{"id":11,"date":"2015-04-13T12:00:00","date_gmt":"2015-04-13T15:00:00","guid":{"rendered":"http:\/\/projects.upei.ca\/mer\/?page_id=11"},"modified":"2022-08-26T17:18:31","modified_gmt":"2022-08-26T20:18:31","slug":"table-of-contents-2","status":"publish","type":"page","link":"https:\/\/projects.upei.ca\/mer\/table-of-contents-2\/","title":{"rendered":"Table of Contents"},"content":{"rendered":"

Methods in Epidemiologic Research<\/b>\u00a0<\/a>is a comprehensive text covering the key principles and methods used in epidemiologic research. It is written primarily for researchers and graduate students in epidemiology, but the material is equally applicable to those in related disciplines (eg<\/em> public health).<\/p>\n

Chapters 1 through 6 focus on basic epidemiologic principles. Chapters 7\u2013<\/span>11 focus on study design issues for observational studies and controlled trials. There has been much discussion over the past decade about the need for epidemiologists to thoroughly report their research findings (and by doing so this will help ensure high-quality study designs in the future) and we have cited the summary recommendations in these chapters.<\/span><\/p>\n

Chapters 14\u2013<\/span>19 cover a range of multivariable models. Chapter 19 (Modelling Survival Data) attempts to provide a comprehensive coverage of the most commonly used methods in the analysis of time-to-event data.<\/span><\/p>\n

Chapters 20\u2013<\/span>23 deal with the issue of clustered data, including a thorough description of methods for analysing repeated measures data. Chapters 24\u2013<\/span>30 cover a range of specialised topics including: Bayesian methods (Chapter 24\u2014<\/span>contributed by Henrik Stryhn in collaboration with William Browne), two chapters on presenting and analysing spatial data (Chapter 25 and 26\u2014<\/span>contributed by Javier Sanchez and Dirk Pfeiffer), an introduction to infectious disease epidemiology (Chapter 27\u2014<\/span>contributed by Graham Medley in collaboration with Ian Dohoo), and meta-analysis (Chapter 28).<\/span><\/p>\n

Supplementary materials for this text will all be made available at upei.ca\/mer<\/a>. These will include datasets, computer programs for all examples presented (initially Stata \u201cdo files\u201d with the expectation that programs for other statistical packages will be added later).<\/span><\/p>\n

All of the datasets used in these examples are described in the text (Chapter 31) and are available through upei.ca\/mer<\/a>. Virtually all of the examples have been analysed using the statistical program Stata™\u2014<\/span>a program which provides a unique combination of statistical and epidemiological tools and which we use extensively in our teaching.<\/span><\/p>\n

TABLE OF CONTENTS<\/a><\/p>\n

1. INTRODUCTION AND CAUSAL CONCEPTS<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n
1.1<\/td>\nIntroduction<\/td>\n2<\/td>\n<\/tr>\n
1.2<\/td>\nA brief history of multiple causation concepts<\/td>\n2<\/td>\n<\/tr>\n
1.3<\/td>\nA brief history of scientific inference<\/td>\n6<\/td>\n<\/tr>\n
1.4<\/td>\nKey components of epidemiologic research<\/td>\n9<\/td>\n<\/tr>\n
1.5<\/td>\nSeeking causes<\/td>\n10<\/td>\n<\/tr>\n
1.6<\/td>\nModels of causation<\/td>\n11<\/td>\n<\/tr>\n
1.7<\/td>\nCounterfactual concepts of causation for a single exposure<\/td>\n18<\/td>\n<\/tr>\n
