Epidemiology and biostatistics : an introduction to clinical research / Bryan Kestenbaum ; editors, Kathryn L. Adeney, Noel S. Weiss ; contributing author, Abigail B. Shoben.

Format:

Book

Author/Creator:

Kestenbaum, Bryan.

Contributor:

Adeney, Kathryn L.

Weiss, Noel S., 1941-

Shoben, Abigail B.

Language:

English

Subjects (All):

Biometry.

Epidemiologic Methods.

Biostatistics.

Medical Subjects:

Epidemiologic Methods.

Biostatistics.

Physical Description:

xiii, 242 pages : illustrations ; 24 cm

Place of Publication:

Dordrecht ; New York : Springer, [2009]

Summary:

This text provides students and other health professionals with the knowledge necessary to interpret clinical research articles, design clinical studies, and learn essential epidemiological concepts in an expedient and concise manner. Fundamental concepts are presented in a highly succinct format suitable for students with no previous background in epidemiology or statistics.

Drawing on teaching experience and student feedback, the author has created a text that attempts to recreate the perspective of learning epidemiology and biostatistics for the first time. The text serves as a rapid, intensive course in clinical research methods that can be used by students taking the required epidemiology course, residency and fellowship programs for physicians entering the clinical research portion of their training, and clinical research programs in fields such as nursing and nutrition that need an intensive course in clinical research design.

Contents:

