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

Preface xi

About the Authors xv

1 Introduction 1

1.1 What Is a PRO Measure? 1

1.2 Development of a PRO Measure 4

1.2.1 Concept Identification 4

1.2.1.1 Literature and Instrument Review 5

1.2.1.2 Patient- Centered Input 6

1.2.2 Item Development 9

1.2.3 Cognitive Interviews 11

1.2.4 Additional Considerations 12

1.2.5 Documentation of Development Process with Conceptual Framework 13

1.3 Psychometric Validation 15

1.3.1 Psychometric Evaluation Data 16

1.3.2 Psychometric Properties 17

1.3.2.1 Distributional Characteristics 19

1.3.2.2 Measurement Model Structure 20

1.3.2.3 Reliability 22

1.3.2.4 Construct Validity 23

1.3.2.5 Ability to Detect Change 24

1.3.2.6 Interpretation 25

1.4 Learning Through Simulations 26

1.5 Summary 27

References 28

2 Validation Workflow 35

2.1 Clinical Trials as a Data Source for Validation 35

2.2 Validation Workflow for Single- Item Scales 39

2.3 Confirmatory Validation Workflow for Multi- item Multi- domain Scales
43

2.4 Validation Flow for a New Multi- item Multi- domain Scale 45

2.4.1 New Scale with Known Conceptual Framework 45

2.4.2 New Scale with Unknown Measurement Structure 47

2.5 Cross- Sectional Studies and Field Tests 48

2.6 Summary 49

References 49

3 An Assessment of Classical Test Theory and Item Response Theory 51

3.1 Overview of Classical Test Theory 52

3.1.1 Basics 52

3.1.2 Illustration 52

3.1.3 Another Look 53

3.2 Person- Item Maps 55

3.2.1 CTT Revisited 55

3.2.2 Note on IRT 56

3.2.3 Implementation of Person- Item Maps 58

3.2.4 CTT- Based Scoring vs. IRT- Based Scoring 69

3.3 Summary 78

References 80

4 Reliability 83

4.1 Reproducibility/TestRetest 85

4.1.1 Measurement Error Model 85

4.1.2 Two Time Points 87

4.1.3 Random- Effects Model for ICC Estimation 90

4.1.4 TestRetest Reliability Assessment in the Context of Clinical Studies
95

4.1.4.1 Pre- Treatment/Pre- Baseline Data 95

4.1.4.2 Post- Baseline Data 97

4.1.4.3 Time Period Between Observations 101

4.1.5 Spearman- Brown Prophecy Formula 104

4.1.6 Domain Score TestRetest vs. Item TestRetest 109

4.1.7 Observer- Based and Interviewer- Based Scales 111

4.1.8 Uncovering True Relationship Between Measurements 113

4.1.8.1 Accounting for Measurement Error 113

4.1.8.2 Measurement Error Model with Two Observations 122

4.2 Cronbachs Alpha 129

4.2.1 Likert- Type Scales 129

4.2.2 Dichotomous Items 139

4.3 Summary 148

References 148

5 Construct Validity and Criterion Validity 151

5.1 Exploratory Factor Analyses 153

5.1.1 Modeling Assumptions 153

5.1.2 Exploratory Factor Analysis Implementation 159

5.1.3 Evaluating the Number of Factors and Factor Loadings 165

5.1.3.1 Scree Plot 165

5.1.3.2 Correlated Latent Factors 168

5.1.3.3 Parallel Analysis with Reduced Correlation Matrix 171

5.1.3.4 Factor Loadings 175

5.2 Confirmatory Factor Analyses 179

5.2.1 Confirmatory Factor Analysis Model 179

5.2.2 Confirmatory Factor Analysis Model Implementation 183

5.2.3 Confirmatory Factor Analysis with Domains Represented by a Single Item
192

5.2.4 Second- Order Confirmatory Factor Analysis 204

5.2.4.1 Implementation of the Model with at Least Three First- Order Latent
Domains 204

5.2.4.2 Implementation of the Model with Two First- Order Latent Domains
207

5.2.5 Formative vs. Reflective Model 213

5.2.6 Bifactor Model 219

5.2.7 Confirmatory Factor Analysis Using Polychoric Correlations 227

5.3 Convergent and Discriminant Validity 231

5.3.1 Convergent and Discriminant Validity Assessment 231

5.3.2 Convergent and Discriminant Validity Evaluation in a Clinical Study
232

5.4 Known- Groups Validity 237

5.5 Criterion Validity 242

5.6 Summary 247

References 248

6 Responsiveness and Sensitivity 251

6.1 Ability to Detect Change 252

6.1.1 Definitions and Concepts 252

6.1.2 Ability to Detect Change Analysis Implementation 255

6.1.3 Correlation Analysis to Support Ability to Detect Change 263

6.1.4 Deconstructing Correlation Between Changes 268

6.2 Sensitivity to Treatment 270

6.2.1 What Is the Sensitivity to Treatment? 270

6.2.2 Concurrent Estimation of the Treatment Effects for a Multi- Domain
Scale 273

6.2.2.1 Assessment of the Treatment Effect for a Single Domain 273

6.2.2.2 Assessment of the Treatment Effects for a Multi- Domain Scale 279

6.3 Summary 292

References 293

7 Interpretation of Patient- Reported Outcome Findings 295

7.1 Meaningful Within- Patient Change 296

7.1.1 Definitions and Concepts 296

7.1.2 Anchor- Based Method to Assess Meaningful Within- Patient Change 298

7.1.3 Cumulative Distribution Functions to Supplement Anchor- Based Methods
310

7.2 Clinical Important Difference 315

7.2.1 Meaningful Within- Patient Change Versus Between- Group Difference
315

7.2.2 Anchor- Based Method to Assess Clinically Important Difference 316

7.3 Responder Analyses and Cumulative Distribution Functions 320

7.3.1 Treatment Effect Model 320

7.3.2 MWPC Application: A Responder Analysis 323

7.3.3 Using CDFs for Interpretation of Results 325

7.4 Summary 331

References 332

Index 335

ANDREW G. BUSHMAKIN earned his M.S. in applied mathematics and physics from the National Research Nuclear University (former Moscow Engineering Physics Institute, Moscow, Russia). He has more than 20 years of experience in mathematical modeling and data analysis. He is a director of biostatistics in the Statistical Research and Data Science Center at Pfizer Inc. He has co-authored numerous articles and presentations on topics ranging from mathematical modeling of neutron physics processes to patient-reported outcomes, as well as several monographs.

JOSEPH C. CAPPELLERI earned his M.S. in statistics from the City University of New York, Ph.D. in psychometrics from Cornell University, and M.P.H. in epidemiology from Harvard University. He is an Executive Director of biostatistics in the Statistical Research and Data Science Center at Pfizer Inc. As an adjunct professor, he has served on the faculties of Brown University, University of Connecticut, and Tufts Medical Center. He has delivered numerous conference presentations and has published extensively on clinical and methodological topics. He is a fellow of the American Statistical Association and recipient of the ISPOR Avedis Donabedian Outcomes Research Lifetime Achievement Award.