 | Summaries and Customer Reviews are supplied by Amazon.com | Medicine deals with treatments that work often but not always, so treatment success must be based on probability. Statistical methods lift medical research from the anecdotal to measured levels of probability. This book presents the common statistical methods used in 90% of medical research, along with the underlying basics, in two parts: a textbook section for use by students in health care training programs, e.g., medical schools or residency training, and a reference section for use by practicing clinicians in reading medical literature and performing their own research. The book does not require a significant level of mathematical knowledge and couches the methods in multiple examples drawn from clinical medicine, giving it applicable context.
* Easy-to-follow format incorporates medical examples, step-by-step methods, and check yourself exercises
* Two-part design features course material and a professional reference section
* Chapter summaries provide a review of formulas, method algorithms, and check lists
* Companion site links to statistical databases that can be downloaded and used to perform the exercises from the book and practice statistical methods
New in this Edition:
* New chapters on: multifactor tests on means of continuous data, equivalence testing, and advanced methods
* New topics include: trial randomization, treatment ethics in medical research, imputation of missing data, and making evidence-based medical decisions
* Updated database coverage and additional exercises
* Expanded coverage of numbers needed to treat and to benefit, and regression analysis including stepwise regression and Cox regression
Thorough discussion on required sample size
|
Average Customer Rating:   not much depth but fairly good examples Covering topics first for non statisticians and then for the more mathematically inclined wastes space that could be used for more depth. The text is reasonably clear, but I was surprised at some of the statistical errors, like (regarding chi square) "it is valid only if the expected value of every cell is at least 1 and the minimum count of 5 appears in every cell". Note: this incorrectly refers the minimum of the observed values of at least 5. Compare that to Shott's textbook "No expected frequency should be less than 1, and no more than 20% of the expected frequencies should be less than 5" or Altman's book "80% of the cells should have expected values of at least 5" or Feinstein's book "Some authors have said the X2 test is not valid if the expected value of any cell is below 5..." . This is just an example of how you need to read this book carefully, as there are several statistical errors present. Furthermore, the superficial description of some of the tests would make actual use of the tests for your own research problematic. kind of o.k. It is kind of o.k.
very difficult to find some particular thing
so many details and information which nobody needs excellent book If you attend the course of statistics, this 's one of my recommedation. Good book and applicable. different approach to teaching biostatistics in health sciences 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.
Excellent source of information for the non-biostatician Comprehensive, succinct with relevant examples that make the teaching points clear to even a non-biostatician like myself.
I have used this book in helping me to design clinical studies and it was extraordinarily useful! Thank you Dr Riffenburgh. |  |
Select button not working? 
