Ahrens / Pigeot Handbook of Epidemiology

1. Auflage 2007
ISBN: 978-3-540-26577-1
Verlag: Springer-Verlag
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 2300 Seiten

This 3-volume reference covers the entire field of epidemiology, from statistical methods and study design, to specialized areas such as molecular epidemiology, and applications in clinical medicine and health services research. This updated edition of the Handbook of  Epidemiology adds 22 new chapters on: History of Epidemiological Methods and Concepts, Cluster Randomized Trials, Internet-Based Epidemiology, Misclassification, Sensitivity Analysis and Bias Analysis,  Emergency and Disaster Health Surveillance, Statistical Inference, Data Management in Epidemiology, Visual Display of Quantitative Information, Bayesian Methods in Epidemiology, Generalized Estimating Equations, Directed Acyclic Graphs, Life Course Epidemiology, Molecular Epidemiology, Physical Activity, Radiation Epidemiology, Epidemiology of  Obesity, Epidemiology of Respiratory Allergies and Asthma, Epidemiology of  Dental Diseases, Epidemiology of Digestive Diseases, Psychiatric Disorders, Epidemiology of  Diabetes.

All other chapters are extensively revised from  the 1st edition. This is a reference for epidemiological researchers and graduate students in public health.

Autoren/Hrsg.

