Missing data in clinical studies:
Missing Data in Clinical Studies provides a comprehensive account of the problems arising when data from clinical and related studies are incomplete, and presents the reader with approaches to effectively address them. The text provides a critique of conventional and simple methods before moving on...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Buch |
| Sprache: | English |
| Veröffentlicht: |
Chichester [u.a.]
Wiley
2007
|
| Schriftenreihe: | Statistics in practice
|
| Schlagworte: | |
| Online-Zugang: | Inhaltsverzeichnis |
| Zusammenfassung: | Missing Data in Clinical Studies provides a comprehensive account of the problems arising when data from clinical and related studies are incomplete, and presents the reader with approaches to effectively address them. The text provides a critique of conventional and simple methods before moving on to discuss more advanced approaches. The authors focus on practical and modeling concepts, providing an extensive set of case studies to illustrate the problems described. Provides a practical guide to the analysis of clinical trials and related studies with missing data. Examines the problems caused by missing data, enabling a complete understanding of how to overcome them. Presents conventional, simple methods to tackle these problems, before addressing more advanced approaches, including sensitivity analysis, and the MAR missingness mechanism. Illustrated throughout with real-life case studies and worked examples from clinical trials. Details the use and implementation of the necessary statistical software, primarily SAS. Missing Data in Clinical Studies has been developed through a series of courses and lectures. Its practical approach will appeal to applied statisticians and biomedical researchers, in particular those in the biopharmaceutical industry, medical and public health organisations. Graduate students of biostatistics will also find much of benefit. |
| Beschreibung: | XX, 504 S. Ill., graph. Darst. |
| ISBN: | 0470849819 9780470849811 |
Internformat
MARC
| LEADER | 00000nam a2200000 c 4500 | ||
|---|---|---|---|
| 001 | BV022205242 | ||
| 003 | DE-604 | ||
| 005 | 20141201 | ||
| 007 | t | ||
| 008 | 061228s2007 ad|| |||| 00||| eng d | ||
| 020 | |a 0470849819 |c (hbk.) : £55.00 |9 0-470-84981-9 | ||
| 020 | |a 9780470849811 |9 978-0-470-84981-1 | ||
| 035 | |a (OCoLC)74967088 | ||
| 035 | |a (DE-599)BVBBV022205242 | ||
| 040 | |a DE-604 |b ger |e rakwb | ||
| 041 | 0 | |a eng | |
| 049 | |a DE-20 |a DE-19 | ||
| 050 | 0 | |a R853.C55 | |
| 082 | 0 | |a 610.724 |2 22 | |
| 084 | |a XF 3500 |0 (DE-625)152766: |2 rvk | ||
| 100 | 1 | |a Molenberghs, Geert |e Verfasser |4 aut | |
| 245 | 1 | 0 | |a Missing data in clinical studies |c Geert Molenberghs ; Michael G. Kenward |
| 264 | 1 | |a Chichester [u.a.] |b Wiley |c 2007 | |
| 300 | |a XX, 504 S. |b Ill., graph. Darst. | ||
| 336 | |b txt |2 rdacontent | ||
| 337 | |b n |2 rdamedia | ||
| 338 | |b nc |2 rdacarrier | ||
| 490 | 0 | |a Statistics in practice | |
| 520 | 3 | |a Missing Data in Clinical Studies provides a comprehensive account of the problems arising when data from clinical and related studies are incomplete, and presents the reader with approaches to effectively address them. The text provides a critique of conventional and simple methods before moving on to discuss more advanced approaches. The authors focus on practical and modeling concepts, providing an extensive set of case studies to illustrate the problems described. Provides a practical guide to the analysis of clinical trials and related studies with missing data. Examines the problems caused by missing data, enabling a complete understanding of how to overcome them. Presents conventional, simple methods to tackle these problems, before addressing more advanced approaches, including sensitivity analysis, and the MAR missingness mechanism. Illustrated throughout with real-life case studies and worked examples from clinical trials. Details the use and implementation of the necessary statistical software, primarily SAS. Missing Data in Clinical Studies has been developed through a series of courses and lectures. Its practical approach will appeal to applied statisticians and biomedical researchers, in particular those in the biopharmaceutical industry, medical and public health organisations. Graduate students of biostatistics will also find much of benefit. | |
| 650 | 4 | |a Médecine - Recherche - Méthodes statistiques | |
| 650 | 4 | |a Observations manquantes (Statistique) | |
| 650 | 4 | |a Études cliniques - Méthodologie | |
| 650 | 4 | |a Clinical Trials as Topic | |
| 650 | 4 | |a Clinical trials |x Statistical methods | |
