Product Description

This is a book for statistical practitioners, particularly those who design and analyze studies for survival and event history data. Its goal is to extend the toolkit beyond the basic triad provided by most statistical packages: the Kaplan-Meier estimator, log-rank test, and Cox regression model. Building on recent developments motivated by counting process and martingale theory, it shows the reader how to extend the Cox model to analyze multiple/correlated event data using marginal and random effects (frailty) models. It covers the use of residuals and diagnostic plots to identify influential or outlying observations, assess proportional hazards and examine other aspects of goodness of fit. Other topics include time-dependent covariates and strata, discontinuous intervals of risk, multiple time scales, smoothing and regression splines, and the computation of expected survival curves. A knowledge of counting processes and martingales is not assumed as the early chapters provide an introduction to this area. The focus of the book is on actual data examples, the analysis and interpretation of the results, and computation. The methods are now readily available in SAS and S-Plus and this book gives a hands-on introduction, showing how to implement them in both packages, with worked examples for many data sets. The authors call on their extensive experience and give practical advice, including pitfalls to be avoided.

Product Details

• Amazon Sales Rank: #636427 in Books
• Published on: 2001-08-24
• Original language: English
• Number of items: 1
• Binding: Hardcover
• 350 pages

Review

From the reviews:

TECHNOMETRICS

"I would be curious to know how many people bought the book by Fleming and Harrington (1991) and by Anderson, Borgan, Gill, and Keiding (1993) when they first appeared hoping for an introduction to the subject. Instead we all should have saved our money and waited fir this volume by Therneau and Grambsch…This book can serve as a useful reference for statistical practitioners who encounter survival data and for researchers who want to update their knowledge in modern survival analysis…The writing style is light and almost humorous in many places. We get the feeling that the authors had a lot of fun writing this book. If only it was available a decade ago."

STATISTICAL METHODS IN MEDICAL RESEARCH

"In short, this is an exciting book, which introduces and illustrates some recent developments in surviavl analysis. The authors maintain an informal and good-humoures style, making the book very easy to read, and insist on a hands-on approach which encourages the reader to re-work examples. I would recommend it to any statistician analysing survival data."

SHORT BOOK REVIEWS

"The authors ... have laid out for us the wealth of their practical experience at all levels; the numerical aspects; computer algorithms; evaluation of different methods and connections between them; possible pitfalls; and interpretation of the results. Remarkable insights abound.
This book completes that of P. Hougaard by giving much detail on actual fitting of the models discussed by him.
It will serve two audiences: the busy practitioner who has not had time to catch up with martingale theory and counting processes and the graduate student who has just completed such a course and who needs to be introduced to the practicalities and judgements needed in data analysis. It is likely to become a well-thumbed copy on the statistician's desk and statistical practice will be the better for it."

STATISTICS IN MEDICINE

"I use S-PLUS in my own applied work and when testing my methodological research. Therefore, I came to this book with high expectations. I was not disappointed. The book is an invaluable resource for all researchers who use SAS and/or S-PLUS in their applied work, and who want to improve their skills in analyzing survival and event history data."

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION

"…I highly recommend [this book] to statisticians analyzing survival data with S-PLUS and SAS."

"Analogous development for Cox’s regression model – stimulated partially by the counting process theory – has taken place over the last quarter of a century. This book provides a well-organized and extensive collection of these methods. … The book is reasonably self-contained. It brings the fruits of the counting process-based methods to the common analyst … . A number of biostatistical data sets have been used for the purpose of illustrative analysis. … I think the book achieves more than its stated objective." (Debasis Sengupta, Sankhya, Vol. 65 (4), 2003)

"This book models survival data, mainly in terms of the Cox regression model and its extensions … . The text is fluently written in the style of a medium-level oral presentation which makes the book well readable and its contents well understandable … . Difficult theoretical concepts are explained in an easy, yet instructive way … . I consider this book as a most valuable source for beginners … and I warmly recommend this book as introductory reading, guidance, and reference for practical work." (Jochen Mau, Metrika, February, 2003)

"This book presents a state-of-the-art overview on modeling survival data. … examples underline the enormous flexibility and potential of the discussed models. … Appendices giving short tutorials into the statistical packages SAS and A-Plus as well as selected data sets will be very useful for most readers. … The well chosen examples, their analysis and interpretation underline the experience of the authors and makes the book a must-have for every biostatistician." (Heinz J. Skala, Statistical Papers, Vol. 44 (2), 2003)

"‘This is a book for statistical practitioners who analyse survival and event history data … .’ it is also for those who do have the theoretical knowledge already but want to know how to implement it in S-Plus or SAS … . analysis of any event data is outlined very clearly and illustrated by a lot of examples. Therefore, this book is recommended to everyone who deals with any kind of event data … . valuable guide on how to do survival/event analysis … ." (Matthias Wahl, ISCB Newsletter, December, 2003)

