Now, we begin our discussion following the Part I Introduction on the concept of the “preference cascade” and how doctors’ information or knowledge may be limited in a such a way that constrains their practice of medicine to a narrowly defined conventional wisdom.
The first and most obvious question we must ask is whether or not physicians read research articles? Or, do they only listen to the pharmaceutical reps who visit their offices? Surely, with all the continuing medical education (CME) requirements physicians must abide by, these professionals are frequently exposed to learning opportunities at conferences, grand rounds, journal clubs as well as the reading of general journals or field-specific journal articles. Unfortunately, this exposure may be insufficient to change physician practices, because the content delivered may simply reinforce current beliefs in the practice of medicine. If you regularly attend heart conferences where the speakers are fully vested in the role of statins in the treatment of cardiovascular health and disease, then you will not hear discussions on the counter-arguments (available for decades) as to how statins may not be as potent as advertised or about their disturbing side-effects related to muscle aches and pains or cognitive issues, particularly memory.
This phenomenon, known as confirmation bias, is rife within the medical community. Eventually, this form of bias may prove instrumental in the formation of a popular or conventional wisdom on an issue, after which it becomes more difficult to find balanced views, which in this case would have suggested statin therapy might need to be more precision-targeted in its use. In contrast, currently, widespread support for blanket use of statins is emerging on the medical landscape. And, once this wholesale use of statins becomes engraved in stone, it usually takes years or decades for the findings to be overturned. With the addition of heavy promotion or advertisement of drug capabilities, conventional wisdom becomes further entrenched.
To reiterate from prior posts, I am in no way anti-drug or even anti-Big Pharma, because I recognize the enormous benefits we have all gained by the advancements in the pharmaceutical industry. However, this confirmation bias leading to conventional wisdom is a weakness of humans (both physicians and patients) who may be too easily influenced by advertising, promotions or other authority figures like leading research experts who promote a particular drug solution as the best answer to their health problems. In such situations, most of us are prone to trusting the good will and efforts of those who should possess the best and most current information. But, as I think you will see as we continue the discussion, there is a competition for ideas that is not necessarily and certainly not always based on what would appear to be hard science.
Which brings us to the next level in how information is transferred to physicians: what can the average physician gain in reading the research literature if they do not develop the critical skills to understand all the distinctions between weak and strong research designs? Eventually, they must rely on someone else to bring them up to speed if it takes just too much time to form an individual opinion. And, if topics of potential interest are not covered in the journals they read, then they would never receive exposure to ideas that may have persuaded them to reevaluate their practices. If all the journals are talking about the positive benefits of statins, then why wouldn’t the physician presume that statins are probably a useful drug to prescribe and would presume the risk/benefit ratio is quite low.
Closer to home, think of all the doctors who write prescriptions for sleeping pills, because these physicians have been exposed to hundreds of research articles that describe the use of sleeping pills in tens of thousands of research patients with a primary complaint of insomnia? Through whatever funding mechanisms might be available, insomnia research using sleeping pills is exponentially more common than studies looking at sleep hygiene or cognitive-behavioral therapy. Any primary care physician who keeps seeing more titles of published papers linking prescription sedatives with insomnia will eventually think and sense that this data must be conclusive evidence. If this view is the predominant one about insomnia, then sleeping pills must be a proper as well as a safe treatment. Voila, a conventional wisdom is born!
In the context of hundreds of non-sleep journals publishing articles on sleeping pills, can you imagine how often such a non-sleep journal might publish articles on UARS? It’s probably not a big fat zero, but it’s close to nothing. Examining all the major internal medicine and family medicine journals, it is less rare to see sleep-disordered breathing covered in general, but the concept of UARS is almost never addressed. Even in specialty journals on cardiovascular health where it is becoming common to see articles on sleep apnea and sometimes insomnia, rarely would you see the topic of UARS.
Adding to what might be called a negative confirmation bias, we must bring into our equation the role of sleep-dedicated institutions like the American Academy of Sleep Medicine that publishes two of the most widely read sleep-specific journals. When general physicians read research papers, but they themselves are not trained or experienced in conducting research, then these professionals are likely to be strongly influenced by the opinions of a guild of like-minded specialist professionals such as a large group of sleep doctors and researchers weighing in on a topic. For example, if the journals SLEEP and Journal of Clinical Sleep Medicine do not routinely publish information on RERAs and UARS, and if what they publish is relatively dismissive of these concepts, then any doctor and especially any sleep doctor who is busy operating a clinical practice will be much more likely to defer to the judgment of these thought-leaders in the elite positions of authority.
In fact, this phenomenon occurred recently when we published an article on RERAs in the Journal of Clinical Sleep Medicine. The article was not originally accepted, because the editors believed it was more commentary than research. Yet, the data provided was quite straightforward and concerning. We looked at all the article published in JCSM from 2006 to 2012 and narrowed our search down to breathing-related articles with objective data (number of breathing events recorded in each study). Of the 219 final qualifying articles, only 36 or 16.4% mentioned RERAs either formally by definition or informally with terms like “snoring arousals.” Thus, in all the other papers published in those years (83.6%), the highest standard of calculation for the Respiratory Disturbance Index (apneas + hypopneas + RERAs) was not included in the papers, because the authors only measured the apneas and hypopneas to provide the Apnea-Hypopnea Index. If leading sleep journals do not recognize and systematically promote the value of RERAs and UARS, then you can see how much bias would develop against these concepts. No surprise, this bias would greatly affect sleep professionals looking for clinical practice pearls when reading sleep journals.
To reiterate the point from Part I, if I am communicating with sleep doctors around the country who do not score RERAS and do not even use NCPT technology to measure RERAS, then it is a highly implausible scenario that the condition UARS would be on the radarscope of most sleep centers in the USA. True, I am just one doctor, but I believe I have communicated with more than 100 sleep physicians on this topic and rarely do I hear much interest in the UARS diagnosis. In fact, the AASM at one point went out of its way to downgrade UARS by saying it should be subsumed within the category of OSA. Yet, some sleep researchers and clinicians have rejected this position, at least in so far as their continued research might warrant or as their continued efforts in clinical practice reveal the necessity for evaluating RERAs and UARS when patients do not show classic OSA. But these few sleep professionals, whether they are researchers or clinicians, cannot overcome the heavy influences of a guild that wants to erase the concepts of RERAs and UARS from their nosology.
As you can see, despite Dr. Guilleminault’s research team’s discoveries of UARS in 1982 in children and in 1991 in adults along with several other researchers re-confirming the concept of flow limited breathing events (the earlier term for RERAs) throughout the 1990s, institutional forces in the field of sleep medicine itself have led to contradictory information. This information influence whether or not sleep professionals see value in testing and treating UARS. This problem is epitomized by the conflictual statements in the clinical guidelines of the AASM to not require the scoring of RERAs on diagnostic PSG, that is, it is optional to score RERAs. Yet, RERAs must (as in are mandated to) be treated when conducting a titration study in the treatment of a sleep breathing disorder. These two concepts—written in complete contradiction of each other—can be read in the same guidelines paper from the AASM.
Since publishing our paper on this contradiction as well as detailing the lack of published articles on RERAs and UARS, there has been no action taken by the AASM to rectify the problem. For all we know, the AASM does not see a problem, albeit Dr. Nancy Collop wrote a commentary about our piece and at least noted the discrepancy and the necessity to treat all breathing events, no matter what you called the breathing event. With this lack of attention to the RERAs/UARS controversy, it would seem unlikely that the preference cascade is going to evolve any time soon, but there are still more influences in our society that may move things along unexpectedly or further impede physician awareness.