Atnaujinkite slapukų nuostatas

Absolute Risk: Methods and Applications in Clinical Management and Public Health [Kietas viršelis]

5.00/5 (2 ratings by Goodreads)
(National Cancer Institute, Rockville, Maryland, USA), (National Cancer Institute, Rockville, Maryland, USA)
Kitos knygos pagal šią temą:
Absolute Risk: Methods and Applications in Clinical Management and Public HealthDidesnis paveikslėlis
Kitos knygos pagal šią temą:
Pastovi nuoroda: https://www.kriso.lt/db/9781466561656.html
Raktažodžiai:
Absolute Risk: Methods and Applications in Clinical Management and Public Health provides theory and examples to demonstrate the importance of absolute risk in counseling patients, devising public health strategies, and clinical management. The book provides sufficient technical detail to allow statisticians, epidemiologists, and clinicians to build, test, and apply models of absolute risk.

Features:











Provides theoretical basis for modeling absolute risk, including competing risks and cause-specific and cumulative incidence regression





Discusses various sampling designs for estimating absolute risk and criteria to evaluate models





Provides details on statistical inference for the various sampling designs





Discusses criteria for evaluating risk models and comparing risk models, including both general criteria and problem-specific expected losses in well-defined clinical and public health applications





Describes many applications encompassing both disease prevention and prognosis, and ranging from counseling individual patients, to clinical decision making, to assessing the impact of risk-based public health strategies





Discusses model updating, family-based designs, dynamic projections, and other topics

Ruth M. Pfeiffer is a mathematical statistician and Fellow of the American Statistical Association, with interests in risk modeling, dimension reduction, and applications in epidemiology. She developed absolute risk models for breast cancer, colon cancer, melanoma, and second primary thyroid cancer following a childhood cancer diagnosis.

Mitchell H. Gail developed the widely used "Gail model" for projecting the absolute risk of invasive breast cancer. He is a medical statistician with interests in statistical methods and applications in epidemiology and molecular medicine. He is a member of the National Academy of Medicine and former President of the American Statistical Association.

Both are Senior Investigators in the Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health.

Recenzijos

"Written by two leading experts in the field, this book provides a comprehensive overview of absolute risk, including both theoretical basis and clinical implications before and after the disease diagnosis. Equipped with sufficient technical details on the estimation and inference of absolute risk aswell as a range of real examples, this book is targeted toward a broad audience, including epidemiologists, clinicians, and statisticians. While a few other books on theoretical aspects of absolute risk are available in the literature, the book by Pfeiffer and Gail treats absolute risk from several new angles . . ." ~Journal of the American Statistical Association

"The book by Pfeiffer and Gail leads us into the higher statistical levels of predicting the medical future. The main focus is on the concept of the absolute risk of an event because this has a clinically meaningful interpretation for the individual person. The much more commonly reported hazard ratios of health research do not provide a directly useful number for the single subject...The examples are about the real world (mostly cancer research), and the mathematics provide all the formula for building a wellcalibrated absolute risk model and the validation study...The book contains a lot of material which is very difficult to find elsewhere, for example, on family studies, handling of missing data, and landmark analysis with time-dependent covariates. Overall, I found the book to provide a very complete documentation of a highly important subject. The authors are to be thanked for their thoroughness and congratulated for their work, which should be useful for many realworld applications of absolute risk." ~Biometrics "Written by two leading experts in the field, this book provides a comprehensive overview of absolute risk, including both theoretical basis and clinical implications before and after the disease diagnosis. Equipped with sufficient technical details on the estimation and inference of absolute risk aswell as a range of real examples, this book is targeted toward a broad audience, including epidemiologists, clinicians, and statisticians. While a few other books on theoretical aspects of absolute risk are available in the literature, the book by Pfeiffer and Gail treats absolute risk from several new angles . . ." ~Journal of the American Statistical Association