| 650 | 4 | |a Data Interpretation, Statistical | |
| 650 | 4 | |a Missing observations (Statistics) | |
| 650 | 4 | |a Research Design | |
| 650 | 4 | |a Statistics as Topic |x methods | |
| 650 | 0 | 7 | |a Klinisches Experiment |0 (DE-588)4164223-5 |2 gnd |9 rswk-swf |
| 655 | 7 | |0 (DE-588)4056995-0 |a Statistik |2 gnd-content | |
| 689 | 0 | 0 | |a Klinisches Experiment |0 (DE-588)4164223-5 |D s |
| 689 | 0 | |5 DE-604 | |
| 700 | 1 | |a Kenward, Michael G. |d 1956- |e Verfasser |0 (DE-588)1030402035 |4 aut | |
| 856 | 4 | 2 | |m HBZ Datenaustausch |q application/pdf |u http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=015416640&sequence=000002&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA |3 Inhaltsverzeichnis |
| 999 | |a oai:aleph.bib-bvb.de:BVB01-015416640 | ||
Datensatz im Suchindex
| _version_ | 1804136182444982272 |
|---|---|
| adam_text | Contents
Preface xv
Acknowledgements xix
I Preliminaries 1
1 Introduction 3
1.1 From Imbalance to the Field of Missing Data Research 3
1.2 Incomplete Data in Clinical Studies 5
1.3 MAR, MNAR. and Sensitivity Analysis H
1.4 Outline of the Book 9
2 Key Examples 11
2.1 Introduction 11
2.2 The Vorozole Study 12
2.3 The Orthodontic Growth Data 12
2.4 Mastitis in Dairy Cattle 14
2.5 The Depression Trials 14
2.6 The Fluvoxamine Trial 17
2.7 The Toenail Data IX
2.8 Age Related Macular Degeneration Trial 20
2.9 The Analgesic Trial 22
2.10 The Slovenian Public Opinion Survey 24
3 Terminology and Framework 27
3.1 Modelling Incompleteness 27
3.2 Terminology 29
3.3 Missing Data Frameworks 30
3.4 Missing Data Mechanisms 51
3.5 Ignorability 5 3
3.6 Pattern Mixture Models 34
vii
viii Contents
II Classical Techniques and the Need for Modelling 39
4 A Perspective on Simple Methods 41
4.1 Introduction 41
4.1.1 Measurement model 41
4.1.2 Method lor handling missingness 42
4.2 Simple Methods 42
4.2.1 Complete case analysis 42
4.2.2 Imputation methods 43
4.2.3 Last observation carried forward 45
4.3 Problems with Complete Case Analysis and Last Observation Carried
Forward 47
4.4 Using the Available Cases: a Frequentist versus a Likelihood Perspective 50
4.4.1 A bivariate normal population 50
4.4.2 An incomplete contingency table 52
4.5 Intention to Treat 5 3
4.6 Concluding Remarks 54
5 Analysis of the Orthodontic Growth Data 5 5
5.1 Introduction and Models 55
5.2 The Original. Complete Data 56
5.3 Direct Likelihood 57
5.4 Comparison of Analyses 59
5.5 Example SAS Code for Multivariate Linear Models 62
5.6 Comparative Power under Different Covariance Structures 6 3
5.7 Concluding Remarks 65
6 Analysis of the Depression Trials 67
6.1 View 1: Longitudinal Analysis 68
6.2 Views 2a and 2b and All versus Two Treatment Arms 72
III Missing at Random and Ignorability 75
7 The Direct Likelihood Method 77
7.1 Introduction 77
7.2 Ignorable Analyses in Practice 78
7.3 The Linear Mixed Model 79
7.4 Analysis of the Toenail Data 82
7.5 The Generalized Linear Mixed Model 85
7.6 The Depression Trials 90
7.7 The Analgesic Trial 91
8 The Expectation Maximization Algorithm 93
8.1 Introduction 9 3
8.2 The Algorithm 94
8.2.1 The initial step 94
Contents ix
8.2.2 The E step 95
8.2.3 The M step 95
8.3 Missing Information 95
8.4 Rate of Convergence 96
8.5 EM Acceleration 97
8.6 Calculation of Precision Estimates 98
8.7 A Simple Illustration 98
8.8 Concluding Remarks 103
9 Multiple Imputation 105
9.1 Introduction 105
9.2 The Basic Procedure 105
9.3 Theoretical Justification 107
9.4 Inference under Multiple Imputation 108
9.5 Efficiency 109
9.6 Making Proper Imputations 110
9.7 Some Roles for Multiple Imputation 115
9.8 Concluding Remarks 117
10 Weighted Estimating Equations 119
10.1 Introduction 119
10.2 Inverse Probability Weighting 120
10.3 Generalized Estimating Equations for Marginal Models 123
10.3.1 Marginal models for non normal data 123
10.3.2 Generalized estimating equations 12 3
10.3.3 A method based on linearization 124
10.4 Weighted Generalized Estimating Equations 126
10.5 The Depression Trials 126
10.6 The Analgesic Trial 128
10.7 Double Robustness 130
10.8 Concluding Remarks 133
11 Combining GEE and MI 135
11.1 Introduction 135
11.2 Data Generation and Fitting 136
11.2.1 The Bahadur model 136
11.2.2 A transition model 137
11.3 MI GEE and Mi Transition 137
11.4 An Asymptotic Simulation Study 137
11.4.1 Design 138
11.4.2 Results 139
11.5 Concluding Remarks 142
12 Likelihood Based Frequentist Inference 145
12.1 Introduction 145
12.2 Information and Sampling Distributions 147
12.3 Bivariate Normal Data 149
12.4 Bivariate Binary Data 153
12.5 Implications for Standard Software 156
x Contents
12.6 Analysis of the Fluvoxamine Trial 158
12.7 The Muscatine Coronary Risk Factor Study 160
12.8 The Crepeau Data 161
