Transcript: Health data design for Medicare's 2026 surgical rules
An AI-generated audio overview (Google Gemini), transcribed by machine on Blaine's Mac (Whisper). The audio was checked against primary sources; what it gets wrong is listed at https://blainomd.com/learn#audio-fhir-native-2026-surgical

What if the secret to surviving the massive new 2026 surgical regulations wasn't a better scalpel, but a better database architecture?
Welcome to today's Deep Dive.
Our mission today is to take you inside the technical framework of a musculoskeletal marketplace called Surgeon Access.
Yeah, it's a really fascinating framework to look into.
Right, because we are going to explore how a back-end database standard called FHIR is turning new regulatory hurdles into a massive clinical and financial advantage.
Because, you know, the landscape is shifting radically right now.
I mean, we are moving away from traditional fee-for-service models and right into outcome-aligned payments.
So to survive that shift, the underlying technology has to fundamentally change how it handles patient data.
Exactly. And the Centers for Medicare and Medicaid Services, or CMS, is forcing this evolution right now with two major 2026 models.
First, you have the XSS model, which pays a $180 annual upfront amount, specifically tied to outcomes for chronic musculoskeletal pain.
Right. And then there's the TEAM model on top of that.
Yeah, the TEAM model. That one is mandatory, and it creates strict 30-day post-op accountability for acute surgeries, like joint replacements.
And because surgeons are suddenly on the hook for that entire 30-day post-op window, they desperately need a way to track patients at home.
Which is why the 2026 Remote Therapeutic Monitoring, or RTM, updates are the missing puzzle piece here.
They completely overhauled those billing codes.
Under the new rules, providers can bill around $40, $40.08 to be exact, for capturing just 2 to 15 days of patient data.
Which is huge, because it drops that old 16-day minimum requirement.
Right. Think of it like this. The old RTM rules were like being locked into an inflexible 30-day gym contract just to use the treadmill once.
But the new 2026 rules are like paying for a highly targeted two-week fitness sprint right when you actually need it.
And automating that two-week sprint is highly lucrative.
I mean, if a clinic can successfully track those short bursts of data without adding extra administrative work, it can generate up to $6,600 in monthly revenue for just a 100-patient caseload.
Wow, $6,600.
But tracking those short data bursts without drowning in paperwork requires a massive tech upgrade, right?
Because legacy systems just break under that kind of pressure.
Oh, absolutely. They crumble.
So that's why Surgeon Access uses Health Samurai's Aidbox platform.
Aidbox treats FHIR, fast healthcare interoperability resources, not just as some external plug-in, but as its foundational native data model.
It utilizes Postgresql JSONB storage.
Which basically means it stores the data as flexible documents rather than rigid traditional database tables.
Okay, but let's unpack this for a second. Why not just use a traditional relational database and, you know, slap a FHIR translation layer on top later?
I mean, that seems like the standard way everyone builds software.
Well, translation layers force complex, messy healthcare data into preset, rigid columns.
So if a patient's unique symptom or recovery metric doesn't perfectly fit a column, that data just gets dropped or completely corrupted.
Oh, wow. So you lose actual clinical data.
Exactly. But by being FHIR native, Aidbox reads the data exactly as it is.
It also creates automatic history tables for every single resource.
So it acts as an automatic immutable ledger.
It never actually overwrites the old data when someone updates a patient's file.
Right. It archives it instantly.
You get a perfect tamper-proof audit trail of the patient's entire recovery journey.
And I imagine that it's crucial for compliance.
It is because that perfect trail is what shields clinics from the potential 15% payment penalties CMS can enforce under these new accountability models.
Here's where it gets really interesting, though.
Because Aidbox structures the data natively and perfectly without those translation errors, it unlocks the ability to deploy advanced AI.
Yes. And it does it without massive engineering friction.
Surgeon Access integrates Claude for healthcare using the Model Context Protocol, or MCP.
Using MCP is like giving Claude a direct, secure pipeline into the hospital's locked filing cabinet, rather than forcing a human to manually copy and paste patient notes into a chat window.
It's like having a tireless robotic medical resident.
And this direct pipeline automates the drafting of letters of medical necessity.
I mean, it turns an eight-hour administrative nightmare of hunting through years of physical therapy logs and imaging results into a task that takes just minutes.
Wait, minutes? That's insane.
Yeah.
But you obviously can't just let an AI write legal medical records unsupervised.
No, definitely not.
Which is why the system relies on strict human-in-the-loop safeguards.
A licensed clinician must review and attest to the AI's work.
Right. To prevent confident AI hallucinations from ever making it into the official legal medical record.
But exactly. It keeps the speed but adds that necessary human checkpoint.
So what does this all mean for you?
Surgeon Access proves that surviving the 2026 musculoskeletal market requires a metadata-driven FHIR native architecture.
Yeah. You really need a system built from the ground up to actually understand healthcare data, not just warehouse it.
The database architecture is the hidden engine here.
It makes the targeted fitness sprint of patient monitoring possible.
And it allows the AI to operate with total precision.
Which leaves us with one final thought to chew on.
If AI can seamlessly read our perfectly structured FHIR native clinical history to optimize our surgical recovery,
what happens when insurance companies utilize those exact same predictive models to evaluate our risk before we even step foot in a hospital?