Family Practice Vol. 19, No. 2, 213
© Oxford University Press 2002
Book Review |
Making sense of statistics in healthcare.
Anna Hart. (170 pages, £19.95.) Radcliffe Medical Press Ltd, 2001. ISBN 1-85775-472-7.
Applied medical statistician who has collaborated extensively with clinicians. University Lecturer in Medical Statistics in the Department of Primary Health Care, University of Oxford
This highly attractive book manages to combine statistical accuracy and clarity with an easy and informal style, while covering a useful and relevant range of material. The author is Principal Lecturer in Statistics at the University of Central Lancashire, and has clearly grappled successfully with the problem of how to teach basic statistical concepts in an interesting and accessible way while demanding no more than everyday numeracy.
In addition to the conventional topics found in introductory statistics texts—charts and graphs, measures of average and variability, confidence intervals and hypothesis testing—this book includes a section on the concepts used in evidence-based medicine, such as odds ratios, relative risks and numbers needed to treat. These are concepts that anyone now working in health care or medical research needs to be familiar with. The copious use of figures as well as the cartoon drawings add to the accessibility of the book, and it is clear that a lot of thought has been put into the layout. It is a pity that some errors in the figures were not spotted before publication, and one figure has been duplicated, which at one point made very confusing reading.
The book is unusual in including a chapter on liaising with a statistician. The chapter starts with a long list of comments not to make, from "It's a very simple question—it won't take you long to work it out" to "I don't know what keys to press, but I know that's what you're good at"! This section will cheer up harassed medical statisticians everywhere, but its more serious purpose is to explain that statistics encompasses far more than merely analysing data (a common misconception) and to clarify just how much a statistician can contribute to the success of a research project, if involved right at the start. In fact, the book emphasizes throughout that making sense of statistics does not just mean knowing the definition of statistical terms or how to carry out a hypothesis test. It includes realizing the overwhelming importance of good study design, always having a common-sense approach to problems and understanding that statistical decisions have an important ethical dimension. Also, it demands continuing critical appraisal (including critical self-appraisal) in conducting and evaluating research.
If you are already comfortable with the central ideas of statistics and want to know more about the detail of various methods, then this is not the book for you. But as an introductory text, or a friendly reference if your grasp of statistical concepts is rusty, then go out and buy it.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||