Gaps in Sleep Medical Knowledge: Part VII (Confusing Compliance with Outcomes)

The big push for compliance data (the hours a patient uses the PAP device) has mostly come from the insurance companies who sought to create a metric to determine which OSA/UARS patients should be eligible for coverage. Some years ago as sleep apnea research was blossoming, the question arose as to how many hours of sleep with a PAP machine would be needed to gain some degree of tangible benefit. Unfortunately, the question was answered by “begging the question,” that is, making the premise about the number of hours of sleep instead of recognizing the correct metric would be closer to something like the number of high quality hours of sleep.

Notable sleep experts in the field communicated with Medicare officials or other insurance carrier administrators who had been asking this question, because the latter were confused about their ability to determine what constituted a legitimate user of PAP therapy in contrast to the obvious non-user who might be letting his or her PAP device rest comfortably at night in the confines of a bedroom closet. Understandably, the insurance carriers would not be expected to reimburse the durable medical equipment (DMEs) companies if a patient never used the device. So, the question was reasonable in asking if there were a minimum number of hours to use it. And, the answer they received was on the order of 4 hours for at least 5 nights per week.

If you multiply 4 hours X 5 nights, the result is 20 hours per week; but, this amount was never acknowledged as an acceptable scientific metric, and there was not any formal attempt to prove its utility. Instead, the rule was first adopted by Medicare, and then several leading insurance carriers followed suit over the years. As discussed in prior posts, the inherent flaw in this metric is the obvious one where someone uses a device 3.50 hours per night for 7 nights per week, totaling 24.50 hours, but does not qualify for coverage under the Medicare guidelines. Unquestionably, many patients use a PAP device for less than 4 hours per night and gain substantial benefits, presumably the types of benefits (decreased sleepiness and insomnia) that would ultimately provide cost-savings to the insurance company in comparison to what they lose from costs related to untreated OSA/UARS. Yet, this logic has never been officially embraced by Medicare or any other insurance carrier, which means there are many patients who fit what we have called “sub-threshold adherence” in our recent paper in Respiratory Care, and these patients are clearly gaining benefits from their less than insurance-certifying number of hours of use.

It remains both an oddity and a gap that the AASM has not addressed this issue, but instead this institution simply appears to go along with the insurance carriers as if there is no recourse. Yet, sub-threshold adherence research has already been published prior to our work, and the AASM should be marshalling this information into a position paper that directly calls out the insurance companies for not paying attention to the scientific evidence demonstrating a lesser number of hours still provides benefit. Moreover, the AASM should be calling for research in this area to better guide insurance carriers as well as the field of sleep medicine, which in time would lead to greater precision in the way we measure how patients respond to PAP devices. 

These problems, regrettably, worsen when a so-called standard with no scientific rationale becomes entrenched in a field of medicine and induces non-medical professionals (e.g. insurance administrators, government regulators and staff bureaucrats) to routinely misinterpret these rules. Nowhere is this flawed strategy more evident than in OSA/UARS treatment where policies conflate compliance with outcomes. And, this point also refers back to the mistaken belief that the number of hours of sleep is equivalent to the number of hours of high quality sleep, which we will return to soon enough.

By presuming a patient gains benefit by using the device 4 or even 5, 6, 7, or 8 hours per night is actually borderline irrational, given the fact there are innumerable patients doing just that yet still complain of being tired or sleepy during the day. In other words, despite their use of the device, they suffer from persistent nonrestorative sleep, and this poor quality of sleep is pronounced enough that no additional hours of sleep will decrease such patients’ complaints of tiredness or sleepiness during the day.

As you could imagine, if you were an administrator or bureaucrat with little knowledge of sleep medicine and knew only how to read a chart that lists the numbers of hours slept, it might easily escape your attention to consider other clinical aspects in the patient’s medical records you were reviewing. This scenario is repeatedly played out as discussed previously in dealing with prior authorizations, where the entire process is slanted toward the compliance numbers.

When talking with administrators we must explain a patient’s need for additional treatment steps and care, such as retitrations or attempts at a different type of PAP device, but during these discussions it is no rare occurrence that their immediate response is something like, "but the patient is already compliant with PAP?”

To which, we might reply, “So?” And, then I would pause to see if the individual on the other end of the phone line recognized the limited perspective he or she was bringing to the dialogue. Next, further explanation on the concept of outcomes must be presented. Sometimes, these concepts must be developed through a Socratic method, because individual administrators may be so indoctrinated to look only at compliance hours and so programmed to find ways to reject the requests of sleep specialists, they need to be challenged in provocative ways to get across the point. Even after these bureaucrats comprehend that the patient is suffering from persistent daytime fatigue and sleepiness or insomnia, they try to turn the conversation back to the number of hours of sleep as in, “well, the patient is already compliant, so maybe they just need to sleep more…what about a sleeping pill?” Soon, they bring up a pre-programmed list of alternative explanations, beyond any sleep-related issues like mask leak (which is a favorite because they think solving mask leak solves all sleep apnea patients’ problems and thus retitrations would never be indicated), such as depression, diabetes or some other co-occurring medical illness or another favorite, stress, as in maybe the patient is getting as good a response as possible for now and should continue on for several months and check back. These sample responses are highly representative of the ways in which many of these administrators think, and among the insurance companies' own physician directors who are also trained to respond to our appeals, this style of thinking may only be slightly more advanced, especially when the physician director is board certified in another specialty unrelated to sleep.

