Association of midlife air pollution exposures and residential road proximity with incident dementia: The Atherosclerosis Risk in Communities (ARIC) study.

BACKGROUND: Increasing evidence links higher air pollution exposures to increased risk of cognitive impairment. While midlife risk factors are often most strongly linked to dementia risk, few studies have considered associations between midlife roadway proximity or ambient air pollution exposure and incident dementia decades later, in late life. OBJECTIVES: Our objective was to determine if midlife exposures to ambient air pollution or roadway proximity are associated with increased risk of dementia in the Atherosclerosis Risk in Communities (ARIC) study over up to 29 years of follow-up. METHODS: Our eligible sample included Black and White ARIC participants without dementia at Visit 2 (1990-1992). Participants were followed through Visit 7 (2018-2019), with dementia status and onset date defined based on formal dementia ascertainment at study visits, informant interviews, and surveillance efforts. We used adjusted Weibull survival models to assess the associations of midlife ambient air pollution and road proximity with incident dementia. RESULTS: The median age at baseline (1990-1992, Visit 2) of the 12,700 eligible ARIC participants was 57.0 years; 56.0% were female, 24.2% were Black, and 78.9% had at least a high school education. Over up to 29 years of follow-up, 2511 (19.8%) persons developed dementia. No associations were found between ambient air pollutants and proximity to major roadways with risk of incident dementia. In exploratory analyses, living closer to roadways in midlife increased dementia risk in individuals younger at baseline and those without midlife hypertension, and there was evidence of increased risk of dementia with increased midlife exposure to NOx, several PM2.5 components, and trace metals among those with diabetes in midlife. CONCLUSIONS: Midlife exposure to ambient air pollution and midlife roadway proximity was not associated with dementia risk over decades of follow-up. Further investigation to explore potential for greater susceptibility among specific subgroups identified here is needed.

A comparison of PM2.5 exposure estimates from different estimation methods and their associations with cognitive testing and brain MRI outcomes.

BACKGROUND: Reported associations between particulate matter with aerodynamic diameter ≤2.5 µm (PM2.5) and cognitive outcomes remain mixed. Differences in exposure estimation method may contribute to this heterogeneity. OBJECTIVES: To assess agreement between PM2.5 exposure concentrations across 11 exposure estimation methods and to compare resulting associations between PM2.5 and cognitive or MRI outcomes. METHODS: We used Visit 5 (2011-2013) cognitive testing and brain MRI data from the Atherosclerosis Risk in Communities (ARIC) Study. We derived address-linked average 2000-2007 PM2.5 exposure concentrations in areas immediately surrounding the four ARIC recruitment sites (Forsyth County, NC; Jackson, MS; suburbs of Minneapolis, MN; Washington County, MD) using 11 estimation methods. We assessed agreement between method-specific PM2.5 concentrations using descriptive statistics and plots, overall and by site. We used adjusted linear regression to estimate associations of method-specific PM2.5 exposure estimates with cognitive scores (n = 4678) and MRI outcomes (n = 1518) stratified by study site and combined site-specific estimates using meta-analyses to derive overall estimates. We explored the potential impact of unmeasured confounding by spatially patterned factors. RESULTS: Exposure estimates from most methods had high agreement across sites, but low agreement within sites. Within-site exposure variation was limited for some methods. Consistently null findings for the PM2.5-cognitive outcome associations regardless of method precluded empirical conclusions about the potential impact of method on study findings in contexts where positive associations are observed. Not accounting for study site led to consistent, adverse associations, regardless of exposure estimation method, suggesting the potential for substantial bias due to residual confounding by spatially patterned factors. DISCUSSION: PM2.5 estimation methods agreed across sites but not within sites. Choice of estimation method may impact findings when participants are concentrated in small geographic areas. Understanding unmeasured confounding by factors that are spatially patterned may be particularly important in studies of air pollution and cognitive or brain health.

The Associations of Dietary Copper With Cognitive Outcomes.

