A Guide to Sample Size for Animal-based Studies

A Guide to Sample Size for Animal-based Studies Understand a foundational area of experimental design with this innovative reference Animal-based research is an essential part of basic and preclinical research, but poses a unique set of experimental design challenges. The most important of these are the 3Rs − Replacement, Reduction and Refinement − the principles comprising the ethical framework for humane animal-based studies. However, many researchers have difficulty navigating the design trade-offs necessary to simultaneously minimize animal use, and produce scientific information that is both rigorous and reliable. A Guide to Sample Size for Animal-based Studies meets this need with a thorough, accessible reference work to the subject. This book provides a straightforward systematic approach to “rightsizing” animal-based experiments, with sample size estimates based on the fundamentals of statistical thinking: structured research questions, variation control and appropriate design of experiments. The result is a much-needed guide to planning animal-based experiments to ensure scientifically valid and reliable results. This book offers: Step-by-step guidance in diverse methods for approximating and refining sample size Detailed treatment of research topics specific to animal-based research, including pilot, feasibility and proof-of-concept studies Sample size approximation methods for different types of data − binary, continuous, ordinal, time to event − and different study types − description, comparison, nested designs, reference interval construction and dose-response studies Numerous worked examples, using real data from published papers, together with SAS and R code A Guide to Sample Size for Animal-based Studies is a must-have reference for preclinical and veterinary researchers, as well as ethical oversight committees and policymakers.

Preface vii Acknowledgements viii Part I. What is Sample Size? 1 The Sample Size Problem in Animal-Based Research 3 2 Sample Size Basics 9 3 Ten Strategies to Increase Information (and Reduce Sample Size) 17 Part II. Sample Size for Feasibility and Pilot Studies 4 Why Pilot Studies? 35 5 Operational Pilot Studies: ‘Can It Work?’ 47 6 Empirical and Translational Pilots 57 7 Feasibility Calculations: Arithmetic 81 8 Feasibility: Counting Subjects 89 Part III. Sample Size for Description 9 Descriptions and Summaries 103 10 Confidence Intervals and Precision 111 11 Prediction Intervals 127 12 Tolerance Intervals 133 13 Reference Intervals 143 Part IV. Sample Size for Comparison 14 Sample Size and Hypothesis Testing 155 15 A Bestiary of Effect Sizes 167 16 Comparing Two Groups: Continuous Outcomes 181 17 Comparing Two Groups: Proportions 189 18 Time-to-Event (Survival) Data 199 19 Comparing Multiple Factors 211 20 Hierarchical or Nested Data 233 21 Ordinal Data 249 22 Dose-Response Studies 257 Index 267