Muutke küpsiste eelistusi

Cluster Randomised Trials [Kõva köide]

3.67/5 (6 hinnangut Goodreads-ist)
(Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA), (London School of Hygiene & Tropical Medicine, UK)
  • Formaat: Hardback, 338 pages, kõrgus x laius: 234x156 mm, kaal: 658 g, 22 Illustrations, black and white
  • Sari: Chapman & Hall/CRC Biostatistics Series
  • Ilmumisaeg: 12-Jan-2009
  • Kirjastus: Chapman & Hall/CRC
  • ISBN-10: 1584888164
  • ISBN-13: 9781584888161
Teised raamatud teemal:
  • Kõva köide
  • Hind: 112,59 €*
  • * saadame teile pakkumise kasutatud raamatule, mille hind võib erineda kodulehel olevast hinnast
  • See raamat on trükist otsas, kuid me saadame teile pakkumise kasutatud raamatule.
  • Kogus:
  • Lisa ostukorvi
  • Tasuta tarne
  • Lisa soovinimekirja
Cluster Randomised TrialsSuurem pilt
  • Formaat: Hardback, 338 pages, kõrgus x laius: 234x156 mm, kaal: 658 g, 22 Illustrations, black and white
  • Sari: Chapman & Hall/CRC Biostatistics Series
  • Ilmumisaeg: 12-Jan-2009
  • Kirjastus: Chapman & Hall/CRC
  • ISBN-10: 1584888164
  • ISBN-13: 9781584888161
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781584888161.html
Märksõnad:
Cluster Randomised Trials discusses the design, conduct, and analysis of health trials that randomise groups of individuals to different treatments. It explores the advantages of cluster randomization, with special attention given to evaluating the effects of interventions against infectious diseases.













Avoiding unnecessary mathematical detail, the book covers basic concepts underlying the use of cluster randomisation, such as direct, indirect, and total effects. The authors also present an array of design issues in cluster randomised trials (CRTs), including strategies for minimizing contamination effects, the use of stratification and restricted randomisation to improve balance between treatment arms, special methods for sample size calculation, and alternatives to the simplest two-arm CRT. After covering analytical methods for CRTs, such as regression methods, the authors examine ethical issues, trial monitoring, interim analyses, reporting, and interpretation.











Although the book mainly focuses on medical and public health applications, it shows that the rigorous evidence of intervention effects provided by CRTs has the potential to inform public policy in a wide range of other areas. The book encourages readers to apply the methods to their own trials, reproduce the analyses presented, and explore alternative approaches.

Arvustused

an informative and readable text on the design, conduct, and analysis of cluster randomized trials (cRCTs). the book is rich with examples providing practical illustrations of concepts This book by Hayes and Moulton adds new insights into the field I would recommend this book highly This book would also provide an excellent resource for those who deliver courses on clinical trials and would be an excellent addition to the reference library of any trials unit. Marion K. Campbell, Clinical Trials, 2011



This comprehensive book covers all the main topics associated with design, conduct, and analysis of cluster randomised trials (CRTs). It follows a logical structure, which is easy to read and avoids excessive mathematical detail, making the book more accessible to wider readership. One of the great advantages of this useful handy book is that it does not just follow the theory; research-based examples aid understanding and bridge theory with practical applications. In summary, this book is an excellent introduction to the field of CRTs. It explains the methodology and conduct of CRTs in an accessible way. In addition, theory is complemented by practical examples. On this basis, I would highly recommend the book to all interested parties. If the aim of this book is to help readers to understand and apply a range of methods for the design and analysis of CRTs, then it is successful and accomplishes it well. Pharmaceutical Statistics, 2011



the authors have achieved their aim in providing a suitable introduction to even the more advanced issues for everyone involved in the planning, conduct and analysis of these studies. ISCB News #49, June 2010



This is a well-written book which I definitely recommend, especially as a textbook for graduate or postgraduate level studies, or as a reference book for researchers working on the design or analysis of CRTs. Journal of the Royal Statistical Society, Series A, Volume 173, Issue 1, January 2010









highly recommend this book for its unique and very important strengths. In particular, Hayes and Moulton should be congratulated for their focus on infectious diseases, a research topic which has received limited attention from statisticians interested in randomized trials. The broader discussion of direct and indirect effects of intervention is also very valuable. Neil Klar, Journal of Biopharmaceutical Statistics, 2010, Issue 1



The authors point out that the CRT is relatively new and that, although the topic is covered here pretty comprehensively, it is still an active research area. Its difficult to think of any important issue or aspect that is not discussed here, and at length and in depth. There is no heavy mathematics so the material is accessible to a wide range of readers. International Statistical Review (2009), 77, 2