1.8<\/td>\nExperimental versus observational evidence of causation<\/td>\n22<\/td>\n<\/tr>\n
1.9<\/td>\nConstructing a causal diagram<\/td>\n23<\/td>\n<\/tr>\n
1.10<\/td>\nCausal criteria<\/td>\n25<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
2. SAMPLING<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
2.1<\/td>\nIntroduction<\/td>\n35<\/td>\n<\/tr>\n
2.2<\/td>\nNon-probability sampling<\/td>\n36<\/td>\n<\/tr>\n
2.3<\/td>\nProbability sampling<\/td>\n39<\/td>\n<\/tr>\n
2.4<\/td>\nSimple random sample<\/td>\n39<\/td>\n<\/tr>\n
2.5<\/td>\nSystematic random sample<\/td>\n40<\/td>\n<\/tr>\n
2.6<\/td>\nStratified random sample<\/td>\n40<\/td>\n<\/tr>\n
2.7<\/td>\nCluster sampling<\/td>\n41<\/td>\n<\/tr>\n
2.8<\/td>\nMultistage sampling<\/td>\n42<\/td>\n<\/tr>\n
2.9<\/td>\nTargeted (risk-based) sampling<\/td>\n43<\/td>\n<\/tr>\n
2.10<\/td>\nAnalysis of survey data<\/td>\n44<\/td>\n<\/tr>\n
2.11<\/td>\nSample-size determination<\/td>\n48<\/td>\n<\/tr>\n
2.12<\/td>\nSampling to detect disease<\/td>\n55<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
3. QUESTIONNAIRE DESIGN<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n
3.1<\/td>\nIntroduction<\/td>\n62<\/td>\n<\/tr>\n
3.2<\/td>\nDesigning the question<\/td>\n64<\/td>\n<\/tr>\n
3.3<\/td>\nOpen question<\/td>\n65<\/td>\n<\/tr>\n
3.4<\/td>\nClosed question<\/td>\n65<\/td>\n<\/tr>\n
3.5<\/td>\nWording the question<\/td>\n69<\/td>\n<\/tr>\n
3.6<\/td>\nStructure of questionnaires<\/td>\n69<\/td>\n<\/tr>\n
3.7<\/td>\nPre-testing questionnaires<\/td>\n70<\/td>\n<\/tr>\n
3.8<\/td>\nValidation<\/td>\n71<\/td>\n<\/tr>\n
3.9<\/td>\nResponse Rate<\/td>\n71<\/td>\n<\/tr>\n
3.10<\/td>\nData-coding and editing<\/td>\n72<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
4. MEASURES OF DISEASE FREQUENCY<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
4.1<\/td>\nIntroduction<\/td>\n78<\/td>\n<\/tr>\n
4.2<\/td>\nCount, proportion, odds and rate<\/td>\n78<\/td>\n<\/tr>\n
4.3<\/td>\nIncidence<\/td>\n79<\/td>\n<\/tr>\n
4.4<\/td>\nCalculating risk<\/td>\n80<\/td>\n<\/tr>\n
4.5<\/td>\nCalculating incidence rates<\/td>\n81<\/td>\n<\/tr>\n
4.6<\/td>\nRelationship between risk and rate<\/td>\n83<\/td>\n<\/tr>\n
4.7<\/td>\nPrevalence<\/td>\n84<\/td>\n<\/tr>\n
4.8<\/td>\nMortality statistics<\/td>\n85<\/td>\n<\/tr>\n
4.9<\/td>\nOther measures of disease frequency<\/td>\n85<\/td>\n<\/tr>\n
4.10<\/td>\nStandard errors and confidence intervals<\/td>\n87<\/td>\n<\/tr>\n
4.11<\/td>\nStandardisation of risks and rates<\/td>\n89<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
5. SCREENING AND DIAGNOSTIC TESTS<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
5.1<\/td>\nIntroduction<\/td>\n96<\/td>\n<\/tr>\n
5.2<\/td>\nAttributes of the test per se<\/td>\n96<\/td>\n<\/tr>\n
5.3<\/td>\nThe ability of a test to detect disease or health<\/td>\n104<\/td>\n<\/tr>\n
5.4<\/td>\nPredictive values<\/td>\n107<\/td>\n<\/tr>\n
5.5<\/td>\nInterpreting test results that are measured on a continuous scale<\/td>\n109<\/td>\n<\/tr>\n
5.6<\/td>\nUsing multiple tests<\/td>\n115<\/td>\n<\/tr>\n
5.7<\/td>\nEvaluation of diagnostic tests<\/td>\n117<\/td>\n<\/tr>\n
5.8<\/td>\nEvaluation when there is no gold standard<\/td>\n121<\/td>\n<\/tr>\n
5.9<\/td>\nOther considerations in test evaluation<\/td>\n125<\/td>\n<\/tr>\n
5.10<\/td>\nSample size requirements<\/td>\n127<\/td>\n<\/tr>\n