"This book provides an excellent comprehensive basis for researchers or advanced courses devoted to the development and assessment of absolute risk models. Ruth Pfeiffer and Mitchell Gail have a long history of active and successful research in the field of risk prediction modeling, the first publication of what has become known as the Gail-Model for breast cancer risk prediction having appeared over 25 years ago. This background allows them to present a broad overview of various model situations and modeling approaches together with various real-life data examples. It is a pleasure to see that assumptions and inference are treated with mathematical stringency in all addressed topics. The mathematical framework is introduced, motivated, and translated into a clinically meaningful context using worked examples, so as to give access to mathematically less experienced readers. ~Biometric Journal

"The book by Pfeiffer and Gail leads us into the higher statistical levels of predicting the medical future. The main focus is on the concept of the absolute risk of an event because this has a clinically meaningful interpretation for the individual person. The much more commonly reported hazard ratios of health research do not provide a directly useful number for the single subject...The examples are about the real world (mostly cancer research), and the mathematics provide all the formula for building a wellcalibrated absolute risk model and the validation study...The book contains a lot of material which is very difficult to find elsewhere, for example, on family studies, handling of missing data, and landmark analysis with time-dependent covariates. Overall, I found the book to provide a very complete documentation of a highly important subject. The authors are to be thanked for their thoroughness and congratulated for their work, which should be useful for many realworld applications of absolute risk." ~Biometrics

