Ryan Radecki, MD MS

If you're waiting for the frontier models to navigate your EHR and complete physician-level tasks – don't hold your breath. But, that dawn is just over the horizon. #FOAMed evidencetriage.com/p/can-the-chat…

Yes, a Google symptom checker embedded into the FitBit platform generated some valid differential diagnoses. Beyond that, the data is messy enough as to limit many conclusions. #FOAMed evidencetriage.com/p/the-fitbit-s…

Ivermectin Is for scabies nejm.org/doi/10.1056/NE…

@TheSGEM @DrNeilStone @DrNice2026 Best RCT ever.

@DrNeilStone @DrNice2026 We reviewed the parachute RCT on the SGEM. thesgem.com/2020/02/sgem28… @DrNice2026

A new parachute design comes out Should it be tested to compare it against a) The old parachute design b) No parachute at all Now do you see the problem with demanding placebo controlled trials for every vaccine??

@clairevo @stripe @ow ... but the next frontier is really: "we build this tool in pretendworld, now how do we integrate it with our existing digital infrastructure?" That's the next point of production friction.

@clairevo @stripe @ow I've only seen kiro.dev We had plenty of non-techies playing with it to ... some success ... in a limited project space as a test. I do agree the future is additional wrappers around the coding tools.

While your team is just getting started vibe coding prototypes, the team at @stripe is vibe coding a full vibe coding platform. @ow is a design manager at Stripe who built Protodash - an internal AI prototyping tool that lets designers and PMs spin up real, clickable prototypes in minutes. In this ep he walks through: - why generic AI tools produce "blurple slop" that doesn't match your design system - how he built a full prototyping studio on top of dev boxes with cursor + cc - why more PMs than designers now use Protodash - the design review mode that lets teams ship feedback direct in the app Plus, we talk about getting to the ultimate dream: demos, not memos. ty to our sponsors! 🧠 @Celigo - Intelligent automation built for AI 💻 @cursor_ai - The best way to code with AI Full episode on yt: youtu.be/hQFEAZK__q0

@clairevo @Anthropologie @AnthropicAI @claudeai @ChatGPTapp @midjourney @OpenAI TBF, if there was a company store that made "spinner words" hats, I wouldn't be sad.

The @Anthropologie x @AnthropicAI crossover no one asked for but I made anyway. @claudeai design @ChatGPTapp image 2 + @midjourney @OpenAI codex anthropic-linen-sage.vercel.app

@DrToddLee @ABsteward

@ABsteward I tell people all the time that nothing beats high dose PO amox and they look at me like I gave five heads.

In vitro activity of β-lactams (amoxicillin piperacillin, imipenem, meropenem and ceftobiprole)against contemporary Enterococcus faecalis clinical isolates France Amox: 100% S PiP: 91% S MER: 97.5% S MER: 75.5% S Cefto: 86% S link.springer.com/article/10.100…

The media hype has decreed: emergency medicine is a dead specialty, and your Friendly Neighborhood LLM is ready to replace you. Just another challenger bringing the disrespect to our specialty .... #FOAMed evidencetriage.com/p/the-emergenc…

Don’t follow/subscribe to convicted sexual predators.

@DrToddLee … thanks for doing it!

@emlitofnote Thanks for highlighting our work.

If you're prescribing Paxlovid to prevent serious complications, the population in whom that might be a valid consideration is shrinking rapidly. If you're chasing theoretical advantages to viral load and symptom resolution, well, good luck. #FOAMed evidencetriage.com/p/paxlovid-is-…

Tired: amoxicillin-clavulanic acid for acute sinusitis Wired: no antibiotics Baby steps: amoxicillin Amazing how we can spend our whole lives talking about antibiotic stewardship and see little, if any, progress .... #FOAMed evidencetriage.com/p/dont-use-amo…

@elipwilber @DrToddLee @JAMA_current These stewardship tools aren't useful at the point of empiric therapy – they're still just part of the potential de-escalation toolkit. Everyone gets blasted if there's sufficient resistance around.

@DrToddLee @JAMA_current 100% agree. We’ve done this before: pubmed.ncbi.nlm.nih.gov/32374822/. These tests can be useful stewardship tools….but they’re not going to move the needle on patient outcomes unless the background rates of resistance go up without a corresponding change in empiric therapy.

