Book Review

By Barrie M Margetts and Michael Nelson
Oxford: Oxford University Press, 1997; xv + 451 pp;
ISBN 0-19-262739-2 (paperback); $79.95 (CAN)

Design Concepts in Nutritional Epidemiology is a valuable addition to the burgeoning field of nutritional epidemiology. The newly released second edition has added chapters on qualitative and sociological measures, anthropometric measures, gene-nutrient interactions and cross-sectional studies. Chapters from the first edition have been revised to take recent developments into account. There is a brief, five-page introduction that touches on the importance of nutritional epidemiology, general progress that has been made in the field and the role of nutritional epidemiology in public health nutrition. This section has been substantially revised from the first edition in order to "bring a more practical focus to the theoretical concepts around study design."

This second edition of the book has been used as a course text for the European postgraduate summer course in public health nutritional epidemiology, and it is divided into three parts. Part A (chapters 1-4) deals with concepts in study design; Part B (chapters 5-11) discusses problems inherent in the measurement and interpretation of a wide variety of variables relevant to nutritional epidemiology; and Part C (chapters 12-16) describes the application of various research designs to studies of nutritional epidemiology. Design Concepts in Nutritional Epidemiology, differs greatly from the current standard text, Willett's Nutritional Epidemiology,¹ in that it does not assume any prior knowledge of epidemiologic methods.

Chapter 1 provides an overview of the principles of nutritional epidemiology. It covers, in part, the objectives of nutritional epidemiologic research, types of epidemiologic studies and measures, and the interpretation of epidemiologic research. Readers familiar with the general concepts of epidemiology may wish to read quickly through this chapter, focusing most of their attention on the examples of nutritional epidemiology. The second chapter covers the design, planning and evaluation of nutritional epidemiologic studies. Chapter 3 considers issues of sampling, study size and power as they relate to different types of epidemiologic studies. It succeeds as a straightforward introduction; however, researchers seeking more depth will need to consult additional texts. "Covariate measurement errors in nutritional epidemiology: effects and remedies" is the title of Chapter 4. While the potential problems are well covered, the remedies, unfortunately, seem too theoretical or out of the reach of most researchers. Those looking for simple, easy-to-implement solutions will be disappointed. A useful feature of Part A is that readers are guided toward additional information on whatever they are reading, should such information exist in subsequent parts of the book.

Part B begins with a comprehensive and easy-to-read chapter focusing on the estimation of nutrient intakes from food consumption surveys using food composition tables. Included in this chapter are references to World Wide Web home pages where the reader can go for additional information. Ways of assessing food consumption and nutrient intake are discussed in Chapter 6. The first section of this chapter looks at the strengths and weaknesses of various methods for assessing household diets, while the second section outlines measures of diet in individuals. The latter section includes a short, interesting look at the problem of underreporting of energy intake and why energy adjustment is sometimes necessary. Chapter 7 presents the reasoning behind the use of various kinds of biomarkers (blood, urine, fecal, etc.) as well as a detailed section on biomarkers for different minerals, vitamins and lipids. Researchers and other health professionals involved in the use of biomarkers will find this chapter invaluable, and at 71 pages, it is the longest chapter in the book. The validation of dietary assessment (Chapter 8) and sociodemographic and psychosocial variables (Chapter 9) are subsequently examined. Chapter 10 reviews appropriate anthropometric measurements and general problems associated with such measurements. Part B concludes by providing approaches to gene-nutrient interaction research and insight into its possible role in nutritional epidemiology.

Chapter 12 opens the last section of the text by presenting a thorough examination of the strengths and weaknesses of ecological studies, including ways of collecting and analyzing ecological data. Chapter 13 is similarly well done and very practical in nature. It provides some useful guidance for readers looking for help in designing a cross-sectional study or for those trying to choose the most appropriate dietary assessment method to use for their study. A short chapter follows on cohort studies, highlighted by examples from different phases of the life cycle and a checklist for planning and analyzing such studies. While the next chapter (15) provides a good summary of the ins and outs of case-control studies, it would be considerably better served if it had greater application to nutritional epidemiology. The final chapter of the text focuses on the design and analysis of experimental studies.

Individual chapters were written by subject area experts, with 24 authors contributing in total. Although a rather uniform style of writing was achieved throughout the text, more effort could have gone into developing standard purpose and conclusion sections for each chapter. In addition, because most of the text was written by people in the United Kingdom, the book sometimes takes on a British slant, particularly in the section dealing with existing sources of nutritional data. However, these limitations don't significantly detract from the overall usefulness of the book.

Design Concepts in Nutritional Epidemiology is a valuable resource for epidemiologists who lack a strong background in nutrition and would be a welcome addition to the bookshelf of anyone involved in nutritional epidemiology.

Reference

1. Willett W. Nutritional epidemiology. Oxford: Oxford University Press, 1990.

Larry Ellison
Behavioural Risk Assessment, Cancer Bureau
Laboratory Centre for Disease Control
Health Canada, Tunney's Pasture
Address Locator: 0601C1
Ottawa, Ontario  K1A 0L2