Tree of Life - the first-known sketch by Charles Darwin of an evolutionary tree describing the relationships among groups of organisms (Cambridge University Library).

Tree of Life – the first-known sketch by Charles Darwin of an evolutionary tree describing the relationships among groups of organisms (Cambridge University Library).

The Merriam-Webster Dictionary defines science as:

Knowledge about or study of the natural world based on facts learned through experiments and observation.


Knowledge as distinguished from ignorance or misunderstanding.

While this should distinguish science from pseudoscience, those who practice the latter often lay claim to the same definition. But one of the major differences between science and pseudoscience is that science advances through constant rejection and revision of prior models and hypotheses as new evidence is produced; it evolves. This is the antithesis of pseudoscience. At the heart of pseudoscience-based medicine (PBM) is dogma and belief. It clings to its preconceptions and never changes in order to improve. It thrives on the intransigence of its belief system, and rejects threats to its dogma. Despite the constant claims by peddlers of pseudoscience that SBM practitioners are closed-minded, we know that, in fact, PBM is the ultimate in closed-minded belief. Of course, those of us who claim to practice SBM aren’t always quick to adopt new evidence. We sometimes continue practices that may once have been the standard of care but are no longer supported by the best available evidence, or perhaps may even be contradicted by the latest evidence. Often this is a byproduct of habituated practice and a failure to keep current with the literature. While this is certainly a failure of modern medicine, it is not an inevitable outcome. It is not emblematic of the practice of medicine, as it is with PBM. When medicine is science-based, it strives for continual improvement based on modifications around emerging evidence.

I would like to highlight three recent examples of how science-based pediatric medicine has made dramatic changes based on emerging evidence. One interesting aspect of these particular changes to best-practice recommendations is that they all lead to a more conservative approach to care; that is, they lead to less or even no treatment compared to prior recommendations. This is yet another significant distinction between SBM and PBM; PBM never does nothing. It always claims to have a treatment or a cure. It is another hallmark of PBM, and often what draws people to it. It is an endless fount of hope when “traditional” medicine has nothing much to offer.

Before I discuss these examples, I would like to diverge briefly to illustrate the hierarchy of evidence we use to come to evidence-based decisions or recommendations. I do this because I will be referring below to guidelines and recommendations published by academic organizations. As can be seen on the pyramid of evidence below, “expert opinion” is located at the bottom due to its high susceptibility to bias. It is after all, just an opinion. Published guidelines and consensus statements vary greatly in their location on the pyramid of evidence. They can also be highly susceptible to the bias and opinions of the committee or panel members who author them. Even then, when good studies are few and far between, these forms of expert opinion may be the best evidence we have to go on. Fortunately, many professional medical organizations, such as the American Academy of Pediatrics (AAP), use an EBM approach to their guidelines and recommendations that is stricter and more transparent than they used to be, and include standardized documentation of the levels of evidence used and the strengths of recommendations made, as I alluded to above. So while these guidelines and recommendations are not meant to be used in cookbook fashion and applied absolutely to all cases, they can be quite helpful in expanding our understanding of issues that may be quite complex.


Ear infections – primum non nocere

A middle ear infection (acute otitis media, AOM) is one of the most common treatable conditions seen by pediatricians. Not long ago, the standard of care for any child diagnosed with an ear infection would be treatment with antibiotics. In 2013, a growing, cumulative body of evidence led the American Academy of Pediatrics (AAP) to revise the evidence-based clinical practice guideline it had previously published in 2009. For the first time in North America, the AAP presented pediatricians with good evidence to support an “observation option” with no antibiotic treatment for selected children with confirmed ear infections. The select groups for whom this option applies are children 6 months or older with one-sided ear infections and children 2 years and older with one or two sided ear infections provided they do not have severe signs or symptoms (i.e. they have mild ear pain and fever < 39°C/102.2°F). It also applies only to children without health problems that might predispose them to more serious infection. The guideline states (the action statement for the older group is given here as an example):

The clinician should either prescribe antibiotic therapy or offer observation with close follow-up based on joint decision-making with the parent(s)/caregiver for AOM (bilateral or unilateral) in children 24 months or older without severe signs or symptoms (ie, mild otalgia for less than 48 hours, temperature less than 39°C [102.2°F]). When observation is used, a mechanism must be in place to ensure follow-up and begin antibiotic therapy if the child worsens or fails to improve within 48 to 72 hours of onset of symptoms. (Evidence Quality: Grade B, Rec Strength: Recommendation).