5.11<\/td>\nHerd-level testing<\/td>\n128<\/td>\n<\/tr>\n
5.12<\/td>\nUse of pooled samples<\/td>\n130<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
6. MEASURES OF ASSOCIATION<\/a><\/legend>\n\n\n\n\n\n\n\n\n
6.1<\/td>\nIntroduction<\/td>\n140<\/td>\n<\/tr>\n
6.2<\/td>\nMeasures of association<\/td>\n141<\/td>\n<\/tr>\n
6.3<\/td>\nMeasures of effect<\/td>\n144<\/td>\n<\/tr>\n
6.4<\/td>\nStudy design and measures of association<\/td>\n147<\/td>\n<\/tr>\n
6.5<\/td>\nHypothesis testing and confidence intervals<\/td>\n147<\/td>\n<\/tr>\n
6.6<\/td>\nMultivariable estimation of measures of association<\/td>\n152<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
7. INTRODUCTION TO OBSERVATIONAL STUDIES<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n
7.1<\/td>\nIntroduction<\/td>\n\u00a0<\/td>\n156<\/td>\n<\/tr>\n
7.2<\/td>\nA unified approach to study design<\/td>\n\u00a0<\/td>\n159<\/td>\n<\/tr>\n
7.3<\/td>\nDescriptive studies<\/td>\n\u00a0<\/td>\n161<\/td>\n<\/tr>\n
7.4<\/td>\nObservational studies<\/td>\n\u00a0<\/td>\n162<\/td>\n<\/tr>\n
7.5<\/td>\nCross-sectional studies<\/td>\n\u00a0<\/td>\n164<\/td>\n<\/tr>\n
7.6<\/td>\nEstimating incidence from one or more cross-sectional studies<\/td>\n\u00a0<\/td>\n168<\/td>\n<\/tr>\n
7.7<\/td>\nInferential limitations of cross-sectional studies<\/td>\n\u00a0<\/td>\n169<\/td>\n<\/tr>\n
7.8<\/td>\nRepeated cross-sectional versus cohort studies<\/td>\n\u00a0<\/td>\n170<\/td>\n<\/tr>\n
7.9<\/td>\nReporting of observational studies<\/td>\n\u00a0<\/td>\n171<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
8. COHORT STUDIES<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n
8.1<\/td>\nIntroduction<\/td>\n180<\/td>\n<\/tr>\n
8.2<\/td>\nSelecting the study group<\/td>\n182<\/td>\n<\/tr>\n
8.3<\/td>\nThe exposure<\/td>\n186<\/td>\n<\/tr>\n
8.4<\/td>\nDisease as exposure<\/td>\n190<\/td>\n<\/tr>\n
8.5<\/td>\nEnsuring exposed and non-exposed groups are comparable<\/td>\n190<\/td>\n<\/tr>\n
8.6<\/td>\nFollow-up period<\/td>\n191<\/td>\n<\/tr>\n
8.7<\/td>\nMeasuring the outcome<\/td>\n191<\/td>\n<\/tr>\n
8.8<\/td>\nAnalysis<\/td>\n192<\/td>\n<\/tr>\n
8.9<\/td>\nReporting of cohort studies<\/td>\n194<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
9. CASE-CONTROL STUDIES<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
9.1<\/td>\nIntroduction<\/td>\n202<\/td>\n<\/tr>\n
9.2<\/td>\nThe study base<\/td>\n202<\/td>\n<\/tr>\n
9.3<\/td>\nThe case series<\/td>\n205<\/td>\n<\/tr>\n
9.4<\/td>\nPrinciples of control selection<\/td>\n207<\/td>\n<\/tr>\n
9.5<\/td>\nSelecting controls in risk-based designs<\/td>\n207<\/td>\n<\/tr>\n
9.6<\/td>\nSelecting controls in rate-based designs<\/td>\n209<\/td>\n<\/tr>\n
9.7<\/td>\nOther sources of controls<\/td>\n214<\/td>\n<\/tr>\n
9.8<\/td>\nThe number of controls per case<\/td>\n215<\/td>\n<\/tr>\n
9.9<\/td>\nThe number of control groups<\/td>\n215<\/td>\n<\/tr>\n
9.10<\/td>\nExposure and covariate assessment<\/td>\n216<\/td>\n<\/tr>\n
9.11<\/td>\nKeeping the cases and controls comparable<\/td>\n216<\/td>\n<\/tr>\n
9.12<\/td>\nAnalysis of case-control data<\/td>\n217<\/td>\n<\/tr>\n
9.13<\/td>\nReporting guidelines for case-control studies<\/td>\n218<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
10. HYBRID STUDY DESIGNS<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n
