Jonathan I. Levy

Science and decisions: advancing risk assessment.

At the request of the U.S. Environmental Protection Agency (EPA), the National Research Council (NRC) recently completed a major report, Science and Decisions: Advancing Risk Assessment, that is intended to strengthen the scientific basis, credibility, and effectiveness of risk assessment practices and subsequent risk management decisions. The report describes the challenges faced by risk assessment and the need to consider improvements in both the technical analyses of risk assessments (i.e., the development and use of scientific information to improve risk characterization) and the utility of risk assessments (i.e., making assessments more relevant and useful for risk management decisions). The report tackles a number of topics relating to improvements in the process, including the design and framing of risk assessments, uncertainty and variability characterization, selection and use of defaults, unification of cancer and noncancer dose-response assessment, cumulative risk assessment, and the need to increase EPAs capacity to address these improvements. This article describes and summarizes the NRC report, with an eye toward its implications for risk assessment practices at EPA.

Ozone Exposure and Mortality: An Empiric Bayes Metaregression Analysis

Background: Results from time-series epidemiologic studies evaluating the relationship between ambient ozone concentrations and premature mortality vary in their conclusions about the magnitude of this relationship, if any, making it difficult to estimate public health benefits of air pollution control measures. We conducted an empiric Bayes metaregression to estimate the ozone effect on mortality, and to assess whether this effect varies as a function of hypothesized confounders or effect modifiers. Methods: We gathered 71 time-series studies relating ozone to all-cause mortality, and we selected 48 estimates from 28 studies for the metaregression. Metaregression covariates included the relationship between ozone concentrations and concentrations of other air pollutants, proxies for personal exposure–ambient concentration relationships, and the statistical methods used in the studies. For our metaregression, we applied a hierarchical linear model with known level-1 variances. Results: We estimated a grand mean of a 0.21% increase (95% confidence interval = 0.16–0.26%) in mortality per 10-μg/m3 increase of 1-hour maximum ozone (0.41% increase per 10 ppb) without controlling for other air pollutants. In the metaregression, air-conditioning prevalence and lag time were the strongest predictors of between-study variability. Air pollution covariates yielded inconsistent findings in regression models, although correlation analyses indicated a potential influence of summertime PM2.5. Conclusions: These findings, coupled with a greater relative risk of ozone in the summer versus the winter, demonstrate that geographic and seasonal heterogeneity in ozone relative risk should be anticipated, but that the observed relationship between ozone and mortality should be considered for future regulatory impact analyses.

Synergistic Effects of Traffic-Related Air Pollution and Exposure to Violence on Urban Asthma Etiology

Background Disproportionate life stress and consequent physiologic alteration (i.e., immune dysregulation) has been proposed as a major pathway linking socioeconomic position, environmental exposures, and health disparities. Asthma, for example, disproportionately affects lower-income urban communities, where air pollution and social stressors may be elevated. Objectives We aimed to examine the role of exposure to violence (ETV), as a chronic stressor, in altering susceptibility to traffic-related air pollution in asthma etiology. Methods We developed geographic information systems (GIS)–based models to retrospectively estimate residential exposures to traffic-related pollution for 413 children in a community-based pregnancy cohort, recruited in East Boston, Massachusetts, between 1987 and 1993, using monthly nitrogen dioxide measurements for 13 sites over 18 years. We merged pollution estimates with questionnaire data on lifetime ETV and examined the effects of both on childhood asthma etiology. Results Correcting for potential confounders, we found an elevated risk of asthma with a 1-SD (4.3 ppb) increase in NO2 exposure solely among children with above-median ETV [odds ratio (OR) = 1.63; 95% confidence interval (CI), 1.14–2.33)]. Among children always living in the same community, with lesser exposure measurement error, this association was magnified (OR = 2.40; 95% CI, 1.48–3.88). Of multiple exposure periods, year-of-diagnosis NO2 was most predictive of asthma outcomes. Conclusions We found an association between traffic-related air pollution and asthma solely among urban children exposed to violence. Future studies should consider socially patterned susceptibility, common spatial distributions of social and physical environmental factors, and potential synergies among these. Prospective assessment of physical and social exposures may help determine causal pathways and critical exposure periods.

Impact of residential nitrogen dioxide exposure on personal exposure: an international study.

Nitrogen dioxide (NO2) concentrations were measured during two-day winter periods in indoor and outdoor environments, and these concentrations were compared with simultaneously measured personal exposures in 18 cities in 15 countries around the world. Information was also gathered on activity patterns and household characteristics in order to determine the influences of these factors on personal exposures. All NO2 measurements were taken using passive filter badges. Personal exposures were found to vary greatly among the array of cities, with mean concentrations ranging between 11.0 ppb and 51.5 ppb. Personal NO2 exposures were more strongly correlated with indoor concentrations (r = 0.75) than with outdoor concentrations (r = 0.57) when all countries were considered simultaneously. Use of a gas stove in the home was the dominant activity influencing NO2 concentrations, with a 67% increase in mean personal NO2 exposure and an increase in indoor-outdoor ratios from 0.7 to 1.2 for participants using gas stoves, although preliminary evidence indicates the importance of combustion space heaters as well. These associations indicate the global nature of the correlation between personal NO2 exposures and indoor NO2 sources such as gas stoves or space heaters, demonstrating that this relationship is not dependent on country-specific parameters.

Using CALPUFF to evaluate the impacts of power plant emissions in Illinois: model sensitivity and implications

Air pollution emissions from older fossil-fueled power plants are often much greater than emissions from newer facilities, in part because older plants are exempt from modern emission standards required of new plants under the Clean Air Act. To quantify potential health benefits of emission reductions, there is a need to apply atmospheric dispersion models that can estimate the incremental contributions of power plants to ambient concentrations with reasonable accuracy over long distances. We apply the CALPUFF atmospheric dispersion model with meteorological data derived from NOAA’s Rapid Update Cycle model to a set of nine power plants in Illinois to evaluate primary and secondary particulate matter impacts across a grid in the Midwest. In total, the population-weighted annual average concentration increments associated with current emissions are estimated to be 0.04m gm 3 of primary fine particulate matter (PM2.5), 0.13m gm 3 of secondary sulfate particles, and 0.10m gm 3 of secondary nitrate particles (maximum impacts of 0.3, 0.2, and 0.2m gm 3 , respectively). The aggregate impact estimates are moderately insensitive to parametric assumptions about chemicalmechanism, wet/dry deposition, background ammonia concentrations, and size of the receptor region, with the largest uncertainties related to nitrate particles and long-range transport issues. Additional uncertainties may be associated with inherent limitations of CALPUFF, but it appears likely that the degree of uncertainty in atmospheric modeling will not dominate the total uncertainty associated with health impact or benefit estimation. Although the annual average concentration increments from a limited number of sources are relatively small, the large population affected by long-range transport and the number of power plant sources around the US imply potentially significant public health impacts using standard epidemiological assumptions. Our analysis demonstrates an approach that is applicable in any setting where source controls are being evaluated from a public health or benefit-cost perspective. r 2002 Elsevier Science Ltd. All rights reserved.

Ranking cancer risks of organic hazardous air pollutants in the United States.

Background In this study we compared cancer risks from organic hazardous air pollutants (HAPs) based on total personal exposure summed across different microenvironments and exposure pathways. Methods We developed distributions of personal exposure concentrations using field monitoring and modeling data for inhalation and, where relevant, ingestion pathways. We calculated risks for a nonoccupationally exposed and nonsmoking population using U.S. Environmental Protection Agency (EPA) and California Office of Environmental Health and Hazard Assessment (OEHHA) unit risks. We determined the contribution to risk from indoor versus outdoor sources using indoor/outdoor ratios for gaseous compounds and the infiltration factor for particle-bound compounds. Results With OEHHA’s unit risks, the highest ranking compounds based on the population median are 1,3-butadiene, formaldehyde, benzene, and dioxin, with risks on the order of 10−4–10−5. The highest risk compounds with the U.S. EPA unit risks were dioxin, benzene, formaldehyde, and chloroform, with risks on a similar order of magnitude. Although indoor exposures are responsible for nearly 70% of risk using OEHHA’s unit risks, when infiltration is accounted for, inhalation of outdoor sources contributed 50% to total risk, on average. Additionally, 15% of risk resulted from exposures through food, mainly due to dioxin. Conclusions Most of the polycyclic aromatic hydrocarbon, benzene, acetaldehyde, and 1,3-butadiene risk came from outdoor sources, whereas indoor sources were primarily responsible for chloroform, formaldehyde, and naphthalene risks. The infiltration of outdoor pollution into buildings, emissions from indoor sources, and uptake through food are all important to consider in reducing overall personal risk to HAPs.

