Comparing Bayesian Spatial Conditional Overdispersion and the Besag–York–Mollié Models: Application to Infant Mortality Rates

Mapping the risk of respiratory infections using suburban district areas in a large city in Colombia

BACKGROUND

Acute respiratory infections (ARI) in Cúcuta -Colombia, have a comparatively high burden of disease associated with high public health costs. However, little is known about the epidemiology of these diseases in the city and its distribution within suburban areas. This study addresses this gap by estimating and mapping the risk of ARI in Cúcuta and identifying the most relevant risk factors.

METHODS

A spatial epidemiological analysis was designed to investigate the association of sociodemographic and environmental risk factors with the rate of ambulatory consultations of ARI in urban sections of Cúcuta, 2018. The ARI rate was calculated using a method for spatial estimation of disease rates. A Bayesian spatial model was implemented using the Integrated Nested Laplace Approximation approach and the Besag-York-Mollié specification. The risk of ARI per urban section and the hotspots of higher risk were also estimated and mapped.

RESULTS

A higher risk of IRA was found in central, south, north and west areas of Cúcuta after adjusting for sociodemographic and environmental factors, and taking into consideration the spatial distribution of the city's urban sections. An increase of one unit in the percentage of population younger than 15 years; the Index of Multidimensional Poverty and the rate of ARI in the migrant population was associated with a 1.08 (1.06-1.1); 1.04 (1.01-1.08) and 1.25 (1.22-1.27) increase of the ARI rate, respectively. Twenty-four urban sections were identified as hotspots of risk in central, south, north and west areas in Cucuta.

CONCLUSION

Sociodemographic factors and their spatial patterns are determinants of acute respiratory infections in Cúcuta. Bayesian spatial hierarchical models can be used to estimate and map the risk of these infections in suburban areas of large cities in Colombia. The methods of this study can be used globally to identify suburban areas and or specific communities at risk to support the implementation of prevention strategies and decision-making in the public and private health sectors.

Small area disease mapping of cancer incidence in British Columbia using Bayesian spatial models and the smallareamapp R Package

INTRODUCTION

There is an increasing interest in small area analyses in cancer surveillance; however, technical capacity is limited and accessible analytical approaches remain to be determined. This study demonstrates an accessible approach for small area cancer risk estimation using Bayesian hierarchical models and data visualization through the smallareamapp R package.

MATERIALS AND METHODS

Incident lung (N = 26,448), female breast (N = 28,466), cervical (N = 1,478), and colorectal (N = 25,457) cancers diagnosed among British Columbia (BC) residents between 2011 and 2018 were obtained from the BC Cancer Registry. Indirect age-standardization was used to derive age-adjusted expected counts and standardized incidence ratios (SIRs) relative to provincial rates. Moran's I was used to assess the strength and direction of spatial autocorrelation. A modified Besag, York and Mollie model (BYM2) was used for model incidence counts to calculate posterior median relative risks (RR) by Community Health Service Areas (CHSA; N = 218), adjusting for spatial dependencies. Integrated Nested Laplace Approximation (INLA) was used for Bayesian model implementation. Areas with exceedance probabilities (above a threshold RR = 1.1) greater or equal to 80% were considered to have an elevated risk. The posterior median and 95% credible intervals (CrI) for the spatially structured effect were reported. Predictive posterior checks were conducted through predictive integral transformation values and observed versus fitted values.

RESULTS

The proportion of variance in the RR explained by a spatial effect ranged from 4.4% (male colorectal) to 19.2% (female breast). Lung cancer showed the greatest number of CHSAs with elevated risk (Nwomen = 50/218, Nmen = 44/218), representing 2357 total excess cases. The largest lung cancer RRs were 1.67 (95% CrI = 1.06-2.50; exceedance probability = 96%; cases = 13) among women and 2.49 (95% CrI = 2.14-2.88; exceedance probability = 100%; cases = 174) among men. Areas with small population sizes and extreme SIRs were generally smoothed towards the null (RR = 1.0).

