Introductory Medical Statistics, 3rd edition / Edition 3

by Richard F. Mould

Overview

Introductory Medical Statistics, now in its third edition, is an introductory textbook on basic statistical techniques. It is written for physicians, surgeons, radiation oncologists, medical physicists, radiographers, hospital administrators, medical statisticians in training, biochemists, and other professionals allied to medicine. It is suitable as a teaching text for clinicians working towards their professional examinations. It is also suitable for Maters degree courses in medical physics.

The third edition has been extensively revised and expanded to include:

Clinical trial design and analysis]

Multivariate analysis

Cox proportional hazards model

McNemar, Wicoxon, Mann-Whitney, Kruskal-Wallis, Mantel-Haenszel, and Kappa tests

Kaplan-Meier survival rates

Sensitivity and Specificity

Specification of treatment success, cure, and quality of life

Risk specification

Case-control and cohort epidemiological studies

Glossary of terms

The major change has been the advent of personal computing, so people rely on the power of their machine, and its software to number crunch. What is missing is that the software may not use the appropriate statistical error standard - Dick Mould

Product Details

ISBN-13:	9780750305136
Publisher:	Taylor & Francis
Publication date:	01/01/1998
Series:	Series in Medical Physics and Biomedical Engineering
Edition description:	REVISED
Pages:	422
Product dimensions:	6.00(w) x 9.00(h) x (d)

Table of Contents

Data presentation. Describing cuves and distributions. The normal distribution curve. Introduction to sampling, errors and accuracy. Introduction to probability. Binomial probabilities. Poisson probabilities. Introduction to statistical significance. The chi-squared test. The Fisher exact probability test. The t-test. Difference between proportions for independent and for non-independent (McNemar's test for paired proportions) samples. Wilcoxon, Mann-Whitney and sign tests. Survival rate calculations. The logrank and Mantel-Haenszel tests. Regression and correlation. Analysis of variance. Multivariate analysis: the Cox proportional hazards model. Sensitivity and specificity. Clinical trials. Cancer treatment success, cure and quality of life. Risk specification with emphasis on ionising radiation. Types of epidemiological study: case-control, cohort and cross-sectional. Glossary of rates and ratios: terminology in vital statistics. References. Index.

Editorial Reviews

Reviewer: Raymond G. Hoffmann, PhD (Medical College of Wisconsin)
Description: This is an introduction to statistics for physicians, fellows, and medical residents. It has a comprehensive survey of statistical methods and includes a substantial amount of clinical epidemiology that should make it useful for the novice and experienced researcher. Much of this material was not present in the previous edition.
Purpose: The book is intended to instruct physicians who need to know and understand statistics. A book like this is needed, since most busy clinicians need a reference that they can use for designing a study or evaluating a research paper. Many of the current books are written only by statisticians and do not have the clinical insight of this author. Few are as comprehensive. In some cases, the author still has an excess of computational methods that are no longer needed. For example, he presents both the Wilcoxon and Mann-Whitney methods (which are quite different) for computing the non-parametric two sample test for means.
Audience: This book is intended to appeal not only to radiation oncologists, radiologists, and medical physicists, but also to other medical specialists who need to know and understand statistics in the current age of computer software for statistical computation. The author has hit his target well. He includes many clinical examples that help illustrate where each statistical method is useful. The author is certainly knowledgeable about the use of statistics, but just as the statistical authors of comparable books need a clinical co-author, this author needs a statistician co-author to correct the many typographical and conceptual errors. For example, there is no discussion of the unequal variance t-test; yet nearly every statistical program presents the results of both of these computations separately and expects the user to choose the correct one.
Features: This book could be used in a self-study or remote education course. It includes an introduction to probability distributions, some useful non-parametric tests, linear regression, survival analysis including the proportional hazards model, an extensive discussion of clinical trial methodology, and examples of the different types of epidemiologic studies. It has a particularly unique discussion of the definition of cure in the case of cancer as well as the use of quality of life measures in evaluating therapeutic alternatives. As is not surprising given the author's medical physics background, there is a comprehensive discussion of the concept of risk in the context of ionizing radiation in the environment (Hiroshima and Chernobyl). It is comprehensive and includes many historical anecdotes to add to the statistics and epidemiology. There are also many humorous anecdotes and cartoons which educate as well as lighten the learning process. However, the misprints (even in the section on combination of errors which should be the bread and butter of a medical physicist), the mysterious graphs of critical values of some of the test statistics, and the nonstandard notation (e.g., for combinations and permutations) are more than annoying.
Assessment: This book has so much potential that it is a shame to see it marred by misprints, errors, and unevenness in the quality of the statistics. The author has a good writing style, and the additions since the previous edition have made this a much stronger book. It could have been used to teach fellows as well as specialty physicians such as those in preventive medicine who need to know more statistics; an area that has few appropriate texts. One comparative text is Altman's Practical Statistics for Medical Researchers, (Chapman and Hall) but it is much more difficult for a physician to read and understand. The examples in the preventive section are more clinical and always have the flavor of being real data or based on real data. Another comparative text would be Dawson-Sanders and Trapp's Basic and Clinical Biostatistics (Lange), whose authors are a combination of a physician and statistician. The clinical examples in Dawson-Sanders are discussed in more depth at the beginning of the chapter than in this book, but the topics are not as comprehensive. Dawson-Sanders has some of the clinical epidemiology that this book has; Altman has less clinical epidemiology than either of these two texts.

Doody's Review Service