Click Here to View the Online Version

Welcome to Plural Community—your newsletter, for your community.

This month, we have held back our case study to bring you two lead articles. The first, from Kathleen Yaremchuk, MD, co-author of the forthcoming Sleep Medicine, highlights the advances in recent years, the greater integration of sleep medicine into the realm of ORL-H&NS as a sub-discipline, societal issues and, particularly, the issue of gaining Board certification from 2011.

Second up is Eiki B. Satake’s piece on the role of statistics in the communication sciences and their disorders. Dr. Satake, co-author of the successful Handbook of Statistical Methods, questions the value of traditional statistical methods in this clinical evidence-based world, and proffers an alternative: the methodology of the eighteenth-century mathematician, Thomas Bayes, and the use of Bayes’ Theorem in considering probability with observational data to arrive at meaningful statistics. You don’t have to be a mathematician...!

Finally, a piece of overt promotion! Whether you are a speech pathologist, audiologist or, particularly, an otolaryngologist, then you’ll be pleased to learn of the forthcoming launch of the new, third edition of Raza Pasha’s bestselling pocket book, Otolaryngology-Head & Neck Surgery: Clinical Reference Guide.

The Evolution of Sleep Medicine

Kathleen Yaremchuk, MD, MSA, ABSM

Interest in sleep by writers, poets, and physicians has existed since the time of the ancient Egyptians (1300 BC) when opium was used to induce sleep. Hippocrates put forward his theory of sleep in 400 BC in Corpus Hippocraticum. In 1834 Robert MacNish described in The Philosophy of Sleep, “In proportion as man exceeds all other animals in the excellency of his physical organization, and in intellectual capability, we shall find in him the various phenomenon of sleep are exhibited in greater regularity and perfection. Sleep seems more indispensably requisite to man than to any other creature.”1

The question arises, how did we get from there to where we are today in our knowledge of sleep medicine? The story is that of a long and curving road.

In the Pickwick Papers, Charles Dickens described Joe—the “fat boy” who consumes great quantities of food and constantly falls asleep in any situation at any time of day. A clamor of incessant knocking besieged Mr. Pickwick’s lodgings. Once opened, the doorway revealed a “wonderfully fat boy” who stood upright, “his eyes closed as if in sleep,” his expression one of “calmness and repose.” Asked his business, he said nothing, but “nodded once, and seemed … to snore feebly,” immobile through three repetitions of the question. Then, as the door was about to close on him, he “suddenly opened his eyes, winked several times, sneezed once, and raised his hand as if to repeat the knocking.”2 Joe’s sleep problem is the origin of the term Pickwickian syndrome which ultimately led to the subsequent description of obstructive sleep apnea syndrome.

An awareness of sleep pathology developed with the description of narcolepsy in 1880 by Jean Baptiste Edouard Gélineau. Narcolepsy comes from the Greek words narcosis (benumbing) and lepsis (to overtake). Constantin Von Economo reported on post viral “sleeping sickness” or encephalitis lethargica in 1917 when an epidemic spread across the globe, leaving its victims in a statue-like condition, speechless and motionless. No recurrence of the epidemic has since been reported, though isolated cases continue to occur. Pavlov described dogs falling asleep during conditioned reflex experiments in 1930, describing the hypnotic state which existed between the wake state and that of sleep.

The scientific discovery that brought discipline to sleep was that of Hans Berger, a German psychiatrist who recorded electrical activity of the human brain in 1928 that demonstrated when a subject was awake or asleep. The tracings he recorded, were called “electroencephograms” or soon to be known as EEGs.

Dr. Nathaniel Kleitman, a professor of physiology at the University of Chicago, recognized the state known as rapid eye movement (REM) sleep in infants and then in adults. Electrooculograms (EOGs) were developed as tracings that would relieve the observer of the tedious task of watching individuals for eye movements as they slept. The basic sleep cycle was not identified since a technique of sampling was used to document EEGs and EOGs for five minutes at a time. The concept of continuous monitoring was thought to be a colossal waste of paper! However, in 1953 the idea of continuous recording during sleep was published by Aserinsky and Kleitman.4 William Dement and Nathaniel Kleitman perfected the protocol for all night sleep studies in the 1960s allowing continued growth of knowledge for polysomnograms as we know them today.