12.9 Concluding Remarks 161
13 Analysis of the Age Related Macular Degeneration
Trial 163
13.1 Introduction 163
13.2 Direct Likelihood Analysis of the Continuous Outcome 164
13.3 Weighted Generalized Estimating Equations 165
13.4 Direct Likelihood Analysis of the Binary Outcome 167
13.5 Multiple Imputation 168
13.6 Concluding Remarks 170
14 Incomplete Data and SAS 171
14.1 Introduction 171
14.2 Complete Case Analysis 171
14.3 Last Observation Carried Forward 173
14.4 Direct Likelihood 174
14.5 Weighted Estimating Equations 175
14.6 Multiple Imputation 176
14.6.1 The MI procedure for the imputation task 177
14.6.2 The analysis task 178
14.6.3 The inference task 181
14.6.4 The MI procedure to create monotone missingness 182
IV Missing Not at Random 183
15 Selection Models 185
15.1 Introduction 185
15.2 The Diggle Kenward Model for Continuous Outcomes 186
15.3 Illustration and SAS Implementation 188
15.4 An MNAR Dale Model 194
15.4.1 Likelihood function 194
15.4.2 Analysis of the fluvoxamine trial 197
15.4.3 The tinea pedis study 202
15.5 A Model for Non monotone Missingness 204
15.5.1 Analysis of the fluvoxamine trial 207
15.6 Concluding Remarks 212
16 Pattern Mixture Models 215
16.1 Introduction 215
16.2 A Simple Gaussian Illustration 216
16.3 A Paradox 219
16.4 Strategies to Fit Pattern Mixture Models 220
16.5 Applying Identifying Restrictions 221
16.6 Pattern Mixture Analysis of the Vorozole Study 222
Contents xi
16.6.1 Derivations 223
16.6.2 Application to the vorozole study 224
16.7 A Clinical Trial in Alzheimer s Disease 237
16.8 Analysis of the Fluvoxamine Trial 242
16.8.1 Selection modelling 242
16.8.2 Pattern mixture modelling 243
16.8.3 Comparison 246
16.9 Concluding Remarks 246
17 Shared Parameter Models 249
18 Protective Estimation 253
18.1 Introduction 253
18.2 Brown s Protective Estimator for Gaussian Data 254
18.3 A Protective Estimator for Categorical Data 256
18.3.1 Likelihood estimation 260
18.3.2 Pseudo likelihood estimation 263
18.3.3 Variance estimation 264
18.3.4 Analysis of artificial data 269
18.3.5 Analysis of the fluvoxamine trial 270
18.3.6 Presence or absence of colds 274
18.4 A Protective Estimator for Gaussian Data 275
18.4.1 Notation and maximum likelihood 275
18.4.2 Protective estimator 277
18.4.3 The six cities study 279
18.5 Concluding Remarks 282
V Sensitivity Analysis 283
19 MNAR, MAR, and the Nature of Sensitivity 285
19.1 Introduction 285
19.2 Every MNAR Model Has an MAR Bodyguard 286
19.2.1 A bivariate outcome with dropout 289
19.2.2 A trivariate outcome with dropout 290
19.2.3 A bivariate outcome with non monotone missingness 291
19.3 The General Case of Incomplete Contingency Tables 292
19.3.1 A bivariate contingency table with dropout 293
19.3.2 A bivariate contingency table with non monotone missingness 294
19.4 The Slovenian Public Opinion Survey 295
19.4.1 The BRD models 296
19.4.2 Initial analysis 296
19.4.3 BRD analysis 299
19.5 Implications for Formal and Informal Model Selection 302
19.6 Behaviour of the Likelihood Ratio Test for MAR versus MNAR 305
19.6.1 Simulated null distributions 3O6
19.6.2 Performance of bootstrap approaches 307
19.7 Concluding Remarks 311
xii Contents
20 Sensitivity Happens 313
20.1 Introduction 313
20.2 A Range of MNAR Models 314
20.3 Identifiability Problems 320
20.4 Analysis of the Fluvoxamine Trial 322
20.5 Concluding Remarks 327
21 Regions of Ignorance and Uncertainty 329
21.1 Introduction 329
21.2 Prevalence of HIV in Kenya 330
21.3 Uncertainty and Sensitivity 330
21.4 Models for Monotone Patterns 331
21.5 Models for Non monotone Patterns 332
21.6 Formalizing Ignorance and Uncertainty 333
21.7 Analysis of the Fluvoxamine Trial 338
21.7.1 Identified models 339
21.7.2 Sensitivity analysis 341
21.8 Artificial Examples 345
21.9 The Slovenian Public Opinion Survey 348
21.10 Some Theoretical Considerations 351
21.11 Concluding Remarks 351
22 Local and Global Influence Methods 353
22.1 Introduction 353
22.2 Gaussian Outcomes 354
22.2.1 Application to the Diggle Kenward model 356
22.2.2 The special case of three measurements 359
22.3 Mastitis in Dairy Cattle 360
22.3.1 Informal sensitivity analysis 361
22.3.2 Local influence approach 367
22.4 Alternative Local Influence Approaches 373
22.5 The Milk Protein Content Trial 375
22.5.1 Informal sensitivity analysis 377
22.5.2 Formal sensitivity analysis 386
22.6 Analysis of the Depression Trials 398
22.7 A Local Influence Approach for Ordinal Data with Dropout 405
22.8 Analysis of the Fluvoxamine Data 406
22.9 A Local Influence Approach for Incomplete Binary Data 410
22.10 Analysis of the Fluvoxamine Data 411
22.11 Concluding Remarks 415
23 The Nature of Local Influence 417
23.1 Introduction 417
23.2 The Rats Data 418