10.1<\/td>\nIntroduction<\/td>\n224<\/td>\n<\/tr>\n
10.2<\/td>\nCase-crossover studies<\/td>\n224<\/td>\n<\/tr>\n
10.3<\/td>\nCase-case studies<\/td>\n228<\/td>\n<\/tr>\n
10.4<\/td>\nCase-case-control studies<\/td>\n229<\/td>\n<\/tr>\n
10.5<\/td>\nCase-series studies<\/td>\n231<\/td>\n<\/tr>\n
10.6<\/td>\nCase-cohort studies<\/td>\n233<\/td>\n<\/tr>\n
10.7<\/td>\nCase-only studies<\/td>\n235<\/td>\n<\/tr>\n
10.8<\/td>\nTwo-stage sampling designs<\/td>\n237<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
11. CONTROLLED STUDIES<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
11.1<\/td>\nIntroduction<\/td>\n244<\/td>\n<\/tr>\n
11.2<\/td>\nBackground, objectives, and summary trial design<\/td>\n246<\/td>\n<\/tr>\n
11.3<\/td>\nParticipants: the study group<\/td>\n247<\/td>\n<\/tr>\n
11.4<\/td>\nSpecifying the intervention<\/td>\n250<\/td>\n<\/tr>\n
11.5<\/td>\nMeasuring the outcome<\/td>\n251<\/td>\n<\/tr>\n
11.6<\/td>\nSample size<\/td>\n252<\/td>\n<\/tr>\n
11.7<\/td>\nAllocation of study subjects<\/td>\n254<\/td>\n<\/tr>\n
11.8<\/td>\nFollow-up\/compliance<\/td>\n258<\/td>\n<\/tr>\n
11.9<\/td>\nStatistical methods and analysis<\/td>\n259<\/td>\n<\/tr>\n
11.10<\/td>\nConclusions<\/td>\n262<\/td>\n<\/tr>\n
11.11<\/td>\nClinical trial designs for prophylaxis of communicable organisms<\/td>\n262<\/td>\n<\/tr>\n
11.12<\/td>\nReporting of clinical trials<\/td>\n265<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
12. VALIDITY IN OBSERVATIONAL STUDIES<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n
12.1<\/td>\nIntroduction<\/td>\n276<\/td>\n<\/tr>\n
12.2<\/td>\nSelection bias<\/td>\n277<\/td>\n<\/tr>\n
12.3<\/td>\nExamples of selection bias<\/td>\n281<\/td>\n<\/tr>\n
12.4<\/td>\nReducing selection bias<\/td>\n287<\/td>\n<\/tr>\n
12.5<\/td>\nInformation bias<\/td>\n288<\/td>\n<\/tr>\n
12.6<\/td>\nBias from misclassification<\/td>\n290<\/td>\n<\/tr>\n
12.7<\/td>\nValidation studies to correct misclassification<\/td>\n297<\/td>\n<\/tr>\n
12.8<\/td>\nMeasurement error<\/td>\n297<\/td>\n<\/tr>\n
12.9<\/td>\nErrors in surrogate measures of exposure<\/td>\n299<\/td>\n<\/tr>\n
12.10<\/td>\nThe impact of information bias on sample size<\/td>\n299<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
13. CONFOUNDING: DETECTION AND CONTROL<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
13.1<\/td>\nIntroduction<\/td>\n308<\/td>\n<\/tr>\n
13.2<\/td>\nControl of confounding prior to data analysis<\/td>\n311<\/td>\n<\/tr>\n
13.3<\/td>\nMatching on confounders<\/td>\n311<\/td>\n<\/tr>\n
13.4<\/td>\nDetection of confounding<\/td>\n316<\/td>\n<\/tr>\n
13.5<\/td>\nAnalytic control of confounding<\/td>\n322<\/td>\n<\/tr>\n
13.6<\/td>\nMultivariable modelling to control confounding<\/td>\n328<\/td>\n<\/tr>\n
13.7<\/td>\nOther approaches to control confounding and estimate causal effects<\/td>\n328<\/td>\n<\/tr>\n
13.8<\/td>\nPropensity scores for controlling confounding<\/td>\n335<\/td>\n<\/tr>\n
13.9<\/td>\nExternal adjustment and sensitivity analysis for unmeasured confounders<\/td>\n340<\/td>\n<\/tr>\n
13.10<\/td>\nUnderstanding causal relationships<\/td>\n342<\/td>\n<\/tr>\n
13.11<\/td>\nSummary of effects of extraneous variables<\/td>\n351<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
14. LINEAR REGRESSION<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
14.1<\/td>\nIntroduction<\/td>\n360<\/td>\n<\/tr>\n
