Atjaunināt sīkdatņu piekrišanu

E-grāmata: Estimands, Estimators and Sensitivity Analysis in Clinical Trials

(Eli Lilly Cananda, Ontario, Canada), , (Quintiles, Durham, North Carolina, USA),
Citas grāmatas par šo tēmu:
  • Formāts - EPUB+DRM
  • Cena: 52,59 €*
  • * ši ir gala cena, t.i., netiek piemērotas nekādas papildus atlaides
  • Ielikt grozā
  • Pievienot vēlmju sarakstam
  • Šī e-grāmata paredzēta tikai personīgai lietošanai. E-grāmatas nav iespējams atgriezt un nauda par iegādātajām e-grāmatām netiek atmaksāta.
Estimands, Estimators and Sensitivity Analysis in Clinical TrialsPalielināt
Citas grāmatas par šo tēmu:

DRM restrictions

  • Kopēšana (kopēt/ievietot):

    nav atļauts

  • Drukāšana:

    nav atļauts

  • Lietošana:

    Digitālo tiesību pārvaldība (Digital Rights Management (DRM))
    Izdevējs ir piegādājis šo grāmatu šifrētā veidā, kas nozīmē, ka jums ir jāinstalē bezmaksas programmatūra, lai to atbloķētu un lasītu. Lai lasītu šo e-grāmatu, jums ir jāizveido Adobe ID. Vairāk informācijas šeit. E-grāmatu var lasīt un lejupielādēt līdz 6 ierīcēm (vienam lietotājam ar vienu un to pašu Adobe ID).

    Nepieciešamā programmatūra
    Lai lasītu šo e-grāmatu mobilajā ierīcē (tālrunī vai planšetdatorā), jums būs jāinstalē šī bezmaksas lietotne: PocketBook Reader (iOS / Android)

    Lai lejupielādētu un lasītu šo e-grāmatu datorā vai Mac datorā, jums ir nepieciešamid Adobe Digital Editions (šī ir bezmaksas lietotne, kas īpaši izstrādāta e-grāmatām. Tā nav tas pats, kas Adobe Reader, kas, iespējams, jau ir jūsu datorā.)

    Jūs nevarat lasīt šo e-grāmatu, izmantojot Amazon Kindle.

The concepts of estimands, analyses (estimators), and sensitivity are interrelated. Therefore, great need exists for an integrated approach to these topics. This book acts as a practical guide to developing and implementing statistical analysis plans by explaining fundamental concepts using accessible language, providing technical details, real-world examples, and SAS and R code to implement analyses. The updated ICH guideline raises new analytic and cross-functional challenges for statisticians. Gaps between different communities have come to surface, such as between causal inference and clinical trialists, as well as among clinicians, statisticians, and regulators when it comes to communicating decision-making objectives, assumptions, and interpretations of evidence.

This book lays out a path toward bridging some of these gaps. It offers

A common language and unifying framework along with the technical details and practical guidance to help statisticians meet the challenges

A thorough treatment of intercurrent events (ICEs), i.e., postrandomization events that confound interpretation of outcomes and five strategies for ICEs in ICH E9 (R1)

Details on how estimands, integrated into a principled study development process, lay a foundation for coherent specification of trial design, conduct, and analysis needed to overcome the issues caused by ICEs:

A perspective on the role of the intention-to-treat principle

Examples and case studies from various areas

Example code in SAS and R

A connection with causal inference

Implications and methods for analysis of longitudinal trials with missing data

Together, the authors have offered the readers their ample expertise in clinical trial design and analysis, from an industrial and academic perspective.

Recenzijas

"The purpose of this book, which is to promote an integrated understanding of key concepts throughout the drug development process through an example-based approach, is certainly achieved. It is the holistic approach to planning the analysis and the focus on practical implementation that distinguishes this text from others... Overall, I enjoyed reading this book, which is a holistic and complete work and will be useful for researchers in the medical statistics area." - Taras Lukashiv, ISCB News, July 2020