Ahrens, Wolfgang

Pigeot, Iris

Inhaltsverzeichnis

1;Foreword;5
2;Preface;6
3;Table of Contents;8
4;An Introduction to Epidemiology;11
4.1;1 Epidemiology and Related Areas;13
4.1.1;1.1 Definition and Purpose of Epidemiology;13
4.1.2;1.2 Epidemiology in Relation to Other Disciplines;15
4.1.3;1.3 Overview;17
4.2;2 Development of Epidemiology;19
4.2.1;2.1 Historical Background;19
4.2.2;2.2 Milestones in Epidemiological Research ;22
4.2.3;2.3 Methodological Limits;24
4.3;3 Concepts and Methodological Approaches in Epidemiology;26
4.3.1;Concepts;26
4.3.2;Study Designs;27
4.3.3;Data Collection;30
4.4;4 Statistical Methods in Epidemiology;31
4.4.1;Principles of Data Analysis;32
4.4.2;Statistical Thinking;33
4.4.3;Multivariate Analysis;34
4.4.4;Handling of Data Problems;37
4.4.5;Meta-Analysis;38
4.5;5 Applications of Epidemiological Methods and Research Areas in Epidemiology;39
4.5.1;Description of the Spectrum of Diseases;39
4.5.2;Identification of Causes of Disease;39
4.5.3;Application of Epidemiological Knowledge;42
4.5.4;Ethical Aspects;45
4.6;References;46
5;Part I Concepts and Methodological Approaches in Epidemiology;52
5.1;1 Basic Concepts;53
5.1.1;Introduction;55
5.1.2;Causation and Causal Inference;55
5.1.2.1;A General Model of Causation;55
5.1.2.2;Concept of Sufficient Cause and Component Causes;55
5.1.2.3;Strength of Effects;57
5.1.2.4;Interaction Among Causes;58
5.1.2.5;Proportion of Disease Due to Specific Causes;59
5.1.2.6;Induction Period and Latent Period;60
5.1.2.7;Philosophy of Scientific Inference;61
5.1.2.8;Refutationism;62
5.1.2.9;Bayesianism;64
5.1.2.10;Impossibility of Proof;65
5.1.2.11;Causal Inference in Epidemiology;66
5.1.2.12;Causal Criteria;67
5.1.3;Measures of Disease Frequency;68
5.1.3.1;Incidence Time;68
5.1.3.2;Incidence Rate;69
5.1.3.3;Closed and Open Populations;71
5.1.3.4;Steady State;72
5.1.3.5;Interpretation of an Incidence Rate;73
5.1.3.6;Incidence Proportions and Survival Proportions;75
5.1.3.7;Prevalence;76
5.1.3.8;Prevalence, Incidence and Mean Duration;77
5.1.3.9;Utility of Prevalence in Etiologic Research;78
5.1.4;Measures of Effect;79
5.1.4.1;Simple Effect Measures;80
5.1.4.2;Effect MeasureModification;82
5.1.4.3;Relative versus AbsoluteMeasures;83
5.1.4.4;Attributable Fractions;83
5.1.4.5;Population Attributable Fractions and Impact Fractions;86
5.1.4.6;Estimation of Effects;86
5.1.4.7;Measures of Association;87
5.1.5;Confounding;88
5.1.6;Selection Bias;90
5.1.6.1;Self-Selection Bias;90
5.1.6.2;Diagnostic Bias;90
5.1.7;Information Bias;91
5.1.7.1;DifferentialMisclassification;91
5.1.7.2;Nondifferential Misclassification;92
5.1.7.3;Misclassification of Confounders;94
5.1.8;Conclusions;95
5.1.9;References;95
5.2;2 Rates, Risks, Measures of Association and Impact;99
5.2.1;Introduction;101
5.2.2;Incidence and Hazard Rates;101
5.2.2.1;Definition;101
5.2.2.2;Estimability and Basic Principles of Estimation;102
5.2.2.3;Relation with Other Measures;104
5.2.3;Measures of Disease Risk;105
5.2.3.1;Definition;105
5.2.3.2;Range;106
5.2.3.3;Synonyms;106
5.2.3.4;Interpretation and Usefulness;107
5.2.3.5;Properties;109
5.2.3.6;Estimability;109
5.2.3.7;Estimation from Cohort Studies;110
5.2.3.8;Estimation from Population-based;114
5.2.3.9;or Nested Case-Control Studies;114
5.2.3.10;Final Notes and Additional References;119
5.2.4;Measures of Association;119
5.2.4.1;Definitions and General Points;119
5.2.4.2;Usefulness and Interpretation;120
5.2.4.3;Measures Based on Ratios;121
5.2.4.4;Measures Based on Differences;124
5.2.4.5;Estimability;125
5.2.4.6;Estimation;126
5.2.5;Measures of Impact;133
5.2.5.1;Attributable Risk;133
5.2.5.2;Attributable Risk Among the Exposed;147
5.2.5.3;Sequential and Partial Attributable Risks;148
5.2.5.4;Preventable and Prevented Fractions;150
5.2.5.5;Generalized Impact Fraction;151
5.2.5.6;Person-Years of Life Lost;152
5.2.6;Other Topics;153
5.2.6.1;Standardization of Risks and Rates;153
5.2.6.2;Measures Based on Prevalence;155
5.2.7;Conclusions;156
5.2.8;References;156
5.3;3 Descriptive Studies;167
5.3.1;Introduction;168
5.3.1.1;Definitions;168
5.3.1.2;Sources of Data;169
5.3.1.3;Outline of the Chapter;169
5.3.2;Measurement;170
5.3.2.1;Incidence;170
5.3.2.2;Prevalence;171
5.3.2.3;Case Fatality| Survival;171
5.3.2.4;Mortality;172
5.3.2.5;Person-Years of Life Lost (PYLL);174
5.3.2.6;Rates of Disease;175
5.3.2.7;Population at Risk;176
5.3.3;Comparisons Between Populations;177
5.3.3.1;Comparisons Between Two Groups;178
5.3.3.2;Comparisons Between Multiple Groups;181
5.3.4;Study Designs in Descriptive Epidemiology;183
5.3.4.1;Data on Individuals;183
5.3.4.2;Ecological Studies;186
5.3.5;Descriptive Studies;195
5.3.5.1;Personal Characteristics;195
5.3.5.2;Place of Residence;204
5.3.5.3;Combinations;217
5.3.6;Conclusions;229
5.3.7;References;230
5.4;4 Use of Disease Registers;241
5.4.1;Introduction;242
5.4.2;Types of Registers;242
5.4.3;Organisation of Registries;243
5.4.4;Contents and Variables in the Registers;245
5.4.5;The Quality of Data in Registers;246
5.4.6;Study Designs in Register-based Studies;247
5.4.7;Potentials and Limitations of Register- based Studies;248
5.4.8;Examples of the Usefulness;249
5.4.8.1;Environmental Studies;250
5.4.8.2;Occupational Studies and Social Epidemiology;251
5.4.8.3;Survival Analysis;252
5.4.8.4;Surveillance of Drug Effects;253
5.4.8.5;Other Examples of Etiological Research;254
5.4.8.6;Quality Control in Medical Care Using Registers;254
5.4.9;Ethics, Confidentiality and Legislation;255
5.4.10;Conclusions;256
5.4.11;References;257
5.5;5 Cohort Studies;263
5.5.1;Introduction;264
5.5.2;A Brief Historical Perspective on Cohort Studies;264
5.5.3;Conceptual Foundations;266
5.5.3.1;Types of Cohort Studies:;266
5.5.3.2;Concurrent and Non-concurrent Approaches;266
5.5.3.3;Description of Outcome Events in the Cohort;268
5.5.3.4;External Comparisons;272
5.5.3.5;Summary Effects of Exposure;274
5.5.3.6;Internal;275
5.5.3.7;Modelling;275
5.5.3.8;of the Effects of Exposure;275
5.5.3.9;Internal Versus External Comparisons;280
5.5.4;Key Concerns in Cohort Studies;281
5.5.4.1;Selection of the Study Population;281
5.5.4.2;Exposure and Confounders in Cohort Studies;284
5.5.4.3;Determining Outcome Events;287
5.5.5;Ethical Issues;290
5.5.6;Conclusions;291
5.5.7;References;292
5.6;6 Case-Control Studies;297
5.6.1;Introduction;298
5.6.1.1;A Brief History;298
5.6.1.2;Early Methodologic Developments;299
5.6.1.3;Chapter Outline;301
5.6.2;Conceptual Foundations;301
5.6.2.1;Sampling from a Real or Fictitious Cohort;301
5.6.2.2;Incidence Rate Ratios are Estimable from Odds Ratios;303
5.6.2.3;Time-dependent Rates and Exposures;304
5.6.2.4;Cumulative Risk Ratios and Case-Cohort Sampling;304
5.6.2.5;Estimation of Absolute Risks;306
5.6.3;Matching and Stratification;308
5.6.3.1;Consequences of Matching;308
5.6.3.2;Efficiency of Matching;308
5.6.3.3;Overmatching;311
5.6.3.4;When to Match;312
5.6.4;Selection of Subjects;312
5.6.4.1;Selection of Cases;312
5.6.4.2;Selection of Controls;314
5.6.4.3;HowMany Controls per Case?;317
5.6.4.4;HowMany Control Groups?;317
5.6.5;Pitfalls;318
5.6.5.1;Selection Bias;318
5.6.5.2;Adjustments for Selection Bias;320
5.6.5.3;in Study Design and Analysis;320
5.6.5.4;Measurement Error;321
5.6.6;Conclusions;322
5.6.7;References;324
5.7;7 Modern Epidemiologic Study Designs;331
5.7.1;Introduction;332
5.7.2;Case-Cohort Studies;333
5.7.3;Nested Case-Control Studies;337
5.7.3.1;Design Features;337
5.7.3.2;Strengths;338
5.7.3.3;Limitations;338
5.7.4;Case-Crossover Studies;339
5.7.5;Case-Time-Control Studies;344
5.7.6;Case-Specular Studies;346
5.7.7;Genetic Epidemiology Case Only Designs;348
5.7.7.1;Design Features;348
5.7.7.2;Strengths;351
5.7.7.3;Limitations;351
5.7.8;Conclusions;351
5.7.9;References;352
5.8;8 Intervention Trials;355
5.8.1;Introduction;356
5.8.1.1;Guidelines for Reporting of Clinical Trials:;357
5.8.1.2;The CONSORT Statement;357
5.8.2;Therapeutic versus Preventive Trials;360
5.8.2.1;Surrogate Endpoints;361
5.8.2.2;From Observation to Intervention;362
5.8.3;The Origin of Randomized Trials;364
5.8.4;Planning of Trials;364
5.8.5;Definition of the Intervention;365
5.8.6;Factorial Design;367
5.8.7;Definition of Participants;368
5.8.8;Enrollment;370
5.8.9;Randomization;371
5.8.10;Cluster Randomization;372
5.8.11;Trial Outcome(s);373
5.8.12;Follow-up, Exclusions andWithdrawal;374
5.8.13;Conclusions;376
5.8.14;References;376
5.9;9 Confounding and Interaction;381
5.9.1;Introduction;382
5.9.2;Confounding;382
5.9.2.1;Basic Concepts;382
5.9.2.2;Example of Confounding;386
5.9.2.3;Control in the Study Design;387
5.9.2.4;Control in the Analysis;388
5.9.2.5;Assessment of Confounding;388
5.9.2.6;Relationship Between Confounding and Other Biases;391
5.9.3;Interaction;393
5.9.3.1;Basic Concepts;393
5.9.3.2;Example of Effect-measureModification;394
5.9.3.3;Concepts of Interaction;394
5.9.3.4;Additive and Multiplicative Models;396
5.9.4;Detecting Interactions;397
5.9.4.1;Assessment of Joint Effects;399
5.9.5;Conclusions;401
5.9.6;References;402
5.10;10 Epidemiological FieldWork in Population-Based Studies;409
5.10.1;Introduction;411
5.10.2;Asking for Information: What Are the Characteristics of Straightforward Questions and Responses?;411
5.10.2.1;Types of Questions;411
5.10.2.2;Types of Responses;413
5.10.3;How Are Field Study Measures and Questions Organized?;414
5.10.3.1;Length;414
5.10.3.2;Question Order;414
5.10.3.3;Aesthetics and Other Concerns;416
5.10.3.4;Branching Questions or Skip Patterns;416
5.10.4;What Does It Take to Ensure Proper Administration of Field Instruments?;417
5.10.4.1;Self-Administered Questionnaires;417
5.10.4.2;Interviews;418
5.10.5;How Can You Assure a Reliable and Valid Field Measure?;421
5.10.5.1;Pilot Testing;421
5.10.5.2;Reliability and Validity: The Quality of the Measure;422
5.10.5.3;Ensuring Quality: Selecting Ready-to-Use Measures;422
5.10.5.4;Reliability;423
5.10.5.5;Questions to Ask;425
5.10.5.6;About a Published Instrument’s Validity;425
5.10.5.7;Suggested Guidelines For Pilot Testing;426
5.10.6;FieldWork Language and Culture;427
5.10.6.1;Suggested Guidelines for Translation;428
5.10.7;Managing the Data;428
5.10.8;What Are Reasonable Resources?;431
5.10.8.1;WhoWill Do It, andWhat Resources Are Needed?;433
5.10.8.2;Personnel, Time, And Money;433
5.10.9;Conclusions;444
5.10.10;References;445
5.11;11 Exposure Assessment;447
5.11.1;Introduction;448
5.11.2;Definition of Exposure and Exposure Assessment;448
5.11.2.1;Types of Exposure;448
5.11.2.2;True Risk Factor or Surrogate;449
5.11.2.3;Dose versus Exposure;449
5.11.2.4;Selection of Metric;449
5.11.3;Exposure Data;450
5.11.3.1;Measurement Data;451
5.11.3.2;Indirect Exposure Data;453
5.11.4;The Process of Exposure Assessment;456
5.11.4.1;Creating an Exposure Estimate;456
5.11.4.2;Establishing a Level of Classification;457
5.11.4.3;Retrospective Exposure Assessment;459
5.11.4.4;Ecological versus Individual Exposure Assessment;460
5.11.4.5;Dealing with Multiple Exposures;461
5.11.5;Measurement Errors;464
5.11.5.1;Types ofMeasurement Errors;464
5.11.5.2;Sources ofMeasurement Errors;465
5.11.5.3;Quantification of Measurement Errors –;467
5.11.5.4;Reproducibility Studies;467
5.11.5.5;Quantification of Measurement Errors –;468
5.11.5.6;Validation Studies;468
5.11.5.7;Methods for Correcting Measurement Errors;468
5.11.6;Conclusions;469
5.11.7;References;470
5.12;12 Design and Planning of Epidemiological Studies;473
5.12.1;Introduction;475
5.12.2;Early Planning;475
5.12.2.1;Objectives – the Concept of the Study;475
5.12.2.2;Scientific Background;476
5.12.2.3;The Study Protocol;476
5.12.3;Design;479
5.12.3.1;Study Base;480
5.12.3.2;Cohort Studies;480
5.12.3.3;Case-Control Studies;484
5.12.3.4;The Study Base Principle:;488
5.12.3.5;Selection, Exclusion and Resulting Bias;488
5.12.3.6;Choosing Between Epidemiological Designs;489
5.12.3.7;Statistical Power;491
5.12.4;Measures of Disease Outcome and Exposure Parameters;492
5.12.4.1;Measurement and Classification of Exposure;492
5.12.4.2;Measurement and Classification of Health Outcomes;497
5.12.4.3;Information Bias;498
5.12.5;Confounding;499
5.12.6;Statistical Analysis;500
5.12.7;Practical Issues;501
5.12.7.1;Fund Raising;501
5.12.7.2;Data Management;502
5.12.7.3;Quality Assurance;503
5.12.7.4;Study Conduct: Manual of Operations;504
5.12.7.5;Time Line;505
5.12.7.6;Project Diary;505
5.12.7.7;Ethical Aspects;507
5.12.7.8;Scientific Collaborations and Multi-centre Studies;507
5.12.7.9;Publication;508
5.12.8;Conclusions;509
5.12.9;References;509
5.13;13 Quality Control and Good Epidemiological Practice;513
5.13.1;Introduction;514
5.13.2;The Datascope;516
5.13.3;Quality Considerations in the Planning Phase;519
5.13.3.1;Protocol;519
5.13.3.2;Documentation of Operations and Procedures;521
5.13.3.3;Personnel, Training and Certification;522
5.13.3.4;Data Collection Forms and Instruments;524
5.13.3.5;Planning Response Rate;526
5.13.3.6;Validity and Reliability;527
5.13.3.7;Planning Data Management;533
5.13.3.8;Quality Assurance Committee;542
5.13.3.9;Communications;542
5.13.3.10;Cost of Quality Assurance;543
5.13.3.11;Ethical Considerations;543
5.13.4;Quality Considerations During Study Conduct;544
5.13.4.1;Training and Certification;545
5.13.4.2;Maintenance and Calibration of Equipment;545
5.13.4.3;Implementing Data Management;547
5.13.4.4;Site Visits;548
5.13.5;Quality Considerations After Data Collection;550
5.13.5.1;Reporting Response Rate;550
5.13.5.2;Analysis;551
5.13.5.3;Storage and Retrieval of Data;556
5.13.6;Conclusions;556
5.13.7;References;558
6;Part II Statistical Methods in Epidemiology;568
6.1;1 Sample Size Determination in Epidemiologic Studies;569
6.1.1;Introduction;571
6.1.2;One Group Designs, Inferences About Proportions;571
6.1.2.1;Confidence Intervals for a Single Population Proportion;572
6.1.2.2;Hypothesis Testing for a Single Population Proportion;574
6.1.2.3;Additional Considerations and References;577
6.1.3;Comparison of Two Independent Proportions;577
6.1.3.1;Study Designs, Parameters, Analysis Methods;577
6.1.3.2;Confidence Intervals for the Risk Difference;579
6.1.3.3;Confidence Interval for Relative Risk (Ratio);581
6.1.3.4;Confidence Intervals for the Odds Ratio;582
6.1.3.5;Testing the Difference Between Two Proportions;584
6.1.3.6;Testing the Relative Risk;585
6.1.3.7;Testing the Odds Ratio;586
6.1.3.8;Additional Considerations and References;587
6.1.4;One Group Designs, Inferences About a Single Mean;588
6.1.4.1;Confidence Intervals for a Single Mean;588
6.1.4.2;Hypothesis Testing for a Single Population Mean;590
6.1.5;Comparison of Two Independent Means;592
6.1.5.1;Confidence Intervals;592
6.1.5.2;for the Difference Between Two Means;592
6.1.5.3;Testing the Difference;594
6.1.5.4;Between;594
6.1.5.5;TwoMeans;594
6.1.5.6;( Two-Sample;594
6.1.5.7;Test);594
6.1.5.8;Additional Considerations and References;595
6.1.6;Logistic Regression Models;597
6.1.6.1;Single Dichotomous Covariate;597
6.1.6.2;Single Continuous Covariate;598
6.1.6.3;Adjusting Sample Size for Inclusion of;599
6.1.6.4;Prior Covariates;599
6.1.6.5;( Variance Inflation Factor);599
6.1.6.6;Assessing;600
6.1.6.7;the Adequacy of Data Already Collected;600
6.1.7;Practical Issues in Sample Size Choice;601
6.1.8;Conclusions;602
6.1.9;References;603
6.2;2 General Principles of Data Analysis: Continuous Covariables in Epidemiological Studies;605
6.2.1;Introduction;607
6.2.2;Classical Methods to Analyse Continuous Covariables Based on Contingency Tables;609
6.2.2.1;General Aspects;609
6.2.2.2;Odds Ratio Estimation;610
6.2.2.3;Confounder Adjustment;611
6.2.2.4;Test for Trend;613
6.2.2.5;Concluding Remarks on Classical Methods;614
6.2.3;Regression Models and Risk Functions;614
6.2.4;Methods to Analyse Continuous Covariables in Regression Models;617
6.2.4.1;Categorisation of;617
6.2.4.2;into;617
6.2.4.3;Levels (Method i);617
6.2.4.4;Leaving;618
6.2.4.5;Untransformed (Method ii);618
6.2.4.6;Transformation of;618
6.2.4.7;via a Monotonous Function;618
6.2.4.8;(Method iii);618
6.2.4.9;Use of Additive (Linear) Risk Function (Method iv);619
6.2.4.10;Additional Dichotomous Variable (Method v);620
6.2.4.11;Fractional Polynomials (FP) (Method vi);621
6.2.4.12;A Special Logarithmic Transformation (Method vii);621
6.2.4.13;Conditioning (Method viii);622
6.2.4.14;Generalized Additive Models (GAM’s) (Method ix);624
6.2.4.15;Spline Regression (Method x);624
6.2.5;Examples;625
6.2.6;Conclusions;630
6.2.7;References;632
6.3;3 Regression Methods for Epidemiologic Analysis;635
6.3.1;Introduction;637
6.3.2;Regression Functions;637
6.3.2.1;Frequentist Regression;638
6.3.2.2;Other Concepts of Population;639
6.3.2.3;Binary Regression;639
6.3.2.4;Multiple Regression;640
6.3.2.5;Regression and Causation;641
6.3.2.6;Frequentist versus Bayesian Regression;644
6.3.3;Basic Regression Models;645
6.3.3.1;Model Specification and Model Fitting;645
6.3.3.2;Background Example;647
6.3.3.3;Vacuous Models;647
6.3.3.4;Constant Models;648
6.3.3.5;Linear Risk Models;649
6.3.3.6;Recentering;650
6.3.3.7;Rescaling;650
6.3.3.8;Exponential Risk Models;651
6.3.3.9;Logistic Models;652
6.3.3.10;A Graphical Example;654
6.3.3.11;Other Risk and Odds Models;655
6.3.3.12;Rate Models;656
6.3.3.13;Incidence-Time and Hazard Models;657
6.3.3.14;Trend Models: Univariate Exposure Transforms;659
6.3.3.15;Interpreting Models After Transformation;661
6.3.4;Multiple Regression Models;662
6.3.4.1;Relations Among Multiple-Regression Models;663
6.3.4.2;Product Terms (Statistical Interactions);665
6.3.4.3;Trends and Product Terms;667
6.3.4.4;Interpreting Product-Term Models;668
6.3.4.5;Categorical Regressors;670
6.3.5;Trend Models in Multiple Regression;673
6.3.5.1;Categorical Trends;673
6.3.5.2;Regression with Category Scores;674
6.3.5.3;PowerModels;675
6.3.5.4;Regression Splines;676
6.3.5.5;Models for Trend Variation;678
6.3.6;Extensions of Logistic Models;679
6.3.6.1;Polytomous Logistic Models;679
6.3.6.2;Ordinal Logistic Models;680
6.3.7;Generalized Linear Models;682
6.3.8;Model Searching;684
6.3.8.1;Role of Prior Information;685
6.3.8.2;Selection Strategies;686
6.3.9;Model Fitting;687
6.3.9.1;Residual Distributions;687
6.3.9.2;Overdispersion;688
6.3.9.3;Sample-Size Considerations;689
6.3.10;Model Checking;690
6.3.10.1;Tabular Checks;690
6.3.10.2;Tests of Regression and;691
6.3.10.3;Tests of Fit;692
6.3.10.4;Global Tests of Fit;694
6.3.10.5;Model Diagnostics;696
6.3.10.6;Delta-Beta Analysis;696
6.3.11;Conclusions;697
6.3.12;References;697
6.4;4 Survival Analysis;703
6.4.1;Introduction;704
6.4.2;Basic Concepts;705
6.4.2.1;Walking Backwards Through Time;705
6.4.2.2;Describing and Estimating the Distribution;705
6.4.2.3;of Survival Times;705
6.4.2.4;Describing Registry Data – Standardised Rates;707
6.4.2.5;Exponential Distribution;708
6.4.3;Cohort Studies;709
6.4.3.1;Standardised Mortality Ratio;710
6.4.3.2;Conditional Inference Within a Cohort;710
6.4.3.3;Regression Models for Rare Events;711
6.4.3.4;Life-Tables;716
6.4.3.5;Estimating Survival from Population Registries;717
6.4.4;Competing Risks;718
6.4.5;Comparing Survival Times Between Groups;722
6.4.6;Regression Models for Survival Data;724
6.4.6.1;Accelerated Failure Time Models;724
6.4.6.2;The Cox Model;725
6.4.6.3;CoxModel;727
6.4.6.4;for Excess Risk;727
6.4.6.5;Sampling from the Risk Set;728
6.4.6.6;Aalen Model;729
6.4.6.7;Adjusted Survival Curves;730
6.4.7;Censoring and Truncation;731
6.4.8;Conclusions;734
6.4.9;References;734
6.5;5 Measurement Error;739
6.5.1;Introduction;740
6.5.2;Measurement Error and Its Effects;741
6.5.2.1;Differential and Nondifferential Error,;741
6.5.2.2;and Surrogate Variables;741
6.5.2.3;Error Models;743
6.5.2.4;Measurement Error in the Normal Linear Model;744
6.5.2.5;Multiple Linear Regression;748
6.5.2.6;Nonlinear Regression;749
6.5.2.7;Logistic Regression Example;750
6.5.3;Planning Epidemiologic Studies with Measurement Error;752
6.5.3.1;Methods for Sample Size Calculations;754
6.5.3.2;Planning for Reliability| Validation Data;755
6.5.3.3;Examples and Applications;757
6.5.4;Measurement Error Models and Methods;757
6.5.4.1;Overview;757
6.5.4.2;Regression Calibration;758
6.5.4.3;SIMEX;760
6.5.4.4;Analytic Example;761
6.5.4.5;Graphic Example;761
6.5.4.6;Estimating Equations and Corrected Scores;763
