Statistics from Altmetric.com
Journal clubs are probably the easiest place to get evidence based medicine (EBM) started. Most attendees will be familiar with this being a place for examining papers, and it might even have a regular slot on the timetable. We’ve found that converting a traditional journal club to an evidence-based one improved attendance and interest in the event.1 It seems to have a lasting effect too, with ex-club members recalling the principles of EBM and the key points of critical appraisal two years after leaving the hospital (L Etheridge and H Jepps, personal communication).
An evidence based journal club is split into three uneven sections (when it’s up and running). A question is devised one week, its search results looked at the next, and the week after sees an analysis of the best paper(s). In each session, the first five minutes are used to review the results of a search, and a paper selected. The next 40–45 minutes are used to discuss a paper, and the last 5–10 minutes are used to identify and clarify a clinical question to roll onwards. In the first few weeks, teaching papers and “planted” questions can be used to get the principles in place. It also helps if the group leader can make sure that the initial questions being asked are likely to have an answer—it can be highly dispiriting to have a three week run of “no evidence for this question”.2 The problems you are likely to face when doing this include:
Lack of answers to the questions asked.
Research nihilism—no paper is perfect so no answer can be given.
Access to papers upsetting your timetable.
Staff changes and revisiting the basics.
The best defences to these problems are encapsulated by Baden-Powell’s motto: “Be prepared”. Have to hand the idea that only about 1/9 questions will have a decent answer; and have a few questions up your sleeve to tickle people with. Push the idea of “how good is the study” rather than “how poor is the study”. Keep a store of papers you’d use for teaching to fill in awkward gaps, and to broaden your understanding of other sorts of studies. (You’ll probably find you get lots and lots of therapeutic questions and not very many diagnostic, prognostic, or aetiologic ones.) And to get around the problem of staff changing, try to empower the group to teach itself as it goes along.