Zero studies from clinics like mine đłď¸, "Once" means eleven đ, 1,000 order sets â 350 đŞ
⥠Around the Wards
Nine landmark ambient-scribe studies, zero from a safety-net clinic â Dr. Gigi Magan checked forty-one references against their primary sources and found the evidence base has a hole shaped exactly like her patients.
âOnce dailyâ reads as âeleven dailyâ in Spanish â RxTranâs Sharon Blank argues pharmacist verification of translated labels is permanent infrastructure, not a transitional safeguard.
One hospital took its order-set library from nearly 1,000 to 350 â the fix for duplicate sepsis order sets turned out to be a governance pattern, not a build ticket.
3,500 tubes of blood, 153 terabytes of data â ChronicleBio says it has already found five sub-diseases inside POTS where the biology differs but the symptoms donât â which would explain a decade of trials that went nowhere.
đ§ Podcast: Radio Advisory â â309: Regional health plans attempt a financial turnaroundâ â Advisory Boardâs Jared Landis on why the rate-setting calendar quietly breaks most vendor pitches to regional plans.
đ§ The Curbside
âOur CIO keeps saying weâre in a squeeze. What does that mean for my pitch?â
Short answer: It means your tool is competing against roughly 150 ideas for about ten funded slots, and âcapabilityâ isnât a category that survives that.
What changed: Bill Russell, Drex DeFord and Sarah Richardson just came back from three rooms of academic medical center leaders â CIOs, CMIOs, and revenue cycle â and reported the number that came up in all three: roughly $900 million in projected top-line revenue coming out over three years. The governance process they described: departments generate ~150 ideas, a subcommittee chair filters up 10 to 20, and the organization lands on roughly ten projects it will actually do â published, so everyone can see where they sit.
Builder read / watchout: The subcommittee chairâs filter is âdoes this add revenue or take out cost,â and the second filter is âhow many of my IT hours does this eat.â A CMIO in that room put it as decisions made elsewhere becoming checks they canât cash. If your one-pager doesnât have a cost-takeout number and an IT-hours number on it, your champion is improvising in a five-minute slot â and improvised pitches land at #11.
đ¤ Haters: âThis is just enterprise sales advice with a stethoscope on.â Partly. But the specific thing here is that the ânoâ you get is usually not about your tool. Itâs a resource statement, and resource statements have a shelf life.
đŹ The Big Thing
Nine landmark scribe studies. Zero from a clinic like hers.
Dr. Gigi Magan, a family physician who writes from a safety-net exam room, spent the last month building a lecture on ambient AI scribes for the California Telehealth Resource Center. Forty-one references, each one checked against the primary source.
Then she lined up the settings of every major study: UCLA. Mass General Brigham. Emory. UCSF. Yale. UC Davis. Kaiser Northern California. Penn. Stanford.
Published trials or large studies conducted in a federally qualified health center or community health center: zero. Peer-reviewed studies reporting scribe accuracy stratified by patient race, language, or accent in real clinical use: also zero.
The patients most likely to benefit are the least represented in the evidence and the most exposed to its failure modes.
That second zero is the one that should make you sit up.
We already know the underlying technology has a demographic gradient. The landmark 2020 test of five commercial speech recognition systems found an average word error rate of 0.35 for Black speakers against 0.19 for white speakers â nearly double â with the gap tracing to the acoustic model and thin training data. Magan adds the number that doesnât get quoted: more than twenty percent of Black speakersâ audio was degraded beyond usability, versus under two percent for white speakers.
Vendors have invested heavily since. Nobody has published the check.
Health centers serve more than 31 million people. The visits are long, multilingual, and heavy with social complexity â which means the documentation burden per visit peaks exactly where the evidence stops.
Magan is careful, and I want to be careful the same way: she uses one of these tools every day and isnât going back to typing.
This is not an argument against ambient documentation. Itâs an argument about who gets to generate the evidence.
Hereâs what makes it a builder story instead of an op-ed.
Cardiology has started building the machinery. The AHAâs AI Assessment Lab, powered by Dandelion Health, ran Ultromicsâ EchoGo Heart Failure against roughly 90,000 real-world echocardiograms and published the result with subgroup findings by race and age alongside the accuracy numbers â HFpEF identified up to 263 days earlier than standard care in patients who would otherwise have been missed, wrapped in clinical and economic modeling out to five years.
Thatâs an independent body, independent data, and stratified results published where a procurement committee can read them.
Ambient documentation is deployed far more widely than that algorithm and has no equivalent body doing that work.
So the assessment layer for the most-deployed clinical AI in America is vacant, and the entry requirement is a QI dashboard, not an R01.
Decline rates by language. Note quality by population. Edit burden on interpreter-mediated visits.
None of that needs a grant. It needs someone to decide on day one that equity gets measured instead of assumed.
đ¤ âAbsence of evidence isnât evidence of absence. Youâre fearmongering about a tool that demonstrably works.â She said the same thing more plainly than I would have â she uses it daily, the benefit is real, and the time savings are modest but the attention benefit isnât. The claim isnât that scribes fail on accented speech. The claim is that nobody has looked, in the settings where it would matter most, and âwe assume it improvedâ is not a finding.
đ¤ âFQHCs donât have research infrastructure. Thatâs why the studies are at Stanford.â Right, and thatâs the two-tier outcome writing itself. Somebody has to go first.
