Respiratory symptoms and lung function 8-10 months after community exposure to chlorine gas: a public health intervention and cross-sectional analysis

Background

We implemented a community based interventional health screening for individuals located within one mile of a 54 metric tons release of liquid chlorine following a 16 tanker car train derailment on 6 January, 2005 in Graniteville, South Carolina, USA. Public health intervention occurred 8–10 months after the event, and provided pulmonary function and mental health assessment by primary care providers. Its purpose was to evaluate those exposed to chlorine for evidence of ongoing impairment for medical referral and treatment. We report comparative analysis between self-report of respiratory symptoms via questionnaire and quantitative spirometry results.

Methods

Health assessments were obtained through respiratory symptom and exposure questionnaires, simple spirometry, and physical exam. Simple spirometry was used as the standard to identify continued breathing problems. Sensitivity, specificity, positive and negative predictive values were applied to evaluate the validity of the respiratory questionnaire. We also identified the direction of discrepancy between self-reported respiratory symptoms and spirometry measures. Generalized estimation equations determined prevalence ratios for abnormal spirometry based on the presence of participant persistent respiratory symptoms. Covariate adjustment was made for participant age, sex, race, smoking and educational status.

Results

Two hundred fifty-nine people participated in the Graniteville health screening; 53 children (mean age = 11 years, range: <1-16), and 206 adults (mean age = 50 years, range: 18–89). Of these, 220 (85%) performed spirometry maneuvers of acceptable quality. Almost 67% (n = 147) displayed abnormal spirometry, while 50% (n = 110) reported persistent new-onset respiratory symptoms. Moreover, abnormal spirometry was seen in 65 participants (29%) who did not report any discernible breathing problems. This represented a net 16.8% underreporting of symptoms. Sensitivity and specificity of questionnaire self-report of symptoms were low at 55.8% and 61.6%, respectively. Persistent cough (41%) and shortness of breath (39%) were the most frequently reported respiratory symptoms.

Conclusion

Eight to ten months after acute chlorine exposure, the Graniteville health screening participants under-reported respiratory symptoms when compared to abnormal spirometry results. Sensitivity and specificity were low, and we determined that relying upon the self-report questionnaire was not adequate to objectively assess the lung health of our population following irritant gas exposure.

Characterizing the food retail environment: impact of count, type, and geospatial error in 2 secondary data sources

OBJECTIVE

Commercial listings of food retail outlets are increasingly used by community members and food policy councils and in multilevel intervention research to identify areas with limited access to healthier food. This study quantified the amount of count, type, and geospatial error in 2 commercial data sources.

METHODS

InfoUSA and Dun and Bradstreet were compared with a validated field census and validity statistics were calculated.

RESULTS

Considering only completeness, Dun and Bradstreet data undercounted 24% of existing supermarkets and grocery stores, and InfoUSA, 29%. In addition, considering accuracy of outlet type assignment increased the undercount error to 42% and 39%, respectively. Marked overcount existed as well, and only 43% of existing supermarkets were correctly identified with respect to presence, outlet type, and location.

CONCLUSIONS AND IMPLICATIONS

Relying exclusively on secondary data to characterize the food environment will result in substantial error. Whereas extensive data cleaning can offset some error, verification of outlets with a field census is still the method of choice.

Bayesian semiparametric model with spatially-temporally varying coefficients selection

In spatiotemporal analysis, the effect of a covariate on the outcome usually varies across areas and time. The spatial configuration of the areas may potentially depend on not only the structured random intercept but also spatially varying coefficients of covariates. In addition, the normality assumption of the distribution of spatially varying coefficients could lead to potential biases of estimations. In this article, we proposed a Bayesian semiparametric space-time model where the spatially-temporally varying coefficient is decomposed as fixed, spatially varying, and temporally varying coefficients. We nonparametrically modeled the spatially varying coefficients of space-time covariates by using the area-specific Dirichlet process prior with weights transformed via a generalized transformation. We modeled the temporally varying coefficients of covariates through the dynamic model. We also took into account the uncertainty of inclusion of the spatially-temporally varying coefficients by variable selection procedure through determining the probabilities of different effects for each covariate. The proposed semiparametric approach shows its improvement compared with the Bayesian spatial-temporal models with normality assumption on spatial random effects and the Bayesian model with the Dirichlet process prior on the random intercept. We presented a simulation example to evaluate the performance of the proposed approach with the competing models. We used an application to low birth weight data in South Carolina as an illustration.

Space-time stick-breaking processes for small area disease cluster estimation

We propose a space-time stick-breaking process for the disease cluster estimation. The dependencies for spatial and temporal effects are introduced by using space-time covariate dependent kernel stick-breaking processes. We compared this model with the space-time standard random effect model by checking each model's ability in terms of cluster detection of various shapes and sizes. This comparison was made for simulated data where the true risks were known. For the simulated data, we have observed that space-time stick-breaking process performs better in detecting medium- and high-risk clusters. For the real data, county specific low birth weight incidences for the state of South Carolina for the years 1997-2007, we have illustrated how the proposed model can be used to find grouping of counties of higher incidence rate.