List of Figures
xv
List of Tables
xvii
Symbols xxi
Preface xxiii
1 Introduction
1(10)
1.1 Examples of risk models for disease incidence
2(5)
1.1.1 Breast cancer incidence
2(1)
1.1.1.1 A brief survey of models
2(2)
1.1.1.1 The National Cancer Institute's (NCI's) Breast Cancer Risk Assessment Tool, BCRAT
4(2)
1.1.2 Other models of cancer incidence
6(1)
1.1.3 Framingham Model for incidence of coronary heart disease
7(1)
1.2 Applications of risk models for disease incidence
7(2)
1.3 Prognosis after disease diagnosis
9(1)
1.4 Contents of book
9(2)
2 Definitions and basic concepts for survival data in a cohort without covariates
11(8)
2.1 Basic survival concepts
11(1)
2.2 Choice of time scale: age, time since diagnosis, time since accrual or counseling
12(1)
2.3 Censoring
13(2)
2.3.1 Right censoring
14(1)
2.4 Truncation
15(1)
2.5 Life-table estimator
15(2)
2.5.1 Kaplan-Meier survival estimate
16(1)
2.6 Counting processes and Markov methods
17(2)
3 Competing risks
19(8)
3.1 Concepts and definitions
19(3)
3.2 Pure versus cause-specific hazard functions
22(1)
3.3 Non-parametric estimation of absolute risk
23(4)
4 Regression models for absolute risk estimated from cohort data
27(36)
4.1 Cause-specific hazard regression
27(5)
4.1.1 Estimation of the hazard ratio parameters
29(1)
4.1.2 Non-parametric estimation of the baseline hazard
30(1)
4.1.3 Semi-parametric estimation of absolute risk rm
30(1)
4.1.4 Estimation of a piecewise exponential baseline hazard model
31(1)
4.1.5 Alternative hazard models
32(1)
4.2 Cumulative incidence regression
32(4)
4.2.1 Proportional sub-distribution hazards model
33(2)
4.2.2 Other cumulative incidence regression models
35(1)
4.2.3 Relationship between the cause-specific and the proportional sub-distribution hazards models
36(1)
4.3 Examples
36(7)
4.3.1 Absolute risk of breast cancer incidence
36(4)
4.3.2 Absolute risk of second primary thyroid cancer (SPTC) incidence
40(3)
4.4 Estimating cause-specific hazard functions from sub-samples from cohorts
43(6)
4.4.1 Case-cohort design
45(2)
4.4.2 Nested case-control design
47(2)
4.5 Estimating cause specific hazard functions from cohorts with complex survey designs
49(6)
4.5.1 Example of survey design
49(1)
4.5.2 Data
50(1)
4.5.3 Estimation of hazard ratio parameters and the baseline hazard function
50(1)
4.5.4 Example: absolute risk of cause-specific deaths from the NHANES I and II
51(4)
4.6 Variance estimation
55(8)
4.6.1 Approaches to variance estimation
56(1)
4.6.2 Influence function based variance of the absolute risk estimate from cohort data
57(6)
5 Estimating absolute risk by combining case-control or cohort data with disease registry data
63(12)
5.1 Relationship between attributable risk, composite age-specific incidence, and baseline hazard
63(1)
5.2 Estimating relative risk and attributable risk from case-control data
64(1)
5.3 Estimating relative risk and attributable risk from cohort data
65(1)
5.4 Estimating the cause-specific hazard of the competing causes of mortality, λ2(t;z2)
66(1)
5.5 Some strengths and limitations of using registry data
67(1)
5.6 Absolute risk estimate
67(1)
5.7 Example: estimating absolute risk of breast cancer incidence by combining cohort data with registry data
68(1)
5.8 Variance computations
68(7)
5.8.1 Relative risk parameters and attributable risk estimated from a case-control study
70(2)
5.8.2 Relative risk parameters and attributable risk estimated from a cohort study
72(1)
5.8.3 Variance computation when an external reference survey is used to obtain the risk factor distribution
72(1)
5.8.4 Resampling methods to estimate variance
73(2)
6 Assessment of risk model performance
75(26)
6.1 Introduction
75(1)
6.2 The risk distribution
76(2)
6.2.1 The predictiveness curve
77(1)
6.3 Calibration
78(6)
6.3.1 Definition of calibration and tests of calibration
78(3)
6.3.2 Reasons for poor calibration and approaches to recalibration
81(2)
6.3.3 Assessing calibration with right censored data
83(1)
6.3.4 Assessing calibration on the training data, that is, internal validation
83(1)
6.4 Accuracy measures
84(6)
6.4.1 Predictive accuracy: the Brier score and the logarithmic score
84(1)
6.4.2 Classification accuracy
85(1)
6.4.2.2 Distribution of risk in cases and non-cases
85(1)
6.4.2.2 Accuracy criteria
86(2)
6.4.3 Discriminatory accuracy
88(1)
6.4.4 Extensions of accuracy measures to functions of time and allowance for censoring
89(1)
6.5 Criteria for applications of risk models for screening or high-risk interventions
90(4)
6.5.1 Proportion of cases followed and proportion needed to follow
90(2)
6.5.1.1 Estimation of PCF and PNF
92(2)
6.6 Model assessment based on expected costs or expected utility specialized for a particular application
94(7)
6.6.1 Two health states and two intervention options
95(2)
6.6.2 More complex outcomes and interventions
97(1)
6.6.2.2 Example with four intervention choices
97(1)
