Annals of Occupational Hygiene Advance Access originally published online on August 6, 2004
Annals of Occupational Hygiene 2004 48(6):581; doi:10.1093/annhyg/meh058
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2004 British Occupational Hygiene Society Published by Oxford University Press;
Book Review |
Interpreting Epidemiologic Evidence: Strategies for Study Design and Analysis David A. Savitz. Oxford University Press, Oxford, 2003. ISBN 0-19-510840-X. Price £34.50.
Paediatric Epidemiology Group, University of Leeds, 30/32 Hyde Terrace, Leeds LS16 5HJ, UK
This book is a welcome addition to the epidemiological literature. Its author has an excellent academic pedigree, the book is clearly and logically structured, and the text is comprehensible. The author's range and width of experience in epidemiology makes the manuscript current and relevant. The overall aim of the book is to provide epidemiologists with source material for linking methodological principals to research practice, either at the design or the evaluation stage. Examples from the published literature are scattered throughout the text and generally provide good illustrations of the topic being addressed.
The first chapter gives an insight into ‘the nature of epidemiologic evidence’, and what can and cannot be inferred from results of epidemiologic studies. The section on the relevance and importance of epidemiology to policy makers is useful, and the note of how epidemiology is open to overinterpretation by the media very pertinent. The chapters on bias give a comprehensive coverage of the practical difficulties which are realities of epidemiological research, and offer some solutions in terms of study design. The chapter on bias due to loss of study participants focuses on an important area that does not often receive such attention and it is therefore of value when reviewing published work. A comprehensive examination of confounding, measurement and classification of disease and exposure gives the reader a good solid overview of these topics, whereas the section on random error is more challenging. The final chapter of the book looks at ‘characterisation of conclusions’—an aspect of epidemiology which is often missing from many text books; it offers an overview of the interpretation of epidemiologic evidence and its pitfalls, and is useful to specialist and non-specialist. Overall, this text book is a solid constructive and helpful addition to the literature on epidemiological methods, offering some new insights along with the basics and fundamental methodology. It has particular value for those with some experience in epidemiology, but may be somewhat impenetrable for students and novice epidemiologists.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||