CPAP vs Bilevel: When Medicare Covers E0470 for Sleep Apnea

Updated November 2026. Written by Matteo Grassi, clinical psychologist and cofounder of HANA Health.

A bilevel PAP device without a backup rate (HCPCS E0470) is a machine that delivers a higher pressure when the patient breathes in and a lower one when they breathe out. For obstructive sleep apnea, Medicare covers it only after a CPAP (E0601) has been tried and documented as ineffective, under LCD L33718 as revised in January 2024. A bilevel with a backup rate (E0471) is never covered when OSA is the primary diagnosis.

From the clinical side, the request for "a stronger machine" is usually a request for help with a mask, a pressure or a bad first week. The coverage rules push suppliers to check those things first, and that's not a bad instinct.

Key facts

• LCD L33718 defines a PAP device for OSA as either a CPAP (E0601) or a bilevel without backup rate (E0470), and both share the same initial and continued coverage rules (CMS, LCD L33718).
• E0470 for OSA needs one extra criterion: an E0601 tried and proven ineffective in a facility titration or at home, despite proper mask fitting and pressure settings (CMS, LCD L33718).
• An E0471 billed with a primary diagnosis of OSA is denied as not reasonable and necessary (CMS, LCD L33718).
• Before the switch, the treating practitioner must document that interface fit and CPAP pressure settings were both addressed (CMS, Policy Article A52467).
• CPAP had a 12.5% improper payment rate in the 2024 reporting period, a projected $146.1 million, and 71.2% of PAP errors were insufficient documentation (CMS MLN, CPAP compliance tip).
• A 2026 meta-analysis of 21 studies and 1,462 adults found pressure support added 0.21 hours a night of use versus CPAP, which wasn't statistically significant (Messineo et al., eClinicalMedicine, 2026).

When does Medicare cover a bilevel instead of a CPAP for sleep apnea?

Medicare covers an E0470 for OSA when the patient meets every CPAP criterion and a CPAP has already failed. There's no path straight to bilevel for uncomplicated OSA.

The CPAP criteria come first. Under LCD L33718, the patient needs an in person clinical evaluation by the treating practitioner before the sleep test, a qualifying sleep test, and instruction from the supplier on how to use and care for the equipment. The sleep test qualifies when the AHI or RDI is 15 or more events an hour with at least 30 events, or between 5 and 14 events an hour with at least 10 events plus documented symptoms or comorbidities such as daytime sleepiness, hypertension or a history of stroke. Our guide to Medicare CPAP qualification and sleep test rules walks through those in detail.

Then comes criterion D. An E0601 must have been tried and proven ineffective, based on a therapeutic trial in a facility or in the home. If a supplier bills E0470 for OSA without all four criteria met, the claim is denied as not reasonable and necessary.

What counts as CPAP being "ineffective"?

Ineffective means documented failure to meet therapeutic goals on a CPAP, either during the titration part of a facility study or during home use, despite optimal therapy. The LCD spells out what optimal therapy means: proper mask selection and fitting, and appropriate pressure settings.

Policy Article A52467 turns that into two things the treating practitioner must document before the change:

1. Interface fit and comfort. A suitable mask has been properly fitted, the patient is using it without difficulty, and the same mask will be used with the E0470.
2. CPAP pressure settings. The current pressure stops the patient from tolerating therapy, and lower CPAP pressures were tried but failed to control OSA symptoms, improve sleep quality, or bring the AHI or RDI down to acceptable levels.

Notice what's missing from that list. "The patient doesn't like CPAP" isn't enough on its own. Neither is a supplier form with checkboxes. A52467 says supplier generated forms don't substitute for the practitioner's own medical record.

Does switching to bilevel restart the 90 day trial?

It depends on when the switch happens. The LCD sets out three situations, and each one has its own timeline.

Switching inside the first trial, with more than 30 days left

The original clock keeps running. The clinical re-evaluation still happens between day 31 and day 91 after the CPAP started, and objective adherence on the E0470 has to be shown before day 91. No new in person evaluation or sleep test is needed.

Switching inside the first trial, with less than 30 days left

The window stretches. The re-evaluation and the objective adherence data must both be in place before day 120 after the CPAP started.

Switching after more than three months on CPAP

A new 3 month trial begins on the E0470. The patient needs a new in person clinical evaluation but not a new sleep test. The re-evaluation then falls between day 31 and day 91 after the E0470 starts, with adherence documented during that new trial.

In every case the adherence standard is the one in our Medicare CPAP compliance guide: 4 or more hours a night on 70% of nights during a consecutive 30 day period.

Why isn't E0471 covered for obstructive sleep apnea?

A bilevel with a backup rate (E0471) can deliver breaths on a timed schedule when the patient doesn't breathe on their own. That feature treats central events and hypoventilation, not upper airway collapse. So LCD L33718 states it plainly: if the primary diagnosis is OSA, E0471 isn't reasonable and necessary, and any E0471 billed with an OSA diagnosis is denied.