23.3 Analysis and Sensitivity Analysis of the Rats Data 419
23.4 Local Influence Methods and Their Behaviour 422
23.4.1 Effect of sample size 423
23.4.2 Point wise confidence limits and simultaneous confidence bounds
for the local influence measure 424
Contents xiii
23.4.3 Anomalies in the missingness mechanism 425
23.4.4 Anomalies in the measurement model 428
23.5 Concluding Remarks 430
24 A Latent Class Mixture Model for Incomplete
Longitudinal Gaussian Data 431
24.1 Introduction 431
24.2 Latent Class Mixture Models 431
24.3 The Likelihood Function and Estimation 434
24.3.1 likelihood function 434
24.3.2 Estimation using the EM algorithm 436
24.3.3 The E step 437
24.3.4 The M step 438
24.3.5 Some remarks regarding the EM algorithm 439
24.4 Classification 440
24.5 Simulation Study 441
24.5.1 A simplification of the latent class mixture model 441
24.5.2 Design 442
24.5.3 Results 443
24.6 Analysis of the Depression Trials 446
24.6.1 Formulating a latent class mixture model 446
24.6.2 A sensitivity analysis 449
24.7 Concluding Remarks 450
VI Case Studies 451
25 The Age Related Macular Degeneration Trial 453
25.1 Selection Models and Local Influence 453
25.2 Local Influence Analysis 455
25.3 Pattern Mixture Models 458
25.4 Concluding Remarks 459
26 The Vorozole Study 461
26.1 Introduction 461
26.2 Exploring the Vorozole Data 461
26.2.1 Average evolution 461
26.2.2 Variance structure 464
26.2.3 Correlation structure 465
26.2.4 Missing data aspects 466
26.3 A Selection Model for the Vorozole Study 471
26.4 A Pattern Mixture Model for the Vorozole Study 475
26.5 Concluding Remarks 481
References 483
Index 497
|
| adam_txt |
Contents
Preface xv
Acknowledgements xix
I Preliminaries 1
1 Introduction 3
1.1 From Imbalance to the Field of Missing Data Research 3
1.2 Incomplete Data in Clinical Studies 5
1.3 MAR, MNAR. and Sensitivity Analysis H
1.4 Outline of the Book 9
2 Key Examples 11
2.1 Introduction 11
2.2 The Vorozole Study 12
2.3 The Orthodontic Growth Data 12
2.4 Mastitis in Dairy Cattle 14
2.5 The Depression Trials 14
2.6 The Fluvoxamine Trial 17
2.7 The Toenail Data IX
2.8 Age Related Macular Degeneration Trial 20
2.9 The Analgesic Trial 22
2.10 The Slovenian Public Opinion Survey 24
3 Terminology and Framework 27
3.1 Modelling Incompleteness 27
3.2 Terminology 29
3.3 Missing Data Frameworks 30
3.4 Missing Data Mechanisms 51
3.5 Ignorability 5 3
3.6 Pattern Mixture Models 34
vii
viii Contents
II Classical Techniques and the Need for Modelling 39
4 A Perspective on Simple Methods 41
4.1 Introduction 41
4.1.1 Measurement model 41
4.1.2 Method lor handling missingness 42
4.2 Simple Methods 42
4.2.1 Complete case analysis 42
4.2.2 Imputation methods 43
4.2.3 Last observation carried forward 45
4.3 Problems with Complete Case Analysis and Last Observation Carried
Forward 47
4.4 Using the Available Cases: a Frequentist versus a Likelihood Perspective 50
4.4.1 A bivariate normal population 50
4.4.2 An incomplete contingency table 52
4.5 Intention to Treat 5 3
4.6 Concluding Remarks 54
5 Analysis of the Orthodontic Growth Data 5 5
5.1 Introduction and Models 55
5.2 The Original. Complete Data 56
5.3 Direct Likelihood 57
5.4 Comparison of Analyses 59
5.5 Example SAS Code for Multivariate Linear Models 62
5.6 Comparative Power under Different Covariance Structures 6 3
5.7 Concluding Remarks 65
6 Analysis of the Depression Trials 67
6.1 View 1: Longitudinal Analysis 68
6.2 Views 2a and 2b and All versus Two Treatment Arms 72
III Missing at Random and Ignorability 75
7 The Direct Likelihood Method 77
7.1 Introduction 77
7.2 Ignorable Analyses in Practice 78
7.3 The Linear Mixed Model 79
7.4 Analysis of the Toenail Data 82
7.5 The Generalized Linear Mixed Model 85
7.6 The Depression Trials 90
7.7 The Analgesic Trial 91
8 The Expectation Maximization Algorithm 93
8.1 Introduction 9 3
8.2 The Algorithm 94
8.2.1 The initial step 94
Contents ix
8.2.2 The E step 95
8.2.3 The M step 95
8.3 Missing Information 95
8.4 Rate of Convergence 96
8.5 EM Acceleration 97
8.6 Calculation of Precision Estimates 98
8.7 A Simple Illustration 98
8.8 Concluding Remarks 103
9 Multiple Imputation 105
9.1 Introduction 105
9.2 The Basic Procedure 105
9.3 Theoretical Justification 107
9.4 Inference under Multiple Imputation 108
9.5 Efficiency 109
9.6 Making Proper Imputations 110
9.7 Some Roles for Multiple Imputation 115
9.8 Concluding Remarks 117
10 Weighted Estimating Equations 119
10.1 Introduction 119
10.2 Inverse Probability Weighting 120
10.3 Generalized Estimating Equations for Marginal Models 123
10.3.1 Marginal models for non normal data 123
10.3.2 Generalized estimating equations 12 3
10.3.3 A method based on linearization 124
10.4 Weighted Generalized Estimating Equations 126
