2024
Using Overlap Weights to Address Extreme Propensity Scores in Estimating Restricted Mean Counterfactual Survival Times
Cao Z, Ghazi L, Mastrogiacomo C, Forastiere L, Wilson F, Li F. Using Overlap Weights to Address Extreme Propensity Scores in Estimating Restricted Mean Counterfactual Survival Times. American Journal Of Epidemiology 2024, kwae416. PMID: 39489504, DOI: 10.1093/aje/kwae416.Peer-Reviewed Original ResearchInverse probability of censoring weightingProbability of censoring weightingOverlap weightingCensoring processVariance estimationInterval coverageInverse probability of treatment weightingTarget estimandInverse probabilityBinary outcomesPropensity scoreRMSTProbability of treatment weightingPropensity score weightingEstimationEstimandsLogistic regressionTreatment comparisonsVariance
2023
Designing individually randomized group treatment trials with repeated outcome measurements using generalized estimating equations
Wang X, Turner E, Li F. Designing individually randomized group treatment trials with repeated outcome measurements using generalized estimating equations. Statistics In Medicine 2023, 43: 358-378. PMID: 38009329, PMCID: PMC10939061, DOI: 10.1002/sim.9966.Peer-Reviewed Original ResearchConceptsSample size proceduresConstant treatment effectCorrelation structureSize proceduresMarginal mean modelClosed-form sample size formulaCorrelation parametersSandwich variance estimatorGroup treatment trialsEquation approachExchangeable correlation structureSample size formulaBinary outcomesVariance estimatorEmpirical powerLinear timeMean modelCorrelation matrixDifferent correlation parametersEstimating EquationsSize formulaEquationsSample size calculationDifferent assumptionsProper sample size calculationSample size requirements for testing treatment effect heterogeneity in cluster randomized trials with binary outcomes
Maleyeff L, Wang R, Haneuse S, Li F. Sample size requirements for testing treatment effect heterogeneity in cluster randomized trials with binary outcomes. Statistics In Medicine 2023, 42: 5054-5083. PMID: 37974475, PMCID: PMC10659142, DOI: 10.1002/sim.9901.Peer-Reviewed Original ResearchConceptsSample size proceduresSize proceduresEfficient Monte Carlo approachTreatment effect heterogeneitySample size methodsMonte Carlo approachContinuous effect modifiersBinary outcomesEffect heterogeneityCarlo approachNumerical illustrationsNecessary sample sizeGeneralized linear mixed modelLinear mixed modelsPopular classSample size requirementsStatistical powerAverage treatment effectHeterogeneous treatment effectsSample size calculationMixed modelsSize methodSize calculationSize requirementsCluster Randomized TrialMediation analysis in the presence of continuous exposure measurement error
Cheng C, Spiegelman D, Li F. Mediation analysis in the presence of continuous exposure measurement error. Statistics In Medicine 2023, 42: 1669-1686. PMID: 36869626, PMCID: PMC11320713, DOI: 10.1002/sim.9693.Peer-Reviewed Original ResearchConceptsBody mass indexExposure measurement errorPhysical activityMediation proportionHealth Professionals FollowCardiovascular disease incidenceProfessionals FollowMediation analysisMass indexCardiovascular diseaseLower riskStudy designEffect estimatesValidation study designContinuous exposureBiased effect estimatesTrue exposureMediatorsExposureValidation studyBinary outcomesHealth science studiesOutcomesRiskDisease incidence
2021
Estimating the natural indirect effect and the mediation proportion via the product method
Cheng C, Spiegelman D, Li F. Estimating the natural indirect effect and the mediation proportion via the product method. BMC Medical Research Methodology 2021, 21: 253. PMID: 34800985, PMCID: PMC8606099, DOI: 10.1186/s12874-021-01425-4.Peer-Reviewed Original ResearchConceptsInterval estimatorsApproximate estimatorExact estimatorMultivariate delta methodFinite sample performanceProduct methodNon-negligible biasBinary outcomesRare outcome assumptionExact expressionDelta methodVariance estimationEmpirical performanceEstimatorCommon data typesBootstrap approachBinary mediatorNatural indirect effectSample sizeswdpwr: A SAS macro and an R package for power calculations in stepped wedge cluster randomized trials
