31 of 31 people found the following review helpful:
5.0 out of 5 stars
different approach to teaching biostatistics in health sciences, February 9, 2008
This review is from: Statistics in Medicine (Hardcover)
This very recent book is already getting rave reviews from health professionals particularly in the US Navy where Dr. Riffenburgh is known. I recently met Dr. Riffenburgh at an ASA chapter meeting in San Diego and again at the Joint Statistical Meetings in Indianapolis this past August. The Foreword by Vice Admiral Richard Nelson sets the stage for the book and praises Dr. Riffenburgh.
I am a biostatistician working in the medical device industry. Last year I taught an introductory course to health science majors at Cal State Long Beach. Had Dr. Riffenburgh's book been out I may have used it. In the future if I teach the course again before my own book comes out I would use it.
Dr. Riffenburgh has a wealth of experience working on clinical trials and medical problems that require statistical analysis. He has a good fundamental understanding of statistical methods and a sense from his experience as to what topics are important to medical researchers. Although he was not formally trained as a statistician he understands statistical methods and the subtleties of the subject. He is sympathetic to the students who do not have strong math backgrounds and conveys the ideas clearly to them with the aid of several carefully selected data sets.
The book is very unique in that it is constructed to be two books in one. He wants to meet the needs of students who need to understand the basics of biostatistics so that they can appreciate, understand and evaluate the medical literature. Part I, consisting of chapters 1 through 9 is designed for them. The second need is for a reference text for physicians and other medical researchers who will need to use statistical methods in their work and publications. Chapters 10 through 21 are designed for them. As he states in his preface the text is aimed to be a 30/90 book. By this he means that it covers 30% of the statistical methods in common use but based on survey articles of methods used in medical journals it covers 90% of those techniques. He claims that 82% are covered in Part I and the remaining 8% in Part II.
Part I is an elementary text that is written in a tutorial fashion and introduces concepts through the data examples. It is well written and does a good job of explaining important ideas such as the difference between clinical and statistical significance, the interpretation of confidence intervals and the idea that the classical statistical approach to hypothesis testing is designed to reject the null hypothesis and not to "prove" it. Consequently he warns the readers not to conclude that the null hypothesis is accepted just because a particular test does not reject it.
Part I covers most important elementary topics. Part II repeats much of the material in Part I but presents it a little differently. In these chapters the material is presented in easy to use reference style using a lot more formulae. The level of the second half is a little higher and includes some more advanced topics (the other 8%). Topics covered in Part II but not Part I include sequential methods, meta-analysis, time series methods and survival analysis. Actually survival analysis is briefly discussed in Part I under epidemiological studies.
Another unique feature is the section of chapter summaries at the end of the book. This is sort of a Reader's Digest summary of the book that highlights the key results for those who want a reference source but may not have the patience to read through the material or those who after taking the course want a quick refresher. I have never seen this done before but I think it may have some value.
I believe that Dr. Riffenburgh has succeeded in his goals and I admire the fresh approach that he has taken. It is very difficult to avoid much of the basic mathematics without losing some of the important concepts and foundation to the subject.
Many authors that try, fail miserably. On the other hand authors that include too much mathematics scare off many of the students that the text is targeted for. Riffenburgh with this book and Motulsky with his book "Intuitive Biostatistics" have both carefully crafted a text that succeeds in this goal.
I did give the book four stars instead of five. This is because I think there are a few important topics in biostatistics that are not included or are covered too briefly. These include equivalence testing, group sequential methods, intention-to-treat analysis, methods for handling missing data and longitudinal data analysis.
The author deserves recognition for including important topics such as meta-analyses, non-parametric methods, logistic regression, sensitivity and specificity and survival analysis methods. These topics are often omitted from elementary courses but are all important in biostatistical applications.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
13 of 13 people found the following review helpful:
5.0 out of 5 stars
On R. H. Riffenburgh; Statistics in Medicine, January 24, 2000
This review is from: Statistics in Medicine (Hardcover)
I remember statistics in medical school in most unfavorable terms. That nightmare text has long been relegated to the dusty Zerox paper box in the garage. Now, enter Dr. Riffenburgh's textbook: Statistics in Medicine. This text is user friendly, laid out in a thoughtful fashion, has plenty of examples of how to apply all of those mysterious tests and is complete enough for the hard core bioresearcher. Bravo!
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
12 of 12 people found the following review helpful:
5.0 out of 5 stars
A very fun book, November 19, 2001
This review is from: Statistics in Medicine (Hardcover)
.
What can be said that Dr. Chernik has not already written?
I am learning my statistics for epidemiological purposes and laboratory methods. This book is fun because the descriptions of the tests are succinct and it is replete with examples using medical data. I have been using it as a quick reference for statistical methods that I read about in journal articles, but have not had formal training in or experience with. This book is complementary to by basic stats book. This book is a must for amateur biostatisticians.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