Dietary copper intake may be associated with cognitive decline and dementia. We used data from 10,269 participants of the Atherosclerosis Risks in Communities Study to study the associations of dietary copper intake with 20-year cognitive decline and incident dementia. Dietary copper intake from food and supplements was quantified using food frequency questionnaires. Cognition was assessed using 3 cognitive tests at study visits; dementia was ascertained at study visits and via surveillance. Multiple imputation by chained equations was applied to account for the missing information of cognitive function during follow-up. Survival analysis with parametric models and mixed-effect models were used to estimate the associations for incident dementia and cognitive decline, respectively. During 20 years of follow-up (1996-1998 to 2016-2017), 1,862 incident cases of dementia occurred. Higher intake of dietary copper from food was associated with higher risk of incident dementia among those with high intake of saturated fat (hazard ratio = 1.49, 95% confidence interval: 1.04, 1.95). Higher intake of dietary copper from food was associated with greater decline in language overall (beta = -0.12, 95% confidence interval: -0.23, -0.02). Therefore, a diet high in copper, particularly when combined with a diet high in saturated fat, may increase the risk of cognitive impairment.

Assessment of source-specific health effects associated with an unknown number of major sources of multiple air pollutants: a unified Bayesian approach.

There has been increasing interest in assessing health effects associated with multiple air pollutants emitted by specific sources. A major difficulty with achieving this goal is that the pollution source profiles are unknown and source-specific exposures cannot be measured directly; rather, they need to be estimated by decomposing ambient measurements of multiple air pollutants. This estimation process, called multivariate receptor modeling, is challenging because of the unknown number of sources and unknown identifiability conditions (model uncertainty). The uncertainty in source-specific exposures (source contributions) as well as uncertainty in the number of major pollution sources and identifiability conditions have been largely ignored in previous studies. A multipollutant approach that can deal with model uncertainty in multivariate receptor models while simultaneously accounting for parameter uncertainty in estimated source-specific exposures in assessment of source-specific health effects is presented in this paper. The methods are applied to daily ambient air measurements of the chemical composition of fine particulate matter ([Formula: see text]), weather data, and counts of cardiovascular deaths from 1995 to 1997 for Phoenix, AZ, USA. Our approach for evaluating source-specific health effects yields not only estimates of source contributions along with their uncertainties and associated health effects estimates but also estimates of model uncertainty (posterior model probabilities) that have been ignored in previous studies. The results from our methods agreed in general with those from the previously conducted workshop/studies on the source apportionment of PM health effects in terms of number of major contributing sources, estimated source profiles, and contributions. However, some of the adverse source-specific health effects identified in the previous studies were not statistically significant in our analysis, which probably resulted because we incorporated parameter uncertainty in estimated source contributions that has been ignored in the previous studies into the estimation of health effects parameters.

Part 2. Development of Enhanced Statistical Methods for Assessing Health Effects Associated with an Unknown Number of Major Sources of Multiple Air Pollutants.

