Efnisyfirlit

• Cover
• Also available to buy!
• Preface
• Part 1 Handling data
• 1 Types of data
• Data and statistics
• Categorical (qualitative) data
• Numerical (quantitative) data
• Distinguishing between data types
• Derived data
• Censored data
• 2 Data entry
• Formats for data entry
• Planning data entry
• Categorical data
• Numerical data
• Multiple forms per patient
• Problems with dates and times
• Coding missing values
• 3 Error checking and outliers
• Typing errors
• Error checking
• Handling missing data
• Outliers
• References
• 4 Displaying data diagrammatically
• One variable
• Two variables
• Identifying outliers using graphical methods
• The use of connecting lines in diagrams
• 5 Describing data: the ‘average’
• Summarizing data
• The arithmetic mean
• The median
• The mode
• The geometric mean
• The weighted mean
• 6 Describing data: the ‘spread’
• Summarizing data
• The range
• Ranges derived from percentiles
• The standard deviation
• Variation within- and between-subjects
• 7 Theoretical distributions: the Normal distribution
• Understanding probability
• The rules of probability
• Probability distributions: the theory
• The Normal (Gaussian) distribution
• The Standard Normal distribution
• 8 Theoretical distributions: other distributions
• Some words of comfort
• More continuous probability distributions
• Discrete probability distributions
• 9 Transformations
• Why transform?
• How do we transform?
• Typical transformations
• Part 2 Sampling and estimation
• 10 Sampling and sampling distributions
• Why do we sample?
• Obtaining a representative sample
• Point estimates
• Sampling variation
• Sampling distribution of the mean
• Interpreting standard errors
• SD or SEM?
• Sampling distribution of the proportion
• 11 Confidence intervals
• Confidence interval for the mean
• Confidence interval for the proportion
• Interpretation of confidence intervals
• Degrees of freedom
• Bootstrapping and jackknifing
• Reference
• Part 3 Study design
• 12 Study design I
• Experimental or observational studies
• Defining the unit of observation
• Multicentre studies
• Assessing causality
• Cross-sectional or longitudinal studies
• Controls
• Bias
• Reference
• 13 Study design II
• Variation
• Replication
• Sample size
• Particular study designs
• Choosing an appropriate study endpoint
• References
• 14 Clinical trials
• Treatment comparisons
• Primary and secondary endpoints
• Subgroup analyses
• Treatment allocation
• Sequential trials
• Blinding or masking
• Patient issues
• The protocol
• References
• 15 Cohort studies
• Selection of cohorts
• Follow-up of individuals
• Information on outcomes and exposures
• Analysis of cohort studies
• Advantages of cohort studies
• Disadvantages of cohort studies
• Study management
• Clinical cohorts
• 16 Case–control studies
• Selection of cases
• Selection of controls
• Identification of risk factors
• Matching
• Analysis of unmatched or group-matched case–control studies
• Analysis of individually matched case–control studies
• Advantages of case–control studies
• Disadvantages of case–control studies
• References
• Part 4 Hypothesis testing
• 17 Hypothesis testing
• Defining the null and alternative hypotheses
• Obtaining the test statistic
• Obtaining the P-value
• Using the P-value
• Non-parametric tests
• Which test?
• Hypothesis tests versus confidence intervals
• Equivalence and non-inferiority trials
• References
• 18 Errors in hypothesis testing
• Making a decision
• Making the wrong decision
• Power and related factors
• Multiple hypothesis testing
• References
• Part 5 Basic techniques for analysing data
• 19 Numerical data: a single group
• The problem
• The one-sample t-test
• The sign test
• 20 Numerical data: two related groups
• The problem
• The paired t-test
• The Wilcoxon signed ranks test
• Reference
• 21 Numerical data: two unrelated groups
• The problem
• The unpaired (two-sample) t-test
• The Wilcoxon rank sum (two-sample) test
• Reference
• 22 Numerical data: more than two groups
• The problem
• One-way analysis of variance
• The Kruskal–Wallis test
• References
• 23 Categorical data: a single proportion
• The problem
• The test of a single proportion
• The sign test applied to a proportion
• 24 Categorical data: two proportions
• The problems
• Independent groups: the Chi-squared test
• Related groups: McNemar’s test
• Reference
• 25 Categorical data: more than two categories
• Chi-squared test: large contingency tables
• Chi-squared test for trend
• 26 Correlation
• Pearson correlation coefficient
• Spearman’s rank correlation coefficient
• 27 The theory of linear regression
• What is linear regression?
