Statistics and Research Methods for Acute Care and General Surgeons

Description

The main aim of this book is to offer an easy tool to read a scientific article with greater awareness, to understand and evaluate it more thoroughly, and to better plan research. Today, in the era of evidence-based medicine, both research and daily patient-focused clinical practice are no longer possible without a thorough knowledge of the literature and its continuous updates. Written by surgeons for surgeons, this practical book makes the basic concept of statistics and research methodology easy to understand and apply for young surgeons and researchers, students and residents.   Preface: Why a Statistics Manual in the Series of “Hot Topics in Acute Care Surgery”? Contents Part I: Designing Your Research 1: Study Typology: An Overview 1.1 Introduction 1.2 The Need for Evidence-Based Medicine 1.3 Research Studies 1.4 Primary and Secondary Research Studies 1.4.1 Primary Studies 1.4.1.1 Laboratorial Research 1.4.1.2 Clinical Studies Clinical Observational Studies Case Reports and Case Series Clinical Experimental Studies 1.4.1.3 Epidemiological Research Cross-Sectional Studies Case–Control Studies Cohort Studies Ecological Studies 1.4.2 Secondary Studies 1.4.2.1 Narrative Reviews 1.4.2.2 Systematic Reviews and Meta-Analysis References 2: Diagnostic Studies Made Easy 2.1 Introduction 2.2 Nature of a Diagnostic Study 2.3 The Need for a Gold Standard 2.4 Components of Diagnostic Studies 2.5 Predictive Values 2.6 Prior Probability of the Disease (Prevalence) 2.7 The Likelihood Ratio (LR) 2.8 Receiver Operating Characteristics (ROC) Curves 2.8.1 Choosing a Cut-off Point: The Youden Index 2.9 Common Errors Encountered in Submitted Diagnostic Studies Further Reading 3: Common Pitfalls in Research Design and Its Reporting 3.1 Introduction 3.2 Unclear Research Question 3.3 Lack of Planning (Failing to Plan Is Planning to Fail) 3.4 Using the Wrong Research Tool 3.5 Selecting the Wrong Population 3.6 Addressing the Missing Data 3.7 Correlation and Prediction 3.7.1 Statistical and Clinical Significance 3.8 Reporting of the Data References Part II: Basic Statistical Analysis 4: Introduction to Statistical Method 4.1 Introduction 4.2 The Hypothesis 4.3 The Aim 4.4 The Errors 4.4.1 Type I Error 4.4.2 Type I Error Rate 4.4.3 Type II Error 4.4.4 Statistical Power 4.4.5 Type II Error Rate 4.4.6 Trade-Off between Type I and Type II Errors 4.4.7 Is a Type I or Type II Error Worse? 4.5 Sample Size Calculation 4.6 The P Value 4.6.1 Results and Interpretation 4.7 Bias 4.7.1 Selection Bias 4.7.2 Classification Bias 4.7.3 Confounding Bias 4.7.4 Other Types of Bias References 5: Analyzing Continuous Variables: Descriptive Statistics, Dispersion and Comparison 5.1 Introduction 5.2 Qualitative Variables 5.3 Quantitative Variables 5.3.1 Discrete Variables 5.3.2 Continuous Variables 5.4 Describing Data 5.4.1 Data Distribution 5.4.2 Test for Normality Assessment 5.4.3 Descriptive Measures 5.4.4 Dispersion Measure 5.4.5 Graphical Representations 5.5 Data Comparison: It Is All About Probability 5.5.1 Paired Data vs. Independent Data 5.5.2 Parametric vs. Non-Parametric Statistics 5.5.3 Commonest Tests 5.6 Linear Correlation 5.6.1 Pearson Correlation 5.6.2 Pearson Coefficient Interpretation 5.6.3 Spearman’s Rank Correlation Coefficient Further Reading 6: Analyzing Categorical Variable: Descriptive Statistics and Comparisons 6.1 Introduction 6.2 Confidence Interval of Proportions 6.3 Absolute Risk Reduction