Chen J, Zhou X, Li F, Spiegelman D. swdpwr: A SAS macro and an R package for power calculations in stepped wedge cluster randomized trials. Computer Methods And Programs In Biomedicine 2021, 213: 106522. PMID: 34818620, PMCID: PMC8665077, DOI: 10.1016/j.cmpb.2021.106522.Peer-Reviewed Original ResearchConceptsWedge clusterIntracluster correlation coefficientContinuous outcomesCross-sectional cohortBinary outcomesExchangeable correlation structureWedge designPublic health intervention evaluationsHealth services researchClosed cohort designPower calculationCohort designClosed cohortStudy designIntracluster correlationIntervention evaluationNeeds of investigatorsOutcomesTrialsCohortServices researchInvestigatorsPrevious studiesSWDSample size estimation for modified Poisson analysis of cluster randomized trials with a binary outcome
Li F, Tong G. Sample size estimation for modified Poisson analysis of cluster randomized trials with a binary outcome. Statistical Methods In Medical Research 2021, 30: 1288-1305. PMID: 33826454, PMCID: PMC9132618, DOI: 10.1177/0962280221990415.Peer-Reviewed Original ResearchConceptsSample size formulaExchangeable working correlationExtensive Monte Carlo simulation studySize formulaMonte Carlo simulation studyFinite sample correctionMarginal relative riskCorresponding sample size formulaeSandwich variance estimatorVariable cluster sizesNumber of clustersAsymptotic efficiencySandwich varianceCluster size variabilityRobust sandwich varianceSample size estimationVariance estimatorAnalytical derivationSimulation studyCluster sizePoisson modelCoefficient estimatesFormulaCorrelation coefficient estimatesBinary outcomesMarginal modeling of cluster-period means and intraclass correlations in stepped wedge designs with binary outcomes
Li F, Yu H, Rathouz PJ, Turner EL, Preisser JS. Marginal modeling of cluster-period means and intraclass correlations in stepped wedge designs with binary outcomes. Biostatistics 2021, 23: 772-788. PMID: 33527999, PMCID: PMC9291643, DOI: 10.1093/biostatistics/kxaa056.Peer-Reviewed Original ResearchConceptsPopulation-averaged interpretationFinite sample inferenceMarginal inferenceMarginal meansRigorous justificationBinary outcomesComputational burdenIndividual-level observationsMarginal modelsInterval estimationMarginal modelingCorrelated binary outcomesCluster-period sizesJoint estimationEquationsLinear modelEstimating EquationsSW-CRTsFlexible toolFast pointInferenceEstimationAdditional mappingModelApproach
2019
Properties and pitfalls of weighting as an alternative to multilevel multiple imputation in cluster randomized trials with missing binary outcomes under covariate-dependent missingness
Turner EL, Yao L, Li F, Prague M. Properties and pitfalls of weighting as an alternative to multilevel multiple imputation in cluster randomized trials with missing binary outcomes under covariate-dependent missingness. Statistical Methods In Medical Research 2019, 29: 1338-1353. PMID: 31293199, DOI: 10.1177/0962280219859915.Peer-Reviewed Original Research
2017
An evaluation of constrained randomization for the design and analysis of group‐randomized trials with binary outcomes
Li F, Turner EL, Heagerty PJ, Murray DM, Vollmer WM, DeLong ER. An evaluation of constrained randomization for the design and analysis of group‐randomized trials with binary outcomes. Statistics In Medicine 2017, 36: 3791-3806. PMID: 28786223, PMCID: PMC5624845, DOI: 10.1002/sim.7410.Peer-Reviewed Original ResearchConceptsGroup-level covariatesPossible allocation schemesMonte Carlo simulationsStatistical propertiesRandomization-based testsStatistical issuesCarlo simulationsPrespecified percentageAllocation schemeStatistical testsCandidate allocationsSpaceBinary outcomesAllocation techniquePermutation testPractical limitationsPower lossSchemeSuch designsGroup-randomized trialLarge numberF-testContinuous outcomesCovariate imbalanceInference