Diagnostic Medical Statistics

Author: Yu Chuanhua
Publisher:
Publish Date: 2005-03-01
Features: The importance of medical diagnostic research lies in estimating and comparing the accuracy of diagnostic tests. This book provides a comprehensive set of statistical methods for the design and analysis of medical diagnostic research, including sample size calculation, estimation of diagnostic test accuracy, comparison of accuracy between competing diagnostic tests, and reclassification analysis of diagnostic accuracy data. It also discusses recent developments in correcting verification bias and imperfect gold standard bias, as well as methods for analyzing clustered diagnostic accuracy data and meta-analysis. Through numerous examples, exercises at the end of each chapter, and websites providing relevant programs, the book offers convenience for practical users. The target readers of this book include clinicians interested in diagnostic research, statisticians interested in analyzing diagnostic research data, and statisticians and graduate students aspiring to study diagnostic medical statistical methods. The book is divided into 12 chapters. The first three chapters discuss common indicators of diagnostic accuracy and the research design of diagnostic accuracy. Chapter 1 introduces some statistical issues in diagnostic accuracy research; Chapter 2 defines several commonly used diagnostic accuracy indicators, such as sensitivity, specificity, predictive value, receiver operating characteristic (ROC) curves and their related indicators; and Chapter 3 describes clinical research designs to avoid common diagnostic accuracy biases. Chapters 4 and 5 discuss estimation and hypothesis testing methods for diagnostic accuracy. Chapter 4 describes methods for estimating sensitivity, specificity, predictive value, and ROC curves; Chapter 5 provides methods for comparing the relative accuracy of several competing tests. Chapter 6 clarifies sample size calculation for diagnostic accuracy research. Chapter 7 discusses non-mathematical issues in meta-analysis of diagnostic test research. Chapters 8 to 12 discuss more advanced analytical techniques. Chapter 8 uses regression models to study the influence of patient characteristics on diagnostic test accuracy; Chapter 9 explores methods for comparing several related ROC curves (e.g., multi-reader studies); Chapter 10 provides estimation and inference methods for correcting verification bias; Chapter 11 discusses the correct estimation methods for diagnostic accuracy when using imperfect gold standards; and Chapter 12 describes the statistical methods for meta-analysis in diagnostic test research.