Muutke küpsiste eelistusi

Dynamic Treatment Regimes: Statistical Methods for Precision Medicine [Kõva köide]

, , (North Carolina State University, Raleigh, USA), (North Carolina State University, Raleigh, USA)
  • Formaat: Hardback, 618 pages, kõrgus x laius: 234x156 mm, kaal: 920 g
  • Ilmumisaeg: 10-Dec-2019
  • Kirjastus: Chapman & Hall/CRC
  • ISBN-10: 1498769772
  • ISBN-13: 9781498769778
Teised raamatud teemal:
Dynamic Treatment Regimes: Statistical Methods for Precision MedicineSuurem pilt
  • Formaat: Hardback, 618 pages, kõrgus x laius: 234x156 mm, kaal: 920 g
  • Ilmumisaeg: 10-Dec-2019
  • Kirjastus: Chapman & Hall/CRC
  • ISBN-10: 1498769772
  • ISBN-13: 9781498769778
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781498769778.html
Märksõnad:
Dynamic Treatment Regimes: Statistical Methods for Precision Medicine provides a comprehensive introduction to statistical methodology for the evaluation and discovery of dynamic treatment regimes from data. Researchers and graduate students in statistics, data science, and related quantitative disciplines with a background in probability and statistical inference and popular statistical modeling techniques will be prepared for further study of this rapidly evolving field.

A dynamic treatment regime is a set of sequential decision rules, each corresponding to a key decision point in a disease or disorder process, where each rule takes as input patient information and returns the treatment option he or she should receive. Thus, a treatment regime formalizes how a clinician synthesizes patient information and selects treatments in practice. Treatment regimes are of obvious relevance to precision medicine, which involves tailoring treatment selection to patient characteristics in an evidence-based way. Of critical importance to precision medicine is estimation of an optimal treatment regime, one that, if used to select treatments for the patient population, would lead to the most beneficial outcome on average. Key methods for estimation of an optimal treatment regime from data are motivated and described in detail. A dedicated companion website presents full accounts of application of the methods using a comprehensive R package developed by the authors.

The authors website www.dtr-book.com includes updates, corrections, new papers, and links to useful websites.

Arvustused

"Biostatisticians, those that are professional as well as masters level and PhD level, will find this book useful. It is written by well-known experts who have incredible track records in this field, both methodologically and in designing and implementing/analyzing SMARTs and observational studies to uncover optimal dynamic treatment regimes. The text is rigorous in its statistical definitions and theorems. It is a comprehensive text on the area of dynamic treatment regimes and SMART design. Both those familiar with this area and those new to the area will learn something. They offer some interesting uses of the SMART design (e.g., dose finding and extending beyond 2 stages), that you cannot find in current manuscripts." ~Kelley Kidwell, University of Michigan

"The book will serve as an excellent reference and textbook. I expect I will use the book in my own class, once it is available. Besides being a comprehensive treatment of dynamic treatment regimes, the revision/re-introduction to causal inference, potential outcomes, M-estimators, propensity scores, and related issues is extremely useful." ~Daniel Lizotte, The University of Western Ontario

"Statisticians/biostatisticians directly involved in planning SMARTs would likely find this material useful, as they would have to adapt or extend these methods to particular trials being planned. Also, academic statisticians aiming to get into this field of methodological research would likely find the material as a useful summary of the already extensive literature; however, as the field is fast-moving, this material only serves as a starting point. The authors are not providing a cookbook-style guide to planning a variety of different kind of SMARTs, they provide examples, and enough theoretical background rigorously presented to get started in the area." ~Olli Saarela, University of Toronto

"(Chapters 2-4) are very nice indeed: well written, well structured, informative and interesting. Congratulations to the authors I went to the website, which is beautiful. The authors have put lots of effort into this. ~Robin Henderson, Newcastle University