10.5 The Depression Trials 126
10.6 The Analgesic Trial 128
10.7 Double Robustness 130
10.8 Concluding Remarks 133
11 Combining GEE and MI 135
11.1 Introduction 135
11.2 Data Generation and Fitting 136
11.2.1 The Bahadur model 136
11.2.2 A transition model 137
11.3 MI GEE and Mi Transition 137
11.4 An Asymptotic Simulation Study 137
11.4.1 Design 138
11.4.2 Results 139
11.5 Concluding Remarks 142
12 Likelihood Based Frequentist Inference 145
12.1 Introduction 145
12.2 Information and Sampling Distributions 147
12.3 Bivariate Normal Data 149
12.4 Bivariate Binary Data 153
12.5 Implications for Standard Software 156
x Contents
12.6 Analysis of the Fluvoxamine Trial 158
12.7 The Muscatine Coronary Risk Factor Study 160
12.8 The Crepeau Data 161
12.9 Concluding Remarks 161
13 Analysis of the Age Related Macular Degeneration
Trial 163
13.1 Introduction 163
13.2 Direct Likelihood Analysis of the Continuous Outcome 164
13.3 Weighted Generalized Estimating Equations 165
13.4 Direct Likelihood Analysis of the Binary Outcome 167
13.5 Multiple Imputation 168
13.6 Concluding Remarks 170
14 Incomplete Data and SAS 171
14.1 Introduction 171
14.2 Complete Case Analysis 171
14.3 Last Observation Carried Forward 173
14.4 Direct Likelihood 174
14.5 Weighted Estimating Equations 175
14.6 Multiple Imputation 176
14.6.1 The MI procedure for the imputation task 177
14.6.2 The analysis task 178
14.6.3 The inference task 181
14.6.4 The MI procedure to create monotone missingness 182
IV Missing Not at Random 183
15 Selection Models 185
15.1 Introduction 185
15.2 The Diggle Kenward Model for Continuous Outcomes 186
15.3 Illustration and SAS Implementation 188
15.4 An MNAR Dale Model 194
15.4.1 Likelihood function 194
15.4.2 Analysis of the fluvoxamine trial 197
15.4.3 The tinea pedis study 202
15.5 A Model for Non monotone Missingness 204
15.5.1 Analysis of the fluvoxamine trial 207
15.6 Concluding Remarks 212
16 Pattern Mixture Models 215
16.1 Introduction 215
16.2 A Simple Gaussian Illustration 216
16.3 A Paradox 219
16.4 Strategies to Fit Pattern Mixture Models 220
16.5 Applying Identifying Restrictions 221
16.6 Pattern Mixture Analysis of the Vorozole Study 222
Contents xi
16.6.1 Derivations 223
16.6.2 Application to the vorozole study 224
16.7 A Clinical Trial in Alzheimer's Disease 237
16.8 Analysis of the Fluvoxamine Trial 242
16.8.1 Selection modelling 242
16.8.2 Pattern mixture modelling 243
16.8.3 Comparison 246
16.9 Concluding Remarks 246
17 Shared Parameter Models 249
18 Protective Estimation 253
18.1 Introduction 253
18.2 Brown's Protective Estimator for Gaussian Data 254
18.3 A Protective Estimator for Categorical Data 256
18.3.1 Likelihood estimation 260
18.3.2 Pseudo likelihood estimation 263
18.3.3 Variance estimation 264
18.3.4 Analysis of artificial data 269
18.3.5 Analysis of the fluvoxamine trial 270
18.3.6 Presence or absence of colds 274
18.4 A Protective Estimator for Gaussian Data 275
18.4.1 Notation and maximum likelihood 275
18.4.2 Protective estimator 277
18.4.3 The six cities study 279
18.5 Concluding Remarks 282
V Sensitivity Analysis 283
19 MNAR, MAR, and the Nature of Sensitivity 285
19.1 Introduction 285
19.2 Every MNAR Model Has an MAR Bodyguard 286
19.2.1 A bivariate outcome with dropout 289
19.2.2 A trivariate outcome with dropout 290
19.2.3 A bivariate outcome with non monotone missingness 291
19.3 The General Case of Incomplete Contingency Tables 292
19.3.1 A bivariate contingency table with dropout 293
19.3.2 A bivariate contingency table with non monotone missingness 294
19.4 The Slovenian Public Opinion Survey 295
19.4.1 The BRD models 296
19.4.2 Initial analysis 296
19.4.3 BRD analysis 299
19.5 Implications for Formal and Informal Model Selection 302
19.6 Behaviour of the Likelihood Ratio Test for MAR versus MNAR 305
19.6.1 Simulated null distributions 3O6
19.6.2 Performance of bootstrap approaches 307
19.7 Concluding Remarks 311
xii Contents
20 Sensitivity Happens 313
20.1 Introduction 313
20.2 A Range of MNAR Models 314
20.3 Identifiability Problems 320
20.4 Analysis of the Fluvoxamine Trial 322
20.5 Concluding Remarks 327
21 Regions of Ignorance and Uncertainty 329
21.1 Introduction 329
21.2 Prevalence of HIV in Kenya 330
21.3 Uncertainty and Sensitivity 330
21.4 Models for Monotone Patterns 331
21.5 Models for Non monotone Patterns 332
21.6 Formalizing Ignorance and Uncertainty 333
21.7 Analysis of the Fluvoxamine Trial 338
21.7.1 Identified models 339
21.7.2 Sensitivity analysis 341
21.8 Artificial Examples 345
21.9 The Slovenian Public Opinion Survey 348
21.10 Some Theoretical Considerations 351
21.11 Concluding Remarks 351
22 Local and Global Influence Methods 353
22.1 Introduction 353
22.2 Gaussian Outcomes 354
