Statistical Approaches for Epidemiology: From Concept to Application

✍ Scribed by Amal K. Mitra (editor)

• Publisher: Springer
• Year: 2023
• Tongue: English
• Leaves: 445
• Category: Library

✦ Synopsis

This textbook provides the basic concepts of epidemiology while preparing readers with the skills of applying statistical tools in real-life situations.

Students, in general, struggle with statistical theories and their practical applications. This book makes statistical concepts easy to understand by focusing on real-life examples, case studies, and exercises. It also provides step-by-step guides for data analysis and interpretation using standard statistical software such as SPSS, SAS, R, Python, and GIS as appropriate, illustrating the concepts.

Through the book's 23 chapters, readers primarily learn how to apply statistical methods in epidemiological studies and problem-solving. Among the topics covered:

• Clinical Trials
• Epidemic Investigation and Control
  
• Geospatial Applications in Epidemiology
  
• Survival Analysis and Applications Using SAS and SPSS
  
• Systematic Review and Meta-Analysis: Evidence-based Decision-Making in Public Health
  
• Missing Data Imputation: A Practical Guide
  
• Artificial Intelligence and Machine Learning
• Multivariate Linear Regression and Logistics Regression Analysis Using SAS
  

Each chapter is written by eminent scientists and experts worldwide, including contributors from institutions in the United States, Canada, Bangladesh, India, Hong Kong, Malaysia, and the Middle East.

Statistical Approaches for Epidemiology: From Concept to Application is an all-in-one book that serves as an essential text for graduate students, faculty, instructors, and researchers in public health and other branches of health sciences, as well as a useful resource for health researchers in industry, public health and health department professionals, health practitioners, and health research organizations and non-governmental organizations. The book also will be helpful for graduate students and faculty in related disciplines such as data science, nursing, social work, environmental health, occupational health, computer science, statistics, and biology.

