Evidence Based Medicine: 25 Years Later

We were saddened to learn earlier this year, of the passing of Professor David Sackett. Widely recognized as the father of evidence based medicine, Professor Sackett confronted tough criticism in advancing the cause of evidence based medicine during the early nineties. During his four years at the Centre for Evidence Based Medicine at Oxford, Sackett’s team produced an array of books, articles, curricular and pedagogical practices, and software techniques which remain foundational to EBM's teaching and learning.

Evidence based medicine refers to the practice of incorporating “current best evidence” when determining care for individual patients [1]. Clinicians use their clinical expertise to specify the problem and the evidence necessary to solve it; the evidence itself, however, makes reference to biostatistics and epidemiology [2].More generally, evidence based medicine defends the view that clinicians should use both their clinical expertise and the findings of general clinical research in their practices, and that neither alone is sufficient to provide an appropriate level of care [1].

In a widely-cited paper, Sackett explains the irony of having to combat criticisms of championing a practice that was simultaneously ‘too old hat’ and yet also ‘too revolutionary’ [1]. In celebration of David Sackett, we consider Evidence Based Medicine in the early 1990s, and consider new developments twenty-five years later.  

Evidence Based Medicine (Then)

Undoubtedly, the world of biopharmaceuticals was a different place twenty-five years ago. In 1989 Bristol-Meyers was still in the process of merging with Squibb, Apple and Windows were each celebrating their five year anniversary, and Timothy Berners-Lee had just launched the World Wide Web. (Incidentally, Rain Man had just won best picture at the Oscars and Cytel was about to release StatXact 1.0.)

The concept of Big Data was forever on the cusp, yet before the widespread use of the Internet and other information technology, finding and collating the newest clinical research was a difficult demand on anyone, much less on busy clinical practitioners. However, a good many doctors were taking the time to keep abreast of the latest research and apply a version of evidence based medicine to their clinical practice. According to Sackett and colleagues, about 53% of clinicians in Great Britain already used a version of evidence based medicine when making diagnoses [3].   

The nature of the game was to make this standard practice. Unfortunately, with a near fifty-fifty split within the practice of the establishment, criticism towards evidence based medicine came from several angles. Some said it was too common a practice to be called groundbreaking. It seemed to these critics that practitioners everywhere were already engaging in evidence based medicine, without the need for standardization or a change in pedagogy.

The other fifty percent claimed that evidence based medicine was too revolutionary to become standard medical practice. In particular, there was the worry that doctors did not have enough time to read the latest research. Sackett and his colleagues agreed that a doctor would have to read seventeen articles a day to keep up with clinical research [2]. However, they believed this informational difficulty could be overcome with a publication of sets of abstracts and commentaries, such as those found in the American College of Physicians’ Journal Club, or the journal Evidence Based Medicine, founded by Sackett and colleagues. [2]

 There was further worry that the authority of doctors as clinical practitioners would be tacitly undermined if subsumed under clinical research. Accusations spread of Sackett and his colleagues promoting ‘cookbook’ medicine which limited clinical practice [1], and that they were favored by the establishment for their purported ability to cut costs. [4]

Evidence Based Medicine (Now)

Nowadays it’s difficult to imagine anyone objecting to evidence based medicine on such strident grounds. However, newer directions in evidence based medicine are an exciting category of observation and experiment. Here are exciting new developments within the practice:

• Stratified Medicine: The combination of clinical research and clinical expertise necessary to practice evidence based medicine has revealed the importance of treatment heterogeneity when applying the findings of clinical research to patients. Since traditional randomized clinical trials provide information on the average patient, applying findings to individual patients has furthered the need for population subgroup analysis in clinical research [5] [6].
• Comparative Effectiveness Research: The rise of comparative effectiveness research has created further questions about the relationship between population level empirical evidence and decision-making for individual patients. There is still a reasonable amount of debate regarding where (if anywhere) evidence based medicine and comparative effectiveness research overlap [7].
• Patient Choice and Values: The practice of evidence based medicine continues to raise critical questions about the role of patient choice and preference in determining appropriate treatments. Patients are in some cases willing to accept greater risks than clinicians would on their behalf, even when presented with the same evidence. Furthermore, an understanding of patient preferences is shown to affect patient adherence to various treatment regimen. As a result, the framework of evidence based medicine is better suited to engage patients in the deliberative process of improving their own health. [8]

 [1] Sackett, David L., et al. "Evidence based medicine: what it is and what it isn't."BMJ: British Medical Journal 312.7023 (1996): 71.

[2] Davidoff, Frank, et al. "Evidence based medicine." BMJ: British Medical Journal 310.6987 (1995): 1085.

[3] Sackett, D. L., et al. "Inpatient general medicine is evidence based." The Lancet 346.8972 (1995): 407-410.

[4] Grahame-Smith, David. "Evidence based medicine: Socratic dissent." BMJ: British Medical Journal 310.6987 (1995): 1126.

[5] Kravitz, Richard L., Naihua Duan, and Joel Braslow. "Evidence‐Based Medicine, Heterogeneity of Treatment Effects, and the Trouble with Averages."Milbank Quarterly 82.4 (2004): 661-687.

[6] Kent, David M., and Rodney A. Hayward. "Limitations of applying summary results of clinical trials to individual patients: the need for risk stratification."Jama 298.10 (2007): 1209-1212.

[7] Luce, Bryan R., et al. "EBM, HTA, and CER: clearing the confusion." Milbank Quarterly 88.2 (2010): 256-276.

[8] Haynes, R. Brian, Peter J. Devereaux, and Gordon H. Guyatt. "Clinical expertise in the era of evidence-based medicine and patient choice." Evidence Based Medicine 7.2 (2002): 36-38