Hierarchical Bayesian spatiotemporal analysis of revascularization odds using smoothing splines

Giovani L. Silva, C. B. Dean, Théophile Niyonsenga, Alain Vanasse

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Hierarchical Bayesian models are proposed for over-dispersed longitudinal spatially correlated binomial data. This class of models accounts for correlation among regions by using random effects and allows a flexible modelling of spatiotemporal odds by using smoothing splines. The aim is (i) to develop models which will identify temporal trends of odds and produce smoothed maps including regional effects, (ii) to specify Markov chain Monte Carlo (MCMC) inference for fitting such models, (iii) to study the sensitivity of such Bayesian binomial spline spatiotemporal analyses to prior assumptions, and (iv) to compare mechanisms for assessing goodness of fit. An analysis of regional variation for revascularization odds of patients hospitalized for acute coronary syndrome in Quebec motivates and illustrates the methods developed.

Original languageEnglish
Pages (from-to)2381-2401
Number of pages21
JournalStatistics in Medicine
Volume27
Issue number13
DOIs
Publication statusPublished - 15 Jun 2008
Externally publishedYes

Fingerprint

Spatio-Temporal Analysis
Markov Chains
Smoothing Splines
Odds
Bayes Theorem
Quebec
Acute Coronary Syndrome
Hierarchical Bayesian Model
Model Fitting
Goodness of fit
Markov Chain Monte Carlo
Random Effects
Acute
Spline
Modeling
Model

Cite this

Silva, Giovani L. ; Dean, C. B. ; Niyonsenga, Théophile ; Vanasse, Alain. / Hierarchical Bayesian spatiotemporal analysis of revascularization odds using smoothing splines. In: Statistics in Medicine. 2008 ; Vol. 27, No. 13. pp. 2381-2401.
@article{34b92e40b1784973aa554c42d14f847e,
title = "Hierarchical Bayesian spatiotemporal analysis of revascularization odds using smoothing splines",
abstract = "Hierarchical Bayesian models are proposed for over-dispersed longitudinal spatially correlated binomial data. This class of models accounts for correlation among regions by using random effects and allows a flexible modelling of spatiotemporal odds by using smoothing splines. The aim is (i) to develop models which will identify temporal trends of odds and produce smoothed maps including regional effects, (ii) to specify Markov chain Monte Carlo (MCMC) inference for fitting such models, (iii) to study the sensitivity of such Bayesian binomial spline spatiotemporal analyses to prior assumptions, and (iv) to compare mechanisms for assessing goodness of fit. An analysis of regional variation for revascularization odds of patients hospitalized for acute coronary syndrome in Quebec motivates and illustrates the methods developed.",
keywords = "Binomial data, Disease mapping, Hierarchical Bayesian model, MCMC methods, Spatiotemporal smoothing",
author = "Silva, {Giovani L.} and Dean, {C. B.} and Th{\'e}ophile Niyonsenga and Alain Vanasse",
year = "2008",
month = "6",
day = "15",
doi = "10.1002/sim.3094",
language = "English",
volume = "27",
pages = "2381--2401",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley & Sons",
number = "13",

}

Hierarchical Bayesian spatiotemporal analysis of revascularization odds using smoothing splines. / Silva, Giovani L.; Dean, C. B.; Niyonsenga, Théophile; Vanasse, Alain.

In: Statistics in Medicine, Vol. 27, No. 13, 15.06.2008, p. 2381-2401.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Hierarchical Bayesian spatiotemporal analysis of revascularization odds using smoothing splines

AU - Silva, Giovani L.

AU - Dean, C. B.

AU - Niyonsenga, Théophile

AU - Vanasse, Alain

PY - 2008/6/15

Y1 - 2008/6/15

N2 - Hierarchical Bayesian models are proposed for over-dispersed longitudinal spatially correlated binomial data. This class of models accounts for correlation among regions by using random effects and allows a flexible modelling of spatiotemporal odds by using smoothing splines. The aim is (i) to develop models which will identify temporal trends of odds and produce smoothed maps including regional effects, (ii) to specify Markov chain Monte Carlo (MCMC) inference for fitting such models, (iii) to study the sensitivity of such Bayesian binomial spline spatiotemporal analyses to prior assumptions, and (iv) to compare mechanisms for assessing goodness of fit. An analysis of regional variation for revascularization odds of patients hospitalized for acute coronary syndrome in Quebec motivates and illustrates the methods developed.

AB - Hierarchical Bayesian models are proposed for over-dispersed longitudinal spatially correlated binomial data. This class of models accounts for correlation among regions by using random effects and allows a flexible modelling of spatiotemporal odds by using smoothing splines. The aim is (i) to develop models which will identify temporal trends of odds and produce smoothed maps including regional effects, (ii) to specify Markov chain Monte Carlo (MCMC) inference for fitting such models, (iii) to study the sensitivity of such Bayesian binomial spline spatiotemporal analyses to prior assumptions, and (iv) to compare mechanisms for assessing goodness of fit. An analysis of regional variation for revascularization odds of patients hospitalized for acute coronary syndrome in Quebec motivates and illustrates the methods developed.

KW - Binomial data

KW - Disease mapping

KW - Hierarchical Bayesian model

KW - MCMC methods

KW - Spatiotemporal smoothing

UR - http://www.scopus.com/inward/record.url?scp=44949225191&partnerID=8YFLogxK

U2 - 10.1002/sim.3094

DO - 10.1002/sim.3094

M3 - Article

VL - 27

SP - 2381

EP - 2401

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 13

ER -