14.2<\/td>\nRegression analysis<\/td>\n360<\/td>\n<\/tr>\n
14.3<\/td>\nHypothesis testing and effect estimation<\/td>\n362<\/td>\n<\/tr>\n
14.4<\/td>\nNature of the X-variables<\/td>\n368<\/td>\n<\/tr>\n
14.5<\/td>\nDetecting highly correlated (collinear) variables<\/td>\n374<\/td>\n<\/tr>\n
14.6<\/td>\nDetecting and modelling interaction<\/td>\n376<\/td>\n<\/tr>\n
14.7<\/td>\nCausal interpretation of a multivariable linear model<\/td>\n377<\/td>\n<\/tr>\n
14.8<\/td>\nEvaluating the least squares model<\/td>\n379<\/td>\n<\/tr>\n
14.9<\/td>\nEvaluating the major assumptions<\/td>\n385<\/td>\n<\/tr>\n
14.10<\/td>\nAssessment of individual observations<\/td>\n390<\/td>\n<\/tr>\n
14.11<\/td>\nTime-series data<\/td>\n396<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
15. MODEL-BUILDING STRATEGIES<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n
15.1<\/td>\nIntroduction<\/td>\n402<\/td>\n<\/tr>\n
15.2<\/td>\nSteps in building a model<\/td>\n403<\/td>\n<\/tr>\n
15.3<\/td>\nBuilding a causal model<\/td>\n403<\/td>\n<\/tr>\n
15.4<\/td>\nReducing the number of predictors<\/td>\n404<\/td>\n<\/tr>\n
15.5<\/td>\nThe problem of missing values<\/td>\n408<\/td>\n<\/tr>\n
15.6<\/td>\nEffects of continuous predictors<\/td>\n411<\/td>\n<\/tr>\n
15.7<\/td>\nIdentifying interaction terms of interest<\/td>\n418<\/td>\n<\/tr>\n
15.8<\/td>\nBuilding the model<\/td>\n418<\/td>\n<\/tr>\n
15.9<\/td>\nEvaluate the reliability of the model<\/td>\n423<\/td>\n<\/tr>\n
15.10<\/td>\nPresenting the results<\/td>\n424<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
16. LOGISTIC REGRESSION<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
16.1<\/td>\nIntroduction<\/td>\n430<\/td>\n<\/tr>\n
16.2<\/td>\nThe logistic model<\/td>\n430<\/td>\n<\/tr>\n
16.3<\/td>\nOdds and odds ratios<\/td>\n431<\/td>\n<\/tr>\n
16.4<\/td>\nFitting a logistic regression model<\/td>\n432<\/td>\n<\/tr>\n
16.5<\/td>\nAssumptions in logistic regression<\/td>\n433<\/td>\n<\/tr>\n
16.6<\/td>\nLikelihood ratio statistics<\/td>\n434<\/td>\n<\/tr>\n
16.7<\/td>\nWald tests<\/td>\n436<\/td>\n<\/tr>\n
16.8<\/td>\nInterpretation of coefficients<\/td>\n436<\/td>\n<\/tr>\n
16.9<\/td>\nAssessing interaction and confounding<\/td>\n439<\/td>\n<\/tr>\n
16.10<\/td>\nModel-building<\/td>\n441<\/td>\n<\/tr>\n
16.11<\/td>\nGeneralised linear models<\/td>\n444<\/td>\n<\/tr>\n
16.12<\/td>\nEvaluating logistic regression models<\/td>\n445<\/td>\n<\/tr>\n
16.13<\/td>\nSample size considerations<\/td>\n455<\/td>\n<\/tr>\n
16.14<\/td>\nExact logistic regression<\/td>\n456<\/td>\n<\/tr>\n
16.15<\/td>\nConditional logistic regression for matched studies<\/td>\n456<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
17. MODELLING ORDINAL AND MULTINOMIAL DATA<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n
17.1<\/td>\nIntroduction<\/td>\n462<\/td>\n<\/tr>\n
17.2<\/td>\nOverview of models<\/td>\n462<\/td>\n<\/tr>\n
17.3<\/td>\nMultinomial logistic regression<\/td>\n466<\/td>\n<\/tr>\n
17.4<\/td>\nModelling ordinal data<\/td>\n470<\/td>\n<\/tr>\n
17.5<\/td>\nProportional odds model (constrained cumulative logit model)<\/td>\n471<\/td>\n<\/tr>\n
17.6<\/td>\nAdjacent-category model<\/td>\n475<\/td>\n<\/tr>\n
17.7<\/td>\nContinuation-ratio model<\/td>\n476<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
18. MODELLING COUNT AND RATE DATA<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n
