Geert Molenberghs - Böcker

Many methods for analyzing clustered data exist, all with advantages and limitations in particular applications. Compiled from the contributions of leading specialists in the field, Topics in Modelling of Clustered Data describes the tools and techniques for modelling the clustered data often encountered in medical, biological, environmental, and social science studies. It focuses on providing a comprehensive treatment of marginal, conditional, and random effects models using, among others, likelihood, pseudo-likelihood, and generalized estimating equations methods.The authors motivate and illustrate all aspects of these models in a variety of real applications. They discuss several variations and extensions, including individual-level covariates and combined continuous and discrete outcomes. Flexible modelling with fractional and local polynomials, omnibus lack-of-fit tests, robustification against misspecification, exact, and bootstrap inferential procedures all receive extensive treatment. The applications discussed center primarily, but not exclusively, on developmental toxicity, which leads naturally to discussion of other methodologies, including risk assessment and dose-response modelling.Clearly written, Topics in Modelling of Clustered Data offers a practical, easily accessible survey of important modelling issues. Overview models give structure to a multitude of approaches, figures help readers visualize model characteristics, and a generous use of examples illustrates all aspects of the modelling process.

An important factor that affects the duration, complexity and cost of a clinical trial is the endpoint used to study the treatment’s efficacy. When a true endpoint is difficult to use because of such factors as long follow-up times or prohibitive cost, it is sometimes possible to use a surrogate endpoint that can be measured in a more convenient or cost-effective way. This book focuses on the use of surrogate endpoint evaluation methods in practice, using SAS and R.

Missing data affect nearly every discipline by complicating the statistical analysis of collected data. But since the 1990s, there have been important developments in the statistical methodology for handling missing data. Written by renowned statisticians in this area, Handbook of Missing Data Methodology presents many methodological advances and the latest applications of missing data methods in empirical research.Divided into six parts, the handbook begins by establishing notation and terminology. It reviews the general taxonomy of missing data mechanisms and their implications for analysis and offers a historical perspective on early methods for handling missing data. The following three parts cover various inference paradigms when data are missing, including likelihood and Bayesian methods; semi-parametric methods, with particular emphasis on inverse probability weighting; and multiple imputation methods. The next part of the book focuses on a range of approaches that assess the sensitivity of inferences to alternative, routinely non-verifiable assumptions about the missing data process. The final part discusses special topics, such as missing data in clinical trials and sample surveys as well as approaches to model diagnostics in the missing data setting. In each part, an introduction provides useful background material and an overview to set the stage for subsequent chapters. Covering both established and emerging methodologies for missing data, this book sets the scene for future research. It provides the framework for readers to delve into research and practical applications of missing data methods.

Both humanitarian and commercial considerations have spurred intensive search for methods to reduce the time and cost required to develop new therapies. The identification and use of surrogate endpoints, i.e., measures that can replace or supplement other endpoints in evaluations of experimental treatments or other interventions, is a general strategy that has stimulated both enthusiasm and skepticism. Surrogate endpoints are useful when they can be measured earlier, more conveniently, or more frequently than the "true" endpoints of primary interest. Regulatory agencies around the globe, particularly in the United States, Europe, and Japan, are introducing provisions and policies relating to the use of surrogate endpoints in registration studies. But how can one establish the adequacy of a surrogate? What kind of evidence is needed, and what statistical methods portray that evidence most appropriately? This book offers a balanced account on this controversial topic. The text presents major developments of the last couple of decades, together with a unified, meta-analytic framework within which surrogates can be evaluated from several angles.Methodological development is coupled with perspectives on various therapeutic areas. Academic views are juxtaposed with standpoints of scientists working in the biopharmaceutical industry as well as of colleagues from the regulatory authorities.Tomasz Burzykowski is Assistant Professor of Biostatistics at the Limburgs Universitair Centrum in Belgium. Dr. Burzykowski has published methodological work on the analysis of survey data, meta-analyses of clinical trials, and validation of surrogate endpoints. He is a co-author of numerous papers applying statistical methods to clinical data in different disease areas (cancer, cardiovascular diseases, dermatology, orthodontics).Geert Molenberghs is Professor of Biostatistics at the Limburgs Universitair Centrum in Belgium. Dr. Molenberghs published methodological work on surrogate markers in clinical trials, categorical data, longitudinal data analysis, and on the analysis of non-response in clinical and epidemiological studies. He serves as Joint Editor for Applied Statistics (2001-2004) and is President of the International Biometric Society (2004-2005).He was elected Fellow of the American Statistical Association and received the Guy Medal in Bronze from the Royal Statistical Society. Marc Buyse founded the International Drug Development Institute in 1991. He is Past President of the International Society for Clinical Biostatistics, Past President of the Quetelet Society, and Past Board Member of the Society for Clinical Trials. He is currently the Executive Director of IDDI (International Drug Development Institute) and Associate Professor of biostatistics at the Limburgs Universitair Centrum, Center for Statistics, Diepenbeek, Belgium. He has published extensively in the fields of biostatistics and oncology. His research interests include meta-analysis, surrogate endpoints, statistical detection of fraud, and the design and statistical analysis of clinical trials.