Coverage for E0470 and E0471 outside of OSA sits in a different policy, the Respiratory Assist Devices LCD L33800, revised in June 2025 to align with NCD 240.9 on home ventilation for COPD. It covers these devices for four groups of conditions:

• Restrictive thoracic disorders, such as neuromuscular disease or severe chest wall abnormalities.
• Central or complex sleep apnea, documented on a facility based, attended polysomnogram.
• Hypoventilation syndrome, backed by arterial blood gas and spirometry criteria.
• Chronic respiratory failure from COPD, under NCD 240.9.

Complex sleep apnea is where the two policies meet. Under L33800 it's central apnea that persists or appears when obstructive events are treated on CPAP or E0470, with central events above 50% of the total and a central index of 5 or more an hour. The continued coverage test is different too: for groups covered for an initial three months, the RAD re-evaluation can't happen before day 61, and use must average 4 hours per 24 hour period.

Does bilevel actually improve adherence?

Not much, on average. That's the honest answer from the best current evidence.

A 2026 systematic review and meta-analysis in eClinicalMedicine pooled 21 studies with 1,462 adults with OSA. Pressure support, meaning bilevel or expiratory pressure relief, added 0.21 hours a night of use compared with CPAP, and the difference wasn't statistically significant. The authors set that against a minimal clinically important difference of 0.5 hours a night, so even the point estimate falls short.

Two details are worth knowing. In patients who were already intolerant of CPAP, bilevel added 0.68 hours a night, but the certainty of that evidence was rated very low. And the gap narrowed over time: longer follow up was linked to falling adherence on pressure support compared with CPAP. Air leak was lower on pressure support, which may matter for some patients.

So bilevel is a reasonable tool for the specific patient who can't tolerate the pressure they need. It isn't a fix for low adherence in general. Our post on what improves CPAP adherence covers the interventions with stronger trial support.

What this means for your DME

Treat a bilevel request as a documentation project, not a device swap. Start by fixing the mask and trying lower pressures on the CPAP, because the record has to show both happened anyway. Then make sure the practitioner's note, not your form, says why the CPAP failed.

Track the trial clock from the original CPAP start date. A switch in week six doesn't buy a fresh 90 days, so the adherence download on the E0470 has to land before day 91, or before day 120 if the switch came late. If a patient has been on CPAP for more than three months, schedule the new in person evaluation before you deliver the E0470.

Watch the patients who switch. The evidence says adherence on pressure support can drift down after the first few months, so the same follow up rhythm you'd use for a new CPAP patient still applies. If a patient fails the trial altogether, the rules in our failed 90 day trial guide take over.

HANA helps DMEs and sleep clinics run PAP adherence and resupply outreach between visits. You can see how that works on our sleep and PAP page.

Frequently asked questions

Does a patient need a new sleep test to switch from CPAP to bilevel?

No. Under LCD L33718, switching to E0470 after a failed CPAP trial doesn't require a new sleep test. If the patient has used CPAP for more than three months, they do need a new in person clinical evaluation before the switch.

Can a DME bill E0470 for a patient who simply prefers bilevel?

Not with the KX modifier. Without a documented ineffective CPAP trial, the E0470 doesn't meet criterion D and will be denied as not reasonable and necessary. If the supplier expects a denial, it adds GA with a valid Advance Beneficiary Notice, or GZ without one.

Is a humidifier covered with a bilevel device?

Yes. LCD L33718 covers a non heated (E0561) or heated (E0562) humidifier when the treating practitioner orders it for use with a covered E0470 or E0601.

Can E0471 ever be covered for a patient who also has OSA?

Only when the primary diagnosis that qualifies the patient is one covered by the Respiratory Assist Devices LCD L33800, such as central or complex sleep apnea or hypoventilation syndrome. Those criteria, including facility based testing for central apnea, have to be met on their own terms.

What documentation errors do auditors find on bilevel claims?

CMS gives an example of an E0470 claim recouped because the treating practitioner's record didn't include the sleep test results. Insufficient documentation drove 71.2% of PAP improper payments in the 2024 reporting period.

Sources

1. CMS Medicare Coverage Database. LCD L33718, Positive Airway Pressure (PAP) Devices for the Treatment of Obstructive Sleep Apnea. Revision effective January 2024.
2. CMS Medicare Coverage Database. Policy Article A52467, PAP Devices for the Treatment of Obstructive Sleep Apnea. Revision effective August 2021.
3. CMS Medicare Coverage Database. LCD L33800, Respiratory Assist Devices. Revision effective June 2025.
4. CMS Medicare Learning Network. Continuous Positive Airway Pressure Devices and Accessories, provider compliance tip. 2024 reporting period data.
5. CMS Medicare Learning Network. Respiratory Assist Devices, provider compliance tip. 2024 reporting period data.
6. Messineo L, Bakker JP, Noah WH, White DP. The effect of bilevel ventilation or modified CPAP on treatment adherence in adults with obstructive sleep apnea: a systematic review and meta-analysis. eClinicalMedicine, 2026.

This article is educational and isn't billing, coding, clinical or legal advice. Check current CMS guidance and your DME MAC before you bill. If you want to talk through PAP follow up for your patients, book a call with Matteo.