The Journal of Pediatrics
Volume 150, Issue 4 , Pages 325-326, April 2007

Pediatrics and Evidence-Based Medicine Revisited

  • Robert M. Jacobson, MD

      Affiliations

    • Corresponding Author InformationReprint requests: Robert M. Jacobson, MD, Chair and Professor of Pediatrics, Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905-0001.

Mayo Clinic College of Medicine, Rochester, MN.

Article Outline

Abbreviations: AAP, American Academy of Pediatrics, EBM, Evidence-based medicine

 

In 2006, Belamarich et al1 examined the American Academy of Pediatrics (AAP) policy statements and identified 162 different verbal health advice directives on which pediatricians should counsel parents and patients throughout childhood. They found that none of the reviewed statements included an evidence-based discussion of the efficacy of the suggested advice.1

Sackett et al2 defined evidence-based medicine as “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients” and that “(t)he practice of evidence based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research.” Five years ago, Frohna and Park3 joined in the call for evidence-based pediatrics and authored an Association of Medical School Pediatric Department Chairs commentary published in this journal entitled, “Promoting the use of evidence-based medicine in pediatrics.” The authors call for pediatric departments to embrace evidence-based medicine (EBM)—to teach it and to practice it. One of their specific recommendations was for our training programs to teach the skills and furthermore for our residents and faculty to learn EBM skills side-by-side.

Are we teaching EBM? In a 2005 survey of chief residents from all 192 North American pediatric residencies, with 80% responding, we certainly appear to be doing so.4 Ninety-seven percent of chief residents endorsed that their programs used EBM. Are we teaching it effectively? Only 56% of the chief residents reported they themselves felt confident in their abilities to practice EBM.4 Even more concerning, only 7% of the responding chief residents felt that their program taught EMB effectively.

Are we applying EBM to patient care? In a telephone survey of the pediatricians-on-call at 97 British and Scottish hospitals with pediatric and neonatal units, with 87 responding, the investigators found otherwise.5 Of note, all but 4 of the 87 respondents were still in training. When on-call, despite access to electronic search engines, electronic publications, and EBM databases, only 14% of the pediatricians-on-call used resources to access the published literature and critically appraise it. Instead, they relied on contacting the local pediatrician for such queries.

Too much of medicine and too much of pediatrics is driven by expert opinion, theoretical machinations, and absent data.6 What Belamarich et al1 found with the AAP guidelines is not a condition limited to primary care; it is just the largest and most obvious target. Pediatric subspecialties also suffer from a deficit of evidence-based guidelines and a plethora of expert opinions. Sackett,7, 8 by the way, not only calls for the mandatory retirement of experts but for a general policy of refusing to believe them altogether.

Now, the deficit of evidence is not necessarily for a lack of trying. All too often, the literature searches directed toward answering questions raised in the clinical pediatric subspecialties fail to find clinical studies and especially high-quality studies. In pediatrics, we are often dealing with a deficit of data, an absence of study, and an evidence-less base from which to operate.

Features of pediatrics limit the evidence base in ways that those practicing adult medicine can only imagine. First, many of the diseases that concern us, those with significant morbidity and mortality rates, are relatively rare. In contrast to our colleagues in adult medicine, we cannot easily access large patient numbers to have adequate statistical power for study.

Second, even when we can achieve the numbers, either by the nature of the disease or through networking or collaboration, we face significant ethical and regulatory concerns. Institutional review boards applying the Common Rule 45 CFR 46 hold pediatric studies to a higher standard.9 For studies that exceed minimal risk, issues of direct benefit and relevance to the individual subject become germane. Critical information such as normal values for certain diagnostic tests (particularly in radiographic imaging) are lacking because of restrictions on enrolling normal children in studies that exceed minimal risk. Even for those studies that do not exceed minimal risk, we not only need parental permission for enrollment but, depending on the age of the child or teenager, their assent as well. Our ethical concerns complicate our decisions regarding control or comparative maneuvers as well. Where standard care exists, with or without an evidence base to support it, we are loath to use a placebo control in minors. Although adults may consent to a randomization that may leave them untreated and at risk for worsening of their condition, children pose concerns because they cannot consent.

Third, running clinical trials are more difficult in the pediatric age group. One not only has to recruit parent and child rather than a single individual, one must retain the child through the study procedures, exposure to discomfort and boredom, and the parents often suffer additional costs of participating in the study that adult volunteers do not have—including school attendance, their own work schedules, and care for the siblings.

Fourth, much of pediatric subspecialty care suffers from a therapeutic orphaning syndrome. Again, the volumes of these diseases are small. Pharmaceutical companies focus on adult disease where the numbers are greater. Clinical trials and studies are expensive. The multi-prong approach the Food and Drug Administration has taken to accelerate pediatric drug development10, 11, 12, 13 is, of course, an important effort, as is the National Institutes of Health requirement for pediatric study where appropriate.14

One of the more important passages in Frohna and Park’s3 2002 call-to-arms addressed the problem that teaching and practicing EBM in pediatrics will rapidly reveal this deficit of evidence. The authors wrote that, in such situations, we must not abandon evidence-based pediatrics. Specifically, they wrote that in our efforts to teach EBM, we must not abandon teaching the skills of critical reasoning and the art of medical decision-making, which includes making decisions that deal with uncertainty and incorporate clinical experience and family preference as well. Critical thinking is not something that intelligent individuals have as a gift or an art that one learns by osmosis or through experience. One must study and practice logical and critical thought to obtain and retain that skill. We cannot assign that topic to the reading of a paper, the attendance of a conference, or the assignment to a rotation. In fact, one must practice it every day with each patient encounter regardless of whether the evidence-base is strong, weak, or absent. Callahan15 stressed the importance of supplementing EBM with rules regarding medical decision-making that one learns from mentors and experience: History and physical examination are the cornerstones of diagnosis. Be cautious in assuming “it’s probably nothing.” Trust your intuition. Follow-up when questions remain. Never stop becoming pediatricians. Read the textbook (or something). Be cautious about releasing responsibility for a patient to a consultant. Never underestimate parents’ desire to provide the best for their children. The physician should appreciate the parents’ level of concern. Good physicians can make bad mistakes.

