---

Clinical trials often involve multiple endpoints that are clinically meaningful to determine efficacy and safety of the investigational medicinal product. These multiple outcomes are often measured over scheduled and/or unscheduled visits which add complexity when analyzing these types of data. This book is developed specifically to deal with multiple longitudinal outcomes in the framework of generalized estimating equations and their post-hoc multiplicity adjustment for the control of familywise error rate. The book covers a recent development and its extension of quasi-least squares to the application in clinical trials for the analysis of single and multiple longitudinal outcomes. A user-written SAS software with its complete code and manual are also provided to fit various models described in this book. The book further discusses issues of inflation of type I error when multiple hypotheses are tested simultaneously and provides a number of practical solutions to control the familywise type I error using the gatekeeping strategies that are emerging as an important direction of improvement in the analysis of clinical trial data.

---

Senior Biostatistician at Forest Research Institute, Inc., earned his PhD in Biostatistics at the University of Pennsylvania. He has years of experience in design of experiments and data analysis. He has authored a number of publications, and is also the developer of the %QLS macro for implementation of quasi-least squares in SAS.

---