Bayesian analysis of multi-type recurrent events and dependent termination with nonparametric covariate functions.

Li-An Lin, Sheng Luo, Bingshu E Chen, Barry R Davis
Author Information
  1. Li-An Lin: 1 Department of Biostatistics, The University of Texas School of Public Health, USA.
  2. Sheng Luo: 1 Department of Biostatistics, The University of Texas School of Public Health, USA.
  3. Bingshu E Chen: 2 Department of Public Health Sciences, Queen's University, Canada.
  4. Barry R Davis: 1 Department of Biostatistics, The University of Texas School of Public Health, USA.

Abstract

Multi-type recurrent event data occur frequently in longitudinal studies. Dependent termination may occur when the terminal time is correlated to recurrent event times. In this article, we simultaneously model the multi-type recurrent events and a dependent terminal event, both with nonparametric covariate functions modeled by B-splines. We develop a Bayesian multivariate frailty model to account for the correlation among the dependent termination and various types of recurrent events. Extensive simulation results suggest that misspecifying nonparametric covariate functions may introduce bias in parameter estimation. This method development has been motivated by and applied to the lipid-lowering trial component of the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial.

Keywords

References

  1. J Neurosurg. 2000 Jan;92(1):31-8 [PMID: 10616079]
  2. Biometrics. 1990 Dec;46(4):1071-85 [PMID: 2085626]
  3. Stat Med. 2014 Feb 20;33(4):580-94 [PMID: 24009073]
  4. Kidney Int. 2000 Jul;58(1):353-62 [PMID: 10886582]
  5. Arch Intern Med. 1991 Jun;151(6):1141-7 [PMID: 2043016]
  6. Stat Med. 2000 Jan 15;19(1):13-33 [PMID: 10623910]
  7. Stat Med. 2013 Dec 30;32(30):5366-80 [PMID: 24030899]
  8. Int J Cardiol. 2013 Sep 30;168(2):934-45 [PMID: 23218570]
  9. Biometrics. 2000 Dec;56(4):1016-22 [PMID: 11129456]
  10. Am J Hypertens. 1996 Apr;9(4 Pt 1):342-60 [PMID: 8722437]
  11. Lifetime Data Anal. 2004 Jun;10(2):121-38 [PMID: 15293628]
  12. JAMA. 2002 Dec 18;288(23):2998-3007 [PMID: 12479764]
  13. Stat Med. 2005 Mar 15;24(5):671-91 [PMID: 15558580]
  14. Biostatistics. 2004 Jan;5(1):129-43 [PMID: 14744832]
  15. Stroke. 1982 May-Jun;13(3):290-5 [PMID: 7080120]
  16. Biometrics. 1995 Dec;51(4):1469-82 [PMID: 8589234]
  17. Stat Med. 1988 Nov;7(11):1121-37 [PMID: 3201038]
  18. Stat Med. 2000 Dec 30;19(24):3309-24 [PMID: 11122497]
  19. Biometrics. 2002 Sep;58(3):510-20 [PMID: 12229985]
  20. Biometrics. 2010 Sep;66(3):716-25 [PMID: 19912171]
  21. Circulation. 2008 Nov 11;118(20):2057-62 [PMID: 18955667]
  22. N Engl J Med. 1994 Sep 8;331(10):637-42 [PMID: 7503821]
  23. Biometrics. 1990 Dec;46(4):1005-16 [PMID: 1964808]
  24. Stat Med. 2011 Sep 30;30(22):2683-95 [PMID: 21751230]
  25. Circulation. 2002 Jul 23;106(4):523-7 [PMID: 12135957]
  26. Stat Med. 2011 Nov 10;30(25):3010-23 [PMID: 21786280]
  27. Biom J. 2013 Nov;55(6):866-84 [PMID: 23929494]

Grants

  1. KL2 TR000370/NCATS NIH HHS
  2. R01 NS091307/NINDS NIH HHS
  3. U01 NS043127/NINDS NIH HHS

MeSH Term

Anticholesteremic Agents
Bayes Theorem
Biostatistics
Cardiovascular Diseases
Computer Simulation
Humans
Hypertension
Likelihood Functions
Longitudinal Studies
Markov Chains
Models, Statistical
Monte Carlo Method
Multivariate Analysis
Proportional Hazards Models
Randomized Controlled Trials as Topic
Recurrence
Statistics, Nonparametric

Chemicals

Anticholesteremic Agents

Word Cloud

Created with Highcharts 10.0.0recurrentmodeleventseventterminationdependentnonparametriccovariatefunctionsoccurmayterminalmulti-typeBayesianmultivariatefrailtyMulti-typedatafrequentlylongitudinalstudiesDependenttimecorrelatedtimesarticlesimultaneouslymodeledB-splinesdevelopaccountcorrelationamongvarioustypesExtensivesimulationresultssuggestmisspecifyingintroducebiasparameterestimationmethoddevelopmentmotivatedappliedlipid-loweringtrialcomponentAntihypertensiveLipid-LoweringTreatmentPreventHeartAttackTrialanalysisMarkovChainMonteCarloRecurrenthypertensionjoint

Similar Articles

Cited By