A major difficulty with assessing source-specific health effects is that source-specific exposures cannot be measured directly; rather, they need to be estimated by a source-apportionment method such as multivariate receptor modeling. The uncertainty in source apportionment (uncertainty in source-specific exposure estimates and model uncertainty due to the unknown number of sources and identifiability conditions) has been largely ignored in previous studies. Also, spatial dependence of multipollutant data collected from multiple monitoring sites has not yet been incorporated into multivariate receptor modeling. The objectives of this project are (1) to develop a multipollutant approach that incorporates both sources of uncertainty in source-apportionment into the assessment of source-specific health effects and (2) to develop enhanced multivariate receptor models that can account for spatial correlations in the multipollutant data collected from multiple sites. We employed a Bayesian hierarchical modeling framework consisting of multivariate receptor models, health-effects models, and a hierarchical model on latent source contributions. For the health model, we focused on the time-series design in this project. Each combination of number of sources and identifiability conditions (additional constraints on model parameters) defines a different model. We built a set of plausible models with extensive exploratory data analyses and with information from previous studies, and then computed posterior model probability to estimate model uncertainty. Parameter estimation and model uncertainty estimation were implemented simultaneously by Markov chain Monte Carlo (MCMC*) methods. We validated the methods using simulated data. We illustrated the methods using PM2.5 (particulate matter ≤ 2.5 µm in aerodynamic diameter) speciation data and mortality data from Phoenix, Arizona, and Houston, Texas. The Phoenix data included counts of cardiovascular deaths and daily PM2.5 speciation data from 1995-1997. The Houston data included respiratory mortality data and 24-hour PM2.5 speciation data sampled every six days from a region near the Houston Ship Channel in years 2002-2005. We also developed a Bayesian spatial multivariate receptor modeling approach that, while simultaneously dealing with the unknown number of sources and identifiability conditions, incorporated spatial correlations in the multipollutant data collected from multiple sites into the estimation of source profiles and contributions based on the discrete process convolution model for multivariate spatial processes. This new modeling approach was applied to 24-hour ambient air concentrations of 17 volatile organic compounds (VOCs) measured at nine monitoring sites in Harris County, Texas, during years 2000 to 2005. Simulation results indicated that our methods were accurate in identifying the true model and estimated parameters were close to the true values. The results from our methods agreed in general with previous studies on the source apportionment of the Phoenix data in terms of estimated source profiles and contributions. However, we had a greater number of statistically insignificant findings, which was likely a natural consequence of incorporating uncertainty in the estimated source contributions into the health-effects parameter estimation. For the Houston data, a model with five sources (that seemed to be Sulfate-Rich Secondary Aerosol, Motor Vehicles, Industrial Combustion, Soil/Crustal Matter, and Sea Salt) showed the highest posterior model probability among the candidate models considered when fitted simultaneously to the PM2.5 and mortality data. There was a statistically significant positive association between respiratory mortality and same-day PM2.5 concentrations attributed to one of the sources (probably industrial combustion). The Bayesian spatial multivariate receptor modeling approach applied to the VOC data led to a highest posterior model probability for a model with five sources (that seemed to be refinery, petrochemical production, gasoline evaporation, natural gas, and vehicular exhaust) among several candidate models, with the number of sources varying between three and seven and with different identifiability conditions. Our multipollutant approach assessing source-specific health effects is more advantageous than a single-pollutant approach in that it can estimate total health effects from multiple pollutants and can also identify emission sources that are responsible for adverse health effects. Our Bayesian approach can incorporate not only uncertainty in the estimated source contributions, but also model uncertainty that has not been addressed in previous studies on assessing source-specific health effects. The new Bayesian spatial multivariate receptor modeling approach enables predictions of source contributions at unmonitored sites, minimizing exposure misclassification and providing improved exposure estimates along with their uncertainty estimates, as well as accounting for uncertainty in the number of sources and identifiability conditions.

Association of Gaseous Ambient Air Pollution and Dementia-Related Neuroimaging Markers in the ARIC Cohort, Comparing Exposure Estimation Methods and Confounding by Study Site.

BACKGROUND: Evidence linking gaseous air pollution to late-life brain health is mixed. OBJECTIVE: We explored associations between exposure to gaseous pollutants and brain magnetic resonance imaging (MRI) markers among Atherosclerosis Risk in Communities (ARIC) Study participants, with attention to the influence of exposure estimation method and confounding by site. METHODS: We considered data from 1,665 eligible ARIC participants recruited from four US sites in the period 1987-1989 with valid brain MRI data from Visit 5 (2011-2013). We estimated 10-y (2001-2010) mean carbon monoxide (CO), nitrogen dioxide (NO2), nitrogen oxides (NOx), and 8- and 24-h ozone (O3) concentrations at participant addresses, using multiple exposure estimation methods. We estimated site-specific associations between pollutant exposures and brain MRI outcomes (total and regional volumes; presence of microhemorrhages, infarcts, lacunes, and severe white matter hyperintensities), using adjusted linear and logistic regression models. We compared meta-analytically combined site-specific associations to analyses that did not account for site. RESULTS: Within-site exposure distributions varied across exposure estimation methods. Meta-analytic associations were generally not statistically significant regardless of exposure, outcome, or exposure estimation method; point estimates often suggested associations between higher NO2 and NOx and smaller temporal lobe, deep gray, hippocampal, frontal lobe, and Alzheimer disease signature region of interest volumes and between higher CO and smaller temporal and frontal lobe volumes. Analyses that did not account for study site more often yielded significant associations and sometimes different direction of associations. DISCUSSION: Patterns of local variation in estimated air pollution concentrations differ by estimation method. Although we did not find strong evidence supporting impact of gaseous pollutants on brain changes detectable by MRI, point estimates suggested associations between higher exposure to CO, NOx, and NO2 and smaller regional brain volumes. Analyses of air pollution and dementia-related outcomes that do not adjust for location likely underestimate uncertainty and may be susceptible to confounding bias. https://doi.org/10.1289/EHP13906.