List of Figures
xiii
List of Tables
xv
List of Code Fragments
xix
Preface xxi
Acknowledgments xxiii
Authors xxv
Section I Setting the Stage
1 Introduction
3(6)
1.1 Understanding the Problem
3(2)
1.2 History
5(3)
1.3 Summary
8(1)
2 Why Are Estimands Important?
9(8)
2.1 Motivating Example
9(1)
2.2 How Estimands Help
10(2)
2.3 Considerations for Differing Stakeholders
12(1)
2.4 Considerations for Differing Clinical Situations
13(1)
2.5 Summary
14(3)
Section II Estimands
3 Estimands and How to Define Them
17(18)
3.1 Introduction
17(1)
3.2 Study Development Process Chart
17(2)
3.3 Process for Defining Estimands
19(6)
3.3.1 Introduction
19(1)
3.3.2 Identifying Intercurrent Events
19(3)
3.3.3 Defining the Treatment Regimen of Interest
22(1)
3.3.4 Overview of Strategies for Handling Intercurrent Events
23(2)
3.4 Defining the Estimand
25(1)
3.5 Special Considerations in Defining Estimands
26(3)
3.5.1 Estimands for Safety Outcomes
26(1)
3.5.2 Estimands for Early-Phase Trials
26(1)
3.5.3 Estimands for Scenarios When Treatment and Outcomes Do Not Occur Concurrently
27(1)
3.5.4 Estimands for Quality-of-Life Evaluation in Trials with Many Deaths
27(1)
3.5.5 Estimands for Assessing Non-inferiority
28(1)
3.6 Trial Design and Conduct Considerations
29(3)
3.6.1 Introduction
29(1)
3.6.2 Data Collection and Trial Conduct Considerations
29(2)
3.6.3 Trial Design Considerations
31(1)
3.7 A Note on Missing Data
32(1)
3.8 A Note on the Intention to Treat Principle
32(2)
3.9 Summary
34(1)
4 Strategies for Dealing with Intercurrent Events
35(14)
4.1 Introduction
35(1)
4.2 Treatment Policy Strategy
35(2)
4.3 Composite Strategy
37(2)
4.4 Hypothetical Strategy
39(2)
4.5 Principal Stratification Strategy
41(2)
4.6 While-on-Treatment Strategy
43(1)
4.7 Assumptions Behind the Strategies for Dealing with Intercurrent Events
44(2)
4.7.1 General Assumptions
44(1)
4.7.2 Treatment Policy Strategy Assumptions
44(1)
4.7.3 Composite Strategy Assumptions
45(1)
4.7.4 Hypothetical Strategy Assumptions
45(1)
4.7.5 Principal Stratification Assumptions
45(1)
4.7.6 While-on-Treatment Strategy Assumptions
46(1)
4.8 Risk-Benefit Implications
46(1)
4.9 Summary
46(3)
5 Examples from Actual Clinical Trials in Choosing and Specifying Estimands
49(24)
5.1 Introduction
49(2)
5.2 Example 1: A Proof of Concept Trial in Major Depressive Disorder
51(4)
5.2.1 Background
51(1)
5.2.2 Trial Description
52(1)
5.2.3 Primary Estimand
52(3)
5.3 Example 2: A Confirmatory Trial in Asthma
55(8)
5.3.1 Background
55(1)
5.3.2 Trial Description
56(1)
5.3.3 Primary Estimand
57(4)
5.3.4 Supportive Estimand
61(2)
5.4 Example 3: A Confirmatory Trial in Rheumatoid Arthritis
63(8)
5.4.1 Background
63(1)
5.4.2 Trial Description
64(1)
5.4.3 Estimand for RA Study Design 1
65(4)
5.4.4 Estimand 2 for RA Study Design 2
69(2)
5.5 Summary
71(2)
6 Causal Inference and Estimands
73(16)
6.1 Introduction
73(1)
6.2 Causal Framework for Estimands in Clinical Trials
74(2)
6.2.1 Defining Potential Outcomes
74(1)
6.2.2 Counterfactual Outcomes and Potential Outcomes
75(1)
6.3 Using Potential Outcomes to Define Estimands
76(7)
6.3.1 Defining Estimands
76(3)
6.3.2 Specifying Treatment Changes
79(4)
6.4 Examples of Defining Estimands in the Potential Outcome Framework
83(5)
6.4.1 Introduction
83(1)
6.4.2 Treatment Policy Strategy
83(1)
6.4.3 Composite Strategy
84(1)
6.4.4 Hypothetical Strategy
84(1)
6.4.5 Principal Stratification Strategy
85(1)
6.4.6 While-on-Treatment Strategy
85(1)
6.4.7 Scenarios with Dynamic Treatment Regimens
86(1)
6.4.8 Treatment of Missing Data
87(1)
6.5 Summary
88(1)
7 Putting the Principles into Practice
89(8)
7.1 Introduction
89(1)
7.2 Overview of Process
90(7)
Section III Estimators and Sensitivity
8 Overview of Estimators
97(2)
9 Modeling Considerations
99(12)
9.1 Introduction
99(1)
9.2 Longitudinal Analyses
99(9)
9.2.1 Choice of Dependent Variable and Statistical Test
99(2)
9.2.2 Modeling Covariance (Correlation)
101(1)
9.2.3 Modeling Means Over Time
102(2)
9.2.4 Accounting for Covariates
104(1)
9.2.5 Categorical Data
105(2)
9.2.6 Model Checking and Verification
107(1)
9.3 Summary
108(3)
