Blog

CMS Is Proposing Pricing a Connected Medical Device The Same Way as a Drugstore Blood Pressure Cuff

Financial Data Shows It Won't Work

By Jon-Michial Carter, ChartSpan Founder and Chief Growth Officer, and John Jepsen, ChartSpan Executive Vice President of Finance

Buried in the CY 2027 Medicare Physician Fee Schedule proposed rule is a technical proposal that sounds like accounting housekeeping and is actually a body blow to every small and rural practice running remote patient monitoring. CMS wants to set the payment for supplying and operating a connected RPM device by borrowing the cost of a completely different service: a patient reading numbers off a blood pressure cuff they bought at the pharmacy.

The two things are not the same. The math and real world data proves it. And if this crosswalk is finalized, the practices least able to absorb the loss are the ones who will be forced to walk away from RPM entirely.

What CMS actually proposed

In the proposed rule (CMS-1848-P, published in the Federal Register on July 16, 2026), CMS proposes to value the practice expense inputs for the RPM device-supply code, CPT 99454, by crosswalking them to CPT 99474. It applies the same logic to the setup code, crosswalking CPT 99453 to CPT 99473. CMS says it is doing this because it lacks reliable invoice and pricing data on the devices actually used for remote monitoring.

Here is the problem in one sentence: 99474 is the self-measured blood pressure code.

99474 pays a clinician to review at least a dozen readings a patient collected on their own, on their own home cuff, over a month. There is no device to procure. No cellular or Bluetooth connectivity to provision. No data platform to host. No alerting engine. No fulfillment, no inventory, no returns processing, no patient tech support line, no 16-day transmission compliance tracking. The patient owns the cuff and types in the numbers.

CPT 99454 is different. To bill it, the practice has to put a connected, clinically validated device in the patient's hands and keep it transmitting data automatically for at least 16 days every month. That is a logistics, technology, and compliance operation. Pricing it off a code that assumes none of that exists does not fix CMS's data gap. It just guarantees the number is wrong, and wrong in a very specific direction.

What it actually costs a real practice

To show the size of the gap, it’s important to study a fully loaded cost estimate of what a small independent primary care practice would spend to run the non-clinical side of an RPM program itself, without the purchasing power or shared infrastructure of a national vendor.  We deliberately excluded the clinical monitoring labor, because that time is separately reimbursable under CPT 99457 and 99458. This is only the cost of getting a working device onto a patient and keeping the data flowing.

On a per-patient-per-month basis:

Cost component	PPPM	What it covers
Device hardware	$4.98	Retail device cost of roughly $60, at 1 device per patient for comorbid cohorts, spread over a 12-month device life
Connectivity and data platform	$40.50	Device connectivity, data visualization, hosting, alerting, and workflow integration
Logistics and administration	$23.39	Storage, inventory and procurement labor, kitting and shipping, returns and warranty processing, patient tech support, 16-day compliance tracking, program oversight, and allocated IT and insurance
Total non-clinical cost	$68.87

A few things worth sitting with.

First, the device line assumes retail pricing, roughly $60 per unit, not the discounted price a scaled buyer gets. That is the point. A national vendor buying tens of thousands of units negotiates a volume and direct-manufacturer discount of nearly 28 percent. An independent practice buying a handful of cuffs from Best Buy or Walgreens does not. CMS's own stated goal in this rule is to protect small practices, but small practices are exactly the buyers who pay full freight.

Second, the logistics and administration line alone, $23.39, already eats up about 45 percent of the current 2026 payment for 99454, which is roughly $52.11. That is before a single dollar of device cost or data platform.

Third, and this is the headline: the fully loaded non-clinical cost of furnishing RPM, about $71 per patient per month, already exceeds today's reimbursement of about $52. The service loses money before CMS changes it. 

The crosswalk turns a squeeze into a cliff

Now layer the proposed rule on top. CMS is not proposing to hold 99454 at $52. It is proposing to reprice it using 99474, a code that has historically paid on the order of $15 a month. So CMS would take a service that already costs a small practice about $70 to deliver and reset its payment toward a fraction of that.