Gastaut, Tassinari, and Duron in France and Jung and Kuhlo in Germany described obstructive sleep apnea in 1965—long after the publication of Charles Dickens’ story—but verifying the diagnosis of the character, Joe. Tracheostomy as a treatment for sleep apnea was described in the 1970s at the Stanford Sleep Center, and was strongly resisted by the medical community at the time. How serendipitous that in 1981, the surgical treatment of Uvulopalatopharyngoplasty (UPPP) was described by Dr. Shiro Fujita5 and the mechanical ventilation with continuous positive airway pressure (CPAP) was described by Sullivan6 for treatment of OSA.

The National Center on Sleep Disorders Research was established by the National Institutes of Health (NIH) Revitalization Act of 1993. The plan called for multidisciplinary research to improve health, safety, and productivity by encouraging basic and translational research on sleep and sleep disorders. Gaps in knowledge needed to be answered in regards to daytime sleepiness and the risk associated with it.

As a medical specialty, however, sleep medicine came into the forefront in the last forty years. In 1996 the American Medical Association recognized sleep medicine as a specialty. In 2005, the Accreditation Council on Graduate Medical Education and the American Board of Medical Specialties developed written exams for the board certification in the subspecialty of sleep medicine for physicians board certified in Internal Medicine, Otolaryngology, Psychiatry and Neurology, Family Medicine, and Pediatrics. This confirms the experience that few specialties do not have integral areas of study within the study of sleep.

In his book Sleep, published in 1989, J. Allan Hobson, stated that “More has been learned about sleep in the last 60 years than in the preceding 6,000 years. In this short period of time, researchers have discovered that sleep is a special activity of the brain, controlled by elaborate and precise mechanisms.”6

No truer words have been spoken. Now with our 24/7 society, the importance of sleep and the ill effects of sleep deprivation on society were publicized by the National Commission on Sleep Disorders Research in 1998 and developed a lasting legacy through the National Sleep Foundation committed to education and research.


  1. Mac Nish R.: The Philosophy of Sleep. New York, D. Appleton, 1834.
  2. Dickens, C.: The Pickwick Papers. Penguin Classics, 1837.
  3. Aserinksy, E., Kleitman, N.: Regularly occurring periods of eye motility, and concomitant phenomena, during sleep. Science 1953; 118:11–18.
  4. Fujita S., Conway W., Zorick F,: Surgical correction of anatomic abnormalities in obstructive sleep apnea syndrome. OtoHNS 1981; 89:923–934.
  5. Sullivan C.E., Issa F.G., Berthron-Jones M.: reversal of obstructive sleep apnea by CPAP. Lancet 1981; 1:862–865.
  6. Hobson J.: Sleep. New York, Scientific American Library, 1989.

Kathleen Yaremchuk, MD, serves as Vice President of Clinical Practice Performance for the Henry Ford Health System and is co-editor of Sleep Medicine (Plural Publishing, Inc., 2010).

Moving Forward to Evidence-Based Statistics: What Really Prevents Us?

Eiki B. Satake, PhD

Several years ago, a colleague and I conducted a two-hour research seminar at the American Speech-Language-Hearing Association (AHSA) convention in Chicago. As always, we began with a one-item, “True or False” pop quiz. The vast majority of the participants attending the seminar were clinical professionals and graduate students in speech and language pathology and audiology. The quiz item was as follows:

“True or False. P value = 0.03 means ‘There is a 3% chance that the null hypothesis is correct.’ Or, equivalently, ‘There is a 97% chance that the null hypothesis is incorrect.’”

Approximately 80% of the participants said the statement is “True.” The correct answer, however, is that the statement is False.

The significance here is not that 80% of participants got the answer wrong, but rather the way the concept of P-value is taught and the subsequent likelihood that it will be misinterpreted. In fact, there was a general expression of shock when, in the process of reviewing the substance of the question, that the true definition of P value is much more complex than what they were led to believe.