22.2.1 Application to the Diggle Kenward model 356
22.2.2 The special case of three measurements 359
22.3 Mastitis in Dairy Cattle 360
22.3.1 Informal sensitivity analysis 361
22.3.2 Local influence approach 367
22.4 Alternative Local Influence Approaches 373
22.5 The Milk Protein Content Trial 375
22.5.1 Informal sensitivity analysis 377
22.5.2 Formal sensitivity analysis 386
22.6 Analysis of the Depression Trials 398
22.7 A Local Influence Approach for Ordinal Data with Dropout 405
22.8 Analysis of the Fluvoxamine Data 406
22.9 A Local Influence Approach for Incomplete Binary Data 410
22.10 Analysis of the Fluvoxamine Data 411
22.11 Concluding Remarks 415
23 The Nature of Local Influence 417
23.1 Introduction 417
23.2 The Rats Data 418
23.3 Analysis and Sensitivity Analysis of the Rats Data 419
23.4 Local Influence Methods and Their Behaviour 422
23.4.1 Effect of sample size 423
23.4.2 Point wise confidence limits and simultaneous confidence bounds
for the local influence measure 424
Contents xiii
23.4.3 Anomalies in the missingness mechanism 425
23.4.4 Anomalies in the measurement model 428
23.5 Concluding Remarks 430
24 A Latent Class Mixture Model for Incomplete
Longitudinal Gaussian Data 431
24.1 Introduction 431
24.2 Latent Class Mixture Models 431
24.3 The Likelihood Function and Estimation 434
24.3.1 likelihood function 434
24.3.2 Estimation using the EM algorithm 436
24.3.3 The E step 437
24.3.4 The M step 438
24.3.5 Some remarks regarding the EM algorithm 439
24.4 Classification 440
24.5 Simulation Study 441
24.5.1 A simplification of the latent class mixture model 441
24.5.2 Design 442
24.5.3 Results 443
24.6 Analysis of the Depression Trials 446
24.6.1 Formulating a latent class mixture model 446
24.6.2 A sensitivity analysis 449
24.7 Concluding Remarks 450
VI Case Studies 451
25 The Age Related Macular Degeneration Trial 453
25.1 Selection Models and Local Influence 453
25.2 Local Influence Analysis 455
25.3 Pattern Mixture Models 458
25.4 Concluding Remarks 459
26 The Vorozole Study 461
26.1 Introduction 461
26.2 Exploring the Vorozole Data 461
26.2.1 Average evolution 461
26.2.2 Variance structure 464
26.2.3 Correlation structure 465
26.2.4 Missing data aspects 466
26.3 A Selection Model for the Vorozole Study 471
26.4 A Pattern Mixture Model for the Vorozole Study 475
26.5 Concluding Remarks 481
References 483
Index 497 |
| any_adam_object | 1 |
| any_adam_object_boolean | 1 |
| author | Molenberghs, Geert Kenward, Michael G. 1956- |
| author_GND | (DE-588)1030402035 |
| author_facet | Molenberghs, Geert Kenward, Michael G. 1956- |
| author_role | aut aut |
| author_sort | Molenberghs, Geert |
| author_variant | g m gm m g k mg mgk |
| building | Verbundindex |
| bvnumber | BV022205242 |
| callnumber-first | R - Medicine |
| callnumber-label | R853 |
| callnumber-raw | R853.C55 |
| callnumber-search | R853.C55 |
| callnumber-sort | R 3853 C55 |
| callnumber-subject | R - General Medicine |
| classification_rvk | XF 3500 |
| ctrlnum | (OCoLC)74967088 (DE-599)BVBBV022205242 |
| dewey-full | 610.724 |
| dewey-hundreds | 600 - Technology (Applied sciences) |
| dewey-ones | 610 - Medicine and health |
| dewey-raw | 610.724 |
| dewey-search | 610.724 |
| dewey-sort | 3610.724 |
| dewey-tens | 610 - Medicine and health |
| discipline | Medizin |
| discipline_str_mv | Medizin |
| format | Book |
| fullrecord | <?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>03316nam a2200505 c 4500</leader><controlfield tag="001">BV022205242</controlfield><controlfield tag="003">DE-604</controlfield><controlfield tag="005">20141201 </controlfield><controlfield tag="007">t</controlfield><controlfield tag="008">061228s2007 ad|| |||| 00||| eng d</controlfield><datafield tag="020" ind1=" " ind2=" "><subfield code="a">0470849819</subfield><subfield code="c">(hbk.) : £55.00</subfield><subfield code="9">0-470-84981-9</subfield></datafield><datafield tag="020" ind1=" " ind2=" "><subfield code="a">9780470849811</subfield><subfield code="9">978-0-470-84981-1</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(OCoLC)74967088</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)BVBBV022205242</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-604</subfield><subfield code="b">ger</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1="0" ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="049" ind1=" " ind2=" "><subfield code="a">DE-20</subfield><subfield code="a">DE-19</subfield></datafield><datafield tag="050" ind1=" " ind2="0"><subfield code="a">R853.C55</subfield></datafield><datafield tag="082" ind1="0" ind2=" "><subfield code="a">610.724</subfield><subfield code="2">22</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">XF 3500</subfield><subfield code="0">(DE-625)152766:</subfield><subfield code="2">rvk</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Molenberghs, Geert</subfield><subfield code="e">Verfasser</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Missing data in clinical studies</subfield><subfield code="c">Geert Molenberghs ; Michael G. Kenward</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="a">Chichester [u.a.]