No independent practice can run a program at that structural loss. Most of these costs are fixed. They do not shrink when a practice stays small. They get heavier, because there are fewer patients to spread them across. That is the cruel twist of pricing device supply off a code built for a service with almost no supply cost at all.

And it does not sit in isolation. The same proposed rule requires that RPM be furnished by the practice’s own employees rather than a contracted third-party, permitting payment only when the work is done by direct employees of the billing practice. So CMS is telling practices, in the same rule, to bring RPM in-house and to accept a payment rate that assumes bringing it in-house costs almost nothing. Those two positions cannot both be true.

Who actually loses

The victims of a bad crosswalk are not vendors. They are the solo internist in a rural county, the small primary care group serving a Medicare-heavy panel, the practices that adopted RPM because it let them watch a hypertensive or heart-failure patient between visits and catch problems early. Those practices run on thin margins. They will not subsidize a Medicare service out of pocket. They will drop RPM, and their patients, the ones with high blood pressure, diabetes, and heart failure who benefit most from monitoring, will lose it. That is the opposite of the access CMS says it wants.

Further, taxpayers lose. Medicare has spent years encouraging care that extends beyond the four walls of the physician's office and focuses on proactively managing chronic disease. Remote patient monitoring is one of the few reimbursement mechanisms specifically designed to support that model. The connected devices, cloud infrastructure, compliance tracking, and continuous flow of patient data required under CPT 99454 enable clinicians to identify worsening conditions earlier and intervene before problems escalate. If reimbursement no longer reflects the actual cost of furnishing that infrastructure, practices will reduce or eliminate RPM programs. The result is not simply fewer connected devices and less longitudinal patient data, but also fewer opportunities for early intervention, and a reimbursement policy that undermines Medicare's broader goals of improving outcomes while using healthcare resources more efficiently. 

The fix is straightforward

CMS acknowledges the real problem: it does not have good cost data on RPM devices and operations. The answer to missing data is to go get the data, through invoice submissions, a supplemental cost survey, or engagement with the practices and technology providers who actually run these programs. The answer is not to pin the price of a connected-device program to a home cuff and a manual logbook and call it settled.

The comment window on CMS-1848-P is open until September 14, 2026. Every practice, association, and technology partner that touches remote monitoring should put its real numbers on the record. Not projections. Not vendor economics. The actual, fully loaded cost of doing what CMS is asking practices to do. Because the current proposal is not priced on evidence. It is priced on an analogy that falls apart the moment you look at the invoices.

How to submit a comment

Comments are due by 5:00 p.m. ET on September 14, 2026. The fastest way to be heard is to file electronically:

1. Go to the official docket at https://www.regulations.gov/docket/CMS-2026-2377. This is the docket for the CY 2027 Physician Fee Schedule proposed rule, file code CMS-1848-P.
2. Click the green "Comment" button on the docket page.
3. Enter your comment in the text box, or upload it as an attachment (PDF or Word). Uploading a document is best if you are submitting a detailed, data-backed comment. Reference file code CMS-1848-P and the specific provision, the proposed crosswalk of CPT 99454 to CPT 99474, at the top of your comment.
4. Complete the commenter information fields, choose whether you are commenting as an individual or on behalf of an organization, and submit. You will receive a confirmation and a tracking number. Save it.

If you prefer to comment by mail, CMS accepts written submissions referencing file code CMS-1848-P at the addresses listed in the "ADDRESSES" section of the proposed rule in the Federal Register. Electronic submission is faster and gives you a confirmation record.

One practical note: CMS gives more weight to comments that include specific, sourced cost data than to general opposition. Attach your own invoices, labor figures, and operating costs if you have them.

---

Jon-Michial Carter is the Founder and Chief Growth Officer of ChartSpan. John Jepsen is Executive Vice President of Finance at ChartSpan. The cost estimate referenced here is a first-pass, fully loaded estimate built on published labor benchmarks (BLS 2026), retail device pricing, HHS OIG 2025 RPM audit data on device utilization, and internal operating assumptions. It is intended to inform ChartSpan's public comment on CMS-1848-P and is offered as a discussion framework for the broader remote monitoring community.