DISCUSSION

We present a ready-to-use approach for small area cancer risk estimation and disease mapping using BYM2 and exceedance probabilities. We developed the smallareamapp R package, which provides a user-friendly interface through an R-Shiny application, for epidemiologists and surveillance experts to examine geographic variation in risk. These methods and tools can be used to estimate risk, generate hypotheses, and examine ecologic associations while adjusting for spatial dependency.

The effect of weather and climate on dengue outbreak risk in Peru, 2000-2018: A time-series analysis

Background

Dengue fever is the most common arboviral disease in humans, with an estimated 50-100 million annual infections worldwide. Dengue fever cases have increased substantially in the past four decades, driven largely by anthropogenic factors including climate change. More than half the population of Peru is at risk of dengue infection and due to its geography, Peru is also particularly sensitive to the effects of El Niño Southern Oscillation (ENSO). Determining the effect of ENSO on the risk for dengue outbreaks is of particular public health relevance and may also be applicable to other Aedes-vectored viruses.

Methods

We conducted a time-series analysis at the level of the district-month, using surveillance data collected from January 2000 to September 2018 from all districts with a mean elevation suitable to survival of the mosquito vector (<2,500m), and ENSO and weather data from publicly-available datasets maintained by national and international agencies. We took a Bayesian hierarchical modeling approach to address correlation in space, and B-splines with four knots per year to address correlation in time. We furthermore conducted subgroup analyses by season and natural region.

Results

We detected a positive and significant effect of temperature (°C, RR 1.14, 95% CI 1.13, 1.15, adjusted for precipitation) and ENSO (ICEN index: RR 1.17, 95% CI 1.15, 1.20; ONI index: RR 1.04, 95% CI 1.02, 1.07) on outbreak risk, but no evidence of a strong effect for precipitation after adjustment for temperature. Both natural region and season were found to be significant effect modifiers of the ENSO-dengue effect, with the effect of ENSO being stronger in the summer and the Selva Alta and Costa regions, compared with winter and Selva Baja and Sierra regions.

Conclusions

Our results provide strong evidence that temperature and ENSO have significant effects on dengue outbreaks in Peru, however these results interact with region and season, and are stronger for local ENSO impacts than remote ENSO impacts. These findings support optimization of a dengue early warning system based on local weather and climate monitoring, including where and when to deploy such a system and parameterization of ENSO events, and provide high-precision effect estimates for future climate and dengue modeling efforts.

The theoretical importance of the El Niño Southern Oscillation to infectious disease transmission is widely accepted, however few studies have quantified this effect or its interaction with local environment. Using surveillance data on outbreaks of dengue fever in Peru, we have found evidence that El Niño events increase the risk of dengue outbreaks in the high jungle and coast, and decrease this risk in the low jungle. Our findings are likely generalizable to other viruses which are, like dengue virus, are transmitted by Aedes aegypti mosquitoes, including Zika, chikungunya, and yellow fever. As climate change is expected to increase the frequency of El Niño events, these results indicate that arbovirus outbreaks may also increase, and that El Niño events may be leveraged to predict them.

Patterns Simulations Using Gibbs/MRF Auto-Poisson Models

Pattern analysis is the process where characteristics of big data can be recognized using specific methods. Recognition of the data, especially images, can be achieved by applying spatial models, explaining the neighborhood structure of the patterns. These models can be introduced by Markov random field (MRF) models where conditional distribution of the pixels may be defined by a specific distribution. Various spatial models could be introduced, explaining the real patterns of the data; one class of these models is based on the Poisson distribution, called auto-Poisson models. The main advantage of these models is the consideration of the local characteristics of the image. Based on the local analysis, various patterns can be introduced and models that better explain the real data can be estimated, using advanced statistical techniques like Monte Carlo Markov Chains methods. These methods are based on simulations where the proposed distribution must converge to the original (final) one. In this work, an analysis of a MRF model under Poisson distribution would be defined and simulations would be illustrated based on Monte Carlo Markov Chains (MCMC) process like Gibbs sampler. Results would be illustrated using simulated and real patterns data.

Spatial Analysis on Supply and Demand of Adult Surgical Masks in Taipei Metropolitan Areas in the Early Phase of the COVID-19 Pandemic

This study aimed to assess the gap between the supply and demand of adult surgical masks under limited resources. Owing to the implementation of the real-name mask rationing system, the historical inventory data of aggregated mask consumption in a pharmacy during the early period of the COVID-19 outbreak (April and May 2020) in Taiwan were analyzed for supply-side analysis. We applied the Voronoi diagram and areal interpolation methods to delineate the average supply of customer counts from a pharmacy to a village (administrative level). On the other hand, the expected number of demand counts was estimated from the population data. The relative risk (RR) of supply, which is the average number of adults served per day divided by the expected number in a village, was modeled under a Bayesian hierarchical framework, including Poisson, negative binomial, Poisson spatial, and negative binomial spatial models. We observed that the number of pharmacies in a village is associated with an increasing supply, whereas the median annual per capita income of the village has an inverse relationship. Regarding land use percentages, percentages of the residential and the mixed areas in a village are negatively associated, while the school area percentage is positively associated with the supply in the Poisson spatial model. The corresponding uncertainty measurement: villages where the probability exceeds the risk of undersupply, that is, Pr (RR < 1), were also identified. The findings of the study may help health authorities to evaluate the spatial allocation of anti-epidemic resources, such as masks and rapid test kits, in small areas while identifying priority areas with the suspicion of undersupply in the beginning stages of outbreaks.

The association of wildfire air pollution with COVID-19 incidence in New South Wales, Australia

The 2020 COVID-19 outbreak in New South Wales (NSW), Australia, followed an unprecedented wildfire season that exposed large populations to wildfire smoke. Wildfires release particulate matter (PM), toxic gases and organic and non-organic chemicals that may be associated with increased incidence of COVID-19. This study estimated the association of wildfire smoke exposure with the incidence of COVID-19 in NSW. A Bayesian mixed-effect regression was used to estimate the association of either the average PM₁₀ level or the proportion of wildfire burned area as proxies of wildfire smoke exposure with COVID-19 incidence in NSW, adjusting for sociodemographic risk factors. The analysis followed an ecological design using the 129 NSW Local Government Areas (LGA) as the ecological units. A random effects model and a model including the LGA spatial distribution (spatial model) were compared. A higher proportional wildfire burned area was associated with higher COVID-19 incidence in both the random effects and spatial models after adjustment for sociodemographic factors (posterior mean = 1.32 (99% credible interval: 1.05-1.67) and 1.31 (99% credible interval: 1.03-1.65), respectively). No evidence of an association between the average PM₁₀ level and the COVID-19 incidence was found. LGAs in the greater Sydney and Hunter regions had the highest increase in the risk of COVID-19. This study identified wildfire smoke exposures were associated with increased risk of COVID-19 in NSW. Research on individual responses to specific wildfire airborne particles and pollutants needs to be conducted to further identify the causal links between SARS-Cov-2 infection and wildfire smoke. The identification of LGAs with the highest risk of COVID-19 associated with wildfire smoke exposure can be useful for public health prevention and or mitigation strategies.

Invited Commentary: Predicting Incidence Rates of Rare Cancers-Adding Epidemiologic and Spatial Contexts

There are unique challenges to identifying causes of and developing strategies for prevention of rare cancers, driven by the difficulty in estimating incidence, prevalence, and survival due to small case numbers. Using a Poisson modeling approach, Salmerón et al. (Am J Epidemiol. 2022;191(3):487-498) built upon their previous work to estimate incidence rates of rare cancers in Europe using a Bayesian framework, establishing a uniform prior for a measure of variability for country-specific incidence rates. They offer a methodology with potential transferability to other settings with similar cancer surveillance infrastructure. However, the approach does not consider the spatiotemporal correlation of rare cancer case counts and other, potentially more appropriate nonnormal probability distributions. In this commentary, we discuss the implications of future work from cancer epidemiology and spatial epidemiology perspectives. We describe the possibility of developing prediction models tailored to each type of rare cancer; incorporating the spatial heterogeneity in at-risk populations, surveillance coverage, and risk factors in these predictions; and considering a modeling framework with which to address the inherent spatiotemporal components of these data. We note that extension of this methodology to estimate subcountry rates at provincial, state, or smaller geographic levels would be useful but would pose additional statistical challenges.