|Year : 2015 | Volume
| Issue : 2 | Page : 174-177
Looking back at over 20 years of EBM
Centre for Clinical Epidemiology and Biostatistics, University of Newcastle; Clinical Research Design, IT, and Statistical Support Unit, Hunter Medical Research Institute; Department of General Medicine, John Hunter Hospital, Newcastle, New South Wales, Australia
|Date of Web Publication||16-Dec-2015|
Hunter Medical Research Institute Building, University of Newcastle, Callaghan, New South Wales - 2308
Source of Support: None, Conflict of Interest: None
|How to cite this article:|
Attia J. Looking back at over 20 years of EBM. Arch Med Health Sci 2015;3:174-7
I was just finishing medical school and starting my internship at McMaster University when the whole evidence-based medicine (EBM) movement began. This was explained to us as the explicit and judicious use of evidence in medical decision-making. In our teaching sessions, we were encouraged to question and identify the evidence base for the decisions we made on the wards. Identifying the key studies on which clinical practice was based then necessitated that we develop the skills to critically appraise those studies. The first "Users' Guides to the Medical Literature" appeared in 1993 in the Journal of the American Medical Association (JAMA), with subsequent installments over the following 20 years; this series became our textbooks long before they were updated and formally compiled into a book (now in its third edition  ).
I remember the opposition and thinly-disguised scorn that greeted this movement in its early days. There was an assumption that of course medicine is based on evidence and observation… what else could it be based on? A tongue-in-cheek article appeared in the BMJ describing possible alternatives to evidence-based medicine: 
However, over the years, as evidence was given more and more prominence and the movement encouraged the development of reviews and guidelines, many successes were seen. Tricia Greenhalgh highlights the example of asthma care: The British Thoracic Society was one of the first adopters of guidelines, which led to increased care plans, increased inhaler use, and subsequently, measurably reduced mortality.  Likewise, long-time myths were also busted, for instance digoxin was found to have minimal impact in the management of congestive heart failure (CHF). 
- Eminence-based medicine: The oldest clinician is right.
- Vehemence-based medicine: The one who argues his/her point most strongly is right.
- Eloquence-based medicine: The one with the smoothest tongue is right.
- Providence-based medicine: Do nothing and most patients will improve.
- Confidence-based medicine: The most confident clinician is right (usually the domain of surgeons).
Nevertheless, astute clinicians have increasingly realized that EBM is not the only tool at our disposal. There is a role for compassion, judgment, and experience; anecdote is often the trigger to go down a different route, both clinically and research-wise. These principles have been articulated under the rubric of "narrative-based medicine." 
There are five challenges to EBM that need to be addressed if its approach of inquiry and considered decision-making is to remain relevant to clinical practice in the 21st century, some of which have already been articulated by Greenhalgh. 
| Challenge 1: The Volume of Literature|| |
Allen and Harkins  identified that one 24-h medical take with 18 patients generated 44 diagnoses, and identified 3679 pages of national guidelines relevant to their care, which would require 122 h of reading. Clearly, evidence-based practice for each and every patient is not feasible for any mortal physician with sleep requirements. However, the movement has already risen to this challenge in making "predigested" evidence accessible quickly. For example, resources such as "Best Practice"  provide practical management recommendations and are continuously updated. Electronic resources such as "UpToDate"  are regularly peer-reviewed, and evidence supporting recommendations is explicitly rated using the GRADE system. 
| Challenge 2: Industry has Hijacked the EBM Brand|| |
Because randomized controlled trials (RCTs) are the best way to eliminate confounding, they are judged to be the highest level of evidence (bested only by meta-analyses that combine many RCTs). However, RCTs are expensive and hence have been funded mainly (either solely or in part) by pharmaceutical companies, meaning that this gives the industry a large say in setting the parameters within which the trial operates. This is evidenced in a number of ways:
- Choice of the comparator: For example, in the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial,  testing 80 mg of atorvastatin in the setting of stroke, the comparator was placebo. It could be easily argued that anyone with stroke is likely to be a vasculopath and would potentially benefit from a statin, so that the better control would have been 20 mg or 40 mg of atorvastatin, rather than placebo.