✦ Table of Contents

Preface
Contents
Contributors
Editors and Contributors
About the Editor
Abbreviations
List of Figures
List of Tables
Chapter 1: Descriptive and Analytical Epidemiology
1 Introduction
2 Person, Place, and Time Model
2.1 Person-Related Factors
2.1.1 Age
2.1.2 Sex/Gender
2.1.3 Race/Ethnicity Differences
2.1.4 Place-Related Factors
2.1.5 Time-Related Factors
3 Descriptive Epidemiological Studies (Box 1.1)
3.1 Case Reports
3.2 Case Series
3.3 Descriptive Cohort (Incidence) Studies
3.4 Cross-Sectional Studies
4 Analytical Studies
4.1 Ecological Studies (Correlational Studies)
4.2 Case–Control Studies
4.3 Cohort Studies
4.3.1 Prospective Cohort Studies
4.3.2 Retrospective Cohort Studies
4.4 Experimental Studies
5 Further Practice
References
Chapter 2: Cross-Sectional Study: The Role of Observation in Epidemiological Studies
1 Introduction
2 Epidemiological Studies: Descriptive Versus Analytical
2.1 Descriptive Studies: Case Reports/Case Series
2.2 Descriptive Studies: Correlational Studies as the First Step Toward Etiological Examination
2.3 Descriptive Studies: Cross-Sectional Studies: What Are They?
2.4 Cross-Sectional Study Design
2.4.1 What Are Cross-Sectional Studies Used For?
2.4.2 Examples of Cross-Sectional Studies
2.5 Analyzing Data from Cross-Sectional Studies
2.6 Advantages and Disadvantages of Cross-Sectional Studies
2.7 The Role of Cross-Sectional Studies in Public Health Research and Public Health Policy (Box 2.2)
2.8 Problems
3 Further Practice
References
Chapter 3: Case–Control Study
1 Introduction
2 Method of Case–Control Studies
2.1 Selection of Cases
2.2 Selection of Controls
2.3 Issues in Selecting Controls
2.4 How Many Controls Suitable for Each Case
2.5 Advantages and Disadvantages of Case–Control Studies
2.6 Nested Case–Control Study
3 Example of Case–Control Studies
4 Calculation of Odds Ratio
4.1 Calculation of 95% Confidence Intervals for OR
5 Further Practice
References
Chapter 4: Cohort Studies
1 Introduction
2 Design of a Cohort Study
2.1 Selection of Study Populations
2.2 A Few Points to Consider for Exposure and Comparison Groups
3 Types of Cohort Studies
3.1 Prospective Cohort Studies
3.2 Retrospective Cohort Studies
3.3 Ambi-directional Cohort Study
4 Advantages and Disadvantages of Cohort Studies
5 Examples of Cohort Studies
5.1 The Framingham Heart Study
5.2 Cardiovascular Health Study
6 Bias in Cohort Studies
6.1 Selection Bias
6.2 Information Bias
7 Measure and Analysis in Cohort Studies
7.1 Incidence Rate
7.2 Calculating Person-Time
7.3 Risk Ratio (Relative Risk)
7.4 Rate Ratio
7.5 Attributable Risk
8 Further Practice
References
Chapter 5: Epidemiological Measures
1 Introduction
2 Crude Numbers and Rates
2.1 Rates, Ratios, and Proportions
3 Measures Used in Vital Statistics
3.1 Birth Data
3.2 Mortality Data
4 Morbidity Measures
5 Useful Online Resources
6 Further Practice: Case Study
References
Chapter 6: Clinical Trials
1 Introduction
2 What Constitutes a Clinical Trial?
3 Study Design
3.1 Superiority, Equivalence, and Noninferiority Trials
3.2 Parallel Group, Crossover, and Factorial Study Designs
4 Sample Size Determination
5 Essential Elements of a Randomized Clinical Trial
5.1 Standardized Study Protocol and Registration of a Study
5.2 Hypothesis and Outcome Measures
5.3 Equipoise Between the Intervention and Comparison Group
5.4 Blinding (Masking) of Study Interventions
5.5 Randomization and Concealment of Allocation
5.6 Methods of Randomization
5.6.1 Community or Cluster Randomization
5.6.2 Defining the Population Enrolled in the Study
6 Trial Organization and Management
6.1 Study Sponsor
6.2 Trial Management Group
6.3 Data Monitoring and Safety Boards (DMSB, or Data Monitoring Committee)
6.4 Trial Steering Committee
7 Data Analysis and Reporting
8 Ethical Considerations and Informed Consent
9 Further Practice
References
Chapter 7: Screening
1 Introduction
