Article Text

Download PDFPDF

  1. N Barber2
  1. 2Centre for Practice and Policy, School of Pharmacy, 29–39 Brunswick Square, London WC1N 1AX, UK; nick.barber{at}

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


    Why did Tolstoy write “War and Peace”? To illustrate a point of political philosophy—that we often ascribe triumphs to one person (in this case, the Russian General Kutuzov who defeated Napoleon at Tarutino)—and, in doing so, we lose the complex welter of activities and contributing factors, including luck, that made the outcome as it was.1

    I was, perhaps curiously, reminded of this when reading the paper by Kaushal and Bates in this issue of QSHC.2 The authors have provided a valuable function in giving a pragmatic summary of where the USA currently stands with respect to the effects, and potential effects, of information technology (IT) on medication safety. They have differentiated the studies with respect to quality, informed us of what is happening in the areas where the evidence base is scant (but the activity great), and given some indication of the political forces driving change.

    Has the Kutuzov of IT vanquished our Napoleon of medication errors? I think not. How good is IT in the USA, and how good in the rest of the world? The answer is a resounding “don't know”. The rumbling juggernaut of IT is gaining momentum, but the technology assessment programme is lagging behind. We continue to make judgements on technologies implemented at pilot sites that seethe with committed able enthusiasts, but we often fail to evaluate the next stages of roll out. Consequently we do not know why the technology works as it does, where it does. We do not know the extent to which findings are generalisable to different settings; we cannot answer the question “Will it work for me?”. The next phases of technology assessment are, as the authors show, urgently required. There is a pressing need for “realistic evaluation”.3

    IT and robotics have great potential but so, too, do humans. In her paper in this issue of QSHC Dean4 identifies the issues of reducing prescribing errors using human, rather than computer, systems. There is a need first to identify errors, and then for a system to bring them to the attention of the prescriber and others so that learning and improvement can take place. A significant problem is that individuals who make an error are often ignorant of that fact, so others have to identify it and feed back to the prescriber; this, in turn, can lead to interdisciplinary tensions unless the organisational culture is right. The paper usefully identifies the broad issues; the challenge is to create the human systems, particularly in the diffuse structure of primary care in countries such as the UK. This would seem an area in which IT has great potential to work with human systems, if only we can get it right.