and Number Needed-to-Treat 6.4 Relative Risk and Relative Risk Reduction 6.5 Odds Ratio 6.6 Chi-Squared Test and Fisher’s Exact Test 6.7 Matched Data 6.8 Chi-Squared Test for Trend 6.9 Standardized Differences References Part III: Advanced Statistics 7: Multivariate Analysis 7.1 Introduction 7.1.1 Statistical Models 7.1.2 Different Types of Regressions and Multiple Regression 7.1.3 Example: The Dataset 7.2 Linear Regression Models 7.2.1 Building a Linear Regression Model: It All Comes Down to the Straight Line Equation 7.2.2 Interpreting the Linear Regression Model 7.2.2.1 Interpreting the Parameters of the Model 7.2.2.2 Interpreting 95% Confidence Intervals 7.2.3 Assumptions of Linear Regression 7.3 Multiple Regression 7.3.1 Choice of Predictors 7.3.1.1 Example: Inclusion of Predictors for Multivariable Analysis 7.3.2 Adjustment for Confounders 7.4 Logistic Regression Models 7.4.1 The Logistic Regression Model 7.4.2 Interpreting the Logistic Regression Model: An Example Further Reading 8: Survival Analysis 8.1 Generalities About Time-to-Event Data 8.2 The Variables We Need to Make Analysis: Event and Time 8.3 The Survival Curve and Life Tables: The Kaplan–Meier Method 8.4 Comparing Survival Curves: The Log-Rank Test 8.5 A Regression Model to Assess the Association of Multiple Predictors with a Survival Outcome: The Cox “Proportional Hazards” Model 9: Meta-Analysis 9.1 Introduction 9.2 The Question 9.3 Systematic Review of the Literature 9.4 Meta-analysis Appropriateness: Study Inclusion 9.5 Study Quality Assessment and the Risk of Bias 9.6 Results: Effect Measure 9.6.1 Binary Outcomes/Dichotomous Data 9.6.2 Continuous Data (Also Scale Data or Counts of Events) 9.7 Results: The Forest Plot 9.8 Results: Heterogeneity 9.9 Interpretation of the Results 9.10 Sensitivity Analysis 9.11 Common Mistakes Encountered in Submitted Systematic Review Manuscripts 9.12 Conclusions References 10: Randomized Trials and Case–Control Matching Techniques 10.1 Introduction to Randomized Trials 10.1.1 Ethical Concerns Are Also Related to RCTs 10.1.2 Placebo Effect 10.2 Hypothesis Testing and Sample Size Calculation 10.3 Reporting the Trials 10.4 Randomized Controlled Trials Designs and Techniques 10.5 Strengths of Randomized Trials 10.6 Limitation of Randomized Trials 10.7 Case–Control Studies and Case–Control Matching Techniques 10.8 Propensity Score and Inverse Probability References 11: Difference-In-Difference Techniques and Causal Inference References 12: Machine Learning Techniques 12.1 Machine Learning and Artificial Intelligence 12.2 Machine Learning Terminologies and Concepts 12.2.1 Algorithms, Models, Inputs, and Outputs 12.2.2 Dimensions 12.2.3 Overfitting vs. Underfitting 12.2.4 Bias vs. Variance Tradeoff 12.2.5 Model Flexibility 12.2.6 Feature Selection and Dimension Reduction 12.2.7 Performance Metrics 12.2.8 Training, Validation, and Test Sets 12.2.8.1 Cross-Validation 12.3 Evolution of a Family of Machine Learning Algorithms: From Linear Models to Deep Learning 12.3.1 Supervised Learning 12.3.2 Logistic and Linear Regression 12.3.3 Generalized Additive Models 12.3.4 Deep Learning References 13: Statistical Editor’s Practical Advice for Data Analysis 13.1 Introduction 13.2 What Is the Objective of the Analysis? 13.3 What Is the Type of Data? 13.4 Are the Data Normally Distributed? 13.5 How Many Groups Are Compared? 13.6 What Is the Number of Subjects in Each Group? 13.7 Are the Compared Data Related or Unrelated? 13.8 Which Test to Use? References Further Reading