Bayesian spatial modelling of geostatistical data using INLA and SPDE methods: A case study predicting malaria risk in Mozambique

Moraga, Paula; Dean, C.; Inoue, Joshua; Morawiecki, Piotr; Noureen, Shahzeb Raja; Wang, Fengpei

doi:10.1016/j.sste.2021.100440

Cited by 28 publications

(27 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The RR for a covariate represents the ratio of nodule prevalence when the covariate is x +1 to the nodule prevalence when the covariate is x , holding all other variables constant [ 57 ]. The significance of the estimates was determined as described in Moraga et al [ 58 ]. The association was deemed significant only if both the 95% BCI values were below 0 for negative association and above 0 for positive association.…”

Section: Resultsmentioning

confidence: 99%

“…Hyperparameters defining the SPDE mesh were used to calculate the spatial effect and project the spatial field (S8 Fig) . The spatial effect indicates the intrinsic spatial variability in the prevalence estimates, helping us understand the data's spatial structure [47]. Further, the spatial field also represents the spatial effect that was not accounted for by the covariates included in the model [58]. The mean spatial field is higher in western Ethiopia while it is lower in central Ethiopia and eastern Ethiopia, along with the high standard deviation of the spatial field in the eastern parts.…”

Section: Model Parametersmentioning

confidence: 99%

See 1 more Smart Citation

Geospatial modeling of pre-intervention nodule prevalence of Onchocerca volvulus in Ethiopia as an aid to onchocerciasis elimination

et al. 2022

View full text Add to dashboard Cite

Background Onchocerciasis is a neglected tropical filarial disease transmitted by the bites of blackflies, causing blindness and severe skin lesions. The change in focus for onchocerciasis management from control to elimination requires thorough mapping of pre-control endemicity to identify areas requiring interventions and to monitor progress. Onchocerca volvulus nodule prevalence in sub-Saharan Africa is spatially continuous and heterogeneous, and highly endemic areas may contribute to transmission in areas of low endemicity or vice-versa. Ethiopia is one such onchocerciasis-endemic country with heterogeneous O. volvulus nodule prevalence, and many districts are still unmapped despite their potential for onchocerciasis transmission. Methodology/Principle findings A Bayesian geostatistical model was fitted for retrospective pre-intervention nodule prevalence data collected from 916 unique sites and 35,077 people across Ethiopia. We used multiple environmental, socio-demographic, and climate variables to estimate the pre-intervention prevalence of O. volvulus nodules across Ethiopia and to explore their relationship with prevalence. Prevalence was high in southern and northwestern Ethiopia and low in Ethiopia’s central and eastern parts. Distance to the nearest river (RR: 0.9850, 95% BCI: 0.9751–0.995), precipitation seasonality (RR: 0.9837, 95% BCI: 0.9681–0.9995), and flow accumulation (RR: 0.9586, 95% BCI: 0.9321–0.9816) were negatively associated with O. volvulus nodule prevalence, while soil moisture (RR: 1.0218, 95% BCI: 1.0135–1.0302) was positively associated. The model estimated the number of pre-intervention cases of O. volvulus nodules in Ethiopia to be around 6.48 million (95% BCI: 3.53–13.04 million). Conclusions/Significance Nodule prevalence distribution was correlated with habitat suitability for vector breeding and associated biting behavior. The modeled pre-intervention prevalence can be used as a guide for determining priorities for elimination mapping in regions of Ethiopia that are currently unmapped, most of which have comparatively low infection prevalence.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Model Parametersmentioning

confidence: 99%

Geospatial modeling of pre-intervention nodule prevalence of Onchocerca volvulus in Ethiopia as an aid to onchocerciasis elimination

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Computational Bayesian inference can be achieved largely in one of two ways, either through sampling-based methods like Markov Chain Monte Carlo (MCMC) and deviants or approximately using approximate methods like Variational methods or Laplace approximations like Integrated Nested Laplace Approximation (INLA). INLA, as introduced by 33 , has been shown to be widely applicable to various statistical models; in particular, to the latent Gaussian models class of which disease mapping models are included 34,35,36,37 INLA employs a series of Laplace approximations and numerical integration to perform approximate Bayesian inference through numerically approximating the posterior densities of the latent field and hyperparameters. For data , latent field and hyperparameters , INLA can be summarized as follows:…”

Section: Approximate Inference Using Inlamentioning

confidence: 99%

Joint Quantile Disease Mapping with Application to Malaria and G6PD Deficiency

Alahmadi¹,

Rue²,

Niekerk³

2022

Preprint

View full text Add to dashboard Cite

Statistical analysis based on quantile regression methods is more comprehensive, flexible, and less sensitive to outliers when compared to mean regression methods.When the link between different diseases are of interest, joint disease mapping is useful for measuring directional correlation between them. Most studies study this link through multiple correlated mean regressions. In this paper we propose a joint quantile regression framework for multiple diseases where different quantile levels can be considered. We are motivated by the theorized link between the presence of Malaria and the gene deficiency G6PD, where medical scientist have anecdotally discovered a possible link between high levels of G6PD and lower than expected levels of Malaria initially pointing towards the occurrence of G6PD inhibiting the occurrence of Malaria. This link cannot be investigated with mean regressions and thus the need for flexible joint quantile regression in a disease mapping framework.Our joint quantile disease mapping model can be used for linear and non-linear effects of covariates by stochastic splines, since we define it as a latent Gaussian model.We perform Bayesian inference of this model using the INLA framework embedded in the R software package INLA. Finally, we illustrate the applicability of model by analyzing the malaria and G6PD deficiency incidences in 21 African countries using linked quantiles of different levels.

show abstract

“…As an initial illustration of the method's practical relevance, we here present an incomplete list of recent applications. In the time-period of May-Sep 2021, we find applications of the SPDE-approach to Gaussian fields in astronomy (Levis et al, 2021), health (Mannseth et al, 2021;Scott, 2021;Moses et al, 2021;Bertozzi-Villa et al, 2021;Moraga et al, 2021;Asri and Benamirouche, 2021), engineering (Zhang et al, 2021), theory (Ghattas and Willcox, 2021;Sanz-Alonso and Yang, 2021a;Lang and Pereira, 2021;Bolin and Wallin, 2021), environmetrics Beloconi et al, 2021;Vandeskog et al, 2021a;Wang and Zuo, 2021;Wright et al, 2021;Gómez-Catasús et al, 2021;Valente and Laurini, 2021b;Bleuel et al, 2021;Florêncio et al, 2021;Valente and Laurini, 2021a;Hough et al, 2021), econometrics (Morales and Laurini, 2021;Maynou et al, 2021), agronomy (Borges da Silva et al, 2021), ecology (Martino et al, 2021;Sicacha-Parada et al, 2021;Williamson et al;Bell et al, 2021;Humphreys et al;Xi et al, 2021;Fecchio et al), urban planning (Li, 2021), imaging (Aquino et al, 2021), modelling of forest fires (Taylor et al; Lindenmayer et al), fisheries (Babyn et al, 2021;van Woesik and Cacciapaglia, 2021;Jarvis et al, 2021;…”

Section: Some Recent Applicationsmentioning

confidence: 99%

The SPDE approach for Gaussian and non-Gaussian fields: 10 years and still running

Lindgren¹,

Bolin²,

Rue³

2021

Preprint

View full text Add to dashboard Cite

Gaussian processes and random fields have a long history, covering multiple approaches to representing spatial and spatio-temporal dependence structures, such as covariance functions, spectral representations, reproducing kernel Hilbert spaces, and graph based models. This article describes how the stochastic partial differential equation approach to generalising Matérn covariance models via Hilbert space projections connects with several of these approaches, with each connection being useful in different situations. In addition to an overview of the main ideas, some important extensions, theory, applications, and other recent developments are discussed. The methods include both Markovian and non-Markovian models, non-Gaussian random fields, non-stationary fields and space-time fields on arbitrary manifolds, and practical computational considerations.

show abstract

Bayesian spatial modelling of geostatistical data using INLA and SPDE methods: A case study predicting malaria risk in Mozambique

Cited by 28 publications

References 29 publications

Geospatial modeling of pre-intervention nodule prevalence of Onchocerca volvulus in Ethiopia as an aid to onchocerciasis elimination

Geospatial modeling of pre-intervention nodule prevalence of Onchocerca volvulus in Ethiopia as an aid to onchocerciasis elimination

Joint Quantile Disease Mapping with Application to Malaria and G6PD Deficiency

The SPDE approach for Gaussian and non-Gaussian fields: 10 years and still running

Contact Info

Product

Resources

About