6.5.4.7;Conditional Scores;764
6.5.4.8;Instrumental Variables;765
6.5.4.9;Likelihood Methods;766
6.5.4.10;Survival Analysis;768
6.5.5;Conclusions;769
6.5.6;References;770
6.6;6 Missing Data;777
6.6.1;Introduction;779
6.6.2;Case Studies;780
6.6.2.1;The Vorozole Study;780
6.6.2.2;The Psychiatric Study;788
6.6.3;Simple Ad Hoc Methods;789
6.6.3.1;Complete Case Analysis;790
6.6.3.2;Simple Forms of Imputation;790
6.6.3.3;Discussion of Imputation Techniques;792
6.6.4;A Classic Model for Continuous Longitudinal Data;793
6.6.5;Simple Methods Applied to the Vorozole Study;794
6.6.6;A Framework for HandlingMissing Values;795
6.6.6.1;Continuous Data and Ignorability;798
6.6.7;Selection Models;800
6.6.8;Selection Models Applied to the Vorozole Study;801
6.6.9;Pattern-MixtureModels;805
6.6.10;Pattern-MixtureModels Applied to the Vorozole Study;806
6.6.11;Non-Gaussian Repeated Measures;809
6.6.11.1;Marginal Models;809
6.6.11.2;Random-Effects Models;812
6.6.12;Weighted Generalized Estimating Equations;815
6.6.13;GLMM andWGEE Applied to the Psychiatric Study;816
6.6.14;From MAR to Sensitivity Analysis;817
6.6.15;Selection Models and Local Influence;819
6.6.16;Pattern-MixtureModeling Approach;822
6.6.16.1;Identifying Restriction Strategies;823
6.6.16.2;How to Use Restrictions?;824
6.6.17;Pattern-Mixture Sensitivity Analysis for the Vorozole Study;825
6.6.17.1;Models Based on Identifying Restrictions;825
6.6.18;Conclusions;830
6.6.19;Appendix;832
6.6.20;Pattern-Mixture Modelling;832
6.6.20.1;Identifying Restriction Strategies;832
6.6.21;References;833
6.7;7 Meta-Analysis in Epidemiology;839
6.7.1;Introduction;840
6.7.2;Different Types of Overviews;840
6.7.3;Reasons forMeta-Analysis in Epidemiology;842
6.7.4;Steps in Performing aMeta-Analysis;843
6.7.5;Statistical Analysis;845
6.7.6;Interpretation of the Results of Meta- Analysis of Observational Studies;857
6.7.6.1;Bias;857
6.7.6.2;Confounding;858
6.7.6.3;Heterogeneity;858
6.7.7;Conclusions;859
6.7.8;Appendix;860
6.7.9;Data and Computer Code and Output;860
6.7.10;Elementary Analysis with SAS;861
6.7.11;SAS Code for the Random Effects Model;862
6.7.12;References;863
6.8;8 Geographical Epidemiology;869
6.8.1;Introduction;871
6.8.1.1;The Nature of Geographical Epidemiology;871
6.8.1.2;Scope of the Chapter;871
6.8.1.3;Chapter Contents;872
6.8.2;Statistical Models;873
6.8.2.1;A Statistical Framework;873
6.8.2.2;for Epidemiological Observations;873
6.8.2.3;Statistical Models for Geographical Data;875
6.8.3;Modelling Disease Risk in Relation to Geographically Referenced Factors;876
6.8.3.1;Areal Data;876
6.8.3.2;An Example of the Log-Linear Model for Areal Data;877
6.8.3.3;Calculating the Expectations;878
6.8.3.4;Continuous Data;879
6.8.3.5;Spatial Structure in the Residual Variation;880
6.8.4;Mapping Disease Risk;880
6.8.5;The Detection of Generalised Heterogeneity;884
6.8.5.1;The Assessment of Heterogeneity in Areal Data;884
6.8.5.2;Detecting Heterogeneity in Poisson Data;884
6.8.5.3;Spatial and Non-spatial Analyses;886
6.8.5.4;Heterogeneity Tests Based on the Risk Surface;888
6.8.6;Clustering;889
6.8.6.1;Methods Based on the RRF;889
6.8.6.2;Knox’s Test;890
6.8.6.3;Other Space-Time Clustering Methods;891
6.8.6.4;Case-Only Clustering;892
6.8.6.5;Population Distance;892
6.8.6.6;Choosing Scale Parameters;893
6.8.7;Pre-defined Sources of Risk;893
6.8.7.1;Tests for Concentration of Risk;894
6.8.7.2;Summary of Recommendations;896
6.8.7.3;Example: Childhood Leukaemia;896
6.8.7.4;Around UK Nuclear Installations;896
6.8.8;Conclusions;897
6.8.9;References;898
7;Part III Applications of Epidemiology;902
7.1;III.1 Social Epidemiology;903
7.1.1;Introduction;904
7.1.2;Research Questions;904
7.1.2.1;The Social Determinants of Health;904
7.1.2.2;Health Behaviours;906
7.1.2.3;Material, Economic and Political Determinants of Health;907
7.1.2.4;Life Course;908
7.1.2.5;Social Biology;909
7.1.2.6;Ecological Perspectives;910
7.1.2.7;General Susceptibility to Disease;911
7.1.3;Research Methods;915
7.1.3.1;Applying a Population Perspective;915
7.1.3.2;Better Measures of Exposures;916
7.1.3.3;Better Measures of Health;917
7.1.3.4;Better Measures of the Association;917
7.1.3.5;Between the Social Structure and Health;917
7.1.3.6;Analysing Population Surveys, Birth Cohorts;918
7.1.4;Setting Government Policy Agenda;918
7.1.4.1;The Black and Acheson Reports;919
7.1.4.2;Collating Evidence for Policies Through Intervention;920
7.1.4.3;Studies and Cross National Comparative Studies;920
7.1.5;Conclusions;920
7.1.6;References;921
7.2;III.2 Occupational Epidemiology;927
7.2.1;Introduction;928
7.2.2;Study Designs;929
7.2.2.1;Cross-Sectional Studies;929
7.2.2.2;Cohort Studies;930
7.2.2.3;Case-Control Studies;933
7.2.3;Exposure Assessment;936
7.2.3.1;Statistical and Deterministic Modelling;936
7.2.3.2;Job-exposure Matrices and Job-specific Modules;938
7.2.3.3;Consequences of Errors in Exposure Assessment;939
7.2.4;Special Issues in Occupational Epidemiology;940
7.2.4.1;Confounding;940
7.2.4.2;HealthyWorker Effect;943
7.2.4.3;Dose-Response Analysis;945
7.2.5;Primary Prevention;948
7.2.6;Conclusions;951
7.2.7;References;952
7.3;III.3 Environmental Epidemiology;961
7.3.1;Introduction;962
7.3.1.1;Issues of Environmental Epidemiology;962
7.3.1.2;Concepts of Environmental Epidemiology;964
7.3.1.3;and Toxicology;964
7.3.2;Examples of Research Fields in Environmental Epidemiology;967
7.3.2.1;Outdoor Air Pollution;967
7.3.2.2;Residential Radon;968
7.3.2.3;Non-ionizing Radiation;970
7.3.3;Special Methodological Issues in Environmental Epidemiology;972
7.3.3.1;Principles of Study Design;972
7.3.3.2;Measurements and Exposure Assessment;979
7.3.3.3;Statistical Analysis;986
7.3.4;Conclusions;998
7.3.5;References;1000
7.4;III.4 Nutritional Epidemiology;1009
7.4.1;Introduction;1010
7.4.2;Measurement and Definition of Nutritional Exposures;1011
7.4.2.1;Validation of the Measure of Exposure;1016
7.4.2.2;Measuring Nutritional Exposures in Different Groups;1020
7.4.2.3;Defining Reference Categories and Cut-Points;1021
7.4.3;Methods for Analysis: Adjustment for Confounding by Energy Intake;1023
7.4.3.1;How Many Independent Variables?;1024
7.4.3.2;Macronutrients;1025
7.4.3.3;and Four Methods for Adjusting for Energy Intake;1025
7.4.3.4;Macronutrients: Categorisation Affects the Range;1031
7.4.3.5;of the Relative Risk Estimate;1031
7.4.3.6;Micronutrients;1032
7.4.3.7;Choosing aModel;1032
7.4.4;Organisation and Presentation of Data: Implications forMeta- Analysis and Reviews;1036
7.4.4.1;Choice of Adjustment Model Affects Interpretation;1037
7.4.4.2;of Results in Studies and Meta-Analyses;1037
7.4.5;Nutritional Epidemiology in Public Health Practice;1039
7.4.5.1;Assessing the Usual Intake of a Population;1040