10 Overview of Analyses for Composite Intercurrent Event Strategies
111(6)
10.1 Introduction
111(1)
10.2 Ad Hoc Approaches
111(2)
10.3 Rank-Based and Related Methods
113(2)
10.4 Summary
115(2)
11 Overview of Analyses for Hypothetical Intercurrent Event Strategies
117(10)
11.1 Introduction
117(1)
11.2 Estimators for What Would Have Happened in the Absence of Relevant Intercurrent Events
118(4)
11.2.1 Introduction
118(1)
11.2.2 Likelihood-Based Analyses
119(1)
11.2.3 Multiple Imputation
120(1)
11.2.4 Inverse Probability Weighting
121(1)
11.2.5 Considerations for Categorical and Time-to-Event Data
122(1)
11.3 Estimators for Treatment Policies That Were Not Included in the Trial or Not Followed
122(3)
11.3.1 Introduction
122(1)
11.3.2 Reference-Based Approaches Using Multiple Imputation
123(1)
11.3.3 Considerations for Categorical and Time-to-Event Data
124(1)
11.3.4 Likelihood and Bayesian Approaches to Reference-Based Imputation
125(1)
11.4 Summary
125(2)
12 Overview of Analyses for Principal Stratification Intercurrent Event Strategies
127(4)
12.1 Introduction
127(1)
12.2 Applications
128(2)
12.3 Summary
130(1)
13 Overview of Analyses for While-on-Treatment Intercurrent Event Strategies
131(2)
14 Overview of Analyses for Treatment Policy Intercurrent Event Strategies
133(2)
15 Missing Data
135(8)
15.1 Introduction
135(1)
15.2 Basic Principles
135(2)
15.3 Missing Data Mechanisms
137(2)
15.3.1 General Considerations
137(1)
15.3.2 Considerations for Time-to-Event Analyses
138(1)
15.4 Analytic Considerations
139(2)
15.4.1 General Considerations
139(1)
15.4.2 Considerations When Changing Treatment Is Possible
140(1)
15.5 Inclusive and Restrictive Modeling Approaches
141(1)
15.6 Summary
142(1)
16 Sensitivity Analyses
143(10)
16.1 General Considerations
143(1)
16.2 Supplementary Analyses
144(1)
16.3 Assessing Sensitivity to Missing Data Assumptions
145(2)
16.3.1 Introduction
145(1)
16.3.2 Assessing Sensitivity to Departures from MAR
146(1)
16.4 Sensitivity to Methods of Accounting for Intercurrent Events
147(3)
16.4.1 Introduction
147(1)
16.4.2 Sensitivity Analyses for Hypothetical Strategies
148(1)
16.4.3 Sensitivity Analyses for Composite Strategies
148(1)
16.4.4 Sensitivity Analyses for Principal Stratification Strategies
149(1)
16.4.5 Sensitivity Analyses for While-on-Treatment Strategies
149(1)
16.4.6 Sensitivity Analyses for Treatment Policy Strategies
150(1)
16.5 Summary
150(3)
Section IV Technical Details on Selected Analyses
17 Example Data
153(8)
17.1 Introduction
153(1)
17.2 Details of Example Data Set
153(8)
17.2.1 Complete Data Set
153(1)
17.2.2 Data Set with Dropout
154(7)
18 Direct Maximum Likelihood
161(10)
18.1 Introduction
161(1)
18.2 Technical Details of Likelihood Estimation for Repeated Measures
161(2)
18.3 Factoring the Likelihood Function for Ignorability
163(3)
18.4 Example
166(3)
18.5 Code Fragments
169(1)
18.6 Summary
170(1)
19 Multiple Imputation
171(30)
19.1 Introduction
171(1)
19.2 Technical Details
172(5)
19.3 Example -- Implementing MI
177(8)
19.3.1 Introduction
177(2)
19.3.2 Imputation
179(3)
19.3.3 Analysis
182(1)
19.3.4 Inference
183(1)
19.3.5 Accounting for Non-monotone Missingness
184(1)
19.4 Situations Where MI Is Particularly Useful
185(3)
19.4.1 Introduction
185(1)
19.4.2 Scenarios Where Direct Likelihood Methods Are Difficult to Implement or Not Available
185(1)
19.4.3 Exploiting Separate Steps for Imputation and Analysis
186(1)
19.4.4 Sensitivity Analysis
187(1)
19.5 Example -- Using MI to Impute Covariates
188(2)
19.5.1 Introduction
188(1)
19.5.2 Implementation
188(2)
19.6 Examples -- Using Inclusive Models in MI
190(4)
19.6.1 Introduction
190(1)
19.6.2 Implementation
191(3)
19.7 MI for Categorical Outcomes
194(1)
19.8 Code Fragments
194(5)
19.9 Summary
199(2)
20 Inverse Probability Weighted Generalized Estimated Equations
201(14)
20.1 Introduction
201(1)
20.2 Technical Details -- Generalized Estimating Equations
201(2)
20.3 Technical Details -- Inverse Probability Weighting
203(6)
20.3.1 General Considerations
203(4)
20.3.2 Specific Implementations
207(2)
20.4 Example
209(2)
20.5 Code Fragments
211(2)
20.6 Summary
213(2)
