What Medicare Actually Pays For

A blood pressure reading leaves a cuff in a bedroom at seven in the morning. It arrives somewhere. A screen in an office two towns away adds it to a row of numbers nobody has looked at yet.

Whether anybody is paid to look depends on how many mornings this has happened.

Nobody puts that in a brochure, and it decides more than the device does.

What is remote patient monitoring in Medicare’s terms?#

Billing codes, and it is worth starting there rather than with the hardware. The government’s own description is narrow and physical: remote monitoring of parameters such as weight, blood pressure and pulse oximetry, to monitor a patient’s condition and inform their management.

Notice what is being described. Readings, and somebody managing a patient with them.

Payment is split in two, and the split explains most of the confusion families run into.

One part covers supplying the device and transmitting data over a period. The other covers the professional time spent reviewing what came in and managing the patient.

The device is the cheap part. The paid part is a person reading what the device sent.

Say that again slowly. Not the cuff. Not the app. The reading of it.

Once you see the split, a lot of marketing reads differently. A company offering a free monitor is not being generous with hardware. Hardware was never the expensive item.

There is a second family of codes alongside it, for monitoring how well somebody is sticking to a treatment at home rather than monitoring their body. Three kinds exist: respiratory, musculoskeletal, and cognitive behavioural therapy.

They work the same way. Supply plus attention.

None of this is how monitoring is sold. A brochure leads with the device, because the device is the thing you can photograph, and because the professional time is harder to promise.

Codes lead with the days and the minutes. A payer counts what it can count.

Start with the codes, not the catalogue.

Who is allowed to bill for it?#

Practices bill for it, not device companies, and that is the first thing to establish about any offer you receive. These are physician fee schedule services, billed by clinicians for professional work. A vendor can supply the equipment and the software. Somebody clinical has to be responsible for what arrives.

Ask who that person is, by name and role.

That answer tells you something no brochure will. A practice that has built monitoring properly can say who reviews the data, on what days, and what happens outside those hours.

An arrangement where nobody can answer that is selling you a device with a subscription attached.

Support staff can do a great deal of the work under a clinician’s direction, which is ordinary in medicine and not a warning sign on its own. What matters is whether the responsibility is real and whether it is named.

Payment rules live in two federal books. The Federal Register (FR) carries new rules as they are made. The Code of Federal Regulations (CFR) carries them once they settle.

And the rulebook has a word for these people. It calls them auxiliary personnel.

Its definition sits at 42 CFR 410.26, which sets the conditions for services furnished incident to a physician’s professional services.

Read the supervision definition beside it and the arrangement becomes clear. General supervision means the service is furnished under the physician’s overall direction and control, and the physician’s presence is not required during the performance of the service.

Not required to be there. Required to be responsible.

So that is the design, and it is a reasonable one. It also means the responsibility can be diffuse in practice unless somebody names it, which is why the question is worth asking out loud.

Picture the chain. A cuff on an arm. A phone in a pocket relaying the reading. A dashboard on a desk in an office. A nurse who opens that dashboard on Tuesday morning with a coffee going cold beside it.

Every link in that chain exists. Only the last one is being paid for.

There is a second reason to ask who. Monitoring arrangements can be sold to a practice by a vendor, packaged with the hardware and the software, and the clinical review is the part the practice has to staff itself. That is where these arrangements succeed or quietly fail, and it is not visible from outside.

Here is why the distinction is not academic. A device generates readings whether or not anybody watches them. A reading nobody reads is worse than no reading at all, because everyone assumes it was seen.

Think of a smoke alarm nobody has wired to anything. It still makes a noise. The noise was never the point.

So the question is never whether the technology works. It is whose job it is when the number is bad.

Machines can measure. They cannot decide. Somebody has to.

Ask for the name.

What has to be true before it is paid for?#

Three conditions, and all of them have to hold. This is where a plausible-sounding arrangement quietly fails, because families assume the test is medical need and the actual test is a combination of days, minutes and a clinical relationship.

First, the device has to transmit data on enough days within a thirty day window. The codes are written around counts of days, and different codes describe different bands within that window.

Second, somebody has to spend professional time on it, and that time is counted in blocks of minutes per month.

Third, the patient has to be under the care of the practice doing the managing. Monitoring is a management service and it presumes somebody is managing.

Miss any one and the arrangement may still be useful and may still not be payable.