</subfield><subfield code="b">Wiley</subfield><subfield code="c">2007</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">XX, 504 S.</subfield><subfield code="b">Ill., graph. Darst.</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="490" ind1="0" ind2=" "><subfield code="a">Statistics in practice</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">Missing Data in Clinical Studies provides a comprehensive account of the problems arising when data from clinical and related studies are incomplete, and presents the reader with approaches to effectively address them. The text provides a critique of conventional and simple methods before moving on to discuss more advanced approaches. The authors focus on practical and modeling concepts, providing an extensive set of case studies to illustrate the problems described. Provides a practical guide to the analysis of clinical trials and related studies with missing data. Examines the problems caused by missing data, enabling a complete understanding of how to overcome them. Presents conventional, simple methods to tackle these problems, before addressing more advanced approaches, including sensitivity analysis, and the MAR missingness mechanism. Illustrated throughout with real-life case studies and worked examples from clinical trials. Details the use and implementation of the necessary statistical software, primarily SAS. Missing Data in Clinical Studies has been developed through a series of courses and lectures. Its practical approach will appeal to applied statisticians and biomedical researchers, in particular those in the biopharmaceutical industry, medical and public health organisations. Graduate students of biostatistics will also find much of benefit.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Médecine - Recherche - Méthodes statistiques</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Observations manquantes (Statistique)</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Études cliniques - Méthodologie</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Clinical Trials as Topic</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Clinical trials</subfield><subfield code="x">Statistical methods</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Data Interpretation, Statistical</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Missing observations (Statistics)</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Research Design</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Statistics as Topic</subfield><subfield code="x">methods</subfield></datafield><datafield tag="650" ind1="0" ind2="7"><subfield code="a">Klinisches Experiment</subfield><subfield code="0">(DE-588)4164223-5</subfield><subfield code="2">gnd</subfield><subfield code="9">rswk-swf</subfield></datafield><datafield tag="655" ind1=" " ind2="7"><subfield code="0">(DE-588)4056995-0</subfield><subfield code="a">Statistik</subfield><subfield code="2">gnd-content</subfield></datafield><datafield tag="689" ind1="0" ind2="0"><subfield code="a">Klinisches Experiment</subfield><subfield code="0">(DE-588)4164223-5</subfield><subfield code="D">s</subfield></datafield><datafield tag="689" ind1="0" ind2=" "><subfield code="5">DE-604</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Kenward, Michael G.</subfield><subfield code="d">1956-</subfield><subfield code="e">Verfasser</subfield><subfield code="0">(DE-588)1030402035</subfield><subfield code="4">aut</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="m">HBZ Datenaustausch</subfield><subfield code="q">application/pdf</subfield><subfield code="u">http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=015416640&sequence=000002&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA</subfield><subfield code="3">Inhaltsverzeichnis</subfield></datafield><datafield tag="999" ind1=" " ind2=" "><subfield code="a">oai:aleph.bib-bvb.de:BVB01-015416640</subfield></datafield></record></collection> |
| genre | (DE-588)4056995-0 Statistik gnd-content |
| genre_facet | Statistik |
| id | DE-604.BV022205242 |
| illustrated | Illustrated |
| index_date | 2024-07-02T16:25:12Z |
| indexdate | 2024-07-09T20:52:20Z |
| institution | BVB |
| isbn | 0470849819 9780470849811 |
| language | English |
