Bayesian inference for longitudinal data with non-parametric treatment effects.

Peter Müller, Fernando A Quintana, Gary L Rosner, Michael L Maitland
Author Information
  1. Peter Müller: Department of Mathematics, University of Texas at Austin, Austin, TX 78712, USA.

Abstract

We consider inference for longitudinal data based on mixed-effects models with a non-parametric Bayesian prior on the treatment effect. The proposed non-parametric Bayesian prior is a random partition model with a regression on patient-specific covariates. The main feature and motivation for the proposed model is the use of covariates with a mix of different data formats and possibly high-order interactions in the regression. The regression is not explicitly parameterized. It is implied by the random clustering of subjects. The motivating application is a study of the effect of an anticancer drug on a patient's blood pressure. The study involves blood pressure measurements taken periodically over several 24-h periods for 54 patients. The 24-h periods for each patient include a pretreatment period and several occasions after the start of therapy.

Keywords

References

  1. Clin Cancer Res. 2009 Oct 1;15(19):6250-7 [PMID: 19773379]
  2. Stat Med. 2007 Apr 30;26(9):2088-112 [PMID: 16906554]
  3. Biometrics. 2009 Sep;65(3):762-71 [PMID: 19210742]
  4. Comput Stat Data Anal. 2010 Sep 1;54(9):2172-2186 [PMID: 24363478]
  5. Biometrics. 1998 Sep;54(3):921-38 [PMID: 9750242]
  6. J Comput Graph Stat. 2011 Mar 1;20(1):260-278 [PMID: 21566678]
  7. Biometrics. 2010 Jun;66(2):435-43 [PMID: 19508243]

Grants

  1. M01-RR000055/NCRR NIH HHS
  2. NCI K23CA124802/NCI NIH HHS
  3. R01 CA157458/NCI NIH HHS
  4. P30-CA014599/NCI NIH HHS
  5. CA075981/NCI NIH HHS
  6. 5T32GM007019-31/NIGMS NIH HHS
  7. P30 CA006973/NCI NIH HHS

MeSH Term

Bayes Theorem
Blood Pressure
Blood Pressure Determination
Data Interpretation, Statistical
Humans
Models, Statistical
Time Factors
Treatment Outcome

Word Cloud

Created with Highcharts 10.0.0dataBayesianmodelnon-parametricregressioninferencelongitudinalpriortreatmenteffectproposedrandompartitioncovariatesstudybloodpressureseveral24-hperiodsconsiderbasedmixed-effectsmodelspatient-specificmainfeaturemotivationusemixdifferentformatspossiblyhigh-orderinteractionsexplicitlyparameterizedimpliedclusteringsubjectsmotivatingapplicationanticancerdrugpatient'sinvolvesmeasurementstakenperiodically54patientspatientincludepretreatmentperiodoccasionsstarttherapyeffectsClusteringMixed-effectsNon-parametricRandomRepeatedmeasurement

Similar Articles

Cited By