2024
Predicting positive Clostridioides difficile test results using large-scale longitudinal data of demographics and medication history
Pham A, El-Kareh R, Myers F, Ohno-Machado L, Kuo T. Predicting positive Clostridioides difficile test results using large-scale longitudinal data of demographics and medication history. Heliyon 2024, 11: e41350. PMID: 39958729, PMCID: PMC11825254, DOI: 10.1016/j.heliyon.2024.e41350.Peer-Reviewed Original ResearchArea under the receiver operating characteristic curveMedical historyClostridioides difficile</i> infectionMonths of medical historyPatient's chancesData of demographicsReceiver operating characteristic curveLogistic regression modelsHealth patientsModerate sample sizesPregnant womenHealthcare institutionsClostridioides difficile</i>Antibiotic useOdds ratioNegative casesLarge-scale longitudinal dataFinancial incentivesPositive testLogistic regressionPatientsIncreased susceptibilityCharacteristic curveRegression modelsSignificant covariates
2017
Rare‐variant association tests in longitudinal studies, with an application to the Multi‐Ethnic Study of Atherosclerosis (MESA)
He Z, Lee S, Zhang M, Smith J, Guo X, Palmas W, Kardia S, Ionita‐Laza I, Mukherjee B. Rare‐variant association tests in longitudinal studies, with an application to the Multi‐Ethnic Study of Atherosclerosis (MESA). Genetic Epidemiology 2017, 41: 801-810. PMID: 29076270, PMCID: PMC5696115, DOI: 10.1002/gepi.22081.Peer-Reviewed Original ResearchConceptsMulti-Ethnic Study of AtherosclerosisMulti-Ethnic StudyStudy of atherosclerosisType I error rateRare-variant association testsRare variantsGene-based association testsRare-variant associationsAssociation TestLongitudinal outcomesLongitudinal studyExome sequencing dataMeasurement of blood pressureGenomic regionsSequence dataTrait heritabilitySequencing studiesMeasured outcomesGenetic variantsVariant analysisModerate sample sizesIndividual variantsRobust to misspecificationWithin-subject correlationStatistical power
2006
A practical approach to computing power for generalized linear models with nominal, count, or ordinal responses
Lyles R, Lin H, Williamson J. A practical approach to computing power for generalized linear models with nominal, count, or ordinal responses. Statistics In Medicine 2006, 26: 1632-1648. PMID: 16817148, DOI: 10.1002/sim.2617.Peer-Reviewed Original ResearchConceptsNon-central chi-square approximationChi-square approximationLikelihood ratio statisticModerate sample sizesEstimating conditional powerVariance-covariance matrixNon-continuous outcomesUnconditional powerRatio statisticWald statisticContinuous covariatesOrdinal responsesJoint distributionGeneralized linear modelApproximation powerConditional powerCount outcomesLinear modelNull hypothesisSample sizeStandard softwareSquare approximationResponse probabilityCovariatesApproximation
This site is protected by hCaptcha and its Privacy Policy and Terms of Service apply