Preface xvii
Authors xix
Glossary of Notation xxi
Part A: Basic Concepts
1 Introduction
3
1.1 Randomised Trials
3
1.1.1 Randomising Clusters
4
1.1.2 Some Case Studies
6
1.1.3 Overview of Book
8
2 Variability between Clusters
11
2.1 Introduction
11
2.2 The Implications of Between-cluster Variability: Some Examples
12
2.3 Measures of Between-cluster Variability
15
2.3.1 Introduction
15
2.3.1.1 Binary Outcomes and Proportions
15
2.3.1.2 Event Data and Person-years Rates
15
2.3.1.3 Quantitative Outcomes and Means
16
2.3.2 Coefficient of Variation, k
16
2.3.3 Intracluster Correlation Coefficient, ρ
17
2.3.3.1 Quantitative Outcomes
17
2.3.3.2 Binary Outcomes
18
2.3.3.3 Estimation of ρ
18
2.3.4 Relationship between k and ρ
18
2.4 The Design Effect
19
2.4.1 Binary Outcomes
19
2.4.2 Quantitative Outcomes
21
2.5 Sources of Within-cluster Correlation
22
2.5.1 Clustering of Population Characteristics
22
2.5.2 Variations in Response to Intervention
22
2.5.3 Correlation Due to Interaction between Individuals
23
3 Choosing Whether to Randomise by Cluster
25
3.1 Introduction
25
3.2 Rationale for Cluster Randomisation
25
3.2.1 Type of Intervention
25
3.2.2 Logistical Convenience and Acceptability
26
3.2.3 Contamination
27
3.3 Using Cluster Randomisation to Capture Indirect Effects of Intervention
28
3.3.1 Introduction
28
3.3.2 Effects of an Intervention on Infectiousness
29
3.3.3 Mass Effects of Intervention
31
3.3.4 Direct, Indirect, Total and Overall Effects
33
3.4 Disadvantages and Limitations of Cluster Randomisation
37
3.4.1 Efficiency
37
3.4.2 Selection Bias
37
3.4.3 Imbalances between Study Arms
39
3.4.4 Generalisability
40
Part B: Design Issues
4 Choice of Clusters
45
4.1 Introduction
45
4.2 Types of Cluster
45
4.2.1 Geographical Clusters
45
4.2.1.1 Communities
46
4.2.1.2 Administrative Units
47
4.2.1.3 Arbitrary Geographical Zones
48
4.2.2 Institutional Clusters
49
4.2.2.1 Schools
49
4.2.2.2 Health Units
50
4.2.2.3 Workplaces
50
4.2.3 Smaller Clusters
51
4.2.3.1 Households and Other Small Groups
52
4.2.3.2 Individuals as Clusters
52
4.3 Size of Clusters
53
4.3.1 Introduction
53
4.3.2 Statistical Considerations
53
4.3.3 Logistical Issues
54
4.3.4 Contamination
55
4.3.4.1 Contacts between Intervention and Control Clusters
55
4.3.4.2 Contacts between Intervention Clusters and the Wider Population
55
4.3.4.3 Contacts between Control Clusters and the Wider Population
56
4.3.4.4 Effects of Cluster Size on Contamination
56
4.3.5 Transmission Zones of Infectious Diseases
56
4.4 Strategies to Reduce Contamination
58
4.4.1 Separation of Clusters
58
4.4.2 Buffer Zones
60
4.4.3 The Fried Egg Design
62
4.5 Levels of Randomisation, Intervention, Data Collection and Inference
64
5 Matching and Stratification
65
5.1 Introduction
65
5.2 Rationale for Matching
65
5.2.1 Avoiding Imbalance between Treatment Arms
66
5.2.2 Improving Study Power and Precision
68
5.3 Disadvantages of Matching
70
5.3.1 Loss of Degrees of Freedom
70
5.3.2 Drop-out of Clusters
72
5.3.3 Limitations in Statistical Inference for Matched Trials
74
5.3.3.1 Adjustment for Covariates
74
5.3.3.2 Testing for Variation in Intervention Effect
74
5.3.3.3 Estimation of Intracluster Correlation Coefficient and Coefficient of Variation
75
5.4 Stratification as an Alternative to Matching
75
5.5 Choice of Matching Variables
77
5.5.1 Estimating the Matching Correlation
77
5.5.2 Matching on Baseline Values of Endpoint of Interest
78
5.5.3 Matching on Surrogate Variables
79
5.5.4 Matching on Multiple Variables
79
5.5.5 Matching on Location
80
5.6 Choosing Whether to Match or Stratify
81
5.6.1 Introduction
81
5.6.2 Trials with a Small Number of Clusters
81
5.6.3 Trials with a Larger Number of Clusters
83
6 Randomisation Procedures