Comparison of PM2.5 Air Pollution Exposures and Health Effects Associations Using 11 Different Modeling Approaches in the Women's Health Initiative Memory Study (WHIMS).

BACKGROUND: Many approaches to quantifying air pollution exposures have been developed. However, the impact of choice of approach on air pollution estimates and health-effects associations remains unclear. OBJECTIVES: Our objective is to compare particulate matter with aerodynamic diameter ≤2.5µm (PM2.5) concentrations and resulting health effects associations using multiple estimation approaches previously used in epidemiologic analyses. METHODS: We assigned annual PM2.5 exposure estimates from 1999 to 2004 derived from 11 different approaches to Women's Health Initiative Memory Study (WHIMS) participant addresses within the contiguous US. Approaches included geostatistical interpolation approaches, land-use regression or spatiotemporal models, satellite-derived approaches, air dispersion and chemical transport models, and hybrid models. We used descriptive statistics and plots to assess relative and absolute agreement among exposure estimates and examined the impact of approach on associations between PM2.5 and death due to natural causes, cardiovascular disease (CVD) mortality, and incident CVD events, adjusting for individual-level covariates and climate-based region. RESULTS: With a few exceptions, relative agreement of approach-specific PM2.5 exposure estimates was high for PM2.5 concentrations across the contiguous US. Agreement among approach-specific exposure estimates was stronger near PM2.5 monitors, in certain regions of the country, and in 2004 vs. 1999. Collectively, our results suggest but do not quantify lower agreement at local spatial scales for PM2.5. There was no evidence of large differences in health effects associations with PM2.5 among estimation approaches in analyses adjusted for climate region. CONCLUSIONS: Different estimation approaches produced similar spatial patterns of PM2.5 concentrations across the contiguous US and in areas with dense monitoring data, and PM2.5-health effects associations were similar among estimation approaches. PM2.5 estimates and PM2.5-health effects associations may differ more in samples drawn from smaller areas or areas without substantial monitoring data, or in analyses with finer adjustment for participant location. Our results can inform decisions about PM2.5 estimation approach in epidemiologic studies, as investigators balance concerns about bias, efficiency, and resource allocation. Future work is needed to understand whether these conclusions also apply in the context of other air pollutants of interest. https://doi.org/10.1289/EHP12995.

Nonparametric Bayesian functional selection in 1-M matched case-crossover studies.

The matched case-crossover study has been used in many areas such as public health, biomedical, and epidemiological research for humans, animals, and other subjects with clustered binary outcomes. The control information for each stratum is based on the subject's exposure experience, and the stratifying variable is the individual subject. It is generally accepted that any effects associated with the matching covariates by stratum can be removed in the conditional logistic regression model. However, when there are numerous covariates, it is important to perform variable selection to study the functional association between the variables and the relative risk of diseases or clustered binary outcomes by simultaneously adjusting effect modifications. The methods for simultaneously evaluating effect modifications by matching covariates such as time, as well as performing automatic variable and functional selections under semiparametric model frameworks, are quite limited. In this article, we propose a unified Bayesian approach due to its ability to detect both parametric and nonparametric relationships between the predictors and the relative risk of diseases or binary outcomes, accounting for potential effect modifications by matching covariates such as time, and perform automatic variable and functional selections. We demonstrate the advantages of our approach using simulation study and an epidemiological example of a 1-4 bidirectional case-crossover study.

Predicting latent source-specific PM2.5 pollution from regional sources at unmonitored sites by Bayesian spatial multivariate receptor modeling.