Among patients with #GramNegativeBacteremia, rapid #AntimicrobialSusceptibilityTesting did not improve 30-day clinical outcomes compared with standard testing, despite providing faster guidance for antibiotic escalation or deescalation. #ESCMIDGlobal2026 ja.ma/4mAFzbM

@JSchuurMD The supply pipeline failures still haven't fully manifested – and jet fuel vs. petrol vs. diesel are their own separate ecosystems. We're still finishing off the fumes of the last bits of crude in SE Asia before the refined product really dries up.

Why aren't oil futures trading w/ time decay (theta) like options? Each day that passes w/out opening Strait of Hormuz increases physical oil shortage in import economies, e.g. Asia/Euro & companies directly affected, like euro/asian airlines should see theta drops e.g. $RYAAY

@mcuban "If you are a politician and reading this" assumes they can read! I kid, I kid. The most powerful actors (and those holding the legislative levers) are the ones benefiting from the current system. Overseas, hospitals still have issues – but, at least, not like these ....

The greatest problem in healthcare ? Hospitals, even market dominant hospitals, won’t walk away from the big ins companies that underpay, late pay, clawback, deny claims, waste their time in denial appeals, and require them to pay up to 8 pct of revenue to RCM consultants so they think they are getting what they are owed. Here is the crazy part. The ins companies ARE NOT THE ONES ACTUALLY PAYING THEM on commercial plans. Employers are. 60 pct of employees get their insurance from their self insured employers. The ins carrier is just a middleman that pretends to add value. All the clinical “value” they add, the hospital could do better, for both medical and pharmacy. Most hospitals have no idea whether they make or lose money with their big ins contracts. They are just afraid to lose patient flow. But. They actually know which companies their patients are coming from. They actually know or can find out, how much more the employers are paying the ins company, than what the ins company pays them (the spread, just like in pharmacy ) And to make it worse, those ins companies negotiate their rates as a discount from the “charge master “, which is like WAC in pharmacy. Just a made up list price. Because the hospitals are afraid or too uninformed to walk away from these deals, the hospitals use the inflated charge master prices as the basis to charge uninsured , or out of network , or insured but not covered for their care, at charge master rates. Which of course the patients can’t afford. And it crushes their finances or they go without care I’ll summarize. Employers , and their members , are paying far more than they should to companies they don’t like working with , that effectively rip off both the employer and hospital , and they could eliminate the middlemen if they went directly to to the employer. It’s so simple. Sell your services to the employers that use your services at a price that is less than what nine companies charge for your services and you will make MORE money and employers will save a ton And if they did this, they could dump the chargemaster and reduce the price they bill patients when they are at their most vulnerable But they don’t want to change. And don’t get me started on how much hospitals over pay for drugs and devices because of the GPO deals they do. It’s just stupid. Which in turn leads to the hospital being a bad actor with 340b , facilities fees and afraid of their doctors who demand they pay more for things like glue and implants so they can get vacations. If you are a politician and reading this. Now you know why this is so fucked up and it’s not about capping rates. The insurance companies are smarter than you. They will just move the money to other places. It’s not about giving money to patients. You can’t shop for care from hospitals that are too gutless to walk away from the ins companies that distort all of healthcare economics Go to your local hospitals , particularly those at risk of closing and ask for their profitability by carrier. Fully burdened. Ask how much they spend on RCM and consultants. In many cases they could survive if they ran like a real business and hired execs that could do the work rather than just manage consultants. They could work out contracts in their communities rather than with ins companies and benefit everyone. The middlemen are not needed. Get rid of them

@clairevo beep bop boop bop

i am creative director

@DrToddLee @LucusHumphrey Outstanding! You must be thrilled to get the result from all the hard work. Very impressive.

Thanks coach @LucusHumphrey for a great winter training session and great advice on pacing. Pretty much bang on what we set out to do. About a 6 minute PR on a much harder course and a year older.

Yes, AI scribe note quality has limitations – amplified when a consult occurs in a noisy environment, or the human comparators know they're writing a note for a head-to-head evaluation. #FOAMed evidencetriage.com/p/more-on-ai-s…

@BrooksWalsh Hah! Yep this just showed up in my Pubmed digest this morning ... and I've posted on it!

Faster medical charting with the aid of AI scribes is, well, faster. But what if the quality suffers? "Across all 5 clinical cases, human-generated notes received higher overall [quality ] scores than AI-generated notes" @emlitofnote acpjournals.org/doi/10.7326/AN…