2 Criteria for an Effective Screening Test
2.1 Disease Characteristics
2.1.1 High Morbidity and Mortality
2.1.2 Early Detection Helps in Prognosis
2.1.3 Availability of Treatment
2.2 Ideal Characteristics of a Screening Test
2.2.1 Validity and Reliability
2.3 Sensitivity
2.4 Specificity
2.5 Positive Predictive Value
2.6 Negative Predictive Value
2.7 Agreement of the Test
3 Several Other Terms in Relation to Screening
3.1 Total Preclinical Phase (TPCP)
3.2 Detectable Preclinical Phase (DPCP)
3.3 Lead Time
3.4 Lead-Time Bias
3.5 Receiver Operating Characteristic (ROC) Curve
4 Commonly Used Screening Tests
4.1 Mammogram
4.2 MRI of Breast
4.3 Pap Smear
4.4 Colonoscopy
4.5 Low-Dose Computed Tomography (LDCT)
4.6 PSA for Prostate Cancer
4.7 Genetic Screening
5 Evaluating Screening Tests: An Exercise
6 Selecting a Cutoff Point
7 Further Practice
References
Chapter 8: Surveillance: The Role of Observation in Epidemiological Studies
1 Introduction
2 Approaches to Public Health Surveillance
2.1 Coverage
2.2 Intensity
2.3 Standardization
2.4 Analysis and Interpretation
2.5 Dissemination
2.6 Evaluation
3 Common Terms Used in Public Health Surveillance
4 Types of Public Health Surveillance
4.1 Active Surveillance
4.2 Passive Surveillance
4.3 Syndromic Surveillance
4.4 Integrated Surveillance
4.5 Sentinel Surveillance
5 Process of Public Health Surveillance
6 Public Health Surveillance System
References
Chapter 9: Standardization
1 Introduction
2 Standardization
3 Direct Standardization
4 Indirect Standardization
5 Further Practice
References
Chapter 10: Causal Association
1 Introduction
2 Confounding Effect
3 Hill’s Criteria of Causality
3.1 Strength of Association
3.1.1 Evidence from Longitudinal Follow-Up Studies
3.1.2 Evidence from Case-Control Study
3.1.3 Evidence from Cross-Sectional Study
3.2 Temporality
3.3 Consistency
3.3.1 Example of Two Clinical Trials Having Consistency
3.4 Biological Gradient or Dose-Response
3.5 Specificity
3.6 Plausibility
3.7 Coherence
3.8 Experimentation
3.9 Analogy
4 Rothman’s Causal Pie
4.1 Illustration of Rothman’s Pie
5 Further Practice
References
Chapter 11: Bias, Confounding, and Effect Modifier
1 Introduction
2 Sources of Bias
2.1 Selection Bias
2.1.1 Reasons for Selection Bias
2.1.2 Minimizing Selection Bias
3 Information Bias
3.1 Examples of Information Bias in Epidemiological Studies
3.1.1 Recall Bias
3.1.2 Minimizing Information Bias
4 Confounding Bias
4.1 Assessment of Confounding
4.1.1 Randomization
4.1.2 Restriction
4.1.3 Matching
4.1.4 Stratified Analysis
4.1.5 Multiple Regression Analysis
5 Types of Confounding Effect (Positive, Negative, and Qualitative)
6 Effect Modification
7 Further Practice
References
Chapter 12: Epidemic Investigation and Control
1 Overview
2 Useful Terms
2.1 Epidemic
2.2 Outbreak
2.3 Pandemic
2.4 Endemic
2.5 Hyperendemic
2.6 Cluster
2.7 Epizootic
2.8 Attack Rate
2.9 Secondary Attack Rate
2.10 Case Fatality Rate (CFR)
3 Threshold Level of Outbreak (for Epidemic Prone Diseases)
4 Outbreak Trigger Levels for Responses
4.1 Rapid Response Team (RRT)
5 Steps of an Outbreak Investigation
5.1 Prepare for Field Work
5.2 Establish the Existence of an Outbreak
5.3 Verify the Diagnosis
5.4 Construct a Working Case Definition
5.5 Find Cases Systematically and Record Information
5.6 Perform Descriptive Epidemiology
5.6.1 Time
5.6.2 Place
5.6.3 Person
5.6.4 Epidemic Types
5.7 Develop Hypotheses
5.8 Evaluate Hypotheses
5.8.1 Retrospective Cohort Study
5.8.2 Case–Control Study
5.8.3 Source of Bias
5.9 Compare and Reconcile with Laboratory and/or Environmental Studies
5.10 Implement Control and Prevention Measures
5.10.1 Water- and/or Food-Borne Outbreak
5.10.2 Vector-Borne Outbreak
5.10.3 Respiratory Outbreak
5.10.4 Vaccine Preventable Disease Outbreak
5.11 Initiate or Maintain Surveillance
5.12 Disseminate Findings
5.12.1 Background
5.12.2 Previous History of Similar Epidemic
5.12.3 Investigation Methods