Fine particulate matter (PM2.5) has been a pollutant of main interest globally for more than two decades, owing to its well-known adverse health effects. For developing effective management strategies for PM2.5, it is vital to identify its major sources and quantify how much they contribute to ambient PM2.5 concentrations. With the expanded monitoring efforts established during recent decades in Korea, speciated PM2.5 data needed for source apportionment of PM2.5 are now available for multiple sites (cities). However, many cities in Korea still do not have any speciated PM2.5 monitoring station, although quantification of source contributions for those cities is in great need. While there have been many PM2.5 source apportionment studies throughout the world for several decades based on monitoring data collected from receptor site(s), none of those receptor-oriented studies could predict unobserved source contributions at unmonitored sites. This study predicts source contributions of PM2.5 at unmonitored locations using a recently developed novel spatial multivariate receptor modeling (BSMRM) approach, which incorporates spatial correlation in data into modeling and estimation for spatial prediction of latent source contributions. The validity of BSMRM results is also assessed based on the data from a test site (city), not used in model development and estimation.

The association of long-term exposure to criteria air pollutants, fine particulate matter components, and airborne trace metals with late-life brain amyloid burden in the Atherosclerosis Risk in Communities (ARIC) study.

BACKGROUND: Studies suggest associations between long-term ambient air pollution exposure and outcomes related to Alzheimer's disease (AD). Whether a link exists between pollutants and brain amyloid accumulation, a biomarker of AD, is unclear. We assessed whether long-term air pollutant exposures are associated with late-life brain amyloid deposition in Atherosclerosis Risk in Communities (ARIC) study participants. METHODS: We used a chemical transport model with data fusion to estimate ambient concentrations of PM2.5 and its components, NO2, NOx, O3 (24-hour and 8-hour), CO, and airborne trace metals. We linked concentrations to geocoded participant addresses and calculated 10-year mean exposures (2002 to 2011). Brain amyloid deposition was measured using florbetapir amyloid positron emission tomography (PET) scans in 346 participants without dementia in 2012-2014, and we defined amyloid positivity as a global cortical standardized uptake value ratio ≥ the sample median of 1.2. We used logistic regression models to quantify the association between amyloid positivity and each air pollutant, adjusting for putative confounders. In sensitivity analyses, we considered whether use of alternate air pollution estimation approaches impacted findings for PM2.5, NO2, NOx, and 24-hour O3. RESULTS: At PET imaging, eligible participants (N = 318) had a mean age of 78 years, 56% were female, 43% were Black, and 27% had mild cognitive impairment. We did not find evidence of associations between long-term exposure to any pollutant and brain amyloid positivity in adjusted models. Findings were materially unchanged in sensitivity analyses using alternate air pollution estimation approaches for PM2.5, NO2, NOx, and 24-hour O3. CONCLUSIONS: Air pollution may impact cognition and dementia independent of amyloid accumulation, though whether air pollution influences AD pathogenesis later in the disease course or at higher exposure levels deserves further consideration.

Evaluating the Performance of Low-Cost Air Quality Monitors in Dallas, Texas.

The emergence of low-cost air quality sensors may improve our ability to capture variations in urban air pollution and provide actionable information for public health. Despite the increasing popularity of low-cost sensors, there remain some gaps in the understanding of their performance under real-world conditions, as well as compared to regulatory monitors with high accuracy, but also high cost and maintenance requirements. In this paper, we report on the performance and the linear calibration of readings from 12 commercial low-cost sensors co-located at a regulatory air quality monitoring site in Dallas, Texas, for 18 continuous measurement months. Commercial AQY1 sensors were used, and their reported readings of O3, NO2, PM2.5, and PM10 were assessed against a regulatory monitor. We assessed how well the raw and calibrated AQY1 readings matched the regulatory monitor and whether meteorology impacted performance. We found that each sensor's response was different. Overall, the sensors performed best for O3 (R2 = 0.36-0.97) and worst for NO2 (0.00-0.58), showing a potential impact of meteorological factors, with an effect of temperature on O3 and relative humidity on PM. Calibration seemed to improve the accuracy, but not in all cases or for all performance metrics (e.g., precision versus bias), and it was limited to a linear calibration in this study. Our data showed that it is critical for users to regularly calibrate low-cost sensors and monitor data once they are installed, as sensors may not be operating properly, which may result in the loss of large amounts of data. We also recommend that co-location should be as exact as possible, minimizing the distance between sensors and regulatory monitors, and that the sampling orientation is similar. There were important deviations between the AQY1 and regulatory monitors' readings, which in small part depended on meteorology, hindering the ability of the low-costs sensors to present air quality accurately. However, categorizing air pollution levels, using for example the Air Quality Index framework, rather than reporting absolute readings, may be a more suitable approach. In addition, more sophisticated calibration methods, including accounting for individual sensor performance, may further improve performance. This work adds to the literature by assessing the performance of low-cost sensors over one of the longest durations reported to date.