Land use regression modeling of intra-urban residential variability in multiple traffic-related air pollutants

BackgroundThere is a growing body of literature linking GIS-based measures of traffic density to asthma and other respiratory outcomes. However, no consensus exists on which traffic indicators best capture variability in different pollutants or within different settings. As part of a study on childhood asthma etiology, we examined variability in outdoor concentrations of multiple traffic-related air pollutants within urban communities, using a range of GIS-based predictors and land use regression techniques.MethodsWe measured fine particulate matter (PM2.5), nitrogen dioxide (NO2), and elemental carbon (EC) outside 44 homes representing a range of traffic densities and neighborhoods across Boston, Massachusetts and nearby communities. Multiple three to four-day average samples were collected at each home during winters and summers from 2003 to 2005. Traffic indicators were derived using Massachusetts Highway Department data and direct traffic counts. Multivariate regression analyses were performed separately for each pollutant, using traffic indicators, land use, meteorology, site characteristics, and central site concentrations.ResultsPM2.5 was strongly associated with the central site monitor (R2 = 0.68). Additional variability was explained by total roadway length within 100 m of the home, smoking or grilling near the monitor, and block-group population density (R2 = 0.76). EC showed greater spatial variability, especially during winter months, and was predicted by roadway length within 200 m of the home. The influence of traffic was greater under low wind speed conditions, and concentrations were lower during summer (R2 = 0.52). NO2 showed significant spatial variability, predicted by population density and roadway length within 50 m of the home, modified by site characteristics (obstruction), and with higher concentrations during summer (R2 = 0.56).ConclusionEach pollutant examined displayed somewhat different spatial patterns within urban neighborhoods, and were differently related to local traffic and meteorology. Our results indicate a need for multi-pollutant exposure modeling to disentangle causal agents in epidemiological studies, and further investigation of site-specific and meteorological modification of the traffic-concentration relationship in urban neighborhoods.

Estimating population exposure to power plant emissions using CALPUFF: a case study in Beijing, China

Abstract Epidemiological studies have shown a significant association between ambient particulate matter (PM) exposures and increased mortality and morbidity risk. Power plants are significant emitters of precursor gases of fine particulate matter. To evaluate the public health risk posed by power plants, it is necessary to evaluate population exposure to different pollutants. The concept of intake fraction (the fraction of a pollutant emitted that is eventually inhaled or ingested by a population) has been proposed to provide a simple summary measure of the relationship between emissions and exposure. Currently available intake fraction estimates from developing countries used models that look only at the near field impacts, which may not capture the full impact of a pollution source. This case study demonstrated how the intake fraction of power plant emissions in China can be calculated using a detailed long-range atmospheric dispersion model—CALPUFF. We found that the intake fraction of primary fine particles is roughly on the order of 10−5, while the intake fractions of sulfur dioxide, sulfate and nitrate are on the order of 10−6. These estimates are an order of magnitude higher than the US estimates. We also tested how sensitive the results were to key assumptions within the model. The size distribution of primary particles has a large impact on the intake fraction for primary particles while the background ammonia concentration is an important factor influencing the intake fraction of nitrate. The background ozone concentration has a moderate impact on the intake fraction of sulfate and nitrate. Our analysis shows that this approach is applicable to a developing country and it provides reasonable population exposure estimates.

Evaluation of the Public Health Impacts of Traffic Congestion: A Health Risk Assessment

BackgroundTraffic congestion is a significant issue in urban areas in the United States and around the world. Previous analyses have estimated the economic costs of congestion, related to fuel and time wasted, but few have quantified the public health impacts or determined how these impacts compare in magnitude to the economic costs. Moreover, the relative magnitudes of economic and public health impacts of congestion would be expected to vary significantly across urban areas, as a function of road infrastructure, population density, and atmospheric conditions influencing pollutant formation, but this variability has not been explored.MethodsIn this study, we evaluate the public health impacts of ambient exposures to fine particulate matter (PM2.5) concentrations associated with a business-as-usual scenario of predicted traffic congestion. We evaluate 83 individual urban areas using traffic demand models to estimate the degree of congestion in each area from 2000 to 2030. We link traffic volume and speed data with the MOBILE6 model to characterize emissions of PM2.5 and particle precursors attributable to congestion, and we use a source-receptor matrix to evaluate the impact of these emissions on ambient PM2.5 concentrations. Marginal concentration changes are related to a concentration-response function for mortality, with a value of statistical life approach used to monetize the impacts.ResultsWe estimate that the monetized value of PM2.5-related mortality attributable to congestion in these 83 cities in 2000 was approximately

Uncertainty and variability in health-related damages from coal-fired power plants in the United States.

31 billion (2007 dollars), as compared with a value of time and fuel wasted of