The P value is the probability of obtaining a result as exact observed value or more extreme than the observed result derived from a sample, given that the null hypothesis (Ho) is true. However, a more challenging question is what it really suggests and how it should be used to determine statistical significance. The P value was first proposed by the English mathematical statistician R. A. Fisher in the 1920s. Initially, Fisher’s intention of the use of the P value was to measure the strength of evidence and as an index to show how severely the truth of the null hypothesis (Ho) is contradicted by the actual evidence (sample data). The ground work for traditional statistics, based on inferential reasoning, was laid down early in the twentieth century by such prominent theoreticians as Fisher, Neyman, and Pearson. In the clinical and medical sciences, generations of clinicians and researchers have been educated according to the principles of statistical inference techniques such as P value and NHST. Usually, such methods do not directly address the issue of how accurate the diagnoses are. While providing the foundation for the vast majority of the clinical and experimental studies reported in the scientific literature, several clinicians and researchers have criticized the limitations and myths of P value and NHST (Goodman, 1999, Iversen, 1984).

Since the calculation of a P value is based on the assumption that Ho is true, (1) it cannot be a direct measure of the probability that Ho is false (or, equivalently, that an alternate hypothesis, Ha, is true); and (2) it does not calculate the probability of the hypothesis being true or false, but rather calculates the probability of the occurrence of a sample data. The correct interpretation is that the probability of obtaining an exact result or more extreme result from the sample is 3%, given that Ho is correct.
Before discussing the reason why the P value method does not work well in the clinical science research, we need to consider the philosophy underlying inferential statistics and how it works.

Statistical inference consists both of deductive reasoning and inductive reasoning. Deductive reasoning is exactly what P-value and Null Hypothesis Significance Testing (referred to as NHST) are all about. We begin with the assumption of the null effect (Ho is true) and predict what we will see if Ho is true. For example, we may begin by assuming that a client has no particular fluency disorder. We then evaluate and calculate the likelihood of each of the symptoms associated with the disorders as seen in the client. In short, we go from the status of the disorders (effect) to a set of the symptoms (cause). Symbolically, we can write it as Probability (Symptom, given Disorders). However, the problem with this sort of reasoning is that we cannot use it to make a diagnosis based on what we actually observed. Clinically, this reasoning is not useful.

Also, this deductive method, specifically regarding P value, uses both observed and hypothetical (“more extreme”) data values. Very often, the inclusion of “more extreme” data values leads to a less reliable and valid conclusion. What a clinician really wants to know is the reasoning that goes in the reverse direction; from symptoms (cause) to disorders (effect), and allows her to be able to make a diagnosis based on only what is observed. Symbolically, this process can be written as Probability (Disorders, given Symptoms). In other words, we need inductive reasoning which allows a clinician to make an accurate diagnosis based on what was actually observed. Inductive reasoning measures the strength of evidence more directly and is clearly “evidence-based.”

However, evidence-based statistics must provide a basis for determining the credibility of a clinician’s diagnosis based on the clinical observation. It must not only be clinically useful but also capable of expanding a clinician’s knowledge beyond what they already know. Sadly, traditional statistical methods, such as P value and NHST, fail to accomplish this task. The question that follows is whether there is an alternative statistical method that is suitable for “evidence-based practice.” Fortunately, there is such a method, namely Bayesian statistics.

In recent years, the importance of Bayesian statistical methods has been increasingly recognized and discussed in the field of clinical sciences because of its ability to state prior probabilities, representative of the initial subjective views of clinicians and then to update or revise these beliefs based on the emergence of new data to generate the posterior probability. Unlike the traditional statistics using P-value and NHST, the Bayesian approach works in a cumulative fashion, not in a terminal way. It allows for continuous revision, which allows a clinician to use the knowledge from previous research with newer findings of another research to update an estimate about a true population parameter. This cumulative treatment of data adjusts the statistical error and the amount of confounding variables (aka covariates) to increase the predictability of the final result and determine not only “statistical significance” but also “clinical significance.” In my view, a Bayesian approach is more practical and useful for clinical research because of its inductive and cumulative nature. Certainly, it is more clinically relevant and more closely synchronous to the “Evidence-Based Practice” than the traditional approach.