5.12.4 Data Analysis and Results
5.12.5 Interpretation of Results
5.12.6 Control Measures
5.12.7 Future Preventive Measures
6 “End” of an Outbreak
6.1 Community-Operated Treatment Centers to Control Diarrheal Epidemic in Rural Bangladesh: A Case Study
7 Conclusion
8 Problem Solving
9 Further Practice
References
Chapter 13: Population Projection
1 Introduction
2 What Is Population Projection? How Do Projections Differ from Estimates?
3 Scope and Use of Population Projection
4 Public Health Importance of Population Projection
5 Measurement and Applications with Examples
5.1 Arithmetic Projection
5.2 Geometric Projection
5.3 Exponential Projection
5.3.1 Shrinking Population
5.4 Cohort-Component Method of Population Projection (CCP) [7]
6 Problem Solving
6.1 Problem 1
6.2 Problem 2
6.3 Problem 3
6.4 Problem 4
7 Further Practice
References
Chapter 14: Geospatial Applications in Epidemiology: Location, Location, Location
1 Introduction
2 The Distinction of Spatial Epidemiology
2.1 The Intersection of Spatial Epidemiology and Geography
2.2 Getting to Know Your Neighbor
3 Testing Spatial Dependence
4 Spatial Visualizations
5 Application
5.1 Determining if Residents of Certain Neighborhoods Are More or Less Likely to Be Exposed to Restaurants with Poor Health Scores
6 Further Practice
References
Chapter 15: Survival Analysis and Applications Using SAS and SPSS
1 Introduction
2 The Time-to-Event Data
2.1 A Real-Life Example
3 Censored Data
3.1 Types of Censoring
4 Terminology and Notation
4.1 The Survival Function
4.2 Hazard Function
4.3 Relationship Between h(t) and S(t)
4.4 Kaplan–Meier Estimator
5 Log-Rank Test
6 Other Survival Analysis Methods
6.1 The Cox Proportional Hazards Model
7 Survival Analysis Using SAS and SPSS
8 Further Practice
References
Chapter 16: Systematic Review and Meta-Analysis: Evidence-Based Decision-Making in Public Health
1 Introduction: Evidence-Based Decision-Making
2 Limitations and Important Considerations for Systematic Review and Meta-Analysis
3 Overview of Step-by-Step Procedures
3.1 Criteria for Inclusion and Exclusion of Studies
3.2 Outcome Measures and Displaying Results
3.3 Validity of Meta-Analysis and Conduct of Sensitivity Analysis
4 Practical Examples and Interpretation of Relevant Studies
4.1 Example 1—Physical Activity Equivalent Labeling Versus Calorie Labeling
4.2 Example 2—BMI Impact on Stroke and All-Cause Mortality
4.3 Risk of Bias Assessment
5 Further Practice
References
Chapter 17: Sample Size Estimation
1 Introduction
2 Steps of Hypothesis Testing
2.1 Hypothesis
2.2 Steps of Hypothesis Testing
3 Estimation of Sample Size
3.1 Sample Size for a Cross-Sectional Study
3.2 Sample Size for a Quantitative Study (Using Mean and Standard Deviation)
3.3 Sample Size for a Qualitative Study (Using Two Proportions)
3.4 Sample Size for a Case–Control Study Using Odds Ratio (OR)
3.5 Sample Size for a Cohort Study Using Relative Risk (RR)
3.6 Sample Size Estimation Using G*Power Software
4 Further Practice
References
Chapter 18: Missing Data Imputation: A Practical Guide
1 Introduction
2 Real-Life Example
2.1 Hepatocellular Carcinoma Data
3 Exploring Missingness
4 Classification of Missing Data
4.1 Missing Completely at Random (MCAR)
4.2 Missing at Random (MAR)
4.3 Missing Not at Random (MNAR)
5 Identifying Missing Data Mechanism
5.1 Testing MCAR
5.2 Testing MAR and MNAR
6 Common Methods for Missing Data Handling
6.1 Listwise Deletion
6.2 Pairwise Deletion
6.3 Mean or Median Imputation
6.4 Regression Imputation
6.5 Hot-Deck Imputation
6.5.1 Random Hot-Deck Imputation
6.5.2 Sequential Hot-Deck Imputation
6.5.3 Predicting Mean Matching Imputation
6.6 K-Nearest Neighbor Imputation
6.7 Multiple Imputations
6.7.1 Multiple Imputations Demonstration
7 Further Practice
References
Chapter 19: Artificial Intelligence and Machine Learning
1 Introduction
2 Machine Learning
2.1 Regression
2.2 Classification