6.6.2.2 Multiple outcomes in prevention trials
98(1)
6.6.2.2 Expected cost calculations for outcomes following disease diagnosis
99(2)
7 Comparing the performance of two models
101(18)
7.1 Use of external validation data for model comparison
101(1)
7.2 Data example
102(1)
7.3 Comparison of model calibration
102(2)
7.4 Model comparisons based on the difference in separate model-specific estimates of a criterion
104(8)
7.4.1 Comparisons of predictive accuracy using the Brier and logarithmic scores
105(1)
7.4.2 Classification accuracy criteria based on single risk threshold
105(2)
7.4.3 Comparisons based on the receiver operating characteristic (ROC) curve
107(1)
7.4.4 Integrated discrimination improvement (IDI) and mean risk difference
108(1)
7.4.5 Comparing two risk models based on PCF, PNF, iPCF, or iPNF
109(1)
7.4.6 Comparisons based on expected loss or expected benefit
110(2)
7.5 Joint distributions of risk
112(1)
7.6 Risk stratification tables and reclassification indices
112(5)
7.6.1 Net reclassification improvement (NRI)
115(1)
7.6.2 Extensions of NRI
116(1)
7.7 Concluding remarks
117(2)
8 Building and updating relative risk models
119(16)
8.1 Introductory remarks
119(1)
8.2 Selection of covariates
119(3)
8.3 Missing data
122(2)
8.3.1 Types of missing data
123(1)
8.3.2 Approaches to handling missing data
123(1)
8.4 Updating risk models with new risk factors
124(11)
8.4.1 Estimating an updated relative risk model, rr(X,Z), from case-control data
125(1)
8.4.2 Estimating rr(X, Z) from new data only
125(1)
8.4.3 Incorporating information on rr(X) into rr(X, Z) via likelihood ratio (LR) updating
126(1)
8.4.3.3 Joint estimation of LRD(Z|X)
127(1)
8.4.3.3 Estimating LRD(Z|X) based on fitting separate models for cases (D = 1) and non-cases (D = 0)
127(1)
8.4.3.3 LR updating assuming independence of Z and X (independence Bayes)
128(1)
8.4.3.3 LR updating with multiple markers
128(1)
8.4.4 Joint estimation, logistic model with offset
128(1)
8.4.5 Independence Bayes with shrinkage
129(1)
8.4.6 Updating using constrained maximum likelihood estimation (CML)
129(1)
8.4.7 Simulations
130(2)
8.4.8 Summary
132(3)
9 Risk estimates based on genetic variants and family studies
135(16)
9.1 Introduction
135(1)
9.2 Mendelian models: the autosomal dominant model for pure breast cancer risk
136(2)
9.3 Models that allow for residual familial aggregation to estimate pure breast cancer risk
138(2)
9.3.1 Polygenic risk
138(1)
9.3.2 Models with latent genetic effects: BOADICEA and IBIS
138(2)
9.3.3 Copula models
140(1)
9.4 Estimating genotype-specific absolute risk from family-based designs
140(4)
9.4.1 General considerations
140(1)
9.4.2 Combining relative-risks from family-based case-control studies with population-based incidence rates
141(1)
9.4.3 Kin-cohort design
141(2)
9.4.4 Families with several affected members (multiplex pedigrees)
143(1)
9.5 Comparisons of some models for projecting breast cancer risk
144(4)
9.6 Discussion
148(3)
10 Related topics
151(20)
10.1 Introduction
151(1)
10.2 Prognosis following disease onset
151(1)
10.3 Missing or misclassified information on event type
152(2)
10.4 Time varying covariates
154(6)
10.4.1 Fixed versus time-varying covariates and internal versus external time-varying covariates
154(1)
10.4.2 Joint modeling of covariates and health outcomes, including multistate models
155(3)
10.4.3 Landmark analysis
158(2)
10.5 Risk model applications for counseling individuals and for public health strategies for disease prevention
160(11)
10.5.1 Use of risk models in counseling individuals
160(1)
10.5.1.1 Providing realistic risk estimates and perspective
160(2)
10.5.1.1 More formal risk-benefit analysis for individual counseling
162(1)
10.5.2 Use of risk models in public health prevention
162(1)
10.5.2.2 Designing intervention trials to prevent disease
162(1)
10.5.2.2 Assessing absolute risk reduction in a population from interventions on modifiable risk factors
163(2)
10.5.2.2 Implementing a "high risk" intervention strategy for disease prevention
165(4)
10.5.2.2 Allocating preventive interventions under cost constraints
169(2)
Bibliography 171(22)
Index 193
Ruth M. Pfeiffer is a mathematical statistician and Fellow of the American Statistical Association, with interests in risk modeling, dimension reduction, and applications in epidemiology. She developed absolute risk models for breast cancer, colon cancer, melanoma, and second primary thyroid cancer following a childhood cancer diagnosis.

Mitchell H. Gail developed the widely used "Gail model" for projecting the absolute risk of invasive breast cancer. He is a medical statistician with interests in statistical methods and applications in epidemiology and molecular medicine. He is a member of the National Academy of Medicine and former President of the American Statistical Association.

Both are Senior Investigators in the Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health.