Assessment of inequity in bicyclist crashes using bivariate Bayesian copulas.

INTRODUCTION: Physical activity associated with active transport modes such as bicycling has major health benefits and can help to reduce health concerns related to sedentary lifestyles, such as cardiovascular disease, Type II diabetes, and obesity, as well as risks of colon and breast cancer, high blood pressure, lipid disorders, osteoporosis, depression, and anxiety. However, as a vulnerable user group, bicyclists experience negative health impacts of transportation policies and infrastructure, such as traffic crashes and exposure to air and noise pollution that is disproportionately distributed within low-income and underserved areas. METHOD: This study used aggregated (block-group) bicyclist crash data from Harris County, Texas, to analyze how various equity measures are associated with both fatal and injury (FI) and no injury (property damage only) bicyclist crashes that occurred from 2010 to 2017. We used Bayesian bivariate copula-based random effects regression analysis to evaluate these associations. In contrast to more traditional univariate analysis, this novel methodology can consider the effects of factors of interest across different severity levels or crash types to fully understand their effects and how they may differ across categories. RESULTS: The analysis results indicate that the bicyclist exposure, vehicle exposure, population demographics, population density, the percentage of African-Americans, and households below the poverty level are associated with both FI and PDO bicyclist crashes. CONCLUSIONS: Although more location and context-specific analyses are required, this study's overall results once again conform with the findings and assumptions in bicycling safety literature that the low-income and racially diverse communities are prone to experience more bicyclist crashes. PRACTICAL APPLICATIONS: The findings of this study may have implications for future transportation and planning policies. These findings can be used to guide the policies and strategies targeting the elimination of inequity in transportation-related health concerns.

Characteristics of movers and predictors of residential mobility in the Atherosclerosis Risk in Communities (ARIC) cohort.

Current efforts to characterize movers and identify predictors of moving have been limited. We used the ARIC cohort to characterize non-movers, short-distance movers, and long-distance movers, and employed best subset algorithms to identify important predictors of moving, including interactions between characteristics. Short- and long-distance movers were notably different from non-movers, and important predictors of moving differed based on the distance of the residential move. Importantly, systematic inclusion of interaction terms enhanced model fit and was substantively meaningful. This work has important implications for epidemiologic studies of contextual exposures and those treating residential mobility as an exposure.

Nighttime effectiveness of the pedestrian hybrid beacon, rectangular rapid flashing beacon, and LED-embedded crossing sign.

INTRODUCTION: A large majority of pedestrian fatal crashes occurred during the nighttime. The focus of this research was to identify if the following pedestrian crossing treatments were more or less effective at night: pedestrian hybrid beacon (PHB), rectangular rapid flashing beacon (RRFB), or LED-embedded crossing warning sign (LED-Em). METHOD: For each treatment, two statistical evaluations were used on the staged pedestrian data: ANCOVA models that considered per site mean yield rates and logistic regression that considered the individual driver response to the crossing pedestrian. RESULTS: For the PHB, essentially no difference was found between the very high daytime and nighttime driver yielding values. The research found RRFBs to be more effective at night, and the LED-Em to be more effective during the day. Using the results from the logistic regression evaluation, higher driver yielding was observed at LED-Em sites in the lower speed limit group (30 or 35 mph (48.3 or 56.3 kph), with 2 lanes (rather than 4 lanes), with narrow lanes of 10.5 or 11 ft (3.2 or 3.4 m) widths (rather than 11.5 or 12 ft (3.5 or 3.7 m) widths), and lower hourly volumes. The results from the ANCOVA model for LED-Ems also showed a statistically significant difference for yield lines (higher yielding when present). CONCLUSIONS: This analysis represents the only known study to date on the effectiveness of pedestrian crossing treatments at night. Practical Applications: This study provides additional support for the PHB as a treatment because the PHB was found to be highly effective during the nighttime as well as the daytime. The value of using advance yield lines was also demonstrated. The findings offer a caution regarding the use of the LED-Em treatment on higher speed, higher volume, or wider roads.