1 Measures of Disease Frequency 3

1.1 Importance of Measures of Disease Frequency 5

1.2 Prevalence 5

1.3 Incidence 6

1.4 Relationship Between Prevalence and Incidence 9

1.5 Stratification of Disease Frequency by Person, Place, and Time 9

1.5.1 Disease Frequency Measurements Stratified by Characteristics of Person 10

1.5.2 Disease Frequency Measurements Stratified by Characteristics of Place 10

1.5.3 Disease Frequency Measurements Stratified by Characteristics of Time 11

1.5.4 Disease Frequency Measurements To Complement Experimental Data 11

2 General Considerations in Clinical Research Design 13

2.1 Study Population 14

2.1.1 Definition of the Study Population 14

2.1.2 Choice of Study Population and Generalizability of Study Findings 15

2.1.3 Where to Find Information About the Study Population in a Clinical Research Article 16

2.2 Exposure and Outcome 17

2.2.1 Definition 17

2.2.2 Specifying and Measuring the Exposure and Outcome 18

2.2.3 Where to Find Exposure and Outcome Data in a Clinical Research Article 18

2.3 Interventional Versus Observational Study Designs 19

2.4 Inferring Causation from Association Studies 21

2.4.1 Importance of Distinguishing Causation from Association 21

2.4.2 Factors Favoring an Inference of Causation 22

3 Case Reports and Case Series 25

4 Cross-Sectional Studies 29

5 Cohort Studies 33

5.1 Overview of Cohort Study Design 33

5.2 Ascertainment of Study Data 35

5.2.1 Validity of Measurements 35

5.2.2 Timing of Measurements 36

5.2.3 Uniform Measurements 37

5.2.4 Retrospective Versus Prospective Data Collection 37

5.3 Advantages of Cohort Studies 38

5.3.1 Study of Multiple Outcomes 38

5.3.2 Ability to Discern Temporal Relationship Between Exposure and Outcome 38

5.4 Disadvantages of Cohort Studies 39

5.4.1 Confounding 39

5.4.2 Inability to examine Diseases That Are Rare or Have a Long Latency 39

5.5 Cohort Studies for Evaluating Medication Use 40

5.6 Analysis of Data From Cohort Studies 41

5.6.1 Incidence Proportion Versus Incidence Rate 41

5.6.2 Relative Risk 42

5.6.3 Attributable Risk (also Called "Risk Difference" or "Excess Risk") 44

6 Case-Control Studies 45

6.1 Case-Control Study Design 47

6.1.1 Overview 47

6.1.2 Selection of Cases 48

6.1.3 Selection of Controls 49

6.2 Advantages of Case-Control Studies 51

6.2.1 Case Control Studies Can Be Ideal for the Study of Rare Diseases or Those with a Long Latency 51

6.2.2 Case-Control Studies Allow for the Study of Multiple Exposures 51

6.3 Disadvantages of Case-Control Studies 52

6.3.1 Observational Study Design 52

6.3.2 Recall Bias 52

6.3.3 Case Control Studies only Provide Information Regarding the Relative Risk (Odds) of Disease 53

6.4 Analysis of Case-Control Data 53

6.4.1 Theory of the Odds Ratio 53

6.4.2 Practical Calculation of the Odds Ratio 55

6.4.3 Odds Ratios and Relative Risk 55

6.4.4 Case-Control Studies Cannot Estimate the Actual Incidence of a Disease or Outcome 56

7 Randomized Trials 59

7.1 Rationale for Randomized Trials 59

7.1.1 Kidney Transplant and Mortality 60

7.1.2 Angioplasty versus Fibrinolysis for Patients with Acute Myocardial Infarction 60

7.1.3 Equipoise 61

7.2 Phases of Drug Development 61

7.2.1 Phase I Studies 62

7.2.2 Phase II Studies 62

7.2.3 Phase III/IV Studies 62

7.3 Conduct of Randomized Trials 62

7.3.1 Comparison Group 62

7.3.2 Placebo 63

7.3.3 Block Randomization 64

7.3.4 Biological Versus Clinical Endpoints 65

7.4 Limitations of Randomized Controlled Trials 65

7.4.1 Generalizability of the Study Population 65

7.4.2 Generalizability of the Study Environment 66

7.4.3 Limited Question 67

7.4.4 Limited Clinical Applicability 67

7.4.5 Randomized Design Accounts only for Confounding 68

7.5 Analysis of Randomized Controlled Trial Data 68

7.5.1 Measures of Effect 68

7.5.2 Numbers Needed to Treat/Harm 69

7.5.3 Measures of Effect in Journal Articles 69

7.5.4 Intention-to Treat-Analysis 70

7.5.5 Subgroup Analyses 71

8 Misclassification 75

8.1 Definition of Misclassification 75

8.2 Nondifferential Misclassification 76

8.2.1 Example of Nondifferential Misclassification of the Exposure 76

8.2.2 Definition and Impact of Nondifferential Misclassification of the Exposure 78

8.2.3 Nondifferential Misclassification of the Outcome 81

8.2.4 Definition and Impact of Nondifferential Misclassification of the Outcome 84

8.3 Differential Misclassification 84

8.4 Assessment of Misclassification in Clinical Research Articles 89

9 Introduction to Confounding 91

9.1 Confounding and the Interpretation of Clinical Data 91

9.2 Formal Evaluation of a Potential Confounding Factor 94

9.2.1 Evaluation of a Confounder: Association with Exposure 95

9.2.2 Evaluation of a Confounder: Association with Outcome 95

9.2.3 Evaluation of a Confounder: Not in the Causal Pathway of Association 96

9.2.4 Other Examples of Factors That Reside on the Causal Pathway of Association 98

9.3 Scientifically Meaningful Versus Statistical Associations 98

9.4 Evaluation of a Confounder in Clinical Research Articles 99

9.5 Confounding-by-Indication 100

10 Methods to Control for Confounding 101

10.1 Restriction 102

10.1.1 Method of Restriction 102

10.1.2 Pros and Cons of Restriction as a Means to Control for Confounding 102

10.1.3 Restriction to Control for Confounding-by-Indication 103

