American Journal of Epidemiology Vol. 152, No. 1 : 96
Copyright © 2000 by The Johns Hopkins University School of Hygiene and Public Health
BOOK REVIEW |
Survey Methods in Community Medicine: Epidemiological Research, Programme Evaluation, Clinical Trials. Fifth Edition - By J. H. Abramson and Z. H. Abramson
Division of Clinical Epidemiology University Hospitals of Geneva Rue Micheli-du-Crest 25 CH-1211 Geneva 14, Switzerland
Have you ever dreamed of an epidemiology textbook that is free of 2 x 2 tables, equations (except in endnotes), scientific graphs, or math? Then here is a book for you. Abramson and Abramson have written 400 pages of fine prose on the practicalities of conducting epidemiologic research.
I have not found information in this book that I wouldn't know where to find in other textbooks, but I was essentially struck by the way known epidemiologic concepts and strategies were formulated. The book is clear and easy to read. Not relying on algebra and formulas is a great quality of the text, since it forced the authors to emphasize the literary qualities of their work.
It has been a tendency of recent epidemiology textbooks to translate concepts in mathematical terms. See, for example, the way the concept of "relative risk" is presented in two outstanding texts:
- A ratio measure of association (or effect) is a frequency measure for one (the ith) exposed group (E_I) divided by a comparable frequency measure for an unexposed or reference group (I = 0) (1
, p. 143).
- If D denotes the presence of disease, D– denotes the absence of disease, E denotes the presence of exposure, E– denotes the absence of exposure, and Pr(.) denotes the probability of an event, then the risk ratio may be expressed in terms of conditional probabilities as: risk ratio = Pr(D/E)/Pr(D/E–) (2
, p. 35).
- If D denotes the presence of disease, D– denotes the absence of disease, E denotes the presence of exposure, E– denotes the absence of exposure, and Pr(.) denotes the probability of an event, then the risk ratio may be expressed in terms of conditional probabilities as: risk ratio = Pr(D/E)/Pr(D/E–) (2
These definitions can be compared with the following one given by Abramson and Abramson:
- The strength of an association is a measure of its importance. If anemia is found among 30% of pregnant women and 2% of nonpregnant women, this marked disparity (a difference of 28%, or a ratio of 15) indicates a strong and important relationship between anemia and pregnancy (3, p. 306).
The title of this book tends to be misleading, as the authors acknowledge in the preface, because it doesn't focus on surveys. Indeed, I was somewhat disappointed not to find a detailed chapter on surveillance, particularly surveillance of risk factors, a topic in which there is growing interest and which is key for the development of community-oriented primary care (to which Abramson and Abramson dedicate their chapter 34). To my knowledge, no manual synthesizes the methods of risk factor surveillance relative to data collection, representation, and monitoring, as well as comparison between communities, populations, or countries. The authors may wish to consider this for the next edition. A future edition might also benefit from expansion of the part dedicated to the Internet as a means of communication, a support system for shared databases, and a source of information.
Epidemiology students vary in their ways of thinking. Therefore, epidemiology teachers need to rely on a variety of texts explaining the same concepts and methods from different perspectives. Because it increases the diversity of the material available for teaching epidemiologic methods, Abramson and Abramson's text is a valuable component of the existing set of epidemiology manuals.
NOTES
ISBN 0-443-06163-7, Churchill Livingstone, 1–3 Baxter's Place, Leith Walk, Edinburgh EH1 3HF, United Kingdom (Telephone: 800-545-2522, Fax: 800-568-5136), 1999, 427 pp., $39.00
REFERENCES
- Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic research: principles and quantitative methods. New York, NY: Van Nostrand Reinhold Company, 1982.
- Kelsey JL, Whittemore AS, Evans AS, et al. Methods in observational epidemiology. 2nd ed. New York, NY: Oxford University Press, 1999.
- Abramson JH, Abramson ZH. Survey methods in community medicine. 5th ed. Edinburgh, United Kingdom: Churchill Livingstone, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||