18.1<\/td>\nIntroduction<\/td>\n480<\/td>\n<\/tr>\n
18.2<\/td>\nThe Poisson distribution<\/td>\n481<\/td>\n<\/tr>\n
18.3<\/td>\nPoisson regression model<\/td>\n482<\/td>\n<\/tr>\n
18.4<\/td>\nInterpretation of coefficients<\/td>\n483<\/td>\n<\/tr>\n
18.5<\/td>\nEvaluating Poisson regression models<\/td>\n485<\/td>\n<\/tr>\n
18.6<\/td>\nNegative binomial regression<\/td>\n488<\/td>\n<\/tr>\n
18.7<\/td>\nProblems with zero counts<\/td>\n496<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
19. MODELLING SURVIVAL DATA<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
19.1<\/td>\nIntroduction<\/td>\n502<\/td>\n<\/tr>\n
19.2<\/td>\nNon-parametric analyses<\/td>\n507<\/td>\n<\/tr>\n
19.3<\/td>\nActuarial life tables<\/td>\n507<\/td>\n<\/tr>\n
19.4<\/td>\nKaplan-Meier estimate of survivor function<\/td>\n510<\/td>\n<\/tr>\n
19.5<\/td>\nNelson-Aalen estimate of cumulative hazard<\/td>\n512<\/td>\n<\/tr>\n
19.6<\/td>\nStatistical inference in non-parametric analyses<\/td>\n512<\/td>\n<\/tr>\n
19.7<\/td>\nSurvivor, failure and hazard functions<\/td>\n514<\/td>\n<\/tr>\n
19.8<\/td>\nSemi-parametric analyses<\/td>\n519<\/td>\n<\/tr>\n
19.9<\/td>\nParametric models<\/td>\n536<\/td>\n<\/tr>\n
19.10<\/td>\nAccelerated failure time models<\/td>\n541<\/td>\n<\/tr>\n
19.11<\/td>\nFrailty models and clustering<\/td>\n545<\/td>\n<\/tr>\n
19.12<\/td>\nMultiple outcome event data<\/td>\n551<\/td>\n<\/tr>\n
19.13<\/td>\nDiscrete-time survival analysis<\/td>\n552<\/td>\n<\/tr>\n
19.14<\/td>\nSample sizes for survival analyses<\/td>\n557<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
20. INTRODUCTION TO CLUSTERED DATA<\/a><\/legend>\n\n\n\n\n\n\n\n
20.1<\/td>\nIntroduction<\/td>\n564<\/td>\n<\/tr>\n
20.2<\/td>\nClustering arising from the data structure<\/td>\n564<\/td>\n<\/tr>\n
20.3<\/td>\nEffects of clustering<\/td>\n570<\/td>\n<\/tr>\n
20.4<\/td>\nSimulation studies on the impact of clustering<\/td>\n574<\/td>\n<\/tr>\n
20.5<\/td>\nIntroduction to methods for dealing with clustering<\/td>\n576<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
21. MIXED MODELS FOR CONTINUOUS DATA<\/a><\/legend>\n\n\n\n\n\n\n\n
21.1<\/td>\nIntroduction<\/td>\n588<\/td>\n<\/tr>\n
21.2<\/td>\nLinear mixed model<\/td>\n588<\/td>\n<\/tr>\n
21.3<\/td>\nRandom slopes<\/td>\n594<\/td>\n<\/tr>\n
21.4<\/td>\nContextual effects<\/td>\n598<\/td>\n<\/tr>\n
21.5<\/td>\nStatistical analysis of linear mixed models<\/td>\n601<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
22. MIXED MODELS FOR DISCRETE DATA<\/a><\/legend>\n\n\n\n\n\n\n\n\n
22.1<\/td>\nIntroduction<\/td>\n616<\/td>\n<\/tr>\n
22.2<\/td>\nLogistic regression with random effects<\/td>\n617<\/td>\n<\/tr>\n
22.3<\/td>\nPoisson regression with random effects<\/td>\n621<\/td>\n<\/tr>\n
22.4<\/td>\nGeneralised linear mixed model<\/td>\n623<\/td>\n<\/tr>\n
22.5<\/td>\nStatistical analysis of GLMMs<\/td>\n630<\/td>\n<\/tr>\n
22.6<\/td>\nSummary remarks on analysis of discrete clustered data<\/td>\n639<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
23. REPEATED MEASURES DATA<\/a><\/legend>\n\n\n\n\n\n\n\n
23.1<\/td>\nIntroduction<\/td>\n646<\/td>\n<\/tr>\n
23.2<\/td>\nUnivariate and multivariate approaches to repeated measures data<\/td>\n648<\/td>\n<\/tr>\n
23.3<\/td>\nLinear mixed models with correlation structure<\/td>\n654<\/td>\n<\/tr>\n