Note the statement in parentheses above. All EBM recommendations and guidelines should have a standardized framework with which to evaluate the level of evidence used and the strength of the recommendation made.

The addition of this “watchful-waiting” option was based on several new, well-conducted placebo-controlled RCTs, as well as systematic reviews of the timing of antibiotic therapy and the outcome of untreated ear infections. What these studies demonstrated is that the vast majority of ear infections are self-limited, resolving spontaneously without complications whether treated or not. In other words, “primum non nocere“, the oft-cited tenet of good medical practice to “first, do no harm.”

Though I strive to practice EBM/SBM, I can tell you it isn’t always easy doing nothing. While some parents are grateful for my explanation of the watchful-waiting approach and for sparing their children from potentially unnecessary antibiotics, others find it difficult to understand this approach or are simply not comfortable allowing their children to remain untreated. For these parents treatment may be the preferred course, but at least I have been honest in my explanation of the SBM evidence for and against the different approaches. Widespread use of these EBM guidelines will result in fewer antibiotic-related complications and fewer antibiotic-resistant bacterial infections.

Bronchiolitis – don’t just stand there, do nothing!

Bronchiolitis is a common respiratory tract illness in infants and young children. Caused by several different viruses, the illness begins like any cold with runny nose, low-grade fever, and cough. Inflammation and spasm of the small airways of the lungs produce worsening cough, wheezing, and sometimes difficulty breathing. Bronchiolitis can be relatively minor, similar to a typical cold, or more severe, requiring supplemental oxygen and even hospitalization. Because the illness typically occurs in infants and children under 2 years of age, it can be alarming to parents and even to pediatricians. And because wheezing (a result of swelling and spasm of the medium and small airways) is a common component of the illness, as it is in asthma, an asthma mindset to evaluation and treatment has been the standard of care for these children. Cumulative and emerging evidence, however, has resulted in a new AAP practice guideline that should greatly simplify our approach to this common illness. The new guideline, published in October 2014, includes significant changes to the previous guideline published in 2006. It recommends eliminating evaluative and management strategies that have long been the mainstay of treatment. Recommended for the scrap heap are:

  • viral or laboratory testing (such as testing for specific viral causes)
  • trial of a bronchodilator medicine (to open the airways, typically albuterol)
  • use of epinephrine (to shrink swelling in the airways)
  • chest x-rays
  • chest physiotherapy (pounding on the back to loosen mucus)
  • On the prevention side, the guideline recommends administering palivizumab (an antibody against RSV, the most common cause of bronchiolitis) only to children born at < 29 weeks of gestation, unless they have chronic lung disease of prematurity or significant heart disease. Previous guidelines recommended its use for premature babies of greater gestational age as well.

These procedures were removed from the recommendations because the best available evidence shows no improvement in outcome with the use of any of them. In short, despite the sometimes serious nature of this illness and the distress it can cause to both children and their parents, there is little to nothing science-based medicine can do to treat it. This can make for some anxious and difficult moments in the exam room or the emergency department when trying to reassure parents and help them through this illness. The understandably strong parental need to do something is often not satisfactorily addressed by our rationale for doing nothing. Yet honest and ethical practice dictates that the use of compassion, reassurance, and support is all we can offer much of the time. Again, this is something no peddler of quackery or sCAM treatments will understand, because there is always something in their bag of tricks, even though tricks are all they are.

Introducing food to babies – the changing tides of advice

When I was a pediatric resident in training, there was a decidedly non-evidence-based dogma to the introduction of solids in infancy. I was taught to introduce iron-fortified cereal at 4 months, followed by the slow introduction of foods between 4 and 6 months of age as follows: vegetables (preferably yellow and orange, then green), then fruits, then a mixture of vegetables and fruits, then meats. Peanut and tree nut proteins, as well as egg, milk, and of course honey were to be strictly avoided until at least 1 year of age. It turns out that, except for the honey, which can lead to botulism in infants, none of this is based in good science.