21 Doubly Robust Methods
215(14)
21.1 Introduction
215(1)
21.2 Technical Details
216(3)
21.3 Specific Implementations
219(3)
21.4 Example
222(2)
21.5 Code Fragments
224(2)
21.6 Summary
226(3)
22 Reference-Based Imputation
229(20)
22.1 Introduction
229(1)
22.2 Multiple Imputation-Based Approach
229(6)
22.2.1 Missing at Random
231(1)
22.2.2 Copy Increment from Reference
232(1)
22.2.3 Jump to Reference
233(1)
22.2.4 Copy Reference
234(1)
22.2.5 Example
235(1)
22.3 Group Mean Imputation Using a Likelihood-Based Approach
235(4)
22.4 Bayesian-Based Approach
239(3)
22.5 Considerations for the Variance of Reference-Based Estimators
242(4)
22.6 Code Fragments
246(1)
22.7 Summary
247(2)
Acknowledgment
248(1)
23 Delta Adjustment
249(6)
23.1 Introduction
249(1)
23.2 Technical Details
249(2)
23.3 Example
251(2)
23.4 Code Fragments
253(1)
23.5 Summary
254(1)
24 Overview of Principal Stratification Methods
255(18)
24.1 Introduction
255(1)
24.2 Principal Stratification Based on Intercurrent Events
255(3)
24.3 Principal Stratification Based on Post-randomization Treatment
258(1)
24.4 Principal Stratification Based on the Post-randomization Outcomes
259(5)
24.4.1 Introduction
259(1)
24.4.2 The Case of a Binary Outcome
260(2)
24.4.3 The Case of a Continuous Outcome
262(2)
24.5 Utilizing Baseline Covariates
264(2)
24.5.1 Introduction
264(1)
24.5.2 Predicted Counterfactual Response
265(1)
24.5.3 Strata Propensity Weighted Estimator
266(1)
24.6 Implementing Principal Stratification Strategy via Imputation
266(2)
24.7 Approximation of Survivor Average Causal Effect
268(2)
24.8 Summary
270(3)
Section V Case Studies: Detailed Analytic Examples
25 Analytic Case Study of Depression Clinical Trials
273(1)
25.1 Introduction
273(2)
26 Analytic Case Study Based on the ACTG
275(6)
HIV Trial
281(1)
26.1 Introduction
281(1)
26.2 Study Details
282(2)
26.3 Estimands and Estimators for the Case Study
284(25)
26.3.1 Introduction
284(1)
26.3.2 Estimands for the Continuous Endpoint of Change from Baseline in CD4
284(8)
26.3.3 Estimands Based on the Time-to-Event Outcomes
292(7)
References
299(10)
Glossary 309(2)
Index 311
Geert Molenberghs is Professor of Biostatistics (Hasselt University, KULeuven. He works on surrogate endpoints, longitudinal and incomplete data, was Editor for Applied Statistics, Biometrics, Biostatistics, Wiley Probability & Statistics, and Wiley StatsRef and is Executive Editor of Biometrics. He was President of the International Biometric Society, is Fellow of the American Statistical Association, and received the Guy Medal in Bronze from the Royal Statistical Society. He has held visiting positions at the Harvard School of Public Health.

Ilya Lipkovich is a Sr. Research Advisor at Eli Lilly and Company. He is a Fellow of the American Statistical Association and published on subgroup identification in clinical data, analysis with missing data, and causal inference. He is a frequent presenter at conferences, a co-developer of subgroup identification methods, and a co-author of the book "Analyzing Longitudinal Clinical Trial Data. A Practical Guide."

Bohdana Ratitch is a Principal Research Scientist at Eli Lilly and Company. Bohdana has contributed to research and practical applications of methodologies for causal inference and missing data in clinical trials through active participation in a pharma industry working group, numerous publications, presentations, and co-authoring the book "Clinical Trials with Missing Data: A Guide for Practitioners".

Craig Mallinckrodt holds the rank of Distinguished Biostatistician at Biogen in Cambridge MA. He has extensive experience in all phases of clinical research. His methodology research focuses on longitudinal and incomplete data. He is Fellow of the American Statistical Association, has led several industry working groups on missing and longitudinal data, and received the Royal Statistical Societys award for outstanding contribution to the pharmaceutical industry.
Biežāk uzdotie jautājumi par e-grāmatām Biežāk uzdotie jautājumi par e-grāmatām
Permanent link: https://www.kriso.lv/db/97804299500566e.html
Keywords: