On the use of adaptive spatial weight matrices from disease mapping multivariate analyses

Corpas-Burgos, Francisca; Martínez-Beneito, Miguel A.

doi:10.1007/s00477-020-01781-5

Cited by 12 publications

(23 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To mitigate the mentioned limitations, adaptive CARs have been proposed with applications to analysis of imaging data ( Brezger et al, 2007 ), analysis of correlated data ( Reich and Hodges, 2008 ), and disease mapping ( Lu and Carlin, 2005 , MacNab et al, 2006 ), 2018, 2021a,b, ( Lu et al, 2007 ), Brewer and Nolan 2007, ( Congdon, 2008 , Lee and Mitchell, 2012 ), Rushworth et al (2017) , ( Gao and Bradley, 2019 , Corpas-Burgos and Martinez-Beneito, 2020 ), among others.…”

Section: Adaptive Carsmentioning

confidence: 99%

“…One contains CARs with adaptive spatial (dependence) parameters; they were typically motivated to model locally varying spatial dependencies ( MacNab et al, 2006 , Congdon, 2008 , MacNab, 2018 , MacNab, 2021a ). Another includes CARs of adaptive scale or precision parameters that serve as spatial weights for modelling weighted pair-wise risk interactions and locally varying risk heterogeneities ( Brezger et al, 2007 ), Brewer and Nolan 2007, ( Reich and Hodges, 2008 , Corpas-Burgos and Martinez-Beneito, 2020 ). CARs with adaptive spatial and scale parameters are among the third category.…”

Section: Adaptive Carsmentioning

confidence: 99%

See 1 more Smart Citation

Bayesian disease mapping: Past, present, and future

MacNab

2022

Spatial Statistics

View full text Add to dashboard Cite

On the occasion of the Spatial Statistics’ 10th Anniversary, I reflect on the past and present of Bayesian disease mapping and look into its future. I focus on some key developments of models, and on recent evolution of multivariate and adaptive Gaussian Markov random fields and their impact and importance in disease mapping. I reflect on Bayesian disease mapping as a subject of spatial statistics that has advanced to date, and continues to grow, in scope and complexity alongside increasing needs of analytic tools for contemporary health science research, such as spatial epidemiology, population and public health, and medicine. I illustrate (potential) utility and impact of some of the disease mapping models and methods for analyzing and monitoring communicable disease such as the COVID-19 infection risks during an ongoing pandemic.

show abstract

Section: Adaptive Carsmentioning

confidence: 99%

Section: Adaptive Carsmentioning

confidence: 99%

Bayesian disease mapping: Past, present, and future

MacNab

2022

Spatial Statistics

View full text Add to dashboard Cite

show abstract

“…This modeling proposal belongs to what are known as M-models, which have already been used for the joint study of 21 causes of death [ 17 ]. Nevertheless, during the set-up of this project, we improved the original M-model proposals by allowing them to explicitly consider different spatial variances for each cause of death [ 23 ] and by estimating a suitable spatial weights matrix for our dataset [ 24 ]. The final statistical modeling proposal in our study includes these two improvements.…”

Section: Methodsmentioning

confidence: 99%

Geographical Variability in Mortality in Urban Areas: A Joint Analysis of 16 Causes of Death

Martínez-Beneito

Vergara-Hernández

Botella-Rocamora

et al. 2021

IJERPH

Self Cite

View full text Add to dashboard Cite

The geographical distribution of mortality has frequently been studied. Nevertheless, those studies often consider isolated causes of death. In this work, we aim to study the geographical distribution of mortality in urban areas, in particular, in 26 Spanish cities. We perform an overall study of 16 causes of death, considering that their geographical patterns could be dependent and estimating the dependence between the causes of death. We study the deaths in these 26 cities during the period 1996–2015 at the census tract level. A multivariate disease mapping model is used in order to solve the potential small area estimation problems that these data could show. We find that most of the geographical patterns found show positive correlations. This suggests the existence of a transversal geographical pattern, common to most causes of deaths, which determines those patterns to a higher/lower extent depending on each disease. The causes of death that exhibit that underlying pattern in a more prominent manner are chronic obstructive pulmonary disease (COPD), lung cancer, and cirrhosis for men and cardiovascular diseases and dementias for women. Such findings are quite consistent for most of the cities in the study. The high positive correlation found between geographical patterns reflects the existence of both high and low-risk areas in urban settings, in general terms for nearly all the causes of death. Moreover, the high-risk areas found often coincide with neighborhoods known for their high deprivation. Our results suggest that dependence among causes of death is a key aspect to be taken into account when mapping mortality, at least in urban contexts.

show abstract

“…( 2017 ) noticed that using a first-order contiguity matrix could produce a better model fit than other alternatives such as distance-based or covariate-based neighborhood matrices. Finally, Corpas-Burgos and Martinez-Beneito ( 2020 ) have recently proposed a methodology to define adaptive conditional autoregressive distributions in which the entries of the neighborhood matrix are treated as random variables to be modeled. These authors show that models based on these adaptive neighborhood matrices can provide a better fit than those based on contiguity matrices in a multivariate disease mapping context considering mortality data.…”

Section: Introductionmentioning

confidence: 99%

A comparison of multiple neighborhood matrix specifications for spatio-temporal model fitting: a case study on COVID-19 data

Briz‐Redón¹,

Iftimi

Correcher

et al. 2021

Stoch Environ Res Risk Assess

View full text Add to dashboard Cite

Establishing proper neighbor relations between a set of spatial units under analysis is essential when carrying out a spatial or spatio-temporal analysis. However, it is usual that researchers choose some of the most typical (and simple) neighborhood structures, such as the first-order contiguity matrix, without exploring other options. In this paper, we compare the performance of different neighborhood matrices in the context of modeling the weekly relative risk of COVID-19 over small areas located in or near Valencia, Spain. Specifically, we construct contiguity-based, distance-based, covariate-based (considering mobility flows and sociodemographic characteristics), and hybrid neighborhood matrices. We evaluate the goodness of fit, the overall predictive quality, the ability to detect high-risk spatio-temporal units, the capability to capture the spatio-temporal autocorrelation in the data, and the goodness of smoothing for a set of spatio-temporal models based on each of the neighborhood matrices. The results show that contiguity-based matrices, some of the distance-based matrices, and those based on sociodemographic characteristics perform better than the matrices based on k -nearest neighbors and those involving mobility flows. In addition, we test the linear combination of some of the constructed neighborhood matrices and the reweighting of these matrices after eliminating weak neighbor relations, without any model improvement.

show abstract

On the use of adaptive spatial weight matrices from disease mapping multivariate analyses

Cited by 12 publications

References 27 publications

Bayesian disease mapping: Past, present, and future

Bayesian disease mapping: Past, present, and future

Geographical Variability in Mortality in Urban Areas: A Joint Analysis of 16 Causes of Death

A comparison of multiple neighborhood matrix specifications for spatio-temporal model fitting: a case study on COVID-19 data

Contact Info

Product

Resources

About