Applied Longitudinal Data Analysis for Epidemiology: A Practical Guide

by Jos W. R. Twisk

Overview

This book discusses the most important techniques available for longitudinal data analysis, from simple techniques such as the paired t-test and summary statistics, to more sophisticated ones such as generalized estimating of equations and mixed model analysis. A distinction is made between longitudinal analysis with continuous, dichotomous and categorical outcome variables. The emphasis of the discussion lies in the interpretation and comparison of the results of the different techniques. The second edition includes new chapters on the role of the time variable and presents new features of longitudinal data analysis. Explanations have been clarified where necessary and several chapters have been completely rewritten. The analysis of data from experimental studies and the problem of missing data in longitudinal studies are discussed. Finally, an extensive overview and comparison of different software packages is provided. This practical guide is essential for non-statisticians and researchers working with longitudinal data from epidemiological and clinical studies.

Product Details

ISBN-13:	9781107065352
Publisher:	Cambridge University Press
Publication date:	05/09/2013
Sold by:	Barnes & Noble
Format:	eBook
File size:	13 MB
Note:	This product may take a few minutes to download.

About the Author

Professor Dr Jos W. R. Twisk is a Professor of Applied Biostatistics and Head of the Expertise Centre of Longitudinal Data Analysis. Furthermore, he is a Senior Researcher and Lecturer in the Department of Epidemiology and Biostatistics, Vrije Universiteit Medical Centre and the Department of Health Sciences, Vrije Universteit, Amsterdam.

Table of Contents

Preface; Acknowledgements; 1. Introduction; 2. Study design; 3. Continuous outcome variables; 4. Continuous outcome variables - relationships with other variables; 5. The modelling of time; 6. Other possibilities for modelling longitudinal data; 7. Dichotomous outcome variables; 8. Categorical and 'count' outcome variables; 9. Analysis data from experimental studies; 10. Missing data in longitudinal studies; 11. Sample size calculations; 12. Software for longitudinal data analysis; 13. One step further; References; Index.

Editorial Reviews

Reviewer: Colleen Kummet, MS (University of Iowa College of Public Health)
Description: This is an update of a relevant and accessible book for researchers faced with the wealth and challenges of longitudinal data. Methods for analyzing longitudinal data are presented in an example-based format, complete with statistical syntax and output. The first edition was published in 2003.
Purpose: The purpose is to bridge the communication gap between statisticians and epidemiologists on the complex topic of longitudinal data analysis. Books on this subject are often dense, containing technical notation that is not easily translated to application. The readable explanation of each statistical method, along with the included software syntax, output, and interpretation of results places an understanding of this material within reach of researchers who are not statisticians.
Audience: The book is intended for epidemiological researchers who have some prior knowledge of regression, according to the author. Although it is clear the author has put much effort into making the text readable, the complex nature of longitudinal analysis would require a moderate level of knowledge of statistical modeling techniques. The author has published peer-reviewed literature on topics covered in the book.
Features: In addition to information on traditional methods of longitudinal data analysis, the book also covers more modern methods such as generalized estimating equations (GEE), linear mixed modeling, and longitudinal data analysis for dichotomous outcomes. The topics of GEE and mixed models are thoroughly covered. The inclusion of syntax and software results for each method will help readers learn how to execute these analyses as well as interpret the results. This book would have benefitted from a discussion of measures of model goodness-of-fit such as Akaike Information Criteria (AIC) or quasi-AIC (QIC) as these are valuable in the process of modeling longitudinal data.
Assessment: This book will serve as a useful guide to understanding and executing longitudinal data analyses. There are many books on this subject, but the author of this one succeeds in presenting the explanation of the statistical detail in words, rather than notation, for a broader audience of researchers. The second edition includes new methods since the 2003 edition and revisions of the trials and missing data sections.

Doody's Review Service

Review of the first edition: 'Overall, the book is well written, and the material is rich and carefully organised … the book is a welcome reference manual or practical guide for practitioners of statistical methods in (but not limited to) epidemiological and clinical studies. The book's main value is in its rather comprehensive presentation of a collection of longitudinal data analyses arising from different research questions. In this sense, the book is unique. Some practitioners of statistics may have struggled to learn longitudinal data analysis by reading manuals of software packages. This book is potentially of great benefit to them.' Journal of the American Statistical Association

From the Publisher