Preface xiii
1 Introduction
1(16)
1.1 What Is a Dynamic Treatment Regime?
1(1)
1.2 Motivating Examples
2(5)
1.2.1 Treatment of Acute Leukemias
2(2)
1.2.2 Interventions for Children with ADHD
4(2)
1.2.3 Treatment of HIV Infection
6(1)
1.3 The Meaning of "Dynamic"
7(1)
1.4 Basic Framework
8(4)
1.4.1 Definition of a Dynamic Treatment Regime
8(3)
1.4.2 Data for Dynamic Treatment Regimes
11(1)
1.5 Outline of this Book
12(5)
2 Preliminaries
17(34)
2.1 Introduction
17(2)
2.2 Point Exposure Studies
19(2)
2.3 Potential Outcomes and Causal Inference
21(7)
2.3.1 Potential Outcomes
21(3)
2.3.2 Randomized Studies
24(2)
2.3.3 Observational Studies
26(2)
2.4 Estimation of Causal Effects via Outcome Regression
28(4)
2.5 Review of M-estimation
32(7)
2.6 Estimation of Causal Effects via the Propensity Score
39(7)
2.6.1 The Propensity Score
39(2)
2.6.2 Propensity Score Stratification
41(1)
2.6.3 Inverse Probability Weighting
41(5)
2.7 Doubly Robust Estimation of Causal Effects
46(4)
2.8 Application
50(1)
3 Single Decision Treatment Regimes: Fundamentals
51(48)
3.1 Introduction
51(1)
3.2 Treatment Regimes for a Single Decision Point
52(3)
3.2.1 Class of All Possible Treatment Regimes
52(1)
3.2.2 Potential Outcomes Framework
53(1)
3.2.3 Value of a Treatment Regime
54(1)
3.3 Estimation of the Value of a Fixed Regime
55(8)
3.3.1 Outcome Regression Estimator
56(2)
3.3.2 Inverse Probability Weighted Estimator
58(3)
3.3.3 Augmented Inverse Probability Weighted Estimator
61(2)
3.4 Characterization of an Optimal Regime
63(5)
3.5 Estimation of an Optimal Regime
68(29)
3.5.1 Regression-based Estimation
68(4)
3.5.2 Estimation via A-learning
72(7)
3.5.3 Value Search Estimation
79(8)
3.5.4 Implementation and Practical Performance
87(4)
3.5.5 More than Two Treatment Options
91(6)
3.6 Application
97(2)
4 Single Decision Treatment Regimes: Additional Methods
99(26)
4.1 Introduction
99(1)
4.2 Optimal Regimes from a Classification Perspective
100(7)
4.2.1 Generic Classification Problem
100(1)
4.2.2 Classification Analogy
101(6)
4.3 Outcome Weighted Learning
107(5)
4.4 Interpretable Treatment Regimes via Decision Lists
112(10)
4.5 Additional Approaches
122(2)
4.6 Application
124(1)
5 Multiple Decision Treatment Regimes: Overview
125(60)
5.1 Introduction
125(1)
5.2 Multiple Decision Treatment Regimes
126(6)
5.3 Statistical Framework
132(12)
5.3.1 Potential Outcomes for K Decisions
132(4)
5.3.2 Data
136(4)
5.3.3 Identifiability Assumptions
140(4)
5.4 The g-Computation Algorithm
144(6)
5.5 Estimation of the Value of a Fixed Regime
150(10)
5.5.1 Estimation via g-Computation
150(3)
5.5.2 Inverse Probability Weighted Estimator
153(7)
5.6 Characterization of an Optimal Regime
160(6)
5.7 Estimation of an Optimal Regime
166(17)
5.7.1 Q-learning
166(7)
5.7.2 Value Search Estimation
173(3)
5.7.3 Backward Iterative Implementation of Value Search Estimation
176(5)
5.7.4 Implementation and Practical Performance
181(2)
5.8 Application
183(2)
6 Multiple Decision Treatment Regimes: Formal Framework
185(60)
6.1 Introduction
185(1)
6.2 Statistical Framework
186(19)
6.2.1 Potential Outcomes for K Decisions
186(3)
6.2.2 Feasible Sets and Classes of Treatment Regimes
189(5)
6.2.3 Potential Outcomes for a Fixed k-Decision Regime
194(2)
6.2.4 Identifiability Assumptions
196(9)
6.3 The g-Computation Algorithm
205(4)
6.4 Estimation of the Value a Fixed Regime
209(34)
6.4.1 Estimation via g-Computation
209(1)
6.4.2 Regression-based Estimation
209(14)
6.4.3 Inverse Probability Weighted Estimator