Furthermore, when our literature searches of a clinical question fail to find studies specific to our patients, Philips16 lays out an approach that allows pediatricians to adapt and adopt evidence. First, we should ask if the biologic differences between our patients and those studied really matter. Second, we should ask whether the demographic differences between our patients and those studied will really matter. Third, we should consider the risk for adverse events of those apparently beneficial and feasible therapies. We need to consider the baseline risks in our patients compared with those studied and weight that accordingly. Fourth, we should consider all of the outcomes measured and consider whether these outcomes, including the negative ones, are comprehensively assessed and applicable to our patients. By doing so, Phillips16 argues that we may be able to extend our reach into the pool of evidence and find applicable data.

Finally, part of our difficulty with the deficit of data is our perception that the data are lacking. Searching for studies is methodologically difficult and all too often incomplete. The Cochrane Collaboration has dedicated particular effort to the pediatric database to improve pediatric access to the data that do exist.17 The Cochrane Collaboration now includes a Child Health Field (http://www.cochranechildhealth.ualberta.ca/) that features a trials registry of more than 20,000 clinical trials in children and youth, as well as a new Cochrane review journal dedicated to those who care for children, entitled, Evidence based Child Health: A Cochrane Review Journal.

EBM should appeal to trainees and practitioners of pediatrics especially. We must protect our patients, with their higher degree of vulnerability, from the dangers of available technology.18 Even if a useless treatment was absolutely harmless, given the disparities in funding pediatric care, we must avoid any waste in time, cost, and effort. Also, given the relative rarity of pediatric subspecialty disease, our subspecialties are for the most part limited to academic centers and thus by necessity (and fortunately) remain scholarly fields for all of the practitioners. We are therefore more naturally attracted to seeking a scientific basis for our practice. Still, one does not learn EBM through the research experience but through clinical training and experience. We clearly need to improve the effectiveness of our teaching EBM and the use of EBM in practice.4, 5 We should extend that academic focus of our specialty and subspecialties to include empirical evidence of diagnostic and therapeutic field-effectiveness whenever possible. That the evidence will often be limited or nonexistent is no reason to abandon the technique. In fact, those revelations should drive our quest for investigation and clarifying the role that critical thinking must have in medical decision-making when dealing with uncertainty.

Back to Article Outline

References 

  1. Belamarich PF, Gandica R, Stein RE, Racine AD. Drowning in a sea of advice: pediatricians and American Academy of Pediatrics policy statements. Pediatrics. 2006;118:e964–e978
  2. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidence based medicine: what it is and what it isn’t. BMJ. 1996;312(7023):71–72
  3. Frohna JG, Park SM. Promoting the use of evidence-based medicine in pediatrics. J Pediatr. 2002;141:599–600
  4. Kersten HB, Randis TM, Giardino AP. Evidence-based medicine in pediatric residency programs: where are we now?. Ambulatory Pediatr. 2005;5:302–305
  5. Riordan FA, Boyle EM, Phillips B. Best paediatric evidence; is it accessible and used on-call?. Arch Dis Child. 2004;89:469–471
  6. Evidence-Based Medicine Working Group. Evidence-based medicine: a new approach to teaching the practice of medicine. JAMA. 1992;268:2420–2425
  7. Sackett DL. Second thoughts (Proposals for the health sciences—I. Compulsory retirement for experts). J Chronic Dis. 1983;36:545–547
  8. Sackett DL. Campaign to revitalise academic medicine: don’t believe us. BMJ. 2004;329(7460):294
  9. Protection of Human Subjects. In: Code of Federal Regulations. 2005;
  10. Rodriguez WJ, Roberts R, Murphy D. Current regulatory policies regarding pediatric indications and exclusivity. J Pediatr Gastroenterol Nutr. 2003;37(Suppl 1):S40–S45
  11. Katz R. FDA update. Epilepsy Res. 2006;68:85–94
  12. Breslow LH. The Best Pharmaceuticals for Children Act of 2002: the rise of the voluntary incentive structure and congressional refusal to require pediatric testing. Harvard J Legislation. 2003;40:133–193
  13. Benjamin DK, Smith PB, Murphy MD, Roberts R, Mathis L, Avant D, et al. Peer-reviewed publication of clinical trials completed for pediatric exclusivity. JAMA. 2006;296:1266–1273
  14. Alexander D. Regulation of research with children: the evolution from exclusion to inclusion. J Health Care Law Policy. 2002;6:1–13
  15. Callahan CW. Pediatric perspectives: observations of an older pediatrician: supplementing evidence-based medicine. Pediatr Rev. 2001;22:293–294
  16. Phillips B. Beyond the evidence. Arch Dis Child. 2004;89:286
  17. Waters E. Evidence-based paediatric practice and its relation to the Cochrane Collaboration. J Paediatr Child Health. 2003;39:137
  18. Silverman WA. Medicine’s “therapeutic imperative”—haunted by the potential for tragedy. Paediatr Perinatal Epidemiol. 2004;18:318–319

PII: S0022-3476(06)01259-5

doi:10.1016/j.jpeds.2006.12.044

The Journal of Pediatrics
Volume 150, Issue 4 , Pages 325-326, April 2007

Article Tools