Bayesian analysis of multivariate crash counts using copulas.

There has been growing interest in jointly modeling correlated multivariate crash counts in road safety research over the past decade. To assess the effects of roadway characteristics or environmental factors on crash counts by severity level or by collision type, various models including multivariate Poisson regression models, multivariate negative binomial regression models, and multivariate Poisson-Lognormal regression models have been suggested. We introduce more general copula-based multivariate count regression models with correlated random effects within a Bayesian framework. Our models incorporate the dependence among the multivariate crash counts by modeling multivariate random effects using copulas. Copulas provide a flexible way to construct valid multivariate distributions by decomposing any joint distribution into a copula and the marginal distributions. Overdispersion as well as general correlation structures including both positive and negative correlations in multivariate crash counts can easily be accounted for by this approach. Our copular-based models can also encompass previously suggested multivariate count regression models including multivariate Poisson-Gamma mixture models and multivariate Poisson-Lognormal regression models. The proposed method is illustrated with crash count data of five different severity levels collected from 451 three-leg unsignalized intersections in California.

Exploration of the relationship among roadway characteristics, operating speed, and crashes for city streets using path analysis.

Estimating the speed-crash relationship has long been a focus area of interest in roadway safety analysis. Because of many confounding factors that may influence both speeds and crashes, the relationship cannot be appropriately established without considering the corresponding roadway contexts and accounting for their effects on speeds and crashes. This paper investigates the speed-crash relationship for city streets by jointly modeling speed, roadway characteristics, and crashes using a path analysis approach that has been recently introduced into safety analysis while incorporating a wide range of roadway and traffic related variables and additional speed measures. The results from the coherent path analysis identified multiple speed measures of interest that have a statistically significant association with crashes as well as having intuitive and useful interpretation. The results also supported a positive relationship between speed variability and crash occurrence (i.e., larger spread/variability in operational speed is associated with more crashes).

Investigation of crashes at pedestrian hybrid beacons: Results of a large-scale study in Arizona.

INTRODUCTION: The pedestrian hybrid beacon (PHB) is a traffic control device used at pedestrian crossings. A recent Arizona Department of Transportation research effort investigated changes in crashes for different severity levels and crash types (e.g., rear-end crashes) due to the PHB presence, as well as for crashes involving pedestrians and bicycles. METHOD: Two types of methodologies were used to evaluate the safety of PHBs: (a) an Empirical Bayes (EB) before-after study, and (b) a long-term cross-sectional observational study. For the EB before-after evaluation, the research team considered three reference groups: unsignalized intersections, signalized intersections, and both unsignalized and signalized intersections combined. RESULTS: For the signalized and combined unsignalized and signalized intersection groups, all crash types considered showed statistically significant reductions in crashes (e.g., total crashes, fatal and injury crashes, rear-end crashes, fatal and injury rear-end crashes, angle crashes, fatal and injury angle crashes, pedestrian-related crashes, and fatal and injury pedestrian-related crashes). A cross-sectional study was conducted with a larger number of PHBs (186) to identify relationships between roadway characteristics and crashes at PHBs, especially with respect to the distance to an adjacent traffic control signal. The distance to an adjacent traffic signal was found to be significant only at the α = 0.1 level, and only for rear-end and fatal and injury rear-end crashes. CONCLUSIONS: This analysis represents the largest known study to date on the safety impacts of PHBs, along with a focus on how crossing and geometric characteristics affect crash patterns. The study showed the safety benefits of PHBs for both pedestrians and vehicles. Practical Applications: The findings from this study clearly support the installation of PHBs at midblock or intersection crossings, as well as at crossings on higher-speed roads.

A multiple linear regression model with multiplicative log-normal error term for atmospheric concentration data.