This book provides a comprehensive treatment on modeling approaches for non-Gaussian repeated measures, possibly subject to incompleteness. The authors begin with models for the full marginal distribution of the outcome vector. This allows model fitting to be based on maximum likelihood principles, immediately implying inferential tools for all parameters in the models. At the same time, they formulate computationally less complex alternatives, including generalized estimating equations and pseudo-likelihood methods. They then briefly introduce conditional models and move on to the random-effects family, encompassing the beta-binomial model, the probit model and, in particular the generalized linear mixed model. Several frequently used procedures for model fitting are discussed and differences between marginal models and random-effects models are given attention.The authors consider a variety of extensions, such as models for multivariate longitudinal measurements, random-effects models with serial correlation, and mixed models with non-Gaussian random effects. They sketch the general principles for how to deal with the commonly encountered issue of incomplete longitudinal data.The authors critique frequently used methods and propose flexible and broadly valid methods instead, and conclude with key concepts of sensitivity analysis.Without putting too much emphasis on software, the book shows how the different approaches can be implemented within the SAS software package. The text is organized so the reader can skip the software-oriented chapters and sections without breaking the logical flow.

This book provides a comprehensive treatment of linear mixed models for continuous longitudinal data. Next to model formulation, this edition puts major emphasis on exploratory data analysis for all aspects of the model, such as the marginal model, subject-specific profiles, and residual covariance structure. Further, model diagnostics and missing data receive extensive treatment. Sensitivity analysis for incomplete data is given a prominent place. Several variations to the conventional linear mixed model are discussed (a heterogeity model, condional linear mid models). This book will be of interest to applied statisticians and biomedical researchers in industry, public health organizations, contract research organizations, and academia. The book is explanatory rather than mathematically rigorous. Most analyses were done with the MIXED procedure of the SAS software package, and many of its features are clearly elucidated. How3ever, some other commercially available packages are discussed as well. Great care has been taken in presenting the data analyses in a software-independent fashion.Geert Verbeke is Assistant Professor at the Biostistical Centre of the Katholieke Universiteit Leuven in Belgium. He received the B.S. degree in mathematics (1989) from the Katholieke Universiteit Leuven, the M.S. in biostatistics (1992) from the Limburgs Universitair Centrum, and earned a Ph.D. in biostatistics (1995) from the Katholieke Universiteit Leuven. Dr. Verbeke wrote his dissertation, as well as a number of methodological articles, on various aspects of linear mixed models for longitudinal data analysis. He has held visiting positions at the Gerontology Research Center and the Johns Hopkins University. Geert Molenberghs is Assistant Professor of Biostatistics at the Limburgs Universitair Centrum in Belgium. He received the B.S. degree in mathematics (1988) and a Ph.D. in biostatistics (1993) from the

