A sleep physician reviews some alternatives for dental practice models.

Oral appliance therapy (OAT*) for the treatment of obstructive sleep apnea (OSA) can provide an alternative for patients unable or unwilling to use CPAP and is increasingly becoming a revenue source for dental practices. Although many educational programs for OAT are available, implementation of a successful program is challenging—including the learning curve, proper patient selection, marketing, cost of treatment, expense to patients, reimbursement, adequate follow-up, obtaining sleep testing, creating relationships with physicians, state licensing laws, and the scope of practice for dentists.


Edward D. Michaelson, MD FACP FCCP FAASM

The 2006 American Academy of Sleep Medicine (AASM) guidelines for the treatment of OSA have now been updated by a 7-member task force jointly commissioned by the AASM and the American Academy of Dental Sleep Medicine (AADSM) and the resulting 6 recommendations approved by the board of directors of both organizations (Clinical practice guidelines for treatment of obstructive sleep apnea and snoring with oral appliance therapy: an update for 2015. J Clin Sleep Med. 2015;11(7):773-827). These guidelines are discussed by two of the dentists from the joint AASM-AADSM task force in the July/August 2015 issue of Sleep Review (SR).

It is not the intent or the scope of this article to discuss the indications, selection criteria, or the relative benefits of CPAP and OAT except to acknowledge that there is a place for both (sometimes combined) in the management of patients with OSA. The selection of CPAP or OAT depends on many variables, some of which include identification of patients likely to benefit from OA, intolerance to PAP therapy, availability and skill of providers, cost and reimbursement issues, patient expense, occupational requirements (at least at present), and of course adequate follow-up and outcome evaluation.

One Novel Program

The October 2015 issue of SR published an article discussing the benefits of a collaborative relationship between a sleep disorders program (in this case a hospital-based sleep lab) and a dentist specializing in OAT for the treatment of OSA. These benefits included maximization of therapeutic success and improved business volume. The collaborative program was accomplished via a time-sharing arrangement, allowing the dentist specializing in sleep to fit the OA on-site at the sleep center followed by an attended polysomnogram (PSG) during which the OA was advanced by sleep technicians trained by the dentist—in effect an attended OA titration instead of a CPAP titration. This procedure allows for optimization of OA effectiveness while minimizing discomfort. Similar models are being used by others.

While this appears to be a great model for patient care, there are some limitations to its widespread implementation. In addition to the challenges listed in the first paragraph of this article, some other limitations are prohibitions on the sharing of space with other Medicare providers, Stark laws, anti-kickback statutes, and anti-markup rules related to marketing and cross-referrals—and for which providers should seek legal advice.

Overnight Pulse Oximetry: An Adjunct to Help to Manage Oral Appliance Therapy?

Although not widely used or specifically endorsed by any particular organization or professional society, some dentists are and have been using overnight pulse oximetry (ONPO) to help evaluate the effectiveness of OAT, and many may work together with sleep specialists and/or sleep center programs.

There are however controversies, mostly relating to which aspects of the use of ONPO may or may not be within the scope of dental practice. Nor do the AASM or AADSM specifically support the use of ONPO as a mode of monitoring OAT effectiveness (see www.sleepreviewmag.com/aasm-response-using-overnight-pulse-oximetry-manage-oral-appliance-therapy). Also there are legal, administrative, and ethical issues related to ownership or delivery of the device, analysis of the data, interpretation of the ONPO, and how the results would apply to or be used for the management of OAT. Some dentists may own ONPO devices and (likely fewer) own and/or utilize HST devices or use HST services. Similar issues apply for HSTs and are addressed in more detail throughout this article.

2 Questions—Some Answers

  • Is ONPO a reasonable tool to help manage OAT?
  • Who can, may, or should carry out which aspects of ONPO (and HST) as an adjunct to OAT management?

Let’s consider some possible answers.

My view is that the answer to the first question is yes, and the next section explains the reasoning and builds the case. The second question is more difficult as the issues are more administrative, regulatory, and ethical and are evolving; nevertheless I hope to provide some reasonable perspective.

Is ONPO a Reasonable Tool to Help Manage OAT?

Let me state at the outset that the diagnosis and management of OSA via OAT should be a cooperative effort between a dentist and a qualified physician and/or sleep specialist or sleep center.

There are many similarities between the use of periodic ONPO during APAP therapy and its potential use in OAT. First consider a common-sense approach to the use of ONPO in the management of patients started on APAP based on the results of a HST (ie, the HST —> APAP model). Many payors have accepted that APAP may be started based on HST without an attended PAP titration, and there are data showing no difference in outcomes from the traditional approach (J Clin Sleep Med. 2014;10(12)). The HST —> APAP model should apply to individuals at high risk that meet accepted inclusion criteria for a HST, when that HST documents straightforward uncomplicated OSA. Even though APAP data downloads provide estimates of usage, leak, and occurrence of respiratory events, APAP units typically do not measure oxygen saturation (SpO2). So it makes sense that if APAP is started based on HST without an attended titration, then ONPO should be done during APAP to ensure correction of SpO2. If ONPO indicates that SpO2 is not corrected, then adjusting APAP settings and/or other testing (eg, PAP titration and/or pulmonary function and/or cardiac evaluation) should be considered. Even if respiratory events are minimized or eliminated, a persistent low SpO2 could indicate previously unrecognized lung disease and/or mild congestive heart failure (CHF). Also some have questioned whether AHI is in fact the only measure that determines the severity of OSA.