đ¤ âVendors have this data internally.â Then ask for it.
𧪠Try the interactives:
A â Nine and 1,352 â an animated field of every federally funded health center in America, sorted by the share of patients best served in a language other than English, against the nine academic settings where the ambient-scribe evidence was actually generated. Built with real HRSA health center data.
B â Clinics Like Mine â every HRSA health center grantee in the country plotted at once: panel size against the share of patients best served in another language. 32.3 million patients, 27.8% of them non-English-preferred, and not one published ambient-scribe study run in any of them. Built with real HRSA health center data.
A hospital retired two-thirds of its order sets and called it a governance pattern
John Lee, MD â an emergency physician and Epic consultant â published the second half of his sepsis order-set story: at a health system he documents but doesnât name, a duplicate-sepsis-order-set conflict got resolved by bucketing review by clinical domain, and when the same model was applied system-wide over two years, the library went from nearly 1,000 records to 350.
Dormant sets were retired outright. Active duplicates werenât arbitrated â they were consolidated into one.
Two-thirds of an EHRâs clinical content library was redundant, and nobody could see it until the review was organized by clinical domain instead of by request queue.
The build here isnât software. Itâs the review structure that makes the redundancy visible â and thatâs a thing a clinician can design and an engineer canât.
đĄ 80/20: Before you build a tool that writes order sets, count the ones you have. Pull the list, group by clinical domain rather than by owner, and find the duplicates. If your ratio looks anything like 1,000-to-350, generation was never your problem.
â Two-thirds of a content library was dead weight and it took a two-year human review to see it. Is deduplicating clinical content actually a good first agent task â or is it the one place you most want a human who knows which of two nearly-identical sepsis order sets the night team actually uses?
A trauma surgeon rebuilt his research labâs operating system in twenty minutes
A trauma surgeon mentoring thirteen trainees across fifteen manuscripts replaced years of accumulated lab-management workarounds with Notion plus an agent wired in over MCP connectors. His section heading for the whole build: âBuild the whole thing in twenty minutes.â
Thatâs the whole clinicians.build argument compressed into one weekend project. The scarce input wasnât engineering.
It was knowing what a manuscript pipeline actually needs to track when the first author is a PGY-3 on nights.
đŽ My bet: within a year the interesting artifact from clinician-builders isnât the app â itâs the connector config. The thing worth sharing is a working MCP setup for a specific clinical or academic workflow, and somebody is going to start a registry of them.
ChronicleBio says POTS is already five diseases, not one
ChronicleBio â cofounded by Fidji Simo, Rohit Gupta and Rishi Reddy â has banked more than 3,500 tubes of blood and pulled 153 terabytes of data out of them: 890 draws from 709 patients in its first year across Utah, Arizona, Texas and India, on $15M raised. On August 11 it opens sign-ups for mobile phlebotomy trucks that come to patientsâ homes, free for the first 250 and $400 after, in exchange for their biological data.
The thesis is a clinical-trial thesis, not an AI thesis. In Simoâs words: âthere could actually be five sub-diseases within POTS, and the drug would work for one of them, but not the other four.â She says theyâve already found them â same symptoms, different biology, immune-driven in one group and mitochondrial in another. The sub-diseases donât have names yet.
If theyâre right, the reason these trials keep failing isnât the drugs. Itâs the phenotype.
đ¤ âThis is a biobank with an AI press release stapled to it.â Itâs a biobank, yes. Thatâs the point â the constraint in neuroimmune disease has never been model architecture, itâs that nobody assembled the cohort. Deep phenotyping on a neglected condition is unglamorous and itâs the actual bottleneck. Ask me again in two years whether the five-way split replicates.
đď¸ From the Pods
đď¸ Radio Advisory â â309: Regional health plans attempt a financial turnaroundâ
Advisory Boardâs Jared Landis, talking with host Rae Woods, lays out the mechanic that breaks most vendor pitches to regional plans: 2025âs rates were determined in the first half of 2024, so everything a plan learned during 2024 couldnât touch its 2025 bid â it lands in 2026. Anything you sell that only improves pricing accuracy has a payoff two budget cycles out.
đ Speaker Blindspot: Composition fallacy. Landis argues the turnaround isnât zero-sum â âthey can all improve their pricing, they can all improve their operations.â True for any one plan; not true for all of them at once, in a market heâd already described as stagnant where a plan grows only by taking an account from a competitor and where employers keep shifting to self-funded. He then uses national for-profit payersâ Q2 earnings as a proxy for Blues performance minutes after explaining that the nationals have structurally different scale and margin dynamics.
Victor Hassid, MD, Associate Vice President of Access Strategic Operations at MD Anderson Cancer Center, describes running every intake question through two comprehension checks â whether the administrative team on the phone understands what the clinical team meant, and whether the patient understands what the scheduler says. His line: it doesnât matter what either of them says if the patient doesnât comprehend it.
đ Speaker Blindspot: Goodhartâs law, self-inflicted. His headline result is cutting median time-to-offer from five days to two â and he himself warns that it is âinappropriate and at the same time risky to assess access to care as access to your first appointment,â because you can clear an upstream bottleneck and create a downstream one. Then he proposes industry-wide access KPIs with âpossibly even ranking,â which is the textbook condition for a measure to stop being a good measure.
What are you building this week? Email and tell me (kevin@clinicians.build) â I read every one.
â Kevin