Intended for healthcare professionals


Practitioners of evidence based care

BMJ 2000; 320 doi: (Published 08 April 2000) Cite this as: BMJ 2000;320:954

Not all clinicians need to appraise evidence from scratch but all need some skills

  1. Gordon H Guyatt (guyatt{at},
  2. Maureen O Meade,
  3. Roman Z Jaeschke,
  4. Deborah J Cook,
  5. R Brian Haynes, clinical epidemiologists
  1. Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada L8N 3Z5

    High quality health care implies practice that is consistent with the best evidence. An intuitively appealing way to achieve such evidence based practice is to train clinicians who can independently find, appraise, and apply the best evidence (whom we call evidence based practitioners). Indeed, we ourselves have advocated this approach.1 Now, however, we want to highlight the limitations of this strategy and suggest two complementary alternatives.

    The skills needed to provide an evidence based solution to a clinical dilemma include defining the problem; constructing and conducting an efficient search to locate the best evidence; critically appraising the evidence; and considering that evidence, and its implications, in the context of patients' circumstances and values. Attaining these skills requires intensive study and frequent, time consuming, application.

    After a decade of unsystematic observation of an internal medicine residency programme committed to systematic training of evidence based practitioners,1 we have concluded—consistent with predictions2 —that not all trainees are interested in attaining an advanced level of evidence based medicine skills. Our trainees' responses mirror those of British general practitioners, who often use evidence based summaries generated by others (72%) and evidence based practice guidelines or protocols (84%) but who overwhelmingly (95%) believe that “learning the skills of evidence-based medicine” is not the most appropriate method for “moving … to evidence based medicine.”3

    Because of the amount of time required to make “from scratch” evidence based decisions, evidence based practitioners will often not succeed in reviewing the original literature that bears on a clinical dilemma they face. Thus, two reasons exist why training evidence based practitioners will not, alone, achieve evidence based practice. Firstly, many clinicians will not be interested in gaining a high level of sophistication in using the original literature, and, secondly, those who do will often be short of time in applying these skills.

    In our residency programme we have observed that even trainees who are less interested in evidence based methods develop a respect for, and ability to track down and use, secondary sources of preappraised evidence (evidence based resources) that provide immediately applicable conclusions. Having mastered this restricted set of skills, these trainees (whom we call evidence users) can become highly competent, up to date practitioners who deliver evidence based care. Time limitations dictate that evidence based practitioners also rely heavily on conclusions from preappraised resources. Such resources, which apply a methodological filter to original investigations and therefore ensure a minimal standard of validity, include the Cochrane Library, ACP Journal Club, Evidence-based Medicine, and Best Evidence and an increasing number of computer decision support systems. Thus, producing more comprehensive and more easily accessible preappraised resources is a second strategy for ensuring evidence based care.

    The availability of evidence based resources and recommendations will still be insufficient to produce consistent evidence based care. Habit, local practice patterns, and product marketing may often be stronger determinants of practice. Controlled trials have shown that traditional continuing education has little effect on combating these forces and changing doctors' behaviour.4 On the other hand, approaches that do change targeted clinical behaviours include one to one conversations with an expert, computerised alerts and reminders, preceptorships, advice from opinion leaders, and targeted audit and feedback.5 7 Other effective strategies include restricted drug formularies, financial incentives, and institutional guidelines. Application of these strategies, which do not demand even a rudimentary ability to use the original medical literature and instead focus on behaviour change, thus constitute a third strategy for achieving evidence based care.

    Nevertheless, there remain reasons for ensuring that medical trainees achieve the highest possible skill level in evidence based practice. Firstly, attempts to change doctors' practice will sometimes be directed to ends other than evidence based care, such as increasing specific drug use or reducing healthcare costs. Clinicians with advanced skills in interpreting the medical literature will be able to determine the extent to which these attempts are consistent with the best evidence. Secondly, they will be able to use the original literature when preappraised synopses and evidence based recommendations are unavailable. At the same time, educators, managers, and policymakers should be aware that the widespread availability of comprehensive preappraised evidence based summaries and the implementation of strategies known to change clinicians' behaviour will both be necessary to ensure high levels of evidence based health care.


    We thank the following for their input: Eric Bass, Pat Brill-Edwards, Antonio Dans, Paul Glasziou, Lee Green, Anne Holbrook, Hui Lee, Tom Newman, Andrew Oxman, and Jack Sinclair


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    View Abstract