More detail on the algorithm/decision tree for analyzing and acting on HST results and the use of ONPO as an adjunct to monitor effectiveness of APAP is available in several of my previous SR articles (found in the November 2014, January 2015, and June 2015 issues). Furthermore periodic follow-up ONPO could provide information about effectiveness of PAP therapy on subsequent nights due to night-to-night variability in sleep habits, supine and/or REM sleep, medication, alcohol, caffeine, smoking, and environment that did not occur during an original single-night HST or PSG. Also, it would be unusual to have significant respiratory events during PAP if SpO2 remains normal. Furthermore subjective symptoms are not always reliable and it is always beneficial to complement outcome evaluation with objective measures.

The potential use of periodic ONPO in OAT is analogous to its use with APAP and should provide a reasonable objective estimate of OA effectiveness during the OA advancement period. An OA titration to SpO2 correction could serve as a marker for optimal advancement of the OA and indicate a time for follow-up HST or PSG. This could help prevent the OA from being advanced more than required and minimize potential side effects of the OA, which should improve OA compliance. The flexible scheduling of periodic ONPO would also allow time for a patient to become accustomed to advancement of the OA prior to the next ONPO. A maximally advanced OA with persistently low SpO2 could be an indication for an attended PAP titration.

In the next section I propose a model for using HST and ONPO to help manage OAT, and later in this article I have addressed some additional caveats, legal, and administrative issues that are more applicable to the second question.

A Proposed Dental-Medical Solution

Many dentists who currently or intend to provide OAT do not have the luxury of working closely and directly withinhospital-based or stand-alone (eg, IDTF) sleep testing facility and may not find it easy to address the necessary legal, administrative, and business issues. Nor for various reasons will some patients be inclined to have facility-based attended OA titrations. In this section I have outlined a simple and inexpensive solution using HST and ONPO.

In the November 2014, January 2015, and June 2015 issues of SR I addressed various aspects of the HST —> APAP model and suggested a decision tree for analyzing results of the HST and criteria for starting APAP. One of the main tenets of the algorithm is to obtain a follow-up ONPO recording during APAP to ensure that APAP is correcting SpO2 and if not adjusting the APAP settings, repeating ONPO measurements, and/or considering PAP titration as required.

The HST —> APAP model shares many similarities that can be applied to OAT. A modification of the HST —> APAP algorithm as it applies to OAT is described in the next section and the sidebar. It is intended that this algorithm be flexible and may be adapted to the needs of OAT providers, according to the particular needs of the associated physicians, business infrastructure, and patient populations. The terms high, low, various degrees of severity, and the particular numerical values shown are examples only and should be determined by the physician and dentist based on their overall clinical knowledge of the patient.

HST —> OAT Decision Tree

Prior to HST it is important to screen and identify individuals at high risk for OSA. This may be done by a dentist or physician using subjective questionnaires such as STOP-BANG, Berlin, Epworth Sleepiness Scale (ESS), or physical findings like obesity, neck circumference, and cranio-facial abnormalities (some providers may additionally use ONPO if there is a pattern of periodic desaturations typical of OSA). It is also important to exclude patients with coexisting sleep disorders and severe COPD or CHF.

With the possible exception of very mild OSA and keeping in mind that “AHI” is generally underestimated by HST, it is recommended that patients be offered and encouraged to accept PAP therapy and that OAT be considered if the patient cannot or will not tolerate PAP. With regard to the suggested decision tree, the terms very high or low as applied to AHI and/or SpO2, frequent central apneas, and the arbitrary suggested values for SpO2 and the degree of OSA severity that they represent are intended to be determined by the patient’s physician.

Again some caveats and legal and administrative issues are addressed below.


Click on this sidebar to view decision tree text at a legible size.

It is understood that prospective OAT patients will arrive under different circumstances: Some will have already had HST, PSG, or CPAP titrations and others may be PAP-intolerant or have dismissed PAP therapy without a trial; the patient may “enter” or “exit” the decision tree at any appropriate point.

Patients who have had a HST to qualify for OAT may still need to be referred for PSG or titration prior to or after initiation of OAT. Most of these situations are incorporated into the algorithm but other indications for PSG or PAP titration may include: suspected sleep disorders other than OSA, OAT is started based on results of the HST and symptoms persist after a reasonable trial of OAT, if the patient is not able to tolerate OAT, or otherwise at the discretion of the physician or dentist.

Dentists and physicians should also be aware that some patients are having side effects of the OA or continue to have sleep-related symptoms while on OAT. This is particularly important with OAT as, although some newer OA can monitor usage, data downloads indicating respiratory events are not available. In this regard, periodic ONPO can provide an objective monitoring tool.

Who Can, May, or Should Carry Out Which Aspects of ONPO as an Adjunct to OAT Management?