2.3 Clustering
3 Neural Networks and Deep Learning
3.1 NN for Regression
3.2 NN for Classification
3.3 NN for Computer Vision
3.4 NN for NLP
4 Recent Advancements
5 Models’ Interpretability
6 Further Practice
References
Chapter 20: A Step-by-Step Guide to Data Analysis Using SPSS: Iron Study Data
1 Introduction
2 About the Iron Study
2.1 Major Research Questions
2.2 Understanding Variables
2.3 Distribution of Data
2.3.1 Comparing Mean, Median, Mode, and Standard Deviation
2.3.2 Using Histogram (with a Normal Curve), Skewness, and Kurtosis [6]
2.3.3 Boxplot, Stem-and-Leaf, and Q–Q (Quantile–Quantile) Plot
2.4 Shapiro–Wilk Test and Kolmogorov–Smirnov Test of Normality
3 What Tests to Do When: A Practical Guideline
3.1 Comparing Two Means and Independent Groups
3.2 Comparing Two Means, Dependent Groups, or Paired Data
3.3 Compare Three or More Means
3.4 Compare Two or More Qualitative (or Group) Variables
3.5 Association Between Two or More Variables
3.6 Prediction of a Dependent Variable from Multiple Independent Variables
4 Data Analysis Addressing the Research Questions
4.1 Did the Hemoglobin Levels Improve After Treatment in Low-Income Postpartum Women?
4.2 Did the Women in the Three Treatment Groups Differ in Hemoglobin Status After Treatment?
5 Further Practice
References
Chapter 21: Data Analysis Using SPSS: Jackson Heart Study
1 Introduction
2 The Jackson Heart Study
2.1 Variables Used in the Analyses
3 Practical Application of the Analytic Plan Using Jackson Heart Study Data
3.1 Describing Characteristics of Groups Using SPSS [2]
3.1.1 Frequency Tables
3.2 Testing Hypothesis of Difference [3]
3.2.1 Testing Differences Between Two Groups When the Dependent Variable (Outcome) Is a Continuous Variable
3.2.2 Differences Between Three or More Groups When the Dependent Variable (Outcome) Is a Number (or Continuous Variable)
3.3 Testing Hypothesis of Association (Relationship/Correlation) [3]
4 Recap and Review
5 How to Write the Null Hypothesis [4, 5]
6 Further Practice
References
Chapter 22: Multiple Linear Regression and Logistic Regression Analysis Using SAS
1 Introduction
1.1 Linear Regression
1.2 Assumptions for Linear Regression
1.3 Model Fit Assumption Before Applying a Regression Model
1.3.1 Test of Normality: Dependent Variable TR_CRP (True CRP)
1.3.2 Regression Diagnostics [7–9]
Influence Statistics
Multicollinearity
1.3.3 Collinearity Diagnosis and SAS Output
1.3.4 Model Fitness Evaluation
Global F-Test
Hypothesis for Model Fitness
2 Model Selection Criteria for Regression
2.1 SAS Code for Model Selection Procedure [10]
2.2 Partial SAS Output from Model Selection Procedure (Tables 22.7a, 22.7b, and 22.7c)
3 Interaction or Effect Modification
4 Binary Logistic Regression
5 Model Equation for Multiple Logistic Regression
5.1 SAS Code for Multiple Logistic Regression with Interaction
5.2 SAS Code for Binary Logistic Regression
5.3 SAS Code for Logistic Analysis for Multiple Logistic Regression Without Interaction [19]
6 Model Fit Evaluation/Model Fit Statistics
7 SAS Code for ROC
7.1 Model Evaluation for Continuous Variable: Hosmer Lemeshow Test
7.2 SAS Code for Hosmer and Lemeshow Test
8 R Square Statistics
9 Interpretation of Regression Parameter
10 SAS Code for Interaction
11 SAS Code for Trend Analysis
12 SAS Code for Forward, Backward, and Stepwise Logistic Regression
13 Further Practice
References
Chapter 23: Epidemiology of the COVID-19 Pandemic: An Update
1 Introduction
2 Infectiousness, Pathogenicity, and Virulence
3 Risk Factors
4 Complications of COVID-19
4.1 COVID-19 Impact on Mental Health
4.2 Long-COVID or Post-COVID Conditions
4.3 “Brain Fog” After COVID-19 Infection
5 Changing Nature of COVID-19 Due to Mutations and Multiple Variants
6 Assessing the Disease Burden Due to COVID-19
6.1 Use of PYLL: A Case Study
7 Treatment Protocol for COVID-19
8 Vaccines for COVID-19
9 Further Practice
References
Index