7.4.5.2;References for Assessing Dietary Intake in Populations;1042
7.4.5.3;Impact of Under-Reporting of Intake;1045
7.4.5.4;Consequences ofWithin-Person Variability;1045
7.4.5.5;in Other Areas of Public Health Nutrition Practice;1045
7.4.6;Conclusions;1047
7.4.7;References;1048
7.5;III.5 Reproductive Epidemiology;1053
7.5.1;Introduction;1055
7.5.1.1;Reproductive Epidemiology: Reading Instructions;1055
7.5.1.2;Reproductive Health –;1056
7.5.1.3;Specific Epidemiologic Research Problems;1056
7.5.1.4;Pregnancies as Repeated Events:;1059
7.5.1.5;Problems and Design Options;1059
7.5.2;The Case-Parent-Triad Design;1066
7.5.3;Infertility and Subfecundity;1067
7.5.3.1;Measures Used to Describe Fertility;1067
7.5.3.2;Design Options in Studies of TTP;1069
7.5.4;Twins;1072
7.5.5;Measuring Reproductive Failures;1073
7.5.5.1;TheMeasures Used to Describe Mortality;1075
7.5.6;Foetal and Infant Death;1077
7.5.6.1;Perinatal Mortality and Health Care;1081
7.5.7;Foetal Growth and BirthWeight;1082
7.5.7.1;Optimal BirthWeight;1084
7.5.8;Gestational Age: Pre- and Post Term Delivery;1084
7.5.9;Congenital Malformations;1089
7.5.10;Pregnancy Complications;1092
7.5.10.1;Operational Definition;1092
7.5.10.2;Methodological Challenges;1093
7.5.10.3;Gestational Diabetes;1098
7.5.10.4;Pregnancy-Induced Hypertension and Pre-Eclampsia;1100
7.5.11;Delivery Complications;1105
7.5.12;Foetal Origins of Adult Diseases;1107
7.5.13;Sources of Data;1108
7.5.14;Conclusions;1109
7.5.15;References;1110
7.6;III.6 Molecular Epidemiology;1121
7.6.1;Introduction;1122
7.6.2;Critical Analysis of Molecular Epidemiology Studies;1123
7.6.3;Examples;1133
7.6.3.1;Example 1: Reliability of;1133
7.6.3.2;measurement;1133
7.6.3.3;Example 2: Sources of Heterogeneity;1139
7.6.3.4;for DNA Repair Polymorphisms;1139
7.6.3.5;Example 3. Bulky DNA Adducts in Epidemiological;1142
7.6.3.6;Studies: Principles of Meta-Analysis;1142
7.6.4;Ethical Issues in Biomarker Research;1145
7.6.5;Conclusions;1146
7.6.6;References;1147
7.7;III.7 Genetic Epidemiology;1149
7.7.1;Introduction;1150
7.7.2;Study Types;1151
7.7.3;GeneticModels;1153
7.7.3.1;Terminology;1153
7.7.3.2;Mendelian Single Locus Model;1154
7.7.3.3;Linkage;1155
7.7.3.4;Linkage Disequilibrium;1159
7.7.4;Segregation Analysis;1161
7.7.5;Linkage Analysis;1166
7.7.6;Association Analysis;1171
7.7.7;Conclusions;1175
7.7.8;References;1175
7.8;III.8 Clinical Epidemiology;1179
7.8.1;Introduction;1180
7.8.1.1;Brief History of Clinical Epidemiology;1180
7.8.1.2;A Definition of Clinical Epidemiology;1180
7.8.1.3;Clinical Epidemiology and Evidence-based Medicine;1181
7.8.1.4;The History and Philosophy of Evidence-based Medicine;1182
7.8.2;Case Scenario;1183
7.8.3;Formulating a Clinical Question;1184
7.8.4;Diagnosis;1185
7.8.4.1;Establish the Framework;1186
7.8.4.2;for Bayesian Thinking for Diagnosis;1186
7.8.4.3;Choosing the Right Test;1186
7.8.4.4;Likelihood Ratios;1188
7.8.4.5;How to Obtain Pre-Test Probabilities;1192
7.8.4.6;Sensitivity and Specificity;1192
7.8.5;Therapy/Prevention;1194
7.8.5.1;Absolute Risk;1196
7.8.5.2;Absolute Risk Reduction;1196
7.8.5.3;Relative Risk;1196
7.8.5.4;Relative Risk Reduction;1198
7.8.5.5;Odds Ratio;1198
7.8.5.6;The Number Needed to Treat;1199
7.8.5.7;How Clinicians Can Use Confidence Intervals;1200
7.8.5.8;Use of Composite Endpoints;1202
7.8.6;Systematic Reviews;1203
7.8.7;Guidelines;1206
7.8.7.1;Recommendations;1209
7.8.8;Health RelatedQuality of Life Instruments and Their Application in Clinical Studies;1212
7.8.8.1;Generic Instruments;1212
7.8.8.2;Health Profiles;1212
7.8.8.3;Specific Instruments;1213
7.8.9;Integrating Patient Preferences in the Decision Making Process and Resolution of the Clinical Scenario;1215
7.8.9.1;Patient as Decision-Maker: Decision Aids;1218
7.8.9.2;Patient as Provider of Values;1219
7.8.9.3;Likelihood of Help Versus Harm;1222
7.8.9.4;Semistructured Conversation and Resolution;1223
7.8.10;Conclusions;1224
7.8.11;References;1224
7.9;III.9 Pharmacoepidemiology;1235
7.9.1;Introduction;1236
7.9.2;Limitations of Premarketing Clinical Trials;1237
7.9.3;Characteristics of Spontaneous Reporting Systems;1239
7.9.3.1;Description;1239
7.9.3.2;Limitations;1240
7.9.3.3;Statistical Approaches:;1242
7.9.3.4;Reporting Rates and Proportional Reporting Ratio;1242
7.9.4;Sources of Data in Pharmacoepidemiological Research;1243
7.9.4.1;Multipurpose Cohorts;1244
7.9.4.2;Record Linkage Studies;1244
7.9.4.3;Advantages and Limitations;1248
7.9.5;Methodological Approaches for Pharmacoepidemiology Studies;1251
7.9.5.1;Case-Control Studies;1252
7.9.5.2;Cohort Studies;1253
7.9.5.3;Nested Case-Control Studies;1254
7.9.5.4;Case-Crossover Design;1256
7.9.5.5;Case-Time-Control Designs;1257
7.9.6;Some Methodological Challenges;1258
7.9.6.1;Immortal Time Bias in Cohort Studies;1258
7.9.6.2;Confounding by Indication;1260
7.9.6.3;Depletion of Susceptibles;1261
7.9.6.4;Use of Propensity Scores in Pharmacoepidemiology;1262
7.9.7;Drug Utilization Studies;1264
7.9.8;Conclusions;1267
7.9.9;References;1268
7.10;III.10 Screening;1277
7.10.1;Introduction;1278
7.10.1.1;General Principles of Screening;1278
7.10.1.2;Definition;1278
7.10.2;General Principles Governing the Introduction of Screening;1279
7.10.2.1;The Validity of a Screening Test;1283
7.10.2.2;The Acceptability of the Test;1289
7.10.3;The Ethics of Screening;1290
7.10.4;The Population to be Included in Screening Programmes;1291
7.10.5;Diagnosis and Treatment of the Discovered Lesions;1293
7.10.6;Evaluation of the Efficacy of Screening;1294
7.10.7;Organised Screening Programmes;1299
7.10.8;Health Related Quality of Life and Screening;1300
7.10.9;Economics of Screening;1301
7.10.10;Genetic Susceptibility Testing;1302
7.10.11;Surveillance;1302
7.10.12;Responsibility for Screening;1304
7.10.13;Evaluation of the Effectiveness of Screening Programmes;1304
7.10.14;Information Systems for Screening Programmes;1306
7.10.14.1;Goals and Objectives of Information Systems;1307
7.10.14.2;Information System Design;1308
7.10.14.3;Policy Issues and Data Use;1308
7.10.14.4;Components of Information Systems;1309
7.10.14.5;Conclusion on Information Systems;1309
7.10.15;Conclusions;1309
7.10.16;References;1310
7.11;III.11 Community-based Health Promotion;1315
7.11.1;Introduction;1316
7.11.2;Advantages of a Total Community Approach;1317
7.11.3;Underlying Theories;1318
7.11.3.1;Community Organizing Theory;1318
7.11.3.2;The Health Communication-Behavior Change Theory;1318
7.11.4;Methods;1319
7.11.4.1;Initial Steps;1319
7.11.4.2;Planning;1320
7.11.4.3;Implementation;1320
7.11.4.4;The Amount of Intervention Needed;1322
7.11.4.5;Evaluation;1323
7.11.4.6;Additional Methods Needed;1323
7.11.5;History of Comprehensive Community Health Promotion;1324
7.11.5.1;The First Two Examples: the First Decade;1324
7.11.5.2;Early International Diffusion, 1977–1983;1325
7.11.5.3;The Second and Third Decades:;1326
7.11.5.4;Projects Begun in the 1980s and 1990s;1326
7.11.5.5;Community Projects in Other Health Topics;1327
7.11.6;Past Experience Leads to a “Master Plan”;1328
7.11.7;The Cultural Basis for Barriers to Success;1329
7.11.8;Conclusions;1330
7.11.9;References;1331
8;Part IV Research Areas in Epidemiology;1335
8.1;IV.1 Infectious Disease Epidemiology;1337
8.1.1;Introduction;1339
8.1.1.1;The Global Burden of Infectious Diseases;1339
8.1.1.2;The Importance of Infectious Disease Epidemiology;1340
8.1.1.3;for Prevention;1340
8.1.1.4;The Changing Picture;1341
8.1.1.5;of Infectious Disease Epidemiology;1341
8.1.2;New Approaches in Infectious Disease Epidemiology;1343
8.1.2.1;Improved Laboratory Methods;1343
8.1.2.2;Mapping as an Epidemiological Tool;1344
8.1.2.3;Computer Reporting and Software Progress;1345
8.1.3;What Are the Questions to Be Answered?;1345
8.1.4;Surveillance Issues;1346
8.1.4.1;Passive Surveillance;1346
8.1.4.2;Active Surveillance;1346
8.1.4.3;Case Register;1348
8.1.4.4;Sentinel Disease Surveillance;1349
8.1.4.5;Evaluation of a Surveillance System;1349
8.1.4.6;Elements of a Surveillance System;1350
8.1.5;Outbreak Investigations;1356
8.1.5.1;Basic Steps in Outbreak Investigations;1358
8.1.5.2;Types of Outbreaks;1364
8.1.6;Surveys;1365
8.1.6.1;Survey Methods;1366
8.1.6.2;Sampling;1366
8.1.6.3;Community Surveys (House to House Surveys);1366
8.1.7;Program Evaluation;1368
8.1.8;Conclusions;1370
8.1.9;References;1370
8.2;IV.2 Cardiovascular Diseases;1373
8.2.1;Introduction;1374
8.2.2;Scope and Basic Concepts;1374
8.2.2.1;Atherosclerotic and Hypertensive Diseases;1374
8.2.2.2;Cardiovascular Diseases in Epidemiologic Perspective;1375
8.2.2.3;Epidemiologic Methods in Cardiovascular Diseases;1378
8.2.3;TheMajor Atherosclerotic and Hypertensive Diseases: an Epidemiologic Description;1380
8.2.3.1;Overview;1380
8.2.3.2;Illustrations;1381
8.2.4;Risk Factors and Determinants;1388
8.2.4.1;Overview;1388
8.2.4.2;Illustrations;1390
8.2.5;Causation and Prevention;1398
8.2.5.1;Overview;1398
8.2.5.2;Illustrations;1399
8.2.6;Conclusions;1407
8.2.7;References;1410
8.3;IV.3 Cancer Epidemiology;1415
8.3.1;Introduction;1416
8.3.2;Scope and Approaches in Cancer Epidemiology;1416
8.3.3;The Global Burden of Cancer;1418
8.3.3.1;Tobacco Smoking;1422
8.3.3.2;Use of Smokeless Tobacco Products;1424
8.3.3.3;Dietary Factors;1424
8.3.3.4;Overweight and Obesity;1428
8.3.3.5;Physical Activity;1428
8.3.3.6;Alcohol Drinking;1429
8.3.3.7;Infectious Agents;1430
8.3.3.8;Occupational Exposures;1432
8.3.3.9;Environmental Agents;1434
8.3.3.10;Reproductive Factors;1436
8.3.3.11;Other Lifestyle Factors;1436
8.3.3.12;Hormones;1436
8.3.3.13;Perinatal Factors;1437
8.3.3.14;Ionizing and Non-ionizing Radiation;1437
8.3.3.15;Medical Procedures and Drugs;1438
8.3.3.16;Medical Conditions;1439
8.3.3.17;Genetic Factors;1439
8.3.4;Screening for Cancer;1441
8.3.5;Conclusions;1442
8.3.6;References;1445
8.4;IV.4 Musculoskeletal Disorders;1453
8.4.1;Introduction;1454
8.4.2;Occurrence and Risk Factors;1454
8.4.2.1;Back Disorders;1454
8.4.2.2;Neck Disorders;1459
8.4.2.3;Osteoarthritis;1460
8.4.2.4;Upper Limb Disorders;1465
8.4.2.5;Osteoporosis;1466
8.4.3;Methodological Problems in Epidemiological Research;1468
8.4.3.1;Aspects of Study Design;1468
8.4.3.2;Health Outcome Assessment;1469
8.4.3.3;Exposure Assessment;1472
8.4.4;Conclusions;1474
8.4.5;References;1475
8.5;IV.5 Health Services Research;1483
8.5.1;Introduction;1485
8.5.1.1;Health Services Research Defined;1485
8.5.1.2;The Input–Output Model of Health Care;1486
8.5.1.3;Level of Analysis;1487
8.5.2;Methodological Considerations;1488
8.5.2.1;Study Designs;1489
8.5.2.2;ComplexModels for Data Analysis;1492
8.5.2.3;Data Sources;1492
8.5.2.4;Measurement Error (Misclassification);1496
8.5.2.5;Sampling Issues;1497
8.5.2.6;Confounding and Risk Adjustment;1500
8.5.3;Demand, Need, Utilisation, and Access to Health Care;1501
8.5.3.1;Assessing Health Needs;1502
8.5.3.2;Assessing Utilisation and Access to Services;1503
8.5.4;Financial Resources, Structure, and Organisation of Health Services;1505
8.5.4.1;Allocation of Resources;1509
8.5.4.2;Evaluating Effects;1517
8.5.4.3;of Organisational Characteristics and Change;1517
8.5.5;Process of Health Care: Effectiveness, Appropriateness and Quality;1519
8.5.5.1;Assessing Effectiveness and Appropriateness of Care;1521
8.5.5.2;Assessing Quality of Care: Clinical Practice Performance;1522
8.5.5.3;Examples for Performance Assessment;1525
8.5.6;Outcomes of Health Care;1529
8.5.6.1;Assessing Output and Outcomes of Care;1529
8.5.6.2;Assessing Efficiency of Care;1535
8.5.6.3;Assessing the Outcome of Health Systems;1536
8.5.7;Conclusions;1540
8.5.8;References;1540
8.6;IV.6 Epidemiology in Developing Countries;1555
8.6.1;Introduction;1556
8.6.2;General;1557
8.6.3;The Framework: Health in Developing Countries;1557
8.6.4;Epidemiologic Needs of Developing Countries;1559
8.6.5;Health Information Systems in Developing Countries;1562
8.6.6;Epidemiologic Insights from Health Information Systems;1567
8.6.7;Epidemiologic Sample Surveys;1569
8.6.8;Epidemiologic Studies;1572
8.6.9;Conclusions;1574
8.6.10;The Example India;1575
8.6.11;The Situation in India;1575
8.6.12;Sources of Data;1578
8.6.12.1;Demographic Statistics;1578
8.6.12.2;Health Data;1582
8.6.12.3;Epidemiologic Surveillance System;1586
8.6.13;Tuberculosis Epidemiology;1587
8.6.13.1;Survey Tools for Tuberculosis Infection and Disease;1588
8.6.13.2;Calculating Tuberculosis Burden;1589
8.6.13.3;Evaluation of Epidemiological Changes Through Time;1592
8.6.13.4;Drug Resistance and HIV-TB Co-infection Studies;1593
8.6.14;Conclusions;1594
8.6.15;References;1595
8.6.15.1;References to A;1595
8.6.15.2;References to B;1596
8.7;IV.7 Ethical Aspects of Epidemiological Research;1601
8.7.1;Introduction;1602
8.7.2;Drivers of Awareness of Ethical Aspects in Epidemiology;1603
8.7.2.1;Surge of Automated Databases;1603
8.7.2.2;Integrity and Conflict of Interest;1605
8.7.2.3;in Epidemiological Research;1605
8.7.2.4;Molecular Epidemiology and Genetics;1608
8.7.3;Ethical Principles: Weighing Ethical ‘Benefits’ and ‘Costs’;1610
8.7.3.1;Autonomy;1611
8.7.3.2;Beneficence and Non-maleficence;1611
8.7.3.3;Justice;1612
8.7.3.4;Balancing the Four Principles;1612
8.7.4;Ethical Issues Specific to Epidemiology;1616
8.7.4.1;Unethical Quality of Research;1616
8.7.4.2;Global Bioethics and Inequity;1616
8.7.4.3;Epidemiological Determinism and Preventive Medicine;1617
8.7.4.4;Precautionary Principle and Scientific Evidence;1617
8.7.4.5;Medical Ethics and Epidemiology;1618
8.7.5;Conclusions;1619
8.7.6;References;1620
9;List of Contributors;1623

Leseproben

(p. 558-562)

II. 1 Sample Size Determination in Epidemiologic Studies

Introduction

When planning a research project an epidemiologist must consider how many subjects should be studied.While factors such as available budget certainly present constraints on the maximumnumber of subjects that might actually be included in a study, statistical considerations are extremelyimportant.Toaddress the statistical questions about appropriate sample size, the researchermust .rst specify the study design, the nature of the outcome variable, the aims of the study, the planned analysis method, and the expected results of the study. Is the goal of the study to distinguish between hypotheses about the value of a parameter or function of parameters, or is the goal to provide a con.dence interval estimate of a parameter such as the odds ratio or relative risk? This chapter is organized as follows.

We introduce the issue of how to choose sample size for estimation of a parameter or for a hypothesis test regarding a parameter in the context of one-sample studies in which it is desired to estimate or test a population proportion. We continue on to two-sample studies involving comparisons between two proportions, and one and two-sample studies involving estimation or testing of population means.We conclude with a section on sample size for logistic regression. In this chapter we will provide a brief introduction to power and sample size computation and only address sample size issues for a few of the procedures that are most commonly used in epidemiologic research.

However, we do hope that the reader will gain a sense for what one can accomplish by planning a study with appropriate attention to sample size considerations. A focus on sample size considerations when the study is .rst being planned is critical for the ultimate likelihood that a study proposal is accepted for funding and that the .nal manuscript will be accepted for publication. To ignore the issue of sample size would greatly increase the likelihood of embarking on a costly and time-consuming epidemiologic studywith little likelihood of .nding any de.nitive results.

One Group Designs, Inferences About Proportions

The simplest study design is one in which interest focuses on results for a single group. One is often interested in making inferences about the value of a population proportion. In this section we will illustrate how to choose sample size for the following examples:

. A district medical of.cer seeks to estimate the proportion of children in the district receiving appropriate childhood vaccinations. Assuming a simple randomsample is to be selected froma community, how many children must be studied if the resulting estimate is to fall within 10 percentage points of the true proportion with 95% con.dence?

. Consider the information given in Example 1, only this time we will determine the sample size necessary to estimate the proportion vaccinated in the population to within 10% (not 10 percentage points) of the true value. ,

. During a virulent outbreak of neonatal tetanus, health workers wish to determine whether the rate is decreasing after a period during which it had risen to a level of 150 cases per thousand live births. What sample size is necessary to test the null hypothesis that the population proportion is 0.15 at the 0.05 level if it is desired to have a 90% probability of detecting a decrease to a rate of 100 per thousand if that were the true proportion?