The dissemination of the MIXED procedure in SAS has provided a whole class of statistical models for routine use. We believe that both the ideas be­ hind the techniques and their implementation in SAS are not at all straight­ forward and users from various applied backgrounds, including the phar­ maceutical industry, have experienced difficulties in using the procedure effectively. Courses and consultancy on PROC MIXED have been in great demand in recent years, illustrating the clear need for resource material to aid the user. This book is intended as a contribution to bridging this gap. We hope the book will be of value to a wide audience, including applied statisticians and many biomedical researchers, particularly in the pharma­ ceutical industry, medical and public health research organizations, con­ tract research organizations, and academic departments. This implies that our book is explanatory rather than research oriented and that it empha­ sizes practice rather than mathematical rigor. In this respect, clear guidance and advice on practical issues are the main focus of the text. Nevertheless, this does not imply that more advanced topics have been avoided. Sections containing material of a deeper level have been sign posted by means of an asterisk.

Missing Data in Clinical Studies provides a comprehensive account of the problems arising when data from clinical and related studies are incomplete, and presents the reader with approaches to effectively address them. The text provides a critique of conventional and simple methods before moving on to discuss more advanced approaches. The authors focus on practical and modeling concepts, providing an extensive set of case studies to illustrate the problems described. Provides a practical guide to the analysis of clinical trials and related studies with missing data.Examines the problems caused by missing data, enabling a complete understanding of how to overcome them.Presents conventional, simple methods to tackle these problems, before addressing more advanced approaches, including sensitivity analysis, and the MAR missingness mechanism.Illustrated throughout with real-life case studies and worked examples from clinical trials.Details the use and implementation of the necessary statistical software, primarily SAS.Missing Data in Clinical Studies has been developed through a series of courses and lectures. Its practical approach will appeal to applied statisticians and biomedical researchers, in particular those in the biopharmaceutical industry, medical and public health organisations. Graduate students of biostatistics will also find much of benefit.

During the past three decades, the use of statistical methods in the health sciences has become quite common, and many sophisticated methods have been developed for specific applications and problems. Presented in this work are chapters covering a broad range of recent advances of statistical methods in the health sciences written by internationally renowned experts. The volume is thematically classified into different areas of the health sciences according to the particular focus of the chapters. Specific topics covered include stochastic models in cancer and AIDS studies, statistical methods in genome sequence analysis, bioequivalence studies, randomized clinical trial designs, survival analysis and applications, and statistical methods in quality-of-life. Some chapters are reviews of active research, while others highlight recent developments, new methods, and practical real-world data. This comprehensive reference text will serve the health sciences community, as well as practitioners, researchers, and graduate students in applied probability and biostatistics.

The concepts of estimands, analyses (estimators), and sensitivity are interrelated. Therefore, great need exists for an integrated approach to these topics. This book acts as a practical guide to developing and implementing statistical analysis plans by explaining fundamental concepts using accessible language, providing technical details, real-world examples, and SAS and R code to implement analyses. The updated ICH guideline raises new analytic and cross-functional challenges for statisticians. Gaps between different communities have come to surface, such as between causal inference and clinical trialists, as well as among clinicians, statisticians, and regulators when it comes to communicating decision-making objectives, assumptions, and interpretations of evidence.This book lays out a path toward bridging some of these gaps. It offers¿ A common language and unifying framework along with the technical details and practical guidance to help statisticians meet the challenges¿ A thorough treatment of intercurrent events (ICEs), i.e., postrandomization events that confound interpretation of outcomes and five strategies for ICEs in ICH E9 (R1)¿ Details on how estimands, integrated into a principled study development process, lay a foundation for coherent specification of trial design, conduct, and analysis needed to overcome the issues caused by ICEs:¿ A perspective on the role of the intention-to-treat principle¿ Examples and case studies from various areas¿ Example code in SAS and R¿ A connection with causal inference¿ Implications and methods for analysis of longitudinal trials with missing dataTogether, the authors have offered the readers their ample expertise in clinical trial design and analysis, from an industrial and academic perspective.