23.4<\/td>\nMixed models for discrete repeated measures data<\/td>\n662<\/td>\n<\/tr>\n
23.5<\/td>\nGeneralised estimating equations<\/td>\n665<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
24. INTRODUCTION TO BAYESIAN ANALYSIS<\/a><\/legend>\n\n\n\n\n\n\n\n
24.1<\/td>\nIntroduction<\/td>\n676<\/td>\n<\/tr>\n
24.2<\/td>\nBayesian analysis<\/td>\n676<\/td>\n<\/tr>\n
24.3<\/td>\nMarkov chain Monte Carlo (MCMC) estimation<\/td>\n680<\/td>\n<\/tr>\n
24.4<\/td>\nStatistical analysis based on MCMC estimation<\/td>\n685<\/td>\n<\/tr>\n
24.5<\/td>\nExtensions of Bayesian and MCMC Modelling<\/td>\n689<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
25. ANALYSIS OF SPATIAL DATA: INTRODUCTION AND VISUALISATION<\/a><\/legend>\n\n\n\n\n\n\n
25.1<\/td>\nIntroduction<\/td>\n702<\/td>\n<\/tr>\n
25.2<\/td>\nSpatial data<\/td>\n702<\/td>\n<\/tr>\n
25.3<\/td>\nSpatial data analysis<\/td>\n705<\/td>\n<\/tr>\n
25.4<\/td>\nAdditional topics<\/td>\n711<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
26. ANALYSIS OF SPATIAL DATA<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n
26.1<\/td>\nIntroduction<\/td>\n718<\/td>\n<\/tr>\n
26.2<\/td>\nIssues specific to statistical analysis of spatial data<\/td>\n718<\/td>\n<\/tr>\n
26.3<\/td>\nExploratory spatial analysis<\/td>\n720<\/td>\n<\/tr>\n
26.4<\/td>\nGlobal spatial clustering<\/td>\n728<\/td>\n<\/tr>\n
26.5<\/td>\nLocalised spatial cluster detection<\/td>\n735<\/td>\n<\/tr>\n
26.6<\/td>\nSpace-time association<\/td>\n738<\/td>\n<\/tr>\n
26.7<\/td>\nModelling<\/td>\n742<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
27. CONCEPTS OF INFECTIOUS DISEASE EPIDEMIOLOGY<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n
27.1<\/td>\nIntroduction<\/td>\n754<\/td>\n<\/tr>\n
27.2<\/td>\nInfection vs disease<\/td>\n756<\/td>\n<\/tr>\n
27.3<\/td>\nTransmission<\/td>\n758<\/td>\n<\/tr>\n
27.4<\/td>\nMathematical modelling of infectious disease transmission<\/td>\n760<\/td>\n<\/tr>\n
27.5<\/td>\nMethods of control of infectious disease<\/td>\n763<\/td>\n<\/tr>\n
27.6<\/td>\nEstimating R0 and other parameters<\/td>\n766<\/td>\n<\/tr>\n
27.7<\/td>\nDeveloping more complex models<\/td>\n771<\/td>\n<\/tr>\n
27.8<\/td>\nUsing models<\/td>\n773<\/td>\n<\/tr>\n
27.9<\/td>\nSummary<\/td>\n775<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
28. SYSTEMATIC REVIEWS AND META-ANALYSIS<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
28.1<\/td>\nIntroduction<\/td>\n780<\/td>\n<\/tr>\n
28.2<\/td>\nNarrative reviews<\/td>\n780<\/td>\n<\/tr>\n
28.3<\/td>\nSystematic Reviews<\/td>\n781<\/td>\n<\/tr>\n
28.4<\/td>\nMeta-analysis \u2013 Introduction<\/td>\n785<\/td>\n<\/tr>\n
28.5<\/td>\nFixed- and random-effects models<\/td>\n786<\/td>\n<\/tr>\n
28.6<\/td>\nPresentation of results<\/td>\n789<\/td>\n<\/tr>\n
28.7<\/td>\nHeterogeneity<\/td>\n791<\/td>\n<\/tr>\n
28.8<\/td>\nPublication bias<\/td>\n798<\/td>\n<\/tr>\n
28.9<\/td>\nInfluential studies<\/td>\n801<\/td>\n<\/tr>\n
28.10<\/td>\nOutcome scales and data issues<\/td>\n801<\/td>\n<\/tr>\n
28.11<\/td>\nMeta-analysis of observational studies<\/td>\n804<\/td>\n<\/tr>\n
28.12<\/td>\nMeta-analysis of diagnostic tests<\/td>\n806<\/td>\n<\/tr>\n
28.13<\/td>\nUse of meta-analysis<\/td>\n807<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
29. ECOLOGICAL AND GROUP-LEVEL STUDIES<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n
