Biostatistics for Epidemiology and Public Health Using R

Bertram K. C. (Bert) Chan, PhD, is currently Consulting Biostatistician at the School of Medicine, Department of Preventive Medicine at Loma Linda University.

• ContentsPreface1. INTRODUCTION1.1 Medicine, Preventive Medicine, Public Health, and EpidemiologyMedicinePreventive Medicine and Public HealthPublic Health and EpidemiologyReview Questions for Section 1.11.2 Personal Health and Public HealthPersonal Health Versus Public HealthReview Questions for Section 1.21.3 Research and Measurements in EPDM and PHEPDM: The Basic Science of PHMain Epidemiologic FunctionsThe Cause of DiseasesExposure Measurement in EpidemiologyAdditional IssuesReview Questions for Section 1.31.4 BIOS and EPDMReview Questions for Section 1.4References2. RESEARCH AND DESIGN IN EPIDEMIOLOGY AND PUBLIC HEALTHIntroduction2.1 Causation and Association in Epidemiology and Public HealthThe Bradford-Hill Criteria for Causation and Association in EpidemiologyLegal Interpretation Using EpidemiologyDisease OccurrenceReview Questions for Section 2.12.2 Causation and Inference in Epidemiology and Public HealthRothman’s Diagrams for Sufficient Causation of DiseasesCausal InferencesUsing the Causal CriteriaJudging Scientific EvidenceReview Questions for Section 2.22.3 Biostatistical Basis of InferenceModes of InferenceLevels of MeasurementFrequentist BIOS in EPDMConfidence Intervals in Epidemiology and Public HealthBayesian Credible IntervalReview Questions for Section 2.32.4 BIOS in EPDM and PHApplications of BIOSBIOS in EPDM and PHProcessing and Analyzing Basic Epidemiologic DataAnalyzing Epidemiologic DataUsing REvaluating a Single Measure of OccurrencePoisson Count (Incidence) and Rate DataBinomial Risk and Prevalence DataEvaluating Two Measures of Occurrence—Comparison of Risk: Risk Ratio and Attributable RiskComparing Two Rate Estimates: Rate Ratio rrComparing Two Risk Estimates: Risk Ratio RR and Disease (Morbidity) Odds Ratio DORComparing Two Odds Estimates From Case–Control: The Salk Polio Vaccine Epidemiologic StudyReview Questions for Section 2.4Exercises for Chapter 2Using Probability TheoryDisease Symptoms in Clinical Drug TrialsRisks and Odds in EpidemiologyCase–Control Epidemiologic StudyMortality, Morbidity, and Fertility RatesIncidence Rates in Case-Cohort Survival AnalysisPrevalenceMortality RatesEstimating Sample SizesReferencesAppendix3. DATA ANALYSIS USING R PROGRAMMINGIntroduction3.1 Data and Data ProcessingData CodingData CaptureData EditingImputationsData QualityProducing ResultsReview Questions for Section 3.13.2 Beginning RR and BiostatisticsA First Session Using RThe R EnvironmentReview Questions for Section 3.23.3 R as a CalculatorMathematical Operations Using RAssignment of Values in R and Computations Using Vectors and MatricesComputations in Vectors and Simple GraphicsUse of Factors in R ProgrammingSimple Graphicsx as Vectors and Matrices in BiostatisticsSome Special Functions That Create VectorsArrays and MatricesUse of the Dimension Function dim in RUse of the Matrix Function matrix in RSome Useful Functions Operating on Matrices in RNA: “Not Available” for Missing Values in DatasetsSpecial Functions That Create VectorsReview Questions for Section 3.3Exercises for Section 3.33.4 Using R in Data Analysis in BIOSEntering Data at the R Command PromptThe Function list() and the Making of data.frame() in RReview Questions for Section 3.4Exercises for Section 3.43.5 Univariate, Bivariate, and Multivariate Data AnalysisUnivariate Data AnalysisBivariate and Multivariate Data AnalysisMultivariate Data AnalysisAnalysis of Variance (ANOVA)Review Questions for Section 3.5Exercises for Section 3.5ReferencesAppendix: Documentation for the plot functionGeneric X–Y Plotting4. GRAPHICS USING RIntroductionChoice of SystemPackages4.1 Base (or Traditional) GraphicsHigh-Level FunctionsLow-Level Plotting FunctionsInteracting with GraphicsUsing Graphics ParametersParameters List for GraphicsDevice DriversReview Questions for Section 4.1Exercises for Section 4.14.2 Grid GraphicsThe lattice Package: Trellis GraphicsThe Grid Model for R GraphicsGrid Graphics ObjectsApplications to Biostatistical and Epidemiologic InvestigationsReview Questions for Section 4.2Exercises for Section 4.2References5. PROBABILITY AND STATISTICS IN BIOSTATISTICSIntroduction5.1 Theories of ProbabilityWhat Is Probability?Basic Properties of ProbabilityProbability Computations Using RApplications of Probability Theory to Health SciencesTypical Summary Statistics in Biostatistics: Confidence Intervals, Significance Tests, and Goodness of FitReview Questions for Section 5.1Exercises for Section 5.15.2 Typical Statistical Inference in Biostatistics: Bayesian BiostatisticsWhat Is Bayesian Biostatistics?Bayes’s Theorem in Probability TheoryBayesian Methodology and Survival Analysis (Time-to-Event) Models for Biostatistics in Epidemiology and Preventive MedicineThe Inverse Bayes FormulaModeling in BiostatisticsReview Questions for Section 5.2Exercises for Section 5.2References6. CASE–CONTROL STUDIES AND COHORT STUDIES IN EPIDEMIOLOGYIntroduction6.1 Theory and Analysis of Case–Control StudiesAdvantages and Limitations of Case–Control StudiesAnalysis of Case–Control StudiesReview Questions for Section 6.1Exercises for Section 6.16.2 Theory and Analysis of Cohort StudiesAn Important Application of Cohort StudiesClinical TrialsRandomized Controlled TrialsCohort Studies for Diseases of Choice and Noncommunicable DiseasesCohort Studies and the Lexis Diagram in the Biostatistics of DemographyReview Questions for Section 6.2Exercises for Section 6.2References7. RANDOMIZED TRIALS, PHASE DEVELOPMENT, CONFOUNDING IN SURVIVAL ANALYSIS, AND LOGISTIC REGRESSIONS7.1 Randomized TrialsClassifications of RTs by Study DesignRandomizationBiostatistical Analysis of Data from RTsBiostatistics for RTs in the R EnvironmentReview Questions for Section 7.1Exercises for Section 7.17.2 Phase DevelopmentPhase 0 or Preclinical PhasePhase IPhase IIPhase IIIPharmacoepidemiology: A Branch of EpidemiologySome Basic Tests in Epidemiologic Phase DevelopmentReview Questions for Section 7.2Exercises for Section 7.27.3 Confounding in Survival AnalysisBiostatistical Approaches for Controlling ConfoundingUsing Regression Modeling for Controlling ConfoundingConfounding and CollinearityReview Questions for Section 7.3Exercises for Section 7.37.4 Logistic RegressionsInappropriateness of the Simple Linear Regression When y Is a Categorical Dependent VariableThe Logistic Regression ModelThe LogitLogistic Regression AnalysisGeneralized Linear Models in RReview Questions for Section 7.4Exercises for Section 7.4ReferencesIndex