The homoscedasticity assumption (the variance of the error term is the same across all the observations) is a key assumption in the ordinary linear squares (OLS) solution of a linear regression model. The validity of this assumption is examined for a multiple linear regression model used to determine the source contributions to the observed black carbon concentrations at 12 background monitoring sites across China using a hybrid modeling approach. Residual analysis from the traditional OLS method, which assumes that the error term is additive and normally distributed with a mean of zero, shows pronounced heteroscedasticity based on the Breusch-Pagan test for 11 datasets. Noticing that the atmospheric black carbon data are log-normally distributed, we make a new assumption that the error terms are multiplicative and log-normally distributed. When the coefficients of the multilinear regression model are determined using the maximum likelihood estimation (MLE), the distribution of the residuals in 8 out of the 12 datasets is in good accordance with the revised assumption. Furthermore, the MLE computation under this novel assumption could be proved mathematically identical to minimizing a log-scale objective function, which considerably reduces the complexity in the MLE calculation. The new method is further demonstrated to have clear advantages in numerical simulation experiments of a 5-variable multiple linear regression model using synthesized data with prescribed coefficients and lognormally distributed multiplicative errors. Under all 9 simulation scenarios, the new method yields the most accurate estimations of the regression coefficients and has significantly higher coverage probability (on average, 95% for all five coefficients) than OLS (79%) and weighted least squares (WLS, 72%) methods.

Assessment of mobile source contributions in El Paso by PMF receptor modeling coupled with wind direction analysis.

It is well-known that El Paso is the only border area in Texas that has violated national air quality standards. Mobile source emissions (including vehicle exhaust) contribute significantly to air pollution, along with other sources including industrial, residential, and cross-border. This study aims at separating unobserved vehicle emissions from air-pollution mixtures indicated by ambient air quality data. The level of contributions from vehicle emissions to air pollution cannot be determined by simply comparing ambient air quality data with traffic levels because of the various other contributors to overall air pollution. To estimate contributions from vehicle emissions, researchers employed advanced multivariate receptor modeling called positive matrix factorization (PMF) to analyze hydrocarbon data consisting of hourly concentrations measured from the Chamizal air pollution monitoring station in El Paso. The analysis of hydrocarbon data collected at the Chamizal site in 2008 showed that approximately 25% of measured Total Non-Methane Hydrocarbons (TNMHC) was apportioned to motor vehicle exhaust. Using wind direction analysis, researchers also showed that the motor vehicle exhaust contributions to hydrocarbons were significantly higher when winds blow from the south (Mexico) than those when winds blow from other directions. The results from this research can be used to improve understanding source apportionment of pollutants measured in El Paso and can also potentially inform transportation planning strategies aimed at reducing emissions across the region.

Traffic-related air emissions in Houston: Effects of light-rail transit.

Recent studies have suggested that automobile pollution poses significantly more harmful health impacts than previously realized. Light-rail transit (LRT) is a major type of transportation infrastructure, but there has been little research assessing the air quality effects of LRT based on the actual air pollution data. This study aimed to assess the effects of LRT on automobile-related air emissions in Houston. Specifically, we examined the effects of LRT on key tailpipe pollutants-carbon monoxide and acetylene-as well as other traffic pollution surrogates referred to as BTEX (benzene, toluene, ethylbenzene, and xylene), measured from ambient air monitoring stations. An interrupted time series design and analysis was used to determine the impact of an intervention, where the intervention was the opening of an LRT on January 1, 2004, with two years (2002-2003) of before and two years (2004-2005) of after period data. We found that, after controlling for weather, the opening of the LRT was associated with statistically significant reductions in traffic-related air emissions. Specifically, at the exposure sites, the daily maximum carbon monoxide level was reduced roughly by 24%, and the daily level of toluene was reduced roughly by 60% (33% after accounting for the reduction at the comparison site). Our findings lend support to the air quality benefits of LRT by providing suggestive evidence of positive effects of LRT based on actual air pollution monitoring data. This study's findings also emphasize the importance of developing effective measures to assess traffic-related pollution and call for advanced data collection strategies of additional data, including traffic volume and speed data.