- Use of run-in periods: Run-in periods are used before randomization to ensure that the participant can tolerate the medication and is compliant in attending clinics. This means that those who would usually have had side effects or who would have dropped out are no longer included in the treatment arm, increasing the likelihood of seeing a therapeutic effect if one is there. We can no longer estimate adverse event rates in the general population or estimate a "real world" effect of the intervention.
- Nomination of surrogate endpoints: For example, in trials of agents for osteoporosis, the aim is ultimately to reduce clinical fractures, but most trials use bone mineral density (BMD) as a surrogate. BMD has been accepted as a surrogate for fractures despite evidence showing that only 4% of the variance in fracture risk is explained by BMD. 
- Neglect of other areas of medical evidence apart from therapeutics: Only 10% of questions generated by patient encounters are related to drug choice,  whereas a large fraction of the studies published by major medical journals and reviewed in hospital journal clubs are RCTs, a situation described as "drug dependence."  Neglected areas worthy of research include history taking, physical exam, differential diagnosis, and diagnostic testing.
- Disease mongering: This is the name given to the "medicalization" of symptoms that had previously been considered part of the spectrum of normal life, such as erectile dysfunction, baldness, or premenstrual mood swings. Moynihan et al.  document the role of pharmaceutical companies in funding and driving "community interest groups," in essence creating more demand for each company's own products.
- Influencing guideline development: Choudhry et al.  surveyed a number of guidelines and documented that 87% of authors had some form of interaction with the pharmaceutical industry, 58% had received financial support, and 38% had served as employees or consultants. This represents a clear conflict of interest.
| Challenge 3: Chasing Increasingly Marginal Gains|| |
Because of the business model in the pharmaceutical industry, there is ongoing investment in developing new therapeutics, often in already crowded and well-serviced areas. As an example, let us consider the case of a patient with CHF on diuretics. The addition of an angiotensin-converting-enzyme inhibitor (ACEI) is extremely effective with a number needed to treat (NNT) of 3.  The addition of a beta-blocker in addition to diuretics and ACEI is less effective, with a NNT of 20.  The further addition of an aldosterone antagonist yields diminishing returns, with a NNT of 70.  It is worth discussing whether further investment in new drugs in this area is warranted or if the investment would be better used in other areas of medicine.
| Challenge 4: Applying Population Evidence to the Individual Patient|| |
Data from large RCTs provide a precise and credible estimate of the effect of a therapy on average. It is a different matter to decide whether the individual patient in front of us will be among the percentage who benefit or among those who do not. For example, although there are many RCTs documenting the efficacy of statins, a common question on rounds is the uncertainty regarding the balance of risks and benefits of these drugs in the elderly. The question of individual application is complex and requires some access to actuarial tables. For example, in Australia, an 80-year-old male living in the community has a median survival of another 8.5 years, longer than the 4-5 years in many statin trials. If he has hypertension and high cholesterol, his absolute risk of cardiovascular disease (CVD) over the next 5 years is 30% compared to 15% in a similar man at 60 years of age; hence the NNT drops from 33 to 16, making treatment attractive. This decision, however, also involves some assessment of the patient's values and priorities, the perception of which varies widely between patients and their physicians. For example, when weighing up the risks and benefits of anticoagulation for stroke prevention in atrial fibrillation (AF), Devereaux et al.  found that most physicians would only accept 1-5 bleeds in order to prevent one stroke compared to accepting over 20 bleeds for most patients.
