|
|
||||||||
Book, Software, and Web Site Reviews |
Diagnostica e Ricerca, San Raffaele S.p.A., Scientific Institute, San Raffaele Hospital, Milan, Italy 20132
This book has the ambitious goal of bridging theory and practice in the complex field of reference intervals. The text comprises only 115 pages but includes a CD-ROM that is a virtual gold mine of both data and computer routines. The CD-ROM provides data from NHANES III (27 900 individuals; 33 analytes) in both SAS® and Microsoft Excel® formats, making it extremely useful as a source of information and as a way of evaluating different computational techniques. Moreover, there are the already calculated reference intervals, both nonparametric and with the robust estimator, as explained in the text on the NHANES data and on the Fernald cohort (a group of residents, monitored for 17 years, living near a uranium ore processing plant). Finally, the SAS programs to calculate reference intervals are provided.
The book is small, but the content is rich and represents a summation of the impressive work of the authors in this field over the last 30 years. It illustrates in detail various methods involved in the determination of reference intervals. Beginning with the guidelines developed by the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) and the Clinical and Laboratory Standards Institute (CLSI; formerly NCCLS), the Guide discusses possible alternative approaches to cope with the 2 main problems encountered when attempting to determine reference intervalsnamely, the difficulty of obtaining a sufficient number of individuals for the reference population and the possible presence of outliers.
The book has many good points. The authors have a clear and profound knowledge of the subject (I very much like the concept of "Medical Heritage"), extremely high statistical competence, and a clear understanding of the everyday problems. It gives an ample array of approaches that could be applied successfully to different situations, and it is a valuable source of bibliographic information as well as of formulas and mathematical definitions.
Nonetheless, I believe that the authors only partially reached their goal. The main limitation for such a book, that wants to be a "users guide", is the level of mathematical competence needed. In particular, chapters 5 and 6, but also many other parts of the book, are written more for a statistician than for the average laboratorian. If most of the mathematics and statistical equations were included in specific appendices, this book would have been much easier to read. A second criticism is that it seems more like a collage of previous articles, with some loose connecting sections, rather than a unified treatise. For this reason, it is not easy to derive practical directions on how to approach the determination of reference intervals in everyday practice. The chapter "Case Histories" gives helpful suggestions, but something like an "executive summary" or a flow chart would have been useful. As a matter of fact, chapter 15, "Conclusions", is too basic and does not provide the reader with sufficient guidance to face real-life problems. It is evident that a single solution for any situation does not exist, but from the books title and its back cover page, I expected clearer statements on which way the authors consider the best to follow (or at least the less dangerous).
Apart from these criticisms, the book is a little gem and cannot be overlooked by anyone interested in reference values.
Footnotes
Editors Note: The author is Chair of the IFCC Committee "Reference Intervals and Decision Limits".
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |