Interpretation and Uses of Medical Statistics

Abstract

There are now many books which seek to introduce basic concepts of statistics, and their application to medicine, to readers who have had no formal training in statistical theory or methods.This book has particular merit, witnessed by its reappearance here in a third edition.It is likely to appeal to a general readership through its breadth, humanity and good humour.The problem faced by all such texts is that authors feel constrained to stop short ofdetailed description ofimportant and widely used methods, such as analysis of variance and covariance, believing, probably quite correctly, that these will not be understood by most of the intended general, nonmathematical audience.However, there is concern that, with advanced computing tools so widely available, such an audience will inevitably encounter and have to try to interpret such methods.The text here achieves clarity of advice without avoiding difficult issues.There is a useful set of Appendices-par- ticularly Appendix C which summarises the commonly encountered statistical tests in terms of the situations in which they are applicable, the assumptions underlying their use, the hypotheses tested and the actual methods employed.The treatment is also very practical-a good example is the discussion of risk and the relationship ofexperimental design to the types of questions which can be answered.Chapter 12 on computing is not up to the standard of the rest of the book-it is perhaps slightly misplaced given the stated objectives, as the topics introduced spread far beyond medical statistics.However, overall, it is an excellent and readable book covering a difficult area.