Joint disease mapping of cervical and male oropharyngeal cancer incidence in blacks and whites in South Carolina

Human papillomavirus (HPV) infection is an established causal agent for cervical cancer and a subset of oropharyngeal cancers. It is hypothesized that orogenital transmission results in oral cavity infection. In this paper we explore the geographical association between cervical and male oropharyngeal cancer incidence in blacks and whites in South Carolina using Bayesian joint disease mapping models fit to publicly available data. Our results suggest weak evidence for county-level association between the diseases, and different patterns of joint disease behavior for blacks and whites.

Associations between soil lead concentrations and populations by race/ethnicity and income-to-poverty ratio in urban and rural areas

Lead (Pb) is a well-studied environmental contaminant that has many negative health effects, especially for children. Both racial/ethnic and income disparities have been documented with respect to exposure to Pb in soils. The objectives of this study were to assess whether soil Pb concentrations in rural and urban areas of South Carolina USA, previously identified as having clusters of intellectual disabilities (ID) in children, were positively associated with populations of minority and low-income individuals and children (≤ 6 years of age). Surface soils from two rural and two urban areas with identified clusters of ID were analyzed for Pb and concentrations were spatially interpolated using inverse distance weighted analysis. Population race/ethnicity and income-to-poverty ratio (ITPR) from United States Census 2000 block group data were aerially interpolated by block group within each area. Urban areas had significantly higher concentrations of Pb than rural areas. Significant positive associations between black, non-Hispanic Latino, individuals and children ≤ 6 years of age and mean estimated Pb concentrations were observed in both urban (r = 0.38, p = 0.0007) and rural (r = 0.53, p = 0.04) areas. Significant positive associations also were observed between individuals and children with an ITPR < 1.00 and Pb concentrations, though primarily in urban areas. Racial/ethnic minorities and low ITPR individuals, including children, may be at elevated risk for exposure to Pb in soils.

Variable selection for spatial random field predictors under a Bayesian mixed hierarchical spatial model

A health outcome can be observed at a spatial location and we wish to relate this to a set of environmental measurements made on a sampling grid. The environmental measurements are covariates in the model but due to the interpolation associated with the grid there is an error inherent in the covariate value used at the outcome location. Since there may be multiple measurements made on different covariates there could be considerable uncertainty in the covariate values to be used. In this paper we examine a Bayesian approach to the interpolation problem and also a Bayesian solution to the variable selection issue. We present a series of simulations which outline the problem of recovering the true relationships, and also provide an empirical example.

Imputational modeling of spatial context and social environmental predictors of walking in an underserved community: the PATH trial

BACKGROUND

This study examined imputational modeling effects of spatial proximity and social factors of walking in African American adults.

PURPOSE

Models were compared that examined relationships between household proximity to a walking trail and social factors in determining walking status.

METHODS

Participants (N=133; 66% female; mean age=55 years) were recruited to a police-supported walking and social marketing intervention. Bayesian modeling was used to identify predictors of walking at 12 months.

RESULTS

Sensitivity analysis using different imputation approaches, and spatial contextual effects, were compared. All the imputation methods showed social life and income were significant predictors of walking, however, the complete data approach was the best model indicating Age (1.04, 95% OR: 1.00, 1.08), Social Life (0.83, 95% OR: 0.69, 0.98) and Income <$10,000 (0.10, 95% OR: 0.01, 0.97) were all predictors of walking.

CONCLUSIONS

The complete data approach was the best model of predictors of walking in African Americans.

A Bayesian normal mixture accelerated failure time spatial model and its application to prostate cancer

In the United States, prostate cancer is the third most common cause of death from cancer in males of all ages, and the most common cause of death from cancer in males over age 75. It has been recognized that the incidence of the prostate cancer is high in African Americans, and its occurrence and progression may be impacted by geographical factors. In order to investigate the spatial effects and racial disparities for prostate cancer in Louisiana, in this article we propose a normal mixture accelerated failure time spatial model, which does not require the proportional hazards assumption and allows the multi-model distribution to be modeled. The proposed model is estimated with a Bayesian approach and it can be easily implemented in WinBUGS. Extensive simulations show that the proposed model provides decent flexibility for a variety of parametric error distributions. The proposed method is applied to 2000–2007 Louisiana prostate cancer data set from the Surveillance, Epidemiology and End Results Program. The results reveal the possible spatial pattern and racial disparities for prostate cancer in Louisiana.