85
6.1 Introduction
85
6.2 Restricted Randomisation
86
6.2.1 Basic Principles
86
6.2.2 Using Restricted Randomisation to Achieve Overall Balance
87
6.2.3 Balance Criteria
89
6.2.4 Validity of Restricted Randomisation
94
6.2.5 Restricted Randomisation with More than Two Treatment Arms
99
6.3 Some Practical Aspects of Randomisation
99
6.3.1 Concealment of Allocation
99
6.3.2 Public Randomisation
99
7 Sample Size
105
7.1 Introduction
105
7.2 Sample Size for Unmatched Trials
106
7.2.1 Event Rates
107
7.2.2 Proportions
109
7.2.3 Means
110
7.2.4 Variable Sample Size per Cluster
111
7.2.5 Sample Size Calculations Based on Intracluster Correlation Coefficient
111
7.3 Sample Size for Matched and Stratified Trials
113
7.3.1 Matched Trials
113
7.3.1.1 Event Rates
114
7.3.1.2 Proportions
114
7.3.1.3 Means
114
7.3.2 Stratified Trials
116
7.4 Estimating the Between-cluster Coefficient of Variation
117
7.4.1 Unmatched Trials
117
7.4.1.1 Event Rates
118
7.4.1.2 Proportions
119
7.4.1.3 Means
120
7.4.2 Matched and Stratified Trials
120
7.4.2.1 Event Rates
120
7.4.2.2 Proportions and Means
121
7.5 Choice of Sample Size in each Cluster
121
7.6 Further Issues in Sample Size Calculation
124
7.6.1 Trials with More than Two Treatment Arms
124
7.6.2 Trials with Treatment Arms of Unequal Size
124
7.6.3 Equivalence Trials
125
7.6.4 Power and Precision
126
7.6.5 Assumptions about Intervention Effects
127
8 Alternative Study Designs
129
8.1 Introduction
129
8.2 Design Choices for Treatment Arms
129
8.2.1 Trials with Several Treatment Arms
129
8.2.2 Factorial Trials
130
8.2.2.1 Independent Effects
130
8.2.2.2 Non-independent Effects
132
8.2.3 Crossover Design
135
8.2.4 Stepped Wedge Design
136
8.3 Design Choices for Impact Evaluation
141
8.3.1 Introduction
141
8.3.2 Repeated Cross-sectional Samples
142
8.3.3 Cohort Follow-up
143
Part C: Analytical Methods
9 Basic Principles of Analysis
149
9.1 Introduction
149
9.2 Experimental and Observational Units
149
9.3 Parameters of Interest
151
9.3.1 Event Rates
151
9.3.2 Proportions
153
9.3.2.1 Cluster-specific Odds Ratio
154
9.3.2.2 Population-average Odds Ratio
155
9.3.3 Means
156
9.3.4 More Complex Parameters
157
9.4 Approaches to Analysis
159
9.4.1 Cluster-level Analysis
159
9.4.2 Individual-level Analysis
159
9.5 Baseline Analysis
160
10 Analysis Based on Cluster-level Summaries
163
10.1 Introduction
163
10.2 Point Estimates of Intervention Effects
164
10.2.1 Point Estimates Based on Cluster Summaries
164
10.2.2 Point Estimates Based on Individual Values
165
10.2.3 Using the Logarithmic Transformation
167
10.2.4 Case Studies
168
10.3 Statistical Inference Based on the t Distribution
172
10.3.1 Unpaired t-test
172
10.3.2 Confidence Intervals Based on Cluster Summaries
173
10.3.2.1 Rate Difference
173
10.3.2.2 Rate Ratio
174
10.3.3 Case Studies
174
10.3.4 Using the Logarithmic Transformation
177
10.3.5 The Weighted t-test
178
10.4 Statistical Inference Based on a Quasi-likelihood Approach
179
10.5 Adjusting for Covariates
182
10.5.1 Stage 1: Obtaining Covariate-adjusted Residuals
182
10.5.1.1 Event Rates
183
10.5.1.2 Proportions
183
10.5.1.3 Means
184
10.5.2 Stage 2: Using the Covariate-adjusted Residuals
184
10.5.2.1 Ratio Measures of Effect
184
10.5.2.2 Difference Measures of Effect
185
10.5.3 Case Study
186
10.6 Nonparametric Methods
189
10.6.1 Introduction
189
10.6.2 Rank Sum Test
189
10.6.3 Permutation Tests
190
10.7 Analysing for Effect Modification
194
11 Regression Analysis Based on Individual-level Data
199
11.1 Introduction
199
11.2 Random Effects Models
200
11.2.1 Poisson and Cox Regressions with Random Effects
201
11.2.1.1 Poisson Regression with Random Effects
201
11.2.1.2 Cox Regression with Random Effects
207