• The regression line
• Method of least squares
• Assumptions
• Analysis of variance table
• Regression to the mean
• 28 Performing a linear regression analysis
• The linear regression line
• Drawing the line
• Checking the assumptions
• Failure to satisfy the assumptions
• Outliers and influential points
• Assessing goodness of fit
• Investigating the slope
• Using the line for prediction
• Improving the interpretation of the model
• 29 Multiple linear regression
• What is it?
• Why do it?
• Assumptions
• Categorical explanatory variables
• Analysis of covariance
• Choice of explanatory variables
• Analysis
• Outliers and influential points
• Reference
• 30 Binary outcomes and logistic regression
• Reasoning
• The logistic regression equation
• The explanatory variables
• Assessing the adequacy of the model
• Comparing the odds ratio and the relative risk
• Multinomial and ordinal logistic regression
• Conditional logistic regression
• References
• 31 Rates and Poisson regression
• Rates
• Poisson regression
• 32 Generalized linear models
• Which type of model do we choose?
• Likelihood and maximum likelihood estimation
• Assessing adequacy of fit
• Regression diagnostics
• 33 Explanatory variables in statistical models
• Nominal explanatory variables
• Ordinal explanatory variables
• Numerical explanatory variables
• Selecting explanatory variables
• Interaction
• Collinearity
• Confounding
• 34 Bias and confounding
• Bias
• Confounding
• References
• 35 Checking assumptions
• Why bother?
• Are the data Normally distributed?
• Are two or more variances equal?
• Are variables linearly related?
• What if the assumptions are not satisfied?
• Sensitivity analysis
• References
• 36 Sample size calculations
• The importance of sample size
• Requirements
• Methodology
• Altman’s nomogram
• Quick formulae
• Power statement
• Adjustments
• Increasing the power for a fixed sample size
• References
• 37 Presenting results
• Numerical results
• Tables
• Diagrams
• Presenting results in a paper
• References
• Part 6 Additional chapters
• 38 Diagnostic tools
• Reference intervals
• Diagnostic tests
• 39 Assessing agreement
• Measurement variability and error
• Reliability
• Categorical variables
• Numerical variables
• Reporting guidelines
• References
• 40 Evidence-based medicine
• 1 Formulate the clinical question (PICO)
• 2 Locate the relevant information (e.g. on diagnosis, prognosis or therapy)
• 3 Critically appraise the methods in order to assess the validity (closeness to the truth) of the evidence
• 4 Extract the most useful results and determine whether they are important
• 5 Apply the results in clinical practice
• 6 Evaluate your performance
• References
• 41 Methods for clustered data
• Displaying the data
• Comparing groups: inappropriate analyses
• Comparing groups: appropriate analyses
• Reference
• 42 Regression methods for clustered data
• Aggregate level analysis
• Robust standard errors
• Random effects models
• Generalized estimating equations (GEE)
• References
• 43 Systematic reviews and meta-analysis
• The systematic review
• Meta-analysis
• References
• 44 Survival analysis
• Censored data
• Displaying survival data
• Summarizing survival
• Comparing survival
• Problems encountered in survival analysis
• Reference
• 45 Bayesian methods
• The frequentist approach
• The Bayesian approach
• Diagnostic tests in a Bayesian framework
• Disadvantages of Bayesian methods
• Reference
• 46 Developing prognostic scores
• Why do we do it?
• Assessing the performance of a prognostic score
• Developing prognostic indices and risk scores for other types of data
• Reporting guidelines
• Appendices
• Appendix A: Statistical tables
• Reference
• Appendix B: Altman’s nomogram for sample size calculations (Chapter 36)
• Appendix C: Typical computer output
• Appendix D: Checklists and trial profile from the EQUATOR network and critical appraisal templates
• Equator Network Statements
• Critical Appraisal Templates
• Reference
• Appendix E: Glossary of terms
• Appendix F: Chapter numbers with relevant multiple-choice questions and structured questions from Medical Statistics at a Glance Workbook
• Index
• End User License Agreement