Modern Methods for Epidemiology

Discussing cutting-edge statistical methods for epidemiology and biomedical research that cannot be found in most textbooks, such as latent growth curve models, this volume reflects the fact that recent years have seen a huge rise in desk-top computing power.

Routine applications of advanced statistical methods on real data have become possible in the last ten years because desktop computers have become much more powerful and cheaper. However, proper understanding of the challenging statistical theory behind those methods remains essential for correct application and interpretation, and rarely seen in the medical literature. Modern Methods for Epidemiology provides a concise introduction to recent development in statistical methodologies for epidemiological and biomedical researchers. Many of these methods have become indispensible tools for researchers working in epidemiology and medicine but are rarely discussed in details by standard textbooks of biostatistics or epidemiology. Contributors of this book are experienced researchers and experts in their respective fields. This textbook provides a solid starting point for those who are new to epidemiology, and for those looking for guidance in more modern statistical approaches to observational epidemiology. Epidemiological and biomedical researchers who wish to overcome the mathematical barrier of applying those methods to their research will find this book an accessible and helpful reference for self-learning and research. This book is also a good source for teaching postgraduate students in medical statistics or epidemiology.

Inhalt

Preface.-1 Confounding and Causal path diagrams.-2 Statistical modelling of partially observed data using multiple imputation: principles and practice.-3 Measurement errors in epidemiology.-4 Selection bias in epidemiologic studies.-5 Multilevel modelling.-6 Modelling data that exhibit an excess number of zeros: zero-inflated models and generic mixture models.-7 Multilevel latent class modelling.-8 Bayesian bivariate disease mapping.-9 A multivariate random frailty effects model for multiple spatially dependent survival data.-10 Meta-analysis of observational studies.-11 Directed acyclic graphs and structural equation modelling.-12 Latent growth curve models.-13 Growth mixture modelling for life course epidemiology.-14 G-estimations.-15 Generalised additive models.-16 Regression and classification trees.-17 Statistical interactions and gene-environment joint effects. Index.