The days condition is the one that surprises people. It means a device sitting unused in a drawer for half a month is not merely unhelpful. It changes what can be billed, which changes whether a practice can sustain the service.

Think of a gym membership that only counts if you go a minimum number of times. The equipment is available all month. The arrangement is built around attendance.

None of that is a criticism of the design. A payer needs some measure that the service actually happened, and days of data is a measurable one.

It does mean a family should know the threshold exists.

Nobody mentions it at the start. The device arrives. The instructions cover charging and pairing. The number of days a month it has to be used is not on the box.

Find out how many days a month it needs.

What does the patient actually pay?#

Something recurring, and that is the honest headline rather than a number. Monitoring is billed as a monthly pattern, a device supply code for the period plus time based management codes, so it behaves on a statement like a subscription rather than like a purchase.

Four things follow, and they are the ones worth planning around.

• It appears every month, not once.
• It rises with the professional time spent.
• A quiet month and a bad month cost differently.
• A gap in transmission changes what can be billed.

Number two is counterintuitive. A month in which your relative was unwell, and somebody spent real time on their readings, can cost more than a quiet month.

Look at where that lands. The statement arriving in an envelope in March is a record of how much attention February needed, and nobody explains that when the monitor is set up.

It is the system working as designed rather than a billing error. The payment is for attention, and more attention was given.

What this page will not do is quote you a figure. What a patient owes depends on their coverage, on what else they have, and on arrangements this article cannot see from here.

The useful move is to ask the practice to show you a typical month before you start, and to ask what an expensive month looks like.

Any practice running this properly has both numbers.

Get them to show you a statement.

How is this different from a telehealth visit?#

By classification, and the difference is stated outright rather than left to be inferred. Monitoring is not a telehealth service in Medicare’s terms, which sounds like a technicality and is in fact the single most consequential fact on this page.

This one is published in the Federal Register (FR), the daily record of federal rulemaking.

Its 2026 fee schedule rule answers a comment on exactly this point.

Responding, it clarifies that these services are inherently non face to face. They do not meet the definitions of section 1834(m) of the Act, and you can read that sentence yourself. They fall outside the scope of the definition of a Medicare telehealth service.

And then the sentence you want: they are not subject to section 1834(m).

Read that last clause again. Then consider what lives in section 1834(m). Every telehealth flexibility with an expiry date on it, and there are more of them than you would guess.

One sentence in a payment rule decides which clock a service runs on.

So a lapse in the telehealth extensions would not, by itself, end remote monitoring. Those dates are written into a section this service does not sit under, which is a narrower point than it sounds and a load bearing one.

That is not the same as saying monitoring is safe. Payment rules are rewritten every year, so monitoring changes by annual rulemaking rather than by a cliff.

Annual rulemaking is its own skill to follow, and the habit for it is set out in reading the federal rulebooks.

A family should know which of the two risks they are exposed to, because the rhythms are not the same. The sorting that produces these two regimes is the subject of which rulebook covers the device, and this is that sorting doing real work.

Different section. Different clock. Different risk entirely.

Worth holding onto, because it is the one fact here that changes a decision rather than explaining one.

What should a family ask about it?#

Four questions, and not one of them is about the equipment. The device is the part everyone can evaluate and the part that matters least, and every question below is aimed at the part that is actually being paid for.

First, who reads the data, and how often. A name and a schedule, not a department.

Second, what happens when a reading looks wrong at two in the morning. Listen for a process rather than a reassurance.

Third, what appears on a statement each month, in a normal month and in a bad one.

Fourth, what happens if your relative stops using the device for a stretch, because the codes turn on days and a gap changes what is billable.

Four questions. Ten minutes. Write the answers down.

Write the answers down with the date you were given them, and keep them where the next person can find them. Monitoring arrangements outlive the conversation that set them up, and the person who set it up is not always the person managing it a year later.

One more thing is worth saying, because the technology is genuinely useful and this article has spent a long time on its rules. A monitor that catches a slow drift in weight before anybody notices it in a room is doing something a weekly visit cannot do. That is real, and it is why this is worth getting right rather than avoiding.

None of it works without somebody on the other end. The gap between what a machine can measure and what a person has to decide is the same gap that runs through abundance and unrest arriving together, and it does not close because the measuring got cheaper.

A son checking on a parent two states away can set all of this up in an afternoon. Knowing which four questions to ask is what makes the afternoon enough.

Not one of them is technical. All of them are about people.

Ask about the reading, not the device.