Arch Dis Child 92:644
  • Archimedes

Towards evidence based medicine for paediatricians

  1. Bob Phillips
  1. Evidence-based On Call, Centre for Evidence-based Medicine, University Department of Psychiatry, Warneford Hospital, Headington OX3 7JX, UK; bob.phillips{at}

    Sketch it out

    We’ve all been in (or observed, at least) a situation where a congenital heart defect is being explained to non-medical folk: parents, teachers or social workers. The first thing the doctor tends to reach for is a pen and a blank piece of paper, and starts to sketch a schematic heart.

    This reasonable and rational use of graphics to explain a complex phenomenon hasn’t been tested against textual descriptions (as far as I can tell) and yet it is practised and known to work well.

    Why then, in describing the complex interventions often used in non-drug trials, do we insist on relying on the written word? When you’ve a trial involving provision of increased physical activity in nurseries, coupled with education packs for the parents of the children, applied over 24 weeks, wouldn’t it be easier to look at a drawing which showed who got what, when, and how long it lasted for? This is an idea proposed by Perera et al1 and should be taken into practice immediately by all those who appraise research. When you pick up a paper, pick up a pen and draw out who is doing what, to whom and when; it makes murky heavily written studies become understandable and can highlight aspects to be careful or critical about. (After all, if you can’t draw what happened, you can’t understand it and probably shouldn’t be believing what the paper tells you.) Go on – try it today – flick back three pages and turn the study you find into a work of art.


    • Bob Phillips