Editor's Choice

More research is needed—but what type?

BMJ 2010; 341 doi: http://dx.doi.org/10.1136/bmj.c4662 (Published 25 August 2010) Cite this as: BMJ 2010;341:c4662
  1. Fiona Godlee, editor, BMJ
  1. fgodlee{at}bmj.com

    My move from clinical medicine to the BMJ as a junior editor 20 years ago meant learning a host of new rules. The best of them are still part of our every day on the journal. When judging research, we were taught to ask three questions on behalf of the reader: “do I understand it, do I believe it, do I care?” though not always in that order. When editing we were taught to think of the reader not the author. “If in doubt cut it out” may not be a good maxim for surgeons but it’s an excellent one for editors.

    Adjectives were frowned on and no article was allowed to end with the redundant claim that more research is needed. We wanted something more informative—exactly what types of research are needed, addressing what specific questions?

    The types of research the editors of the BMJ are interested in—on behalf of readers—are those that will benefit patients and will meet the needs of decision makers. Neither idea seems radical, but two articles in this week’s journal explain why these concepts have taken on new importance.

    Geoff Watts is initially baffled by the new emphasis on translational research (doi:10.1136/bmj.c4363). Hasn’t the idea that research findings should be exploited for the benefit of people with diseases always been the point of medical research? But through the people he interviews he warms to his subject. A multidisciplinary team based approach could “unblock the pipeline” of medical innovation, he concludes, though we won’t know if it has succeeded for 10 years or more.

    Bafflement was my initial response to the furore over the push for more comparative effectiveness research in America, when the need for it is so obvious. And even after Sean Tunis and Stephen Pearson have patiently explained what it was all about (doi:10.1136/bmj.c3615), I’m still amazed. But see what you think.

    Elsewhere in the journal our authors identify some specific research questions that need addressing. Next to the summary of new NICE guidelines for managing heart failure, Philip Jong and colleagues use the Uncertainties page to highlight the lack of evidence on how to manage heart failure with preserved ejection fraction (doi:10.1136/bmj.c4202). In their clinical review on evaluating the child presenting with an acute limp, Daniel Perry and Colin Bruce ask what is the best clinical algorithm to distinguish between transient synovitis and septic arthritis (doi:10.1136/bmj.c4250).

    Eleanor Love and colleagues ask, and attempt to answer, how soon couples should try to conceive again after a miscarriage (doi:10.1136/bmj.c3967) and in a linked editorial Julia Shelley explains why this question has proved such a challenge for researchers (doi:10.1136/bmj.c4181). And building on the study by Eric Chen and colleagues on when to stop maintenance treatment after a first episode of psychosis (doi:10.1136/bmj.c4024), Jonathan Craig and colleagues ask why there is so little good evidence on one of the key questions in clinical care: how long to treat chronic conditions (doi:10.1136/bmj.c4102).


    Cite this as: BMJ 2010;341:c4662


    View Abstract