by Craig Mallinckrodt (Author), IlyaLipkovich (Author)
Analyzing Longitudinal Clinical Trial Data: A Practical Guide provides practical and easy to implement approaches for bringing the latest theory on analysis of longitudinal clinical trial data into routine practice.The book, with its example-oriented approach that includes numerous SAS and R code fragments, is an essential resource for statisticians and graduate students specializing in medical research.
The authors provide clear descriptions of the relevant statistical theory and illustrate practical considerations for modeling longitudinal data. Topics covered include choice of endpoint and statistical test; modeling means and the correlations between repeated measurements; accounting for covariates; modeling categorical data; model verification; methods for incomplete (missing) data that includes the latest developments in sensitivity analyses, along with approaches for and issues in choosing estimands; and means for preventing missing data. Each chapter stands alone in its coverage of a topic. The concluding chapters provide detailed advice on how to integrate these independent topics into an over-arching study development process and statistical analysis plan.
Format: Illustrated
Pages: 336
Edition: 1
Publisher: Chapman and Hall/CRC
Published: 11 Jan 2017
ISBN 10: 1498765319
ISBN 13: 9781498765312
This book deals mostly with longitudinal clinical trial data, but also with the related issue of imputing missing data. The book is an excellent resource overall, as it is fairly comprehensive, well referenced, and clear.
~Vance W. Berger, PhD, NIH/NCI/DCP/BRG
Analyzing Longitudinal Clinical Trial Data provides, in a well organized and small format, a fairly easy read that could be helpful for both researchers analyzing longitudinal data collected from clinical trials (or perhaps even observational studies) and instructors teaching undergraduate and graduate courses on clinical trials, longitudinal data, and missing data. The book is divided into four well-structured and complementary sections: background and setting, general modeling strategies and methods, methods for dealing with missing data, and overall guidance (with illustration) for developing a study.
~Journal of the American Statistical Association