223(9)
6.4.4 Augmented Inverse Probability Weighted Estimator
232(7)
6.4.5 Estimation via Marginal Structural Models
239(4)
6.5 Application
243(2)
7 Optimal Multiple Decision Treatment Regimes
245(80)
7.1 Introduction
245(1)
7.2 Characterization of an Optimal Regime
246(12)
7.2.1 ψ-Specific Regimes
246(2)
7.2.2 Characterization in Terms of Potential Outcomes
248(5)
7.2.3 Justification
253(3)
7.2.4 Characterization in Terms of Observed Data
256(2)
7.3 Optimal "Midstream" Regimes
258(4)
7.4 Estimation of an Optimal Regime
262(61)
7.4.1 Q-learning
262(5)
7.4.2 A-learning
267(10)
7.4.3 Value Search Estimation
277(10)
7.4.4 Backward Iterative Estimation
287(15)
7.4.5 Classification Perspective
302(8)
7.4.6 Interpretable Regimes via Decision Lists
310(6)
7.4.7 Estimation via Marginal Structural Models
316(2)
7.4.8 Additional Approaches
318(2)
7.4.9 Implementation and Practical Performance
320(3)
7.5 Application
323(2)
8 Regimes Based on Time-to-Event Outcomes
325(122)
8.1 Introduction
325(1)
8.2 Single Decision Treatment Regimes
326(22)
8.2.1 Statistical Framework
326(4)
8.2.2 Outcome Regression Estimators
330(3)
8.2.3 Inverse Probability of Censoring Regression Estimators
333(3)
8.2.4 Inverse Probability Weighted and Value Search Estimators
336(8)
8.2.5 Discussion
344(4)
8.3 Multiple Decision Treatment Regimes
348(88)
8.3.1 Multiple Decision Regimes
348(2)
8.3.2 Statistical Framework
350(13)
8.3.3 Estimation of the Value of a Fixed Regime
363(46)
8.3.4 Characterization of an Optimal Regime
409(3)
8.3.5 Estimation of an Optimal Regime
412(23)
8.3.6 Discussion
435(1)
8.4 Application
436(1)
8.5 Technical Details
436(11)
9 Sequential Multiple Assignment Randomized Trials
447(68)
9.1 Introduction
447(3)
9.2 Design Considerations
450(25)
9.2.1 Basic SMART Framework, K = 2
450(2)
9.2.2 Critical Decision Points
452(9)
9.2.3 Feasible Treatment Options
461(9)
9.2.4 Interim Outcomes, Randomization, and Stratification
470(5)
9.2.5 Other Candidate Designs
475(1)
9.3 Power and Sample Size for Simple Comparisons
475(13)
9.3.1 Comparing Response Rates
476(5)
9.3.2 Comparing Fixed Regimes
481(7)
9.4 Power and Sample Size for More Complex Comparisons
488(9)
9.4.1 Marginalizing versus Maximizing
488(3)
9.4.2 Marginalizing over the Second Stage
491(2)
9.4.3 Marginalizing with Respect to Standard of Care
493(1)
9.4.4 Maximizing over the Second Stage
494(3)
9.5 Power and Sample Size for Optimal Treatment Regimes
497(15)
9.5.1 Normality-based Sample Size Procedure
499(6)
9.5.2 Projection-based Sample Size Procedure
505(7)
9.6 Extensions and Further Reading
512(3)
10 Statistical Inference
515(56)
10.1 Introduction
515(2)
10.2 Nonsmoothness and Statistical Inference
517(8)
10.3 Inference for Single Decision Regimes
525(28)
10.3.1 Inference on Model Parameters
526(9)
10.3.2 Inference on the Value
535(18)
10.4 Inference for Multiple Decision Regimes
553(16)
10.4.1 Q-learning
554(9)
10.4.2 Value Search Estimation with Convex Surrogates
563(2)
10.4.3 g-Computation
565(4)
10.5 Discussion
569(2)
11 Additional Topics
571(6)
Bibliography 577(18)
Index 595
Anastasios Tsiatis is Gertrude M. Cox Distinguished Professor Emeritus, Marie Davidian is J. Stuart Hunter Distinguished Professor, Shannon Holloway is Senior Research Scholar, and Eric Laber is Goodnight Distinguished Professor, all in the Department of Statistics at North Carolina State University. They have published extensively and are internationally-recognized authorities on methodology for dynamic treatment regimes.