I am of the opinion all these considerations could be posed as reasonable questions and concerns, but I also believe the context does not fit actual patient scenarios the majority of the time, because there are two main reasons any sleep apnea patient is not doing well despite attaining some arbitrary level of hours of sleep: 1) the patient needs more hours of sleep with the PAP device beyond the compliance levels; or, 2) the sleep obtained remains nonrestorative. Both issues require further investigation to attempt to detect previously unrecognized sleep factors or to re-examine known problems such as leg jerks, excessive brain arousal activity, residual breathing events (notably flow limitation), and specific sleep position effects that may worsen breathing or compromise the integrity of the mask seal. This short list of factors is more likely to prove pivotal in preventing a “compliant” patient from achieving good to optimal outcomes. 

From this vantage point, it should be obvious there is sleep and then there is really good quality sleep. Astonishingly, there is no serious discussion of this issue either within insurance circles or to a surprisingly large extent within the field of sleep medicine itself. As more and more sleep facilities feel the pressure from insurance carriers and government regulators, the sleep staff often does not stop to think about these conundrums, or they have been intimidated to avoid investigating the rigorous physiological relationships between the various levels of sleep quality and a patient’s outcomes. I am certain some sleep patients reading this post are astonished by this pronouncement, and I would hope many patients have had much better encounters with their sleep staff than I am describing here. But, due to our extensive second opinion research we know anecdotally among a few thousand patients during the past decade that many of the conversations between sleep professionals and patients did not cover this ground in ways to have helped the doctor fight back against the arbitrary standards of the insurance industry. And, this lack of comprehension and curiosity about the highly relevant distinctions between sleep and really good quality sleep is rife within the field, because too many people who should know better are wasting precious resources obsessing about compliance instead of measuring the actual quality of the slumber in their patients and how the quality is affecting outcomes.

One the most remarkable aspects of this problem is the failure of the AASM to recognize the inherent value of retitration protocols. Through retitrations, it is abundantly clear that a dose-response relationship emerges between pressurized air settings and changes in the breathing events. Instead of capitalizing on this essential fact, there is almost no research targeting this process, because so many sleep experts are wedded to the belief CPAP therapy works for everyone, and CPAP pressure settings never need to be adjusted once a patient starts treatment following say a first titration in the lab. Making matters worse, with the advent of home sleep testing many sleep experts and a fair number of sleep professionals imagine they can make relevant pressure adjustments using data downloads, yet this approach is highly unreliable if the goal is to follow the AASM mandate to eliminate the flow limitation breathing events. Of course, my premise is faulty here, because so many clinical sleep professionals completely ignore the concept of flow limitation and only base results on changes in the apnea-hypopnea index (AHI).

This limited scientific perspective adopted by many sleep professionals will only worsen as older sleep docs retire and younger, new sleep physicians are over-exposed to the home testing model and thereby receive much less training or experience in the use of titration methodology, even for just a first titration. These newer recruits to the field will be tempted to believe their patients are doing well solely based on the numbers on the data download or the attainment of compliant hours, but when confronted with poor outcomes, how capably will they perform to provide top-notch clinical care?

Candidly, there are no easy or simple answers to this problem caused by conflating compliance metrics with patient outcomes. One would think the AASM would be a more patient-friendly institution, but judging by their lack of attention to these matters, it seems uncertain how such an organization will attempt to change the minds of insurance carriers and government regulators, albeit their recent efforts to outline new guidelines for the use of PSG vs HST in the diagnosis of OSA reflects a small step in the right direction. However, neither the draft guidelines nor the supplementary materials make note of the importance of RERAs (the discrete respiratory event measured in UARS) or explain how these more subtle breathing events factor into the use of the PSG vs HST technology. Thus, among those who believe this problem of conflating compliance with outcomes needs more attention, we instead perceive the AASM as neglecting a major contributor to sleep breathing disorders, which will over time further compromise sleep professionals’ abilities to communicate with insurance carriers on the necessity of looking beyond compliance numbers.

This gap is a large one and overlaps considerably with our lack of understanding regarding the definition of normal sleep breathing. Without such a clarifying definition, more patients will suffer as they will be labeled compliant and receive no further attention for their less than optimal outcomes.


Read more from this series:

Gaps in Sleep Medical Knowledge: Part I (Normal Sleep)
Gaps in Sleep Medical Knowledge: Part II (Normal Breathing)
Gaps in Sleep Medical Knowledge: Part III (The Medicare Hypopnea Scoring Maze)
Gaps in Sleep Medical Knowledge: Part IV (The Unsolved Puzzle of UARS)
Gaps in Sleep Medical Knowledge: Part V (Coding Issues for Billing and Reimbursement)
Gaps in Sleep Medical Knowledge: Part VI (Prior Authorizations and Atypical Patients)
Gaps in Sleep Medical Knowledge: Part VII (Confusing Compliance with Outcomes)
Gaps in Sleep Medical Knowledge: Part VIII (Comorbid Sleep Disorders)
Gaps in Sleep Medical Knowledge: Part IX (Miscellaneous Topics)


Barry Krakow MD


Dr Krakow’s 27 years of sleep research have focused on the complex relationship between physiological and psychological sleep disorders. Dr Krakow currently operates private sleep medical center, Maimonides Sleep Arts & Sciences, Ltd., and serves as Classic SleepCare’s paid Medical Director.

Leave a comment

Comments have to be approved before showing up.