paper out today in Nature Digital Medicine on quality improvement (QI) for clinical AI (aka AI-QI). nature.com/articles/s41746-0… @pirracchior 1/4
Bryan Bunning retweeted
A rounding dashboard reduced data gathering time by ~50%.
jmir.org/2022/5/e35981/
"If you've chosen not to wear a mask in
indoor public places recently, now is a good time to start again"
After weeks of going maskless in the Bay Area, recently I chose to mask again until rates die down.
After two years of being the star pupil in terms of Covid case rates, rates in the Bay Area are now fairly high – including highest in California (Marin 45/100K/d; SF 40; Alameda 37; vs. CA overall 23).
@nytimes on why that might be. I'm quoted. nytimes.com/2022/05/17/us/ba…
1
I use wastewater data to inform my mask decision making, which has showed an uptick from early may.
It’s easier, community specific, a leading indicator, and more consistent imo to make such decisions
covid19.sccgov.org/dashboard…
Bryan Bunning retweeted
Replying to @xkcd
"Donate now to help us find a cure for causality. No one should have to suffer through events because of other events."
Bryan Bunning retweeted
Google taps FDA’s former digital health chief for global strategy role statnews.com/2022/05/16/baku… via @statnews @caseymross @KBDeSalvo @_bakulpatel @cristianliu #googlehealth
RCTs won’t ever go away and no amount of brilliant causal inference methodology will change that. Both are useful, though!
“Most importantly, real-world evidence complements evidence from randomized trials and does not replace it.”
Real world evidence: Lessons learned from an FDA pilot show the limits of emulating RCTs - endpts.com/real-world-eviden…
“The second lesson is about progress, generally: Invention is easily overrated, and implementation is often underrated.”
EXECUTION MATTERS!!!
"What if humans invented a technology to save the planet, and then just refused to use it?"
This isn't a riddle. It's reality. From COVID vaccines to nuclear power, Americans keep failing to implement or deploy world-saving inventions. I wrote about why:
theatlantic.com/newsletters/…




