The questions here relate to ownership and/or delivery of the device, scoring of the raw data, data interpretation, reporting the results and to whom, determining the action based on test results, billing for the service, accepting liability, and determination of scope of practice. Dentists who currently or intend to utilize ONPO (and/or HST) as tools to facilitate management of OAT should address these issues. In particular, dentists should review the detailed requirements listed in the local carrier determination (LCD) of their regional Medicare contractor.

Regardless of specific state regulations, I recommend that dentists do not order HSTs as they will be legally required to deal with the results, whether or not OAT is indicated. Also, some payors will not reimburse for therapy or the HST if some or all aspects of the sleep test are not performed in accordance with their requirements. In particular, Medicare requires (prior to the HST) a physician’s order, a documented face-to-face patient-physician encounter in the medical record (including an ESS, as well as specific symptoms and physical findings) and that the HST be interpreted by a board-certified sleep physician. Medicare will not pay for a CPAP machine unless these HST requirements are met and a follow-up face-to-face evaluation with a physician documenting that the HST results have been reviewed with the patient on or prior to the date of the CPAP prescription. The same rules apply to coverage for OAT—as Medicare considers an OA to be durable medical equipment (DME). Medicare also specifies that only a board-certified sleep physician and an enrolled IDTF (if that physician reassigns benefits to the IDTF) can perform and bill globally for a HST. Furthermore the “AADSM Treatment Protocol: Oral Appliance Therapy for Sleep Disordered Breathing: An Update for 2013” indicates that medical assessment must be made by a physician before OAT is initiated and that following diagnosis, the dentist may provide OAT as appropriate with a prescription provided by a physician who has documented a face-to-face patient evaluation. The protocol states that the treatment of primary snoring does not require a physician’s prescription (www.aadsm.org/treatmentprotocol.aspx).

Although dentists could purchase a HST device to use as they deem indicated, they would need to accept liability regarding billing, scope of practice issues, and potential medical malpractice. Interestingly I have discussed this issue informally with both the FDA and the American Chiropractic Association (ACA). The FDA does not appear to have specific requirements for HST or ONPO prescriptions and suggested this is a state licensure issue. The ACA indicated that the scope of practice for a HST order by chiropractors varies from state to state. However according to Medicare no aspect of a HST, including but not limited to delivery and/or pickup of the device, may be performed by a DME supplier. The dentist (a DME supplier) therefore cannot provide HST (in any capacity) to a Medicare patient. Medicare does not specifically address whether dentists can provide ONPO.

The position of the AADSM is that all dentists should follow state licensing laws and is not aware of any state that permits a dentist to order or interpret a diagnostic sleep test. The AADSM December 7, 2012 “Policy Statement on the Diagnosis and Treatment of Obstructive Sleep Apnea” indicates that dental licensing laws and practice acts do not include provisions for the diagnosis of medical diseases and disorders or the treatment of non-dental (italics mine) diseases by dentists without a prescription from a physician.

The issues concerning ONPO are somewhat less clear. At least for qualification of Medicare payment for supplementary oxygen or CPAP-APAP, the pulse oximeter may be owned by and delivered to the patient by the DME provider. However in these cases, a physician’s prescription is required for ONPO, and the data must be uploaded to and analyzed and reported by a qualified physician or enrolled diagnostic testing facility. As ONPO is not used as a qualifier for OAT and while a physician’s prescription is required for ONPO, there do not appear to be specific guidelines regarding a dentist’s ownership and delivery of the device or billing and reimbursement. Whether a dentist could use ONPO informally to monitor the effectiveness of OAT may depend on scope of practice as regulated by the state or recommended by professional societies.


In addition to those listed in the first paragraph of this article, additional challenges include: documentation of medical necessity and face-to-face visits with physicians, decreasing reimbursement, and compliance/adherence to OAT. Another major roadblock to the screening, testing, treatment, and management of OSA not limited to OAT is provider apathy. While most non-sleep physicians and dentists recognize the importance of diagnosis and treatment of OSA many do not routinely screen and may not order testing for OSA even when indicated. This may be due to busy practices, little formal training in sleep medicine, lack of credentials to perform or interpret sleep tests, low reimbursement, onerous record keeping, or being uncomfortable prescribing, managing, monitoring, and documenting adherence to CPAP or OAT, and many physicians are not familiar with OAT as an alternative to CPAP.

In order to facilitate education, relationships, communication and cross-referrals, dentists should consider spearheading the development of provider communities consisting of sleep and non-sleep physicians, sleep testing facilities, HST providers, home medical equipment companies, and patient support groups. This will improve patient care as well as business for all stakeholders by bringing together education, screening, testing, treatment, monitoring, outcome evaluation, and follow-up for OSA.

Edward D. Michaelson, MD FACP FCCP FAASM, is based in Ft Lauderdale, Fla, and a member of Sleep Review’s Editorial Advisory Board.

*In the context of this article, OA will be used for oral appliance, mandibular advancement device (MAD), and mandibular advancement splint (MAS). APAP will mean auto-adjusting CPAP and PAP refers to both APAP and CPAP.