Past, Present, and Future of Statistical Science was commissioned in 2013 by the Committee of Presidents of Statistical Societies (COPSS) to celebrate its 50th anniversary and the International Year of Statistics. COPSS consists of five charter member statistical societies in North America and is best known for sponsoring prestigious awards in statistics, such as the COPSS Presidents’ award. Through the contributions of a distinguished group of 50 statisticians who are past winners of at least one of the five awards sponsored by COPSS, this volume showcases the breadth and vibrancy of statistics, describes current challenges and new opportunities, highlights the exciting future of statistical science, and provides guidance to future generations of statisticians. The book is not only about statistics and science but also about people and their passion for discovery.Distinguished authors present expository articles on a broad spectrum of topics in statistical education, research, and applications. Topics covered include reminiscences and personal reflections on statistical careers, perspectives on the field and profession, thoughts on the discipline and the future of statistical science, and advice for young statisticians. Many of the articles are accessible not only to professional statisticians and graduate students but also to undergraduate students interested in pursuing statistics as a career and to all those who use statistics in solving real-world problems. A consistent theme of all the articles is the passion for statistics enthusiastically shared by the authors. Their success stories inspire, give a sense of statistics as a discipline, and provide a taste of the exhilaration of discovery, success, and professional accomplishment.

In today's healthcare landscape, there is a pressing need for quantitative methodologies that include the patients' perspective in any treatment decision.Handbook of Generalized Pairwise Comparisons: Methods for Patient-Centric Analysis provides a comprehensive overview of an innovative and powerful statistical methodology that generalizes the traditional Wilcoxon-Mann-Whitney test by extending it to any number of outcomes of any type and including thresholds of clinical relevance into a single, multidimensional evaluation.The book covers the statistical foundations of generalized pairwise comparisons (GPC), applications in various disease areas, implications for regulatory approvals and benefit-risk analyses, and considerations for patient-centricity in clinical research. With contributions from leading experts in the field, this book stands as an essential resource for a more holistic and patient-centric assessment of treatment effects.

In today's healthcare landscape, there is a pressing need for quantitative methodologies that include the patients' perspective in any treatment decision.Handbook of Generalized Pairwise Comparisons: Methods for Patient-Centric Analysis provides a comprehensive overview of an innovative and powerful statistical methodology that generalizes the traditional Wilcoxon-Mann-Whitney test by extending it to any number of outcomes of any type and including thresholds of clinical relevance into a single, multidimensional evaluation.The book covers the statistical foundations of generalized pairwise comparisons (GPC), applications in various disease areas, implications for regulatory approvals and benefit-risk analyses, and considerations for patient-centricity in clinical research. With contributions from leading experts in the field, this book stands as an essential resource for a more holistic and patient-centric assessment of treatment effects.

The concepts of estimands, analyses (estimators), and sensitivity are interrelated. Therefore, great need exists for an integrated approach to these topics. This book acts as a practical guide to developing and implementing statistical analysis plans by explaining fundamental concepts using accessible language, providing technical details, real-world examples, and SAS and R code to implement analyses. The updated ICH guideline raises new analytic and cross-functional challenges for statisticians. Gaps between different communities have come to surface, such as between causal inference and clinical trialists, as well as among clinicians, statisticians, and regulators when it comes to communicating decision-making objectives, assumptions, and interpretations of evidence.This book lays out a path toward bridging some of these gaps. It offersA common language and unifying framework along with the technical details and practical guidance to help statisticians meet the challengesA thorough treatment of intercurrent events (ICEs), i.e., postrandomization events that confound interpretation of outcomes and five strategies for ICEs in ICH E9 (R1)Details on how estimands, integrated into a principled study development process, lay a foundation for coherent specification of trial design, conduct, and analysis needed to overcome the issues caused by ICEs:A perspective on the role of the intention-to-treat principleExamples and case studies from various areasExample code in SAS and RA connection with causal inferenceImplications and methods for analysis of longitudinal trials with missing dataTogether, the authors have offered the readers their ample expertise in clinical trial design and analysis, from an industrial and academic perspective.

