Group Sequential Methods with Applications to Clinical Trials

摘要: INTRODUCTION About This Book Why Sequential Methods A Short History of and Group Chapter Organization: Roadmap Bibliography Notes TWO-SIDED TESTS: Two-Sided Tests for Comparing Two Treatments with Normal Response Known Variance Fixed Sample Test Pocock's O'Brien Fleming's Properties Pocock Fleming Other Conclusions GENERAL APPLICATIONS Unified Formulation Applying the Equal Sizes Unequal Increments in Information Linear Models Parametric Binary Data: Proportions The Log-Rank Survival Data t-Tests ONE-SIDED TESTS Introduction Power Family One-Sided Adapting to Whitehead's Triangular WITH EARLY STOPPING UNDER THE NULL HYPOTHESIS Two-Sided, Inner Wedge Double EQUIVALENCE Equivalence Equivalence: Application Comparative Bioavailability Studies Individual Bioequivalence: FLEXIBLE MONITORING: ERROR SPENDING APPROACH Unpredictable Sequences Dependent Timing Analyses Computations Error Spending ANALYSIS FOLLOWING SEQUENTIAL TEST Distribution Theory Point Estimation P-Values Confidence intervals REPEATED CONFIDENCE INTERVALS Example: Difference Means Derived Tests: Use RCIs Aid Early Stopping Decisions Repeated Discussion STOCHASTIC CURTAILMENT Conditional Approach Predictive Parameter-Free Case Study GROUP DISTRIBUTION THEORY Standard Joint Successive Estimates a Parameter Vector Unknown Variance: An Exact One-Sample t-Test General Models: Generalized Connection Analysis BINARY DATA Single Bernoulli Probability Probabilities Odds Ratio Multiple 2 x Tables Case-Control Matched Pair Logistic Regression: Adjusting Covariates SURVIVAL Log Rank Stratified Intervals Hazard Clinical Trial Carcinoma Oropharynx Quantiles INTERNAL PILOT STUDIES: SAMPLE SIZE RE-ESTIMATION Role an Internal Pilot Phase Size Re-estimation MULTIPLE ENDPOINTS Bonferroni Procedure Hotelling Version O'Brien's Based on other Global Statistics Marginal Criteria MULTI-ARMED TRIALS Monitoring Pairwise Comparisons ADAPTIVE TREATMENT ASSIGNMENT Multi-Stage Adaptive Design Time Trends Validity Procedures BAYESIAN APPROACHES Bayesian Paradigm Rules Choice Prior NUMERICAL COMPUTATIONS FOR Basic Calculation Distributions Defined by Functions Following Further Applications Numerical Computation Computer Software