Dataset Information

A BAYESIAN SPATIAL AND TEMPORAL MODELING APPROACH TO MAPPING GEOGRAPHIC VARIATION IN MORTALITY RATES FOR SUBNATIONAL AREAS WITH R-INLA.

ABSTRACT: Hierarchical Bayes models have been used in disease mapping to examine small scale geographic variation. State level geographic variation for less common causes of mortality outcomes have been reported however county level variation is rarely examined. Due to concerns about statistical reliability and confidentiality, county-level mortality rates based on fewer than 20 deaths are suppressed based on Division of Vital Statistics, National Center for Health Statistics (NCHS) statistical reliability criteria, precluding an examination of spatio-temporal variation in less common causes of mortality outcomes such as suicide rates (SRs) at the county level using direct estimates. Existing Bayesian spatio-temporal modeling strategies can be applied via Integrated Nested Laplace Approximation (INLA) in R to a large number of rare causes of mortality outcomes to enable examination of spatio-temporal variations on smaller geographic scales such as counties. This method allows examination of spatiotemporal variation across the entire U.S., even where the data are sparse. We used mortality data from 2005-2015 to explore spatiotemporal variation in SRs, as one particular application of the Bayesian spatio-temporal modeling strategy in R-INLA to predict year and county-specific SRs. Specifically, hierarchical Bayesian spatio-temporal models were implemented with spatially structured and unstructured random effects, correlated time effects, time varying confounders and space-time interaction terms in the software R-INLA, borrowing strength across both counties and years to produce smoothed county level SRs. Model-based estimates of SRs were mapped to explore geographic variation.

SUBMITTER: Khana D

PROVIDER: S-EPMC5839164 | biostudies-literature | 2018 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A BAYESIAN SPATIAL AND TEMPORAL MODELING APPROACH TO MAPPING GEOGRAPHIC VARIATION IN MORTALITY RATES FOR SUBNATIONAL AREAS WITH R-INLA.

Khana Diba D Rossen Lauren M LM Hedegaard Holly H Warner Margaret M

Journal of data science : JDS 20180101 1

Hierarchical Bayes models have been used in disease mapping to examine small scale geographic variation. State level geographic variation for less common causes of mortality outcomes have been reported however county level variation is rarely examined. Due to concerns about statistical reliability and confidentiality, county-level mortality rates based on fewer than 20 deaths are suppressed based on Division of Vital Statistics, National Center for Health Statistics (NCHS) statistical reliabilit ...[more]

PMID: 29520299

Similar Datasets

Project description:ImportanceChildhood obesity is a principal public health concern. Understanding the geographic distribution of childhood obesity can inform the design and delivery of interventions.ObjectiveTo better understand the causes of spatial dependence in rates of childhood obesity across neighborhoods.Design, setting, and participantsThis cohort study used data from a legislatively mandated body mass index screening program for public school children in Arkansas from the 2003-2004 through 2014-2015 academic years. Spatial autoregressive moving average (SARMA) models for panel data were used to estimate spatial dependency in childhood obesity at 2 levels of spatial aggregation. Data were analyzed from August 2017 to February 2018.ExposuresThe SARMA models included geographic fixed effects to capture time-invariant differences in neighborhood characteristics along with controls for the mean age of children and the proportion of children by race/ethnicity, school meal status, and sex.Main outcomes and measuresThe proportion of obese schoolchildren in Arkansas neighborhoods by year, defined at larger (census tract) and smaller (census block group) spatial scales.ResultsThe geographic aggregations were based on 935 800 children with a mean (SD) age of 132 (39) months. Of these children, 51% were male; 65% were white, 21% were black, 10% were Hispanic, 2% were Asian, and the remainder were of other or unidentified race/ethnicity. In models without geographic fixed effects, there was evidence of positive and significant spatial autocorrelation in obesity rates across tracts (ρ = 0.511; 95% CI, 0.469-0.553) and block groups (ρ = 0.569; 95% CI, 0.543-0.595). When geographic fixed effects were included, spatial autocorrelation diminished at the census tract level (ρ = 0.271; 95% CI, 0.147-0.396) and disappeared at the census block group level (ρ = -0.075; 95% CI, -0.264 to 0.114).Conclusions and relevanceBecause block groups are smaller than tracts, children in neighboring block groups were more likely to attend the same schools and interact through neighborhood play. Thus, geographic-based social networks were more likely to span block group boundaries. The lack of evidence of spatial autocorrelation in block group-level models suggests that social contagion may be less important than differences in neighborhood context across space. Caution should be used in interpreting significant spatial autocorrelation as evidence of social contagion in obesity.

Project description:Background Rates of major lower extremity amputation in patients with peripheral artery disease are higher in rural communities with markers of low socioeconomic status, but most Americans live in metropolitan areas. Whether amputation rates vary within US metropolitan areas is unclear, as are characteristics of high amputation rate urban communities. Methods and Results We estimated rates of major lower extremity amputation per 100 000 Medicare beneficiaries between 2010 and 2018 at the ZIP code level among ZIP codes with ≥100 beneficiaries. We described demographic characteristics of high and low amputation ZIP codes, and the association between major amputation rate and 3 ZIP code-level markers of socioeconomic status-the proportion of patients with dual eligibility for Medicaid, median household income, and Distressed Communities Index score-for metropolitan, micropolitan, and rural ZIP code cohorts. Between 2010 and 2018, 188 995 Medicare fee-for-service patients living in 31 391 ZIP codes with ≥100 beneficiaries had a major lower extremity amputation. The median (interquartile range) ZIP code-level number of amputations per 100 000 beneficiaries was 262 (75-469). Though nonmetropolitan ZIP codes had higher rates of major amputation than metropolitan areas, 78.2% of patients undergoing major amputation lived in metropolitan areas. Compared with ZIP codes with lower amputation rates, top quartile amputation rate ZIP codes had a greater proportion of Black residents (4.4% versus 17.5%, P<0.001). In metropolitan areas, after adjusting for clinical comorbidities and demographics, every $10 000 lower median household income was associated with a 4.4% (95% CI, 3.9-4.8) higher amputation rate, and a 10-point higher Distressed Communities Index score was associated with a 3.8% (95% CI, 3.4%-4.2%) higher amputation rate; there was no association between the proportion of patients eligible for Medicaid and amputation rate. These findings were comparable to the associations identified across all ZIP codes. Conclusions In metropolitan areas, where most individuals undergoing lower extremity amputation live, markers of lower socioeconomic status and Black race were associated with higher rates of major lower extremity amputation. Development of community-based tools for peripheral artery disease diagnosis and management targeted to communities with high amputation rates in urban areas may help reduce inequities in peripheral artery disease outcomes.

Project description:BackgroundAt present, dengue control focuses on reducing the density of the primary vector for the disease, Aedes aegypti, which is the only vulnerable link in the chain of transmission. The use of new approaches for dengue entomological surveillance is extremely important, since present methods are inefficient. With this in mind, the present study seeks to analyze the spatio-temporal dynamics of A. aegypti infestation with oviposition traps, using efficient computational methods. These methods will allow for the implementation of the proposed model and methodology into surveillance and monitoring systems.MethodsThe study area includes a region in the municipality of Rio de Janeiro, characterized by high population density, precarious domicile construction, and a general lack of infrastructure around it. Two hundred and forty traps were distributed in eight different sentinel areas, in order to continually monitor immature Aedes aegypti and Aedes albopictus mosquitoes. Collections were done weekly between November 2010 and August 2012. The relationship between egg number and climate and environmental variables was considered and evaluated through Bayesian zero-inflated spatio-temporal models. Parametric inference was performed using the Integrated Nested Laplace Approximation (INLA) method.ResultsInfestation indexes indicated that ovipositing occurred during the entirety of the study period. The distance between each trap and the nearest boundary of the study area, minimum temperature and accumulated rainfall were all significantly related to the number of eggs present in the traps. Adjusting for the interaction between temperature and rainfall led to a more informative surveillance model, as such thresholds offer empirical information about the favorable climatic conditions for vector reproduction. Data were characterized by moderate time (0.29 - 0.43) and spatial (21.23 - 34.19 m) dependencies. The models also identified spatial patterns consistent with human population density in all sentinel areas. The results suggest the need for weekly surveillance in the study area, using traps allocated between 18 and 24 m, in order to understand the dengue vector dynamics.ConclusionsAedes aegypti, due to it short generation time and strong response to climate triggers, tend to show an eruptive dynamics that is difficult to predict and understand through just temporal or spatial models. The proposed methodology allowed for the rapid and efficient implementation of spatio-temporal models that considered zero-inflation and the interaction between climate variables and patterns in oviposition, in such a way that the final model parameters contribute to the identification of priority areas for entomological surveillance.

Dataset Information

A BAYESIAN SPATIAL AND TEMPORAL MODELING APPROACH TO MAPPING GEOGRAPHIC VARIATION IN MORTALITY RATES FOR SUBNATIONAL AREAS WITH R-INLA.

Publications

A BAYESIAN SPATIAL AND TEMPORAL MODELING APPROACH TO MAPPING GEOGRAPHIC VARIATION IN MORTALITY RATES FOR SUBNATIONAL AREAS WITH R-INLA.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets