Intended for healthcare professionals


Academic medicine: time for reinvention

BMJ 2004; 328 doi: (Published 01 January 2004) Cite this as: BMJ 2004;328:48

Research needs researching

  1. Jonathan Grant, associate programme director (jgrant{at},
  2. Steve Hanney, research fellow,
  3. Martin Buxton, professor of health economics
  1. RAND Europe, Grafton House, Cambridge CB5 8DD
  2. Health Economics Research Group (HERG), Brunel University, Uxbridge, Middlesex UB8 3PH

    EDITOR—We agree with Bell that advocating funding for clinical research needs to become evidence based.1

    With colleagues we recently explored the relations between basic research, clinical research, and clinical practice.2 3 In the first report we tried to replicate Comroe and Dripps's seminal study as reported in Science in 1976,4 tracing back from current clinical practice to the knowledge behind the advance.2 Comroe and Dripps concluded that 40% of all research articles judged to be essential for later clinical advance were not clinically oriented at the time of the study and that 62% of key articles reported basic research. Their paper has often been used (albeit at times implicitly) in support of the increased funding for basic biomedical research in the United Kingdom and elsewhere over the past two decades.

    We found we were unable to repeat the study method, thus confirming earlier doubts about it5 and raising questions about its reliability, validity, and applicability, at least for current circumstances.

    In recognising the difficulties in tracing backwards from clinical practice, we undertook a preliminary study that tries to work forwards by tracing the impact of a range of research conducted 20 years ago.3 In this study we explored several methods that could help to identify the benefits from previous research and thus potentially be used to develop an evidence base to support funding decisions for research and development.

    Our work has only begun to scratch the surface of this important but complex issue. Nevertheless, we think that research funding agencies should accept the need for a firm evidence basis for their policies and therefore support researching research, as Smith put it in 1987.5


    • Competing interests The authors are in receipt of research grants from a variety of funding agencies.


    View Abstract