While I have no intention of covering the topic of peanut allergy here, recent studies exploring the effect of introducing peanut to the diet at different ages has the potential to change the way we think about the introduction of this, and other foods, to infants. It was previously felt that the early introduction of foods could predispose children to the development of allergies, eczema, and asthma. In 2003, the AAP recommended delaying cow’s milk until 1 year of age, egg until 2 years of age, and peanut, tree nut, and fish until 3 years of age. In 2008, they withdrew specific recommendations on the timing of food introduction, recognizing the lack of good supporting evidence for any particular recommendation. Since then, evidence has emerged that suggests earlier dietary introduction of peanuts may actually decrease the risk of developing peanut allergy. This observational study, published in 2008, showed that young Israeli infants who were fed a popular soft peanut-puff as a first food snack had a significantly lower prevalence of peanut allergy than a cohort of children of similar heritage being raised in the UK, where peanut protein is not typically given until after 1 year of age. This observation was followed by a larger prospective study known as the LEAP study (Learning Early About Peanut Allergy). This study was designed to look specifically at the effect of introducing peanuts during early infancy compared to waiting until later. The results of the LEAP study were published in March of this year in The New England Journal of Medicine. The study demonstrated an 81% reduction of peanut allergy in children given peanut-containing foods early in infancy, and who had continuous consumption of peanuts, compared to infants who avoided peanuts. While this is an important study with a potentially significant impact, it is also important to understand that the infants studied were part of a group already at high risk for developing allergy due to having either documented eczema or egg allergy. It isn’t clear that these results will be generalizable to a broader population at average risk, though it is reasonable to hypothesize that children with less of a risk might respond at least as well. In any case, given the lack of evidence for delaying the addition of peanuts to the diet, the results of these studies will likely cause many pediatricians to change the way they counsel parents about the introduction of foods to their babies. Still, without additional data, specific answers to questions such as how much peanut to give, at what age, and for how long, will be elusive until further studies are conducted and more data is available. As with any intriguing piece of data, it is always only one piece of a much larger, more complicated puzzle. While the emerging data is fascinating, it is too early to know the answer to these questions just yet. And that is science.


My intended goal for providing the above examples of medicine-in-evolution is just that; to show that science-based medicine evolves. It rejects and tosses out old, inadequately, or incorrectly supported ideas and replaces them with newer ones supported by more or better evidence. It grows and improves. It admits its errors and strives to better itself. When done right, it constantly fights against bias and preconception, and strives, ultimately, to reach toward the truth. This is the essence of science, and distinguishes it from belief and quackery.




Posted by John Snyder

John Snyder, MD, FAAP, is an Assistant Professor of Pediatrics at Tufts University School of Medicine, and a practicing pediatrician at Amherst Pediatrics in Amherst, Massachusetts. Previously, he was Medical Director of the teaching clinic at Baystate Children's Hospital, and before that he was Chief of the Section of General Pediatrics and Medical Director of Pediatric Ambulatory Care at Saint Vincent's Hospital in New York City. Since 1994, Dr Snyder has been active in pediatric resident and medical student education with a particular interest in evidence-based pediatrics. His main area of interest is medical myth and the ways in which parents utilize information in making medical decisions for their children. One area of focus has been the vaccine myth, and he lectures frequently on this subject in both academic and community settings. His other activities have included: contributor to the Gotham Skeptic blog, member of the New York City Skeptics' board of advisors, and expert for ("A New Social Network on Health Founded by America's Top Doctors"). Dr Snyder graduated from Mount Sinai School of Medicine, completing his residency training in pediatrics at The Mount Sinai Hospital in New York City. He is board certified in pediatrics, and is a Fellow of The American Academy of Pediatrics. Dr. Snyder has no ties to industry, and no conflicts of interest regarding any of his writings. Dr. Snyder’s posts for Science-Based Medicine are archived here.