CPAP Compliance Calls the Device Data Should Be Driving

CPAP compliance is not a habit problem the practice is supposed to nag about. It is a coverage requirement with a clock on it, and the clock starts the day the patient takes the device home. Miss the window and the payer stops covering the equipment, the patient stops using it, and the practice loses both the follow-up visit and the patient.

Medicare and most commercial plans condition continued coverage of a positive airway pressure device on documented use during an initial trial period, confirmed at a follow-up visit inside a defined window. The practice does not control whether the patient uses the device. It does control whether anyone notices in time to help.

The usual process is a call list worked when somebody has an afternoon. By the time a coordinator gets to a patient who has been using the device two hours a night, three nights a week, half the window is gone and the conversation has become a rescue rather than a nudge.

Meanwhile the device is reporting nightly. The data that says exactly who is drifting, and how many days remain before the documentation window closes, exists days or weeks before anyone acts on it. That gap between signal and outreach is the whole opportunity.

The clock, not the reminder, is the thing being managed

The design question is not how often to call patients on therapy. It is which patient is at risk of falling outside a coverage requirement, and how many days are left to fix it.

That reframing changes who gets contacted. A patient using the device consistently needs a follow-up appointment booked and nothing else. A patient with a partial-use pattern early in the window needs contact now, while there is still time for the pattern to change. A patient with no recorded use at all needs a different conversation entirely, and often it belongs with clinical staff rather than a coordinator.

It also changes what a successful call looks like. Not a promise to try harder, but a booked follow-up visit inside the window, an equipment issue routed to whoever can fix it, or a documented reason the patient has stopped.

Thresholds differ between Medicare and commercial plans, and the practice’s own re-evaluation cadence sits on top of that. Those rules belong in the configuration, per payer, rather than in a coordinator’s memory of what the rule was last year.

Device data decides the queue

Working an alphabetical list is what makes this program expensive. Working the exceptions is what makes it viable.

The useful segmentation comes straight off the adherence data. Consistent users, patients trending down, patients with a pattern that suggests an equipment problem rather than a motivation problem, and patients showing nothing at all. Each of those gets a different outreach at a different urgency, and three of them are administrative work that never needs to reach clinical staff.

Days remaining in the documentation window is the second axis and it drives priority. A patient at day 20 of a 90-day window with light use is a coaching call. The same patient at day 70 is an urgent scheduling call, because the only remaining fix is getting the visit on the calendar before the window closes.

The practical constraint worth naming early: adherence data usually lives in a manufacturer or supplier portal rather than in athenaOne, and it lags. Any program built on it has to account for the lag rather than treating last night’s number as live, and the schedule of when data actually lands is a question to answer in week one.

What the call handles, and where it stops

The line here is clean, and keeping it clean is what makes the program defensible.

Inside scope: confirming the patient still has the device. Capturing in her own words what is getting in the way. Booking the follow-up visit against the right appointment type. Routing a supply or equipment issue to the DME coordinator or supplier. Checking whether the patient is due for resupply, and confirming the address and insurance on file so a shipment does not bounce.

Outside scope, always: anything about pressure settings, mask type selection, or whether the therapy is working. Those go to clinical staff with the patient’s own description attached and nothing added to it.

The most common real answer to what is getting in the way is mundane and logistical. The mask leaks, the humidifier is loud, the supplies ran out and nobody knew how to order more, the patient moved and the machine is still in a box. Each of those is an administrative fix, and each of them is why an adherence number is low. Collecting them at scale is the thing a practice has never had the hours to do.

The complications that show up in sleep programs specifically

Four recurring ones.

The supplier is a separate company. In many practices the DME supplier is not the practice, so the equipment fix and the visit live in two organizations with two records. The workable pattern is for the practice to own the outreach and the booking, capture the equipment issue with enough detail to be actionable, and hand it across with a tracked task rather than a verbal referral.

The patient who has already stopped and does not want to say so. This is common and the call has to make it easy to say. A documented reason for discontinuation is a better outcome than a fifth unanswered attempt, and it keeps the practice from carrying dead entries on a compliance list for months.

Contact reachability. Sleep patients skew toward working adults who do not answer daytime calls, which makes evening contact and a text-first pattern more effective than the standard business-hours call cadence.

Resupply timing. Eligibility for replacement supplies runs on its own schedule, and a compliance call that also checks resupply eligibility converts one contact into two outcomes. That is the version of this program that pays for itself rather than sitting on a coordinator’s task list.

What to report

The headline number is coverage-window completion: the share of patients who had their follow-up visit documented inside the payer window, month over month. That is the number the program exists to move.

Underneath it, contact-before-day-N. If most outreach happens in the last two weeks of the window, the program is running late even when the completion number looks acceptable, and that will show up as a bad month the first time volume rises.

Then the disposition mix: how many calls produced a booked visit, an equipment task, a resupply order, or a documented discontinuation. A program producing mostly conversations and few of those four is not doing anything the practice can bank.

And the honest one: how many patients on the compliance list have been contacted three or more times with no response. That group is not an outreach problem, it is a list-hygiene problem, and cleaning it up is what keeps the real work visible.

Key Takeaways

Treat compliance as a coverage window with a deadline, not a reminder campaign. The unit of work is days remaining, not days since the last call.
Segment off the adherence data: consistent, trending down, likely equipment issue, no use recorded. Each gets a different outreach and urgency.
Configure payer thresholds and the practice’s re-evaluation cadence per plan rather than relying on staff memory.
Expect device data to sit in a supplier or manufacturer portal and to lag. Design around the lag rather than treating it as live.
Keep the call to logistics: confirm the device, capture the obstacle in the patient’s words, book the visit, route the equipment issue, check resupply eligibility.
Send anything about pressure settings, mask selection, or whether therapy is working to clinical staff, with the patient’s description passed through unchanged.
Make it easy for a patient to say she has stopped. A documented discontinuation beats a fifth unanswered call.
Report coverage-window completion, contact-before-day-N, and the disposition mix rather than call volume.

A sleep program already has the signal it needs to protect both the patient’s coverage and its own follow-up revenue. What it usually lacks is the hours to act on that signal in the part of the window where acting still helps. An AI team can work the adherence exceptions every day, book the follow-up inside athenaOne against the right appointment type, route the mask and supply problems to the people who fix them, and hand the clinical questions to clinical staff with nothing added.

Sources