  1. Goodman, S.N. (2005). Introduction to Bayesian methods 1: measuring the strength of evidence. Clinical Trials, 2, 282–290.
  2. Goodman, S.N. (1999). Toward Evidence-Based Medical Statistics. 1: The P Value Fallacy. Annals of Internal Medicine, 130, 995–1004.
  3. Iversen, G.R. (1984). Bayesian Statistical Inference. Beverly Hills, CA: Sage Publications, Inc.
  4. Kadane, J.B. (1995). Prime Time for Bayes. Controlled Clinical Trials,16, 313–318.
  5. Matthews, R.A.J. (2006). Why should clinicians care about Bayesian methods? http://–16. Retrieved December 12, 2006.
  6. Maxwell, D.L., and Satake, E. (2006). Research and Statistical Methods in Communication Sciences and Disorders. Cliff Park, NY: Thomson Delmar Learning.
  7. Maxwell, D.L., and Satake, E. (2005). Analyzing and Interpreting Single-Subject Data: A Tutorial, 2 Hour Seminar presented at the annual American Speech-Language-Hearing Association Convention, San Diego, CA, November 19.
  8. Salsburg, D. (2001). The Lady Tasting Tea. New York, NY: Henry Holt and Company, LLC.
  9. Satake, E., Jagaroo, V., and Maxwell, D.L. (2008). Handbook of Statistical Methods: Single Subject Design. San Diego, CA: Plural Publishing.
  10. Satake, E., and Maxwell, D.L. (2008). Evidence-Based Statistics for Clinicians & Researchers: Current Problems & Possible Solutions, 2 Hour Seminar presented at the annual American Speech-Language-Hearing Association Convention, Chicago, IL, November 20.
  11. Satake, E. (1994). Bayesian Inference in Polling Technique: 1992 Presidential Polls. Communication Research, 21, 3, 396–407.
Eiki B. Satake, PhD, is Associate Professor of Mathematics in the Department of Communication Sciences and Disorders at Emerson College in Boston, Massachusetts, and co-editor of Handbook of Statistical Methods: Single-Subject Design (Plural Publishing, Inc., 2008).

Listed here are dates through the end of September for new product releases and exhibitions at which you can meet our people and browse our titles.

August 2010

New Releases

Otolaryngology – Head and Neck Surgery: Clinical Reference Guide, 3rd Ed.
Brain-Based Communication Disorders
Preclinical Speech Science Workbook
Clinical Management of Swallowing Disorders Workbook
CHARGE Syndrome
Sleep Medicine


September 2010

September 6-9
The 4th World Voice Congress
Seoul, Korea
More Details Here

September 27-28
Manchester Voice Course
Manchester, England
More Details Here

September 26-29
American Academy of Otolaryngology-Head and Neck Surgery
Boston, Massachusetts
More Details Here

New Releases

Here’s How to Treat Childhood Apraxia of Speech: An Integrated Approach
Communication Development and Disorders for Partners in Service
Building a Research Career
Psycholinguistics: Introduction and Applications
Human Auditory Evoked Potentials

New Releases

Speech-Language Pathology

Preclinical Speech Science Workbook

Clinical Management of Swallowing Disorders Workbook

Brain-Based Communication Disorders


Otolaryngology – Head and Neck Surgery: Clinical Reference Guide, 3rd Ed.


Sleep Medicine

Genetic Syndromes

CHARGE Syndrome


Competition time!
This month’s competition is a free prize drawing for a copy of Satake et al’s Handbook of Statistical Methods. Enter today: All you have to do is email your name and address to pluralcommunity, placing “August Competition” in the subject line. The drawing will take place on August 21, 2010 and the winner will be announced in the September issue of Plural Community.

Conratulations to our July-drawing winner – Kay Meyer! She will receive a copy of Singing and Teaching Singing: A Holistic Approach to Classical Voice by Janice Chapman, AUA, OAM.

Don't forget to follow us on Facebook, Twitter, and check out the Plural Blog for news, updates, information, and feedback on everything going on in the Plural community

Want to submit an article or a case study? Or just want to comment on anything here in Plural Community or elsewhere? Contact us at pluralcommunity anytime.





If you do not wish to receive future emails, click here.
Please allow 90 days for this option to take effect.
(If link does not work, please copy and paste it into your browser’s address bar)[ ].
© 2006-2010 Plural Publishing, Inc. All Rights Reserved. Follow Us on Facebook and Twitter | Plural Blog