A Practical Approach to Quantitative Validation of Patient-Reported Outcomes: A Simulation-based Guide Using SAS

Description

A Practical Approach to Quantitative Validation of Patient-Reported Outcomes A Practical Approach to Quantitative Validation of Patient-Reported Outcomes A Simulation-Based Guide Using SAS In A Practical Approach to Quantitative Validation of Patient-Reported Outcomes, two distinguished researchers, with 50 years of collective research experience and hundreds of publications on patient-centered research, deliver a detailed and comprehensive exposition on the critical steps required for quantitative validation of patient-reported outcomes (PROs). The book provides an incisive and instructional explanation and discussion on major aspects of psychometric validation methodology on PROs, especially relevant for medical applications sponsored by the pharmaceutical industry, where SAS is the primary software, and evaluated in regulatory and other healthcare environments. Central topics include test-retest reliability, exploratory and confirmatory factor analyses, construct and criterion validity, responsiveness and sensitivity, interpretation of PRO scores and findings, and meaningful within-patient change and clinical important difference. The authors provide step-by-step guidance while walking readers through how to structure data prior to a PRO analysis and demonstrate how to implement analyses with simulated examples grounded in real-life scenarios. Readers will also find: A thorough introduction to patient-reported outcomes, including their definition, development, and psychometric validationComprehensive explorations of the validation workflow, including discussions of clinical trials as a data source for validation and the validation workflow for single and multi-item scalesIn-depth discussions of key concepts related to a validation of a measurement scaleSpecial attention is given to the US Food and Drug Administration (FDA) guidance on development and validation of the PROs, which lay the foundation and inspiration for the analytic methods executed A Practical Approach to Quantitative Validation of Patient-Reported Outcomes is a required reference that will benefit psychometricians, statisticians, biostatisticians, epidemiologists, health service and public health researchers, outcome research scientists, regulators, and payers. STATISTICS IN PRACTICE A series of practical books outlining the use of statistical techniques in a wide range of applications areas: HUMAN AND BIOLOGICAL SCIENCES EARTH AND ENVIRONMENTAL SCIENCES INDUSTRY, COMMERCE AND FINANCE Cover Title Page Copyright Page Contents Preface About the Authors Chapter 1 Introduction 1.1 What Is a PRO Measure? 1.2 Development of a PRO Measure 1.2.1 Concept Identification 1.2.1.1 Literature and Instrument Review 1.2.1.2 Patient-Centered Input 1.2.2 Item Development 1.2.3 Cognitive Interviews 1.2.4 Additional Considerations 1.2.5 Documentation of Development Process with Conceptual Framework 1.3 Psychometric Validation 1.3.1 Psychometric Evaluation Data 1.3.2 Psychometric Properties 1.3.2.1 Distributional Characteristics 1.3.2.2 Measurement Model Structure 1.3.2.3 Reliability 1.3.2.4 Construct Validity 1.3.2.5 Ability to Detect Change 1.3.2.6 Interpretation 1.4 Learning Through Simulations 1.5 Summary References Chapter 2 Validation Workflow 2.1 Clinical Trials as a Data Source for Validation 2.2 Validation Workflow for Single-Item Scales 2.3 Confirmatory Validation Workflow for Multi-item Multi-domain Scales 2.4 Validation Flow for a New Multi-item Multi-domain Scale 2.4.1 New Scale with Known Conceptual Framework 2.4.2 New Scale with Unknown Measurement Structure 2.5 Cross-Sectional Studies and Field Tests 2.6 Summary References Chapter 3 An Assessment of Classical Test Theory and Item Response Theory 3.1 Overview of Classical Test Theory 3.1.1 Basics 3.1.2 Illustration 3.1.3 Another Look 3.2 Person-Item Maps 3.2.1 CTT Revisited 3.2.2 Note on IRT 3.2.3 Implementation of Person-Item Maps 3.2.4 CTT-Based Scoring vs. IRT-Based Scoring 3.3 Summary References Chapter 4 Reliability 4.1 Reproducibility/Test–Retest 4.1.1 Measurement Error Model 4.1.2 Two Time Points 4.1.3 Random-Effects Model for ICC Estimation 4.1.4 Test–Retest Reliability Assessment in the Context of Clinical Studies 4.1.4.1 Pre-Treatment/Pre-Baseline Data 4.1.4.2 Post-Baseline Data 4.1.4.3 Time Period Between Observations 4.1.5 Spearman-Brown Prophecy Formula 4.1.6 Domain Score Test–Retest vs. Item Test–Retest 4.1.7 Observer-Based and Interviewer-Based Scales 4.1.8 Uncovering True Relationship Between Measurements 4.1.8.1 Accounting for Measurement Error 4.1.8.2 Measurement Error Model with Two Observations 4.2 Cronbach’s Alpha 4.2.1 Likert-Type Scales 4.2.2 Dichotomous Items 4.3 Summary References Chapter 5 Construct Validity and Criterion Validity 5.1 Exploratory Factor Analyses 5.1.1 Modeling Assumptions 5.1.2 Exploratory Factor Analysis Implementation 5.1.3 Evaluating the Number of Factors and Factor Loadings 5.1.3.1 Scree Plot 5.1.3.2 Correlated Latent Factors 5.1.3.3 Parallel Analysis with Reduced Correlation Matrix 5.1.3.4 Factor Loadings 5.2 Confirmatory Factor Analyses 5.2.1 Confirmatory Factor Analysis Model 5.2.2 Confirmatory Factor Analysis Model Implementation 5.2.3 Confirmatory Factor Analysis with Domains Represented by a Single Item 5.2.4 Second-Order Confirmatory Factor Analysis 5.2.4.1 Implementation of the Model with at Least Three First-Order Latent Domains 5.2.4.2 Implementation of the Model with Two First-Order Latent Domains 5.2.5 Formative vs. Reflective Model 5.2.6 Bifactor Model 5.2.7 Confirmatory Factor Analysis Using Polychoric Correlations 5.3 Convergent and Discriminant Validity 5.3.1 Convergent and Discriminant Validity Assessment 5.3.2 Convergent and Discriminant Validity Evaluation in a Clinical Study 5.4 Known-Groups Validity 5.5 Criterion Validity 5.6 Summary References Chapter 6 Responsiveness and Sensitivity 6.1 Ability to Detect Change 6.1.1 Definitions and Concepts 6.1.2 Ability to Detect Change Analysis Implementation 6.1.3 Correlation Analysis to Support Ability to Detect Change 6.1.4 Deconstructing Correlation Between Changes 6.2 Sensitivity to Treatment 6.2.1 What Is the Sensitivity to Treatment? 6.2.2 Concurrent Estimation of the Treatment Effects for a Multi-Domain Scale 6.2.2.1 Assessment of the Treatment Effect for a Single Domain 6.2.2.2 Assessment of the Treatment Effects for a Multi-Domain Scale 6.3 Summary References Chapter 7 Interpretation of Patient-Reported Outcome Findings 7.1 Meaningful Within-Patient Change 7.1.1 Definitions and Concepts 7.1.2 Anchor-Based Method to Assess Meaningful Within-Patient Change 7.1.3 Cumulative Distribution Functions to Supplement Anchor-Based Methods 7.2 Clinical Important Difference 7.2.1 Meaningful Within-Patient Change Versus Between-Group Difference 7.2.2 Anchor-Based Method to Assess Clinically Important Difference 7.3 Responder Analyses and Cumulative Distribution Functions 7.3.1 Treatment Effect Model 7.3.2 MWPC Application: A Responder Analysis 7.3.3 Using CDFs for Interpretation of Results 7.4 Summary References Index EULA