Missing data affect nearly every discipline by complicating the statistical analysis of collected data. But since the 1990s, there have been important developments in the statistical methodology for handling missing data. Written by renowned statisticians in this area, Handbook of Missing Data Methodology presents many methodological advances and the latest applications of missing data methods in empirical research.Divided into six parts, the handbook begins by establishing notation and terminology. It reviews the general taxonomy of missing data mechanisms and their implications for analysis and offers a historical perspective on early methods for handling missing data. The following three parts cover various inference paradigms when data are missing, including likelihood and Bayesian methods; semi-parametric methods, with particular emphasis on inverse probability weighting; and multiple imputation methods. The next part of the book focuses on a range of approaches that assess the sensitivity of inferences to alternative, routinely non-verifiable assumptions about the missing data process. The final part discusses special topics, such as missing data in clinical trials and sample surveys as well as approaches to model diagnostics in the missing data setting. In each part, an introduction provides useful background material and an overview to set the stage for subsequent chapters. Covering both established and emerging methodologies for missing data, this book sets the scene for future research. It provides the framework for readers to delve into research and practical applications of missing data methods.

This paperback edition is a reprint of the 2000 edition. This book provides a comprehensive treatment of linear mixed models for continuous longitudinal data. Several variations to the conventional linear mixed model are discussed (a heterogeity model, conditional linear mixed models).

Both humanitarian and commercial considerations have spurred intensive search for methods to reduce the time and cost required to develop new therapies. The identification and use of surrogate endpoints, i.e., measures that can replace or supplement other endpoints in evaluations of experimental treatments or other interventions, is a general strategy that has stimulated both enthusiasm and skepticism. Surrogate endpoints are useful when they can be measured earlier, more conveniently, or more frequently than the "true" endpoints of primary interest. Regulatory agencies around the globe, particularly in the United States, Europe, and Japan, are introducing provisions and policies relating to the use of surrogate endpoints in registration studies. But how can one establish the adequacy of a surrogate? What kind of evidence is needed, and what statistical methods portray that evidence most appropriately? This book offers a balanced account on this controversial topic. The text presents major developments of the last couple of decades, together with a unified, meta-analytic framework within which surrogates can be evaluated from several angles. Methodological development is coupled with perspectives on various therapeutic areas. Academic views are juxtaposed with standpoints of scientists working in the biopharmaceutical industry as well as of colleagues from the regulatory authorities.Tomasz Burzykowski is Assistant Professor of Biostatistics at the Limburgs Universitair Centrum in Belgium. Dr. Burzykowski has published methodological work on the analysis of survey data, meta-analyses of clinical trials, and validation of surrogate endpoints. He is a co-author of numerous papers applying statistical methods to clinical data in different disease areas (cancer, cardiovascular diseases, dermatology, orthodontics).Geert Molenberghs is Professor of Biostatistics at the Limburgs Universitair Centrum in Belgium. Dr. Molenberghs published methodological work on surrogate markers in clinical trials, categorical data, longitudinal data analysis, and on the analysis of non-response in clinical and epidemiological studies. He serves as Joint Editor for Applied Statistics (2001-2004) and is President of the International Biometric Society (2004-2005). He was elected Fellow of the American Statistical Association and received the Guy Medal in Bronze from the Royal Statistical Society. Marc Buyse founded the International Drug Development Institute in 1991. He is Past President of the International Society for Clinical Biostatistics, Past President of the Quetelet Society, and Past Board Member of the Society for Clinical Trials. He is currently the Executive Director of IDDI (International Drug Development Institute) and Associate Professor of biostatistics at the Limburgs Universitair Centrum, Center for Statistics, Diepenbeek, Belgium. He has published extensively in the fields of biostatistics and oncology. His research interests include meta-analysis, surrogate endpoints, statistical detection of fraud, and the design and statistical analysis of clinical trials.From the reviews:"A strength of this book is its comprehensive and up-to-date presentation of issues pertinent to the evaluation of surrgoate endpoints...This book makes an important contribution to the clinical trials literature..." Journal of Biopharmaceutical Statistics, 2006"Many of the chapters deal with real-life data examples and studies involving surrogate outcomes, many written by authors who were directly involved in these studies...The editors have written nice background sections...until a more concise manuscript on this topic is written, this book will remain the most important resource for biostatisticians and researchers in this area." Debajyoti Sinha for the Journal of the American Statistical Association, December 2006"This book is a reflection of the ongoing debate on the definition and use of surrogate markers...I see the book as an invitation to join the debate. There is much work to be done and reading the book might inspire many to participate. It will be useful for researchers in this and related fields, such as joint modeling of longitudinal and survival data and multivariate meta-analysis. The book is well organized, is a pleasure to read, and is very well documented with up-to-date references." Hans C. Van Houwelingen for Bioometrics, September 2006

