Article Text

Download PDFPDF

Health Services Research: Avoiding the Pitfalls
  1. T Sheldon
  1. Department of Health Sciences, University of York, York YO10 5DD, UK; tas5{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.

    Edited by Huw T O Davies. London: Quay Books, Mark Allen Publishing Ltd, 2001. £14.99, 168 pp. ISBN 1 85642 195 3

    Health professionals are increasingly expected to refer to research evidence when making policy and practice decisions. However, there is considerable variation in the quality and applicability of research. To use research sensibly it is therefore important not only to access the evidence, but also to judge its adequacy for the question in hand.

    This book collects together a series of articles published in Hospital Medicine since 1998, which seek to provide guidance on the most widely used healthcare research methods. Rather than being a comprehensive textbook, it highlights probably the most important issues such as measurement, representation, association versus causation, and bias (a particularly strong chapter which more logically should come before variation). The book then deals with the most commonly encountered research designs such as trials, systematic reviews, and decision analysis. Here it would have been helpful to explain more clearly the appropriateness of designs to different sorts of questions.

    The treatment of these issues is clear and non-technical. Indeed, at times it is too superficial to be really understandable—for example, a complete novice may not be able to understand the few pages on p values and confidence intervals. Similarly, the concise treatment means that some important aspects have been omitted—for example, the chapter on screening does not deal with the important issue of adverse effects.

    These and other sections would have benefited from some more examples from the literature dealing with real health issues. The editor unfortunately did not take the opportunity of the space offered by a book to rework some of the articles which, of necessity, have been more constrained by space. Strangely, the chapter on benefits is mainly devoted to economic analysis which is about the resources needed to generate benefits. More attention could have been given to the sorts of outcomes that can be measured and their relevance to decision making.

    Despite these weaknesses this is a useful, simple, and easy to read introduction to research methods and the critical appraisal of research. It was written for clinicians and I suspect it will not have much wider appeal for managers and lay readers, which is a pity given the calibre of the editor and other authors. Perhaps a next edition will supplement the text with health services research applied to questions of organisation, management, and service delivery. Those looking for a more rigorous treatment of this material should probably look elsewhere.