| oai_aleph_id | oai:aleph.bib-bvb.de:BVB01-015416640 |
| oclc_num | 74967088 |
| open_access_boolean | |
| owner | DE-20 DE-19 DE-BY-UBM |
| owner_facet | DE-20 DE-19 DE-BY-UBM |
| physical | XX, 504 S. Ill., graph. Darst. |
| publishDate | 2007 |
| publishDateSearch | 2007 |
| publishDateSort | 2007 |
| publisher | Wiley |
| record_format | marc |
| series2 | Statistics in practice |
| spelling | Molenberghs, Geert Verfasser aut Missing data in clinical studies Geert Molenberghs ; Michael G. Kenward Chichester [u.a.] Wiley 2007 XX, 504 S. Ill., graph. Darst. txt rdacontent n rdamedia nc rdacarrier Statistics in practice Missing Data in Clinical Studies provides a comprehensive account of the problems arising when data from clinical and related studies are incomplete, and presents the reader with approaches to effectively address them. The text provides a critique of conventional and simple methods before moving on to discuss more advanced approaches. The authors focus on practical and modeling concepts, providing an extensive set of case studies to illustrate the problems described. Provides a practical guide to the analysis of clinical trials and related studies with missing data. Examines the problems caused by missing data, enabling a complete understanding of how to overcome them. Presents conventional, simple methods to tackle these problems, before addressing more advanced approaches, including sensitivity analysis, and the MAR missingness mechanism. Illustrated throughout with real-life case studies and worked examples from clinical trials. Details the use and implementation of the necessary statistical software, primarily SAS. Missing Data in Clinical Studies has been developed through a series of courses and lectures. Its practical approach will appeal to applied statisticians and biomedical researchers, in particular those in the biopharmaceutical industry, medical and public health organisations. Graduate students of biostatistics will also find much of benefit. Médecine - Recherche - Méthodes statistiques Observations manquantes (Statistique) Études cliniques - Méthodologie Clinical Trials as Topic Clinical trials Statistical methods Data Interpretation, Statistical Missing observations (Statistics) Research Design Statistics as Topic methods Klinisches Experiment (DE-588)4164223-5 gnd rswk-swf (DE-588)4056995-0 Statistik gnd-content Klinisches Experiment (DE-588)4164223-5 s DE-604 Kenward, Michael G. 1956- Verfasser (DE-588)1030402035 aut HBZ Datenaustausch application/pdf http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=015416640&sequence=000002&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA Inhaltsverzeichnis |
| spellingShingle | Molenberghs, Geert Kenward, Michael G. 1956- Missing data in clinical studies Médecine - Recherche - Méthodes statistiques Observations manquantes (Statistique) Études cliniques - Méthodologie Clinical Trials as Topic Clinical trials Statistical methods Data Interpretation, Statistical Missing observations (Statistics) Research Design Statistics as Topic methods Klinisches Experiment (DE-588)4164223-5 gnd |
| subject_GND | (DE-588)4164223-5 (DE-588)4056995-0 |
| title | Missing data in clinical studies |
| title_auth | Missing data in clinical studies |
| title_exact_search | Missing data in clinical studies |
| title_exact_search_txtP | Missing data in clinical studies |
| title_full | Missing data in clinical studies Geert Molenberghs ; Michael G. Kenward |
| title_fullStr | Missing data in clinical studies Geert Molenberghs ; Michael G. Kenward |
| title_full_unstemmed | Missing data in clinical studies Geert Molenberghs ; Michael G. Kenward |
| title_short | Missing data in clinical studies |
| title_sort | missing data in clinical studies |
| topic | Médecine - Recherche - Méthodes statistiques Observations manquantes (Statistique) Études cliniques - Méthodologie Clinical Trials as Topic Clinical trials Statistical methods Data Interpretation, Statistical Missing observations (Statistics) Research Design Statistics as Topic methods Klinisches Experiment (DE-588)4164223-5 gnd |
| topic_facet | Médecine - Recherche - Méthodes statistiques Observations manquantes (Statistique) Études cliniques - Méthodologie Clinical Trials as Topic Clinical trials Statistical methods Data Interpretation, Statistical Missing observations (Statistics) Research Design Statistics as Topic methods Klinisches Experiment Statistik |
| url | http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&local_base=BVB01&doc_number=015416640&sequence=000002&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA |
| work_keys_str_mv | AT molenberghsgeert missingdatainclinicalstudies AT kenwardmichaelg missingdatainclinicalstudies |