The linear mixed model has become the main parametric tool for the analysis of continuous longitudinal data, as the authors discussed in their 2000 book. Without putting too much emphasis on software, the book shows how the different approaches can be implemented within the SAS software package. The authors received the American Statistical Association's Excellence in Continuing Education Award based on short courses on longitudinal and incomplete data at the Joint Statistical Meetings of 2002 and 2004.

This book provides a comprehensive treatment of linear mixed models for continuous longitudinal data. Next to model formulation, this edition puts major emphasis on exploratory data analysis for all aspects of the model, such as the marginal model, subject-specific profiles, and residual covariance structure. Further, model diagnostics and missing data receive extensive treatment. Sensitivity analysis for incomplete data is given a prominent place. Most analyses were done with the MIXED procedure of the SAS software package, but the data analyses are presented in a software-independent fashion.

Past, Present, and Future of Statistical Science was commissioned in 2013 by the Committee of Presidents of Statistical Societies (COPSS) to celebrate its 50th anniversary and the International Year of Statistics. COPSS consists of five charter member statistical societies in North America and is best known for sponsoring prestigious awards in statistics, such as the COPSS Presidents’ award. Through the contributions of a distinguished group of 50 statisticians who are past winners of at least one of the five awards sponsored by COPSS, this volume showcases the breadth and vibrancy of statistics, describes current challenges and new opportunities, highlights the exciting future of statistical science, and provides guidance to future generations of statisticians. The book is not only about statistics and science but also about people and their passion for discovery.Distinguished authors present expository articles on a broad spectrum of topics in statistical education, research, and applications. Topics covered include reminiscences and personal reflections on statistical careers, perspectives on the field and profession, thoughts on the discipline and the future of statistical science, and advice for young statisticians. Many of the articles are accessible not only to professional statisticians and graduate students but also to undergraduate students interested in pursuing statistics as a career and to all those who use statistics in solving real-world problems. A consistent theme of all the articles is the passion for statistics enthusiastically shared by the authors. Their success stories inspire, give a sense of statistics as a discipline, and provide a taste of the exhilaration of discovery, success, and professional accomplishment.

An important factor that affects the duration, complexity and cost of a clinical trial is the endpoint used to study the treatment’s efficacy. When a true endpoint is difficult to use because of such factors as long follow-up times or prohibitive cost, it is sometimes possible to use a surrogate endpoint that can be measured in a more convenient or cost-effective way. This book focuses on the use of surrogate endpoint evaluation methods in practice, using SAS and R.

Many methods for analyzing clustered data exist, all with advantages and limitations in particular applications. Compiled from the contributions of leading specialists in the field, Topics in Modelling of Clustered Data describes the tools and techniques for modelling the clustered data often encountered in medical, biological, environmental, and social science studies. It focuses on providing a comprehensive treatment of marginal, conditional, and random effects models using, among others, likelihood, pseudo-likelihood, and generalized estimating equations methods.The authors motivate and illustrate all aspects of these models in a variety of real applications. They discuss several variations and extensions, including individual-level covariates and combined continuous and discrete outcomes. Flexible modelling with fractional and local polynomials, omnibus lack-of-fit tests, robustification against misspecification, exact, and bootstrap inferential procedures all receive extensive treatment. The applications discussed center primarily, but not exclusively, on developmental toxicity, which leads naturally to discussion of other methodologies, including risk assessment and dose-response modelling.Clearly written, Topics in Modelling of Clustered Data offers a practical, easily accessible survey of important modelling issues. Overview models give structure to a multitude of approaches, figures help readers visualize model characteristics, and a generous use of examples illustrates all aspects of the modelling process.