Bayesian point event modeling in spatial and environmental epidemiology

This paper reviews the current state of point event modeling in spatial epidemiology from a Bayesian perspective. Point event (or case event) data arise when geo-coded addresses of disease events are available. Often, this level of spatial resolution would not be accessible due to medical confidentiality constraints. However, for the examination of small spatial scales, it is important to be capable of examining point process data directly. Models for such data are usually formulated based on point process theory. In addition, special conditioning arguments can lead to simpler Bernoulli likelihoods and logistic spatial models. Goodness-of-fit diagnostics and Bayesian residuals are also considered. Applications within putative health hazard risk assessment, cluster detection, and linkage to environmental risk fields (misalignment) are considered.

Comparing variational Bayes with Markov chain Monte Carlo for Bayesian computation in neuroimaging

In this article, we consider methods for Bayesian computation within the context of brain imaging studies. In such studies, the complexity of the resulting data often necessitates the use of sophisticated statistical models; however, the large size of these data can pose significant challenges for model fitting. We focus specifically on the neuroelectromagnetic inverse problem in electroencephalography, which involves estimating the neural activity within the brain from electrode-level data measured across the scalp. The relationship between the observed scalp-level data and the unobserved neural activity can be represented through an underdetermined dynamic linear model, and we discuss Bayesian computation for such models, where parameters represent the unknown neural sources of interest. We review the inverse problem and discuss variational approximations for fitting hierarchical models in this context. While variational methods have been widely adopted for model fitting in neuroimaging, they have received very little attention in the statistical literature, where Markov chain Monte Carlo is often used. We derive variational approximations for fitting two models: a simple distributed source model and a more complex spatiotemporal mixture model. We compare the approximations to Markov chain Monte Carlo using both synthetic data as well as through the analysis of a real electroencephalography dataset examining the evoked response related to face perception. The computational advantages of the variational method are demonstrated and the accuracy associated with the resulting approximations are clarified.

Prior choice in discrete latent modeling of spatially referenced cancer survival

In this article, we examine the development and use of covariate models where the relation with explanantory covariates is spatially adaptive. In this way space is regarded as an effect modifier. We examine the possibility of discrete groupings of coefficients (clustering of coefficients). Our application is to prostate cancer survival based on the SEER cancer registry for the state of Louisiana, USA. This registry holds individual records linked to vital outcomes and is geo-coded at county level. We examine a range of potential prior distributions for groupings of regression coefficients in application to these data.

A Bayesian latent model with spatio-temporally varying coefficients in low birth weight incidence data

In spatial epidemiology studies, the effects of covariates on adverse health outcomes could vary over space and time so examining the spatio-temporally varying effects is useful. In particular, the association between covariates and health outcomes could have locally different temporal patterns. In this article, we develop a Bayesian spatio-temporal latent model to identify spatial clusters in each of which covariate effects have homogeneous temporal patterns as well as estimate heterogeneous temporal effects of covariates depending on spatial groups. We compare the proposed model to several alternative models to assess the performance of the proposed model in terms of a range of model assessment measures. Low birth weight incidence data in Georgia for the years 1997-2006 are used.

Associations of estimated residential soil arsenic and lead concentrations and community-level environmental measures with mother-child health conditions in South Carolina

We undertook a community-level aggregate analysis in South Carolina, USA, to examine associations between mother-child conditions from a Medicaid cohort of pregnant women and their children using spatially interpolated arsenic (As) and lead (Pb) concentrations in three geographic case areas and a control area. Weeks of gestation at birth was significantly negatively correlated with higher estimated As (r(s) = -0.28, p = 0.01) and Pb (r(s) = -0.26, p = 0.02) concentrations in one case area. Higher estimated Pb concentrations were consistently positively associated with frequency of black mothers (all p < 0.02) and negatively associated with frequency of white mothers (all p < 0.01), suggesting a racial disparity with respect to Pb.

Bayesian latent structure models with space-time-dependent covariates

Spatial-temporal data requires flexible regression models which can model the dependence of responses on space- and time-dependent covariates. In this paper, we describe a semiparametric space-time model from a Bayesian perspective. Nonlinear time dependence of covariates and the interactions among the covariates are constructed by local linear and piecewise linear models, allowing for more flexible orientation and position of the covariate plane by using time-varying basis functions. Space-varying covariate linkage coefficients are also incorporated to allow for the variation of space structures across the geographical location. The formulation accommodates uncertainty in the number and locations of the piecewise basis functions to characterize the global effects, spatially structured and unstructured random effects in relation to covariates. The proposed approach relies on variable selection-type mixture priors for uncertainty in the number and locations of basis functions and in the space-varying linkage coefficients. A simulation example is presented to evaluate the performance of the proposed approach with the competing models. A real data example is used for illustration.