"This book contains a comprehensive description of what already can be found in the well-reputed S library survival 5 … . It gives an introduction to survival analysis, with emphasis on the Cox model, and with several illustrations from the authors’ practice as biostatisticians. … the book ends with … several appendices containing SAS macros, S functions and data sets. The book is a very useful companion for the practitioner of survival analysis and particularly for one who uses the Cox model and survival 5." (Göran Broström, Zentralblatt MATH, Vol. 958, 2001)

"This book presents and illustrates various failure time data analysis techniques … . the context of many of the examples seems quite interesting and each is adequately explained by the authors. … The book is clearly written … . In summary, this work is a valuable addition to the list of books on failure time data analysis. … can be recommended to data analysts wishing to try out some of the more recent failure time analysis developments." (Ross Prentice, SIAM Review, Vol. 43 (3), 2001)

"The monograph provides an excellent overview on a number of introductory and advanced topics concerning the Cox proportional hazards model. … Model features, ideas and alternatives are given … . is clearly and concisely written and theory is illustrated with lot of examples. It aims at the practitioner as it covers all practical problems. … I think that the authors have succeeded in writing an excellent monograph on practical survival analysis. It is recommended for everyone who is working in this research field." (M. H. J. de Bruijne, Kwantitatieve Methoden, Vol. 22 (66), 2001)

From the Publisher
This book presents some of the recent developments in survival methods, along with the appropriate code in S-PLUS and SAS so they can be implemented. The book will be of interest to researchers, practitioners, and graduate students working in the areas of biostatistics, epidemilogy, and medical research.

About the Author
Terry Therneau is Head of the Section of Biostatistics, Mayo Clinic, Rochester, Minnesota. He is actively involved in medical consulting, with emphasis in the areas of chronic liver disease, physical medicine, hematology, and laboratory medicine, and is an author on numerous papers in medical and statistical journals. He wrote two of the original SAS procedures for survival analysis (coxregr and survtest) as well as the majority of the S-Plus survival functions.

Customer Reviews

excellent treatment of modern survival analysis emphasizing the Cox model and extensions of it
Terry Therneau is a research statistician at the Mayo Clinic and Patricia Grambsch is a Professor of Biostatistics at the University of Minnesota. The Cox proportional hazards model has been one of the key methods for analyzing survival data with covariates for the last 25 years. Proportionality is a key assumption that limits its use. There has long been a need to find methods which diagnose when the hazard rates are not proportional and provide alternative methods in such situations. Using the theory of counting processes the authors are able to extend the Cox model to more general situations including multiple/correlated event data using either marginal models or random effects (frailty) models. Time dependent covariates are also covered. Some of the theory of martigales and counting processes is included to make the book self-contained. Generalized residuals are used to identify outlying and influential observations (analogous to ordinary regression) and also to assess the proportional hazards assumption.
Although the topics are advanced and the mathematical level is high the book is designed for practitioners, emphasizing applications and providing numerous examples, many from the authors' experience. Statistical analyses are done in SAS and SPlus. The authors tend to use SAS for data management and analysis and SPlus for diagnostics and other plots. Therneau is an expert programmer who has written much of the necessary software in both systems.

Therneau gave an excellent short course that I attended a couple of years ago at the Joint Statistical Meetings based on a draft of the text. The finished product is as good as I expected.

The appendices include SAS and S-Plus tutorials on survival analysis and provide SAS Macros and S functions to apply the new methodology.

The book is now (December 2008) in its 6th printing which is another testament to its value and popularity and a nice deal at amazon's current price of $87. But O'Quigley's book is out now too. So maybe Terry and Patricia should be thinking about doing a revision if they don't already have one in the works.

Anderson et al for the common man
This text is one of the few to make the work of Andersen et al. (Statistical Models Based on Counting Processes, Springer, 1993) accessible to the average statistician. It has three limitations:
1) fails to mention the use of permutation tests for hypotheses regarding the Nelson-Aalen estimator,
2) fails to cite Good PI, Globally almost most powerful tests for censored data,Nonpar Statist 1992, 1:253-262.
3) fails to deal with multiple dependent events (the most common case).
The text also fails to be prescriptive; one is often left feeling that all tests are equal which simply isn't the case.

One of the best statistics texts available today!
As a biostatistics PhD student I've had to endure many very poorly written textbooks (though there are many good one's too). Not only is this book a great text on applied survival analysis, it's a great piece of statistical writing and should be used as an example for all applied texts. The general approach of introducing the theory followed by examples with SAS/SPlus code makes learning the material easy and fun. I wish all statistics texts were even half this good!