| Challenge 5: Crossing the Evidence-Practice Gap|| |
Even when solid evidence is generated, there is a gap in implementing that evidence consistently in clinical practice. Knowledge in and of itself is not sufficient to generate behavior change among clinicians. Knowledge passes through the filter of attitudes before affecting behavior.  Some clinicians may be cynical nihilists, feeling that patients "never follow instructions anyway," while other may be more curious and reflective and more likely to be early adopters, as seen with technology. Malcolm Gladwell in his book The Tipping Point describes these people as "connectors" or "mavens," people who have a large sphere of influence and whose practice can sway others to change their behavior. Making use of these people as clinical leaders may be one way to promote behavior change in clinical practice. Other barriers to behavior change may include medicolegal concerns in taking up a new practice, or system level variables that hinder the adoption of new practices, such as lack of time and organizational constraints in billing, equipment, and budget. It is worth noting that the evidence-practice gap can work in many directions, such as not doing what we should, e.g., underuse of ACEI, beta-blockers, etc. in various settings, or doing what we shouldn't, e.g., overuse of arthroscopic knee washouts in osteoarthritis, or doing things wrongly, e.g., misuse of opioids in chronic pain. A formal study of these factors in behavior change constitutes the emerging field of implementation science or knowledge translation and should become a fertile area for EBM in the future.
| Conclusion|| |
EBM can rightly be proud of its success in training a generation of clinicians who are more conscious of the evidence base for their decisions and more literate in critiquing the studies that constitute that base. However, we who practice EBM also need to rise to the challenges that have developed over the last 20 years. EBM is not the sole arrow in our quiver. As Osler said, "It is much more important to know what sort of a patient has a disease than what sort of a disease a patient has." And that requires compassion, insight, and a wide variety of skills and competencies.
| References|| |
Guyatt G, Rennie D, Meade MO, Cook DJ. Users′ Guides to the Medical Literature. A Manual for Evidence-Based Clinical Practice. 3 rd
ed. McGraw-Hill Education/Medical; 2014.
Isaacs D, Fitzgerald D. Seven alternatives to evidence based medicine. BMJ 1999;319:1618.
Greenhalgh T, Howick J, Maskrey N; Evidence Based Medicine Renaissance Group. Evidence based medicine: A movement in crisis? BMJ 2014;348:g3725.
Digitalis Investigation Group. The effect of digoxin on mortality and morbidity in patients with heart failure. N Engl J Med 1997;336:525-33.
Greenhalgh T. Narrative based medicine: narrative based medicine in an evidence based world. BMJ 1999;318:323-5.
Allen D, Harkins KJ. Too much guidance? Lancet 2005;365:1768.
Amarenco P, Bogousslavsky J, Callahan A 3 rd
, Goldstein LB, Hennerici M, Rudolph AE, et al
. E Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) Investigators. High-dose atorvastatin after stroke or transient ischemic attack. N Engl J Med 2006;355:549-59.
Sarkar S, Mitlak BH, Wong M, Stock JL, Black DM, Harper KD. Relationships between bone mineral density and incident verebral fracture risk with raloxifene therapy. J Bone Miner Res 2002;17:1-10.
Del Fiol G, Workman TE, Gorman PN. Clinical questions raised by clinicians at the point of care: A systematic review. JAMA 2014;174:710-8.
Cowan J. Drug dependence in a journal club. ACP J Club 2000;132:A23.
Moynihan R, Heath I, Henry D. Selling sickness: The pharmaceutical industry and disease mongering. BMJ 2002;324:886-91.
Choudhry NK, Stelfox HT, Detsky AS. Relationships between authors of clinical practice guidelines and the pharmaceutical industry. JAMA 2002;287:612-7.
Effects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). The CONSENSUS Trial Study Group. N Engl J Med 1987;316:1429-35.
Effect of metoprolol CR/XL in chronic heart failure: Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF). Lancet 1999;353:2001-7.
Zannad F, McMurray JJ, Krum H, van Veldhuisen DJ, Swedberg K, Shi H, et al
.; EMPHASIS-HF Study Group. Eplerenone in patients with systolic heart failure and mild symptoms. N Engl J Med 2011;364:11-21.
Devereaux PJ, Anderson DR, Gardner MJ, Putnam W, Flowerdew GJ, Brownell BF, et al
. Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: Observational study. BMJ 2001;323:1218-22.
Lang ES, Wyer PC, Haynes RB. Knowledge translation: Closing the evidence-to-practice gap. Ann Emerg Med 2007;49:355-63.