11.2.2 Mixed Effects Linear Regression
208
11.2.3 Logistic Regression with Random Effects
213
11.3 Generalised Estimating Equations
219
11.3.1 GEE Models for Binary Data
219
11.3.2 GEE for Other Types of Outcome
221
11.4 Choice of Analytical Method
223
11.4.1 Small Numbers of Clusters
223
11.4.2 Larger Numbers of Clusters
224
11.5 Analysing for Effect Moditication
225
11.6 More Complex Analyses
226
11.6.1 Controlling for Baseline Values
226
11.6.2 Repeated Measures during Follow-up
227
11.6.3 Repeated Episodes
229
12 Analysis of Trials with More Complex Designs
233
12.1 Introduction
233
12.2 Analysis of Pair-matched Trials
233
12.2.1 Introduction
233
12.2.2 Analysis Based on Cluster-level Summaries
234
12.2.3 Adjusting for Covariates
237
12.2.4 Regression Analysis Based on Individual-level Data
241
12.3 Analysis of Stratified Trials
242
12.3.1 Introduction
242
12.3.2 Analysis Based on Cluster-level Summaries
243
12.3.3 Regression Analysis Based on Individual-level Data
250
12.4 Analysis of Other Study Designs
251
12.4.1 Trials with More than Two Treatment Arms
251
12.4.2 Factorial Trials
252
12.4.3 Stepped Wedge Trials
253
Part D: Miscellaneous Topics
13 Ethical Considerations
257
13.1 Introduction
257
13.2 General Principles
257
13.2.1 Beneficence
258
13.2.2 Equity
258
13.2.3 Autonomy
259
13.3 Ethical Issues in Group Allocation
259
13.4 Informed Consent in Cluster Randomised Trials
260
13.4.1 Consent for Randomisation
261
13.4.1.1 Political Authorities
262
13.4.1.2 Village Heads
262
13.4.1.3 Community Representatives
263
13.4.1.4 Medical Practitioners
263
13.4.2 Consent for Participation
264
13.5 Other Ethical Issues
266
13.5.1 Scientific Validity
266
13.5.2 Phased Intervention Designs
266
13.5.3 Trial Monitoring
267
13.6 Conclusion
267
14 Data Monitoring
269
14.1 Introduction
269
14.2 Data Monitoring Committees
270
14.2.1 Review of DMC Responsibilities
270
14.2.2 When Are DMCs Necessary for CRTs?
271
14.2.2.1 Likelihood of Adverse Events
271
14.2.2.2 Seriousness or Severity of Outcome Measures
271
14.2.2.3 Timing of Data Collection
272
14.2.3 Monitoring for Adverse Events
273
14.2.4 Monitoring for Efficacy
274
14.2.5 Monitoring Adequacy of Sample Size
274
14.2.6 Assessing Comparability of Treatment Arms
275
14.2.7 Approving the Analytical Plan
275
14.2.8 Presentation of Data to the DMC
276
14.3 Interim Analyses
277
14.3.1 Introduction
277
14.3.2 Timing of Interim Analyses
277
14.3.3 Stopping Rules
278
14.3.3.1 Event Rates
280
14.3.3.2 Proportions
280
14.3.3.3 Means
280
14.3.4 Disadvantages of Premature Stopping
282
15 Reporting and Interpretation
285
15.1 Introduction
285
15.2 Reporting of Cluster Randomised Trials
285
15.2.1 Overview
285
15.2.1.1 Extended CONSORT Statement
286
15.2.1.2 Publication Bias
286
15.2.2 Reporting of Methods
289
15.2.2.1 Rationale for Cluster Randomisation
289
15.2.2.2 Description of Clusters and Interventions
289
15.2.2.3 Sample Size
290
15.2.2.4 Matching, Stratification and Randomisation
291
15.2.2.5 Blinding and Allocation Concealment
291
15.2.2.6 Definition of Primary Endpoints
292
15.2.2.7 Statistical Methods
293
15.2.3 Reporting of Results
294
15.2.3.1 Flow Diagram
294
15.2.3.2 Baseline Comparisons
294
15.2.3.3 Analysis of Endpoints
295
15.2.3.4 Subgroup Analyses
296
15.2.3.5 Contamination
297
15.2.3.6 Estimates of Between-cluster Variability
297
15.3 Interpretation and Generalisability
298
15.3.1 Interpretation
298
15.3.2 Generalisability
299
15.3.3 Systematic Reviews
300
References 303
Index 309
London School of Hygiene & Tropical Medicine, London, UK Johns Hopkins Bloomberg School of Public Health, Baltimore, University of Kent, UK University of Copenhagen, Denmark Utrecht University, The Netherlands University of California, Berkeley, USA