29.1<\/td>\nIntroduction<\/td>\n814<\/td>\n<\/tr>\n
29.2<\/td>\nRationale for group level studies<\/td>\n815<\/td>\n<\/tr>\n
29.3<\/td>\nTypes of ecologic variable<\/td>\n816<\/td>\n<\/tr>\n
29.4<\/td>\nIssues related to modelling approaches in ecologic studies<\/td>\n817<\/td>\n<\/tr>\n
29.5<\/td>\nThe linear model in the context of ecologic studies<\/td>\n818<\/td>\n<\/tr>\n
29.6<\/td>\nIssues related to inferences<\/td>\n819<\/td>\n<\/tr>\n
29.7<\/td>\nSources of ecologic bias<\/td>\n820<\/td>\n<\/tr>\n
29.8<\/td>\nAnalysis of ecologic data<\/td>\n825<\/td>\n<\/tr>\n
29.9<\/td>\nNon-ecologic group-level studies<\/td>\n826<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
30. A STRUCTURED APPROACH TO DATA ANALYSIS<\/a><\/legend>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
30.1<\/td>\nIntroduction<\/td>\n834<\/td>\n<\/tr>\n
30.2<\/td>\nData-collection sheets<\/td>\n834<\/td>\n<\/tr>\n
30.3<\/td>\nData coding<\/td>\n835<\/td>\n<\/tr>\n
30.4<\/td>\nData entry<\/td>\n835<\/td>\n<\/tr>\n
30.5<\/td>\nKeeping track of files<\/td>\n836<\/td>\n<\/tr>\n
30.6<\/td>\nKeeping track of variables<\/td>\n836<\/td>\n<\/tr>\n
30.7<\/td>\nProgram mode versus interactive processing<\/td>\n837<\/td>\n<\/tr>\n
30.8<\/td>\nData-editing<\/td>\n838<\/td>\n<\/tr>\n
30.9<\/td>\nData verification<\/td>\n839<\/td>\n<\/tr>\n
30.10<\/td>\nData processing\u2014outcome variable(s)<\/td>\n839<\/td>\n<\/tr>\n
30.11<\/td>\nData processing\u2014predictor variables<\/td>\n840<\/td>\n<\/tr>\n
30.12<\/td>\nData processing\u2014multilevel data<\/td>\n840<\/td>\n<\/tr>\n
30.13<\/td>\nUnconditional associations<\/td>\n841<\/td>\n<\/tr>\n
30.14<\/td>\nKeeping track of your analyses<\/td>\n841<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>
<\/legend>\n\n\n\n
\u00a0<\/td>\n\u00a031. DESCRIPTION OF DATASETS<\/a><\/td>\n843<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\u00a0<\/td>\n\u00a0GLOSSARY AND TERMINOLOGY<\/a><\/td>\n861<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/fieldset>\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

Methods in Epidemiologic Research\u00a0is a comprehensive text covering the key principles and methods used in epidemiologic research. It is written primarily for researchers and graduate students in epidemiology, but the material is equally applicable to those in related disciplines (eg … Continue reading →<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"tags":[],"_links":{"self":[{"href":"https:\/\/projects.upei.ca\/mer\/wp-json\/wp\/v2\/pages\/11"}],"collection":[{"href":"https:\/\/projects.upei.ca\/mer\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/projects.upei.ca\/mer\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/projects.upei.ca\/mer\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/projects.upei.ca\/mer\/wp-json\/wp\/v2\/comments?post=11"}],"version-history":[{"count":17,"href":"https:\/\/projects.upei.ca\/mer\/wp-json\/wp\/v2\/pages\/11\/revisions"}],"predecessor-version":[{"id":158,"href":"https:\/\/projects.upei.ca\/mer\/wp-json\/wp\/v2\/pages\/11\/revisions\/158"}],"wp:attachment":[{"href":"https:\/\/projects.upei.ca\/mer\/wp-json\/wp\/v2\/media?parent=11"}],"wp:term":[{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/projects.upei.ca\/mer\/wp-json\/wp\/v2\/tags?post=11"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}