Neighborhood context and incidence of type 1 diabetes: the SEARCH for Diabetes in Youth study

Findings regarding type 1 diabetes mellitus (T1DM) and neighborhood-level characteristics are mixed, with few US studies examining the influence of race/ethnicity. We conducted an ecologic study using SEARCH for Diabetes in Youth Study data to explore the association of neighborhood characteristics and T1DM incidence. 2002-2003 incident cases among youth at four SEARCH centers were included. Residential addresses were geocoded to US Census Tract. Standardized incidence ratios tended to increase with increasing education and median household income. Results from Poisson regression mixed models were similar and stable across race/ethnic groups and population density. Our study suggests a relationship of T1DM incidence with neighborhood-level socioeconomic status, independent of individual-level race/ethnic differences.

Neighborhood level risk factors for type 1 diabetes in youth: the SEARCH case-control study

BACKGROUND

European ecologic studies suggest higher socioeconomic status is associated with higher incidence of type 1 diabetes. Using data from a case-control study of diabetes among racially/ethnically diverse youth in the United States (U.S.), we aimed to evaluate the independent impact of neighborhood characteristics on type 1 diabetes risk. Data were available for 507 youth with type 1 diabetes and 208 healthy controls aged 10-22 years recruited in South Carolina and Colorado in 2003-2006. Home addresses were used to identify Census tracts of residence. Neighborhood-level variables were obtained from 2000 U.S. Census. Multivariate generalized linear mixed models were applied.

RESULTS

Controlling for individual risk factors (age, gender, race/ethnicity, infant feeding, birth weight, maternal age, number of household residents, parental education, income, state), higher neighborhood household income (p = 0.005), proportion of population in managerial jobs (p = 0.02), with at least high school education (p = 0.005), working outside the county (p = 0.04) and vehicle ownership (p = 0.03) were each independently associated with increased odds of type 1 diabetes. Conversely, higher percent minority population (p = 0.0003), income from social security (p = 0.002), proportion of crowded households (0.0497) and poverty (p = 0.008) were associated with a decreased odds.

CONCLUSIONS

Our study suggests that neighborhood characteristics related to greater affluence, occupation, and education are associated with higher type 1 diabetes risk. Further research is needed to understand mechanisms underlying the influence of neighborhood context.

Variable selection for spatial latent predictors under Bayesian spatial model

The problem of variable selection is encountered in model fitting with unobserved spatial predictors at the sites where outcomes are measured. The variability of the interpolated predictors at outcome sites results in potential problems of variable selection and averaging the results across different datasets. A Bayesian spatial model is developed to tackle this issue. By sampling the latent spatial predictors and selecting the spatial and non-spatial predictors using stochastic search variable selection Gibbs sampling algorithm, our approach allows for uncertainty of the predictors including the interpolated predictors. The approach is evaluated and illustrated through a simulated data example and an application to mental retardation and developmental delay in a Medicaid population in South Carolina with samples of soil chemistry.

Evaluation of Bayesian spatio-temporal latent models in small area health data

Health outcomes are linked to air pollution, demographic, or socioeconomic factors which vary across space and time. Thus, it is often found that relative risks in space-time health data have locally different temporal patterns. In such cases, latent modeling is useful in the disaggregation of risk profiles. In particular, spatio-temporal mixture models can help to isolate spatial clusters each of which has a homogeneous temporal pattern in relative risks. In mixture modeling, various weight structures can be used and two situations can be considered: the number of underlying components is known or unknown. In this paper, we compare spatio-temporal mixture models with different weight structures in both situations. In addition, spatio-temporal Dirichlet process mixture models are compared to them when the number of components is unknown. For comparison, we propose a set of spatial cluster detection diagnostics based on the posterior distribution of the weights. We also develop new accuracy measures to assess the recovery of true relative risks. Based on the simulation study, we examine the performance of various spatio-temporal mixture models in terms of proposed methods and goodness-of-fit measures. We apply our models to a county-level chronic obstructive pulmonary disease data set from the state of Georgia.

Spatio-temporal patterning of small area low birth weight incidence and its correlates: a latent spatial structure approach

Low birth weight (LBW) defined as infant weight at birth of less than 2500 g is a useful health outcome for exploring spatio-temporal variation and the role of covariates. LBW is a key measure of population health used by local, national and international health organizations. Yet its spatio-temporal patterns and their dependence structures are poorly understood. In this study we examine the use of flexible latent structure models for the analysis of spatio-temporal variation in LBW. Beyond the explanatory capabilities of well-known predictors, we observe spatio-temporal effects, which are not directly observable using conventional modeling approaches. Our analysis shows that for county-level counts of LBW in Georgia and South Carolina the proportion of black population is a positive risk factor while high-income is a negative risk factor. Two dominant residual temporal components are also estimated. Finally our proposed method provides a better goodness-of-fit to these data than the conventional space–time models.