10.2 Stratification 103

10.2.1 Method of Stratification 103

10.2.2 Pros and Cons of Stratification as a Means to Control for Confounding 105

10.2.3 Stratum-Specific Associations 105

10.3 Matching 106

10.3.1 Method of Matching 106

10.3.2 Pros and Cons of Matching as a Means to Control Confounding 107

10.4 Regression 108

10.5 Randomization 108

10.6 Interpreting Study Results After Adjustment for Confounding 109

10.7 Unadjusted Versus Adjusted Associations: Confounding 109

10.8 Confounding: An Advanced Example 110

11 Effect Modification 113

11.1 Concept of Effect Modification 113

11.2 Synergy Between Exposure variables 114

11.3 Effect Modification Versus Confounding 115

11.4 Evaluation of Effect Modification 116

11.4.1 Epidemiologic Evaluation of Effect Modification 116

11.4.2 Statistical Evaluation of Effect Modification 116

11.5 Effect Modification in Clinical Research Articles 117

11.6 Effect Modification on the Relative and Absolute Scales 118

12 Screening 121

12.1 General Principles of Screening 122

12.2 Qualities of Diseases Appropriate for Screening 122

12.2.1 The Disease should be Important in the Screened Population 122

12.2.2 Early Recognition and Treatment of the Disease Should Prevent Clinical Outcomes 123

12.2.3 The Disease Should have a Preclinical Phase 123

12.3 Qualities of Screening Tests 123

12.3.1 General Qualities 123

12.3.2 Reliability and Validity 123

12.4 Validity of Screening Tests 124

12.4.1 Sensitivity and Specificity 124

12.4.2 Positive and Negative Predictive Value 125

12.4.3 Screening Tests with Continuous Values 129

12.5 Reliability of Screening Tests 132

12.5.1 Variation in Measurement Tools and Within and Individual 132

12.5.2 Measures of Reliability 133

12.6 Types of Bias in Screening Studies 134

12.6.1 Referral Bias 134

12.6.2 Lead Time Bias 135

12.6.3 Length Bias Sampling 136

12.6.4 Overdiagnosis Bias 137

12.7 Association versus Prediction 137

13 Diagnostic Testing 139

13.1 General Considerations in Medical Testing 139

13.2 Likelihood Ratios 143

Biostatistics

14 Summary Measures in Statistics 153

14.1 Types of Variables 153

14.2 Univariate Statistics 154

14.2.1 Histograms 154

14.2.2 Measures of Location and Spread 156

14.2.3 Quantiles 158

14.2.4 Univariate Statistics for Binary Data 159

14.3 Bivariate

Statistics 159

14.3.1 Tabulation Across Categories 159

14.3.2 Correlation 160

14.3.3 Quantile-Continuous Variable Plots 162

15 Introduction to Statistical Inference 163

15.1 Definition of a Population, Sample, and random Sample 163

15.2 Statistical Inference 164

15.3 Generalizability 165

15.4 Confidence Intervals 165

15.5 P-values 168

15.6 Confidence Intervals and p-values in Clinical Research 169

16 Hypothesis Testing 171

16.1 Study Hypothesis and Null Hypothesis 172

16.2 Distribution of Sampling Means 173

16.3 Properties of the Distribution of Sampling Means 174

16.3.1 Normal (Bell-Shaped) Distribution for Reasonably Large Sample Sizes 174

16.3.2 Mean Equal to the Population Mean 175

16.3.3 Spread of the Distribution Related to Population Variation and Sample Size 175

16.3.4 Distribution of Sampling Means: Summary 177

16.4 Conducting the Experiment 177

17 Interpreting Hypothesis Tests 181

17.1 Common Tests of Hypothesis in Clinical Research 181

17.1.1 T-Tests 181

17.1.2 Chi-Square Tests 182

17.1.3 ANOVA Tests 182

17.2 An Imperfect System 183

17.2.1 Type I Errors 183

17.2.2 Type II Errors 184

17.2.3 Power 184

18 Linear Regression 189

18.1 Describing the Association Between Two Variables 189

18.2 Univariate Linear Regression 192

18.2.1 The Linear Regression Equation 192

18.2.2 Residuals and the Sum of Squares 193

18.2.3 Absolute Versus Relative Fit 194

18.3 Interpreting Results from Univariate Regression Equations 195

18.3.1 Interpreting Continuous Covariates 195

18.3.2 Interpreting Binary Covariates 195

18.4 Special Considerations 197

18.4.1 Influential Points 197

18.4.2 Nonlinear Associations 198

18.4.3 Extrapolating the Regression Equation Beyond the Study Data 200

18.5 Multiple Linear Regression 200

18.5.1 Definition of the Multivariate Model 200

18.5.2 Interpreting Results from the Multiple Regression Model 201

18.6 Confounding and Effect Modification in Regression Models 204

18.6.1 Confounding 204

18.6.2 Effect Modification 205

19 Non-Linear Regression 209

19.1 Regression for Ratios 209

19.2 Logistic Regression 211

19.3 Application of Logistic Regression Models 213

20 Survival Analysis 215

20.1 Limitations of Incidence Measures for Evaluating Risk 215

20.1.1 Incidence Measures: Oversimplification of Study Results Over time 216

20.1.2 Incidence Measures: Crude Handling of Participant Dropout 216

20.2 Survival Data 217

20.3 Statistical Testing of Survival Data 219

20.4 Definitions of Events and Censoring 220

20.5 Kaplan-Meier Estimation 221

20.5.1 Kaplan-Meier Estimation of S(t) 221

20.5.2 Kaplan-Meier Estimation of S(t) with Censored Data 222

20.6 Cox's Proportional Hazards Model 224

20.6.1 Description of the Proportional Hazards Model 224

20.6.2 Interpreting Survival Data and the Proportional Hazards Model 227

20.6.3 Survival Versus Hazard Ratio Data 228.

Notes:

Includes bibliographical references (pages 229-231) and indexes.

ISBN:

9780387884325

0387884327

OCLC:

401157485