Evaluating heterogeneity in indoor and outdoor air pollution using land-use regression and constrained factor analysis.

Previous studies have identified associations between traffic exposures and a variety of adverse health effects, but many of these studies relied on proximity measures rather than measured or modeled concentrations of specific air pollutants, complicating interpretability of the findings. An increasing number of studies have used land-use regression (LUR) or other techniques to model small-scale variability in concentrations of specific air pollutants. However, these studies have generally considered a limited number of pollutants, focused on outdoor concentrations (or indoor concentrations of ambient origin) when indoor concentrations are better proxies for personal exposures, and have not taken full advantage of statistical methods for source apportionment that may have provided insight about the structure of the LUR models and the interpretability of model results. Given these issues, the primary objective of our study was to determine predictors of indoor and outdoor residential concentrations of multiple traffic-related air pollutants within an urban area, based on a combination of central site monitoring data; geographic information system (GIS) covariates reflecting traffic and other outdoor sources; questionnaire data reflecting indoor sources and activities that affect ventilation rates; and factor-analytic methods to better infer source contributions. As part of a prospective birth cohort study assessing asthma etiology in urban Boston, we collected indoor and/or outdoor 3-to-4 day samples of nitrogen dioxide (NO2) and fine particulate matter with an aerodynamic diameter or = 2.5 pm (PM2.5) at 44 residences during multiple seasons of the year from 2003 through 2005. We performed reflectance analysis, x-ray fluorescence spectroscopy (XRF), and high-resolution inductively coupled plasma-mass spectrometry (ICP-MS) on particle filters to estimate the concentrations of elemental carbon (EC), trace elements, and water-soluble metals, respectively. We derived multiple indicators of traffic using Massachusetts Highway Department (MHD) data and traffic counts collected outside the residences where the air monitoring was conducted. We used a standardized questionnaire to collect data on home characteristics and occupant behaviors. Additional housing information was collected through property tax records. Ambient concentrations of pollutants as well as meteorological data were collected from centrally located ambient monitors. We used GIS-based LUR models to explain spatial and temporal variability in residential outdoor concentrations of PM2.5, EC, and NO2. We subsequently derived latent-source factors for residential outdoor concentrations using confirmatory factor analysis constrained to nonnegative loadings. We developed LUR models to determine whether GIS covariates and other predictors explain factor variability and thereby support initial factor interpretations. To evaluate indoor concentrations, we developed physically interpretable regression models that explored the relationship between measured indoor and outdoor concentrations, relying on questionnaire data to characterize indoor sources and activities. Because outdoor pollutant concentrations measured directly outside of homes are unlikely to be available for most large epidemiologic studies, we developed regression models to explain indoor concentrations of PM2.5, EC, and NO2 as a function of other, more readily available data: GIS covariates, questionnaire data reflecting both sources and ventilation, and central site monitoring data. As we did for outdoor concentrations, we then derived latent-source factors for residential indoor concentrations and developed regression models explaining variability in these indoor latent-source factors. Finally, to provide insight about the effects of improved characterization of exposures for the results of subsequent epidemiologic investigations, we developed a simulation framework to quantitatively compare the implications of using exposure models derived from validation studies with the use of other surrogate models with varying amounts of measurement error. The concentrations of outdoor PM2.5 were strongly associated with the central site monitor data, whereas EC concentrations showed greater spatial variability, especially during colder months, and were predicted by the length of roadway within 200 m of the home. Outdoor NO2 also showed significant spatial variability, predicted in part by population density and roadway length within 50 m of the home. Our constrained factor analysis of outdoor concentrations produced loadings indicating long-range transport, brake wear and traffic exhaust, diesel exhaust, fuel oil combustion, and resuspended road dust as sources; corresponding LUR models largely corroborated these factor interpretations through covariate significance. For example, long-range transport was predicted by central site PM2.5, and season, brake wear and traffic exhaust and resuspended road dust by traffic and residential density, diesel exhaust by the percentage of diesel traffic on the nearest major road, and fuel oil combustion by population density. Our modeling of the concentrations of indoor pollutants demonstrated substantial variability in indoor-outdoor relationships across constituents, helping to separate constituents dominated by outdoor sources (e.g., S, Se, and V) from those dominated by indoor sources (e.g., Ca and Si). Regression models indicated that indoor PM2.5 was not influenced substantially by local traffic but had significant indoor sources (cooking activity and occupant density), while EC was associated with distance to the nearest designated truck route, and NO2 was associated with both traffic density within 50 m of the home and gas stove usage. Our constrained factor analysis of indoor concentrations helped to separate outdoor-dominated factors from indoor-dominated factors, though some factors appeared to be influenced by both indoor and outdoor sources. Subsequent factor analyses of the indoor-attributable fractions from indoor-outdoor regression models provided generally consistent interpretations of indoor-dominated factors. The use of regression models on indoor factors demonstrated the limited predictive power of questionnaire data related to indoor sources, but reinforced the viability of modeling indoor concentrations of pollutants of ambient origin. In spite of the relatively weak predictive power of some of the indoor-concentration regression models, our epidemiologic simulations illustrated that exposure models with fairly modest R2 values (in the range of 0.3 through 0.4, corresponding with the regression models for PM2.5 and NO2) yielded substantial improvements in epidemiologic study performance relative to the use of exposure proxies that could be applied in the absence of validation studies. In spite of limitations related to sample size and available covariate data, our study demonstrated significant outdoor spatial variability within an urban area in NO2 and in several constituents of airborne particles. LUR techniques combined with constrained factor analysis helped to disentangle the contributions to temporal variability of local, long-range transport, and other sources, ultimately allowing exposures from defined source categories to be investigated in epidemiologic studies. For the indoor residential environment, we demonstrated substantial variability in indoor-outdoor relationships among particle constituents; then, using information from public databases and focused questionnaire data, we were able to predict indoor concentrations for a subset of key pollutants. Constrained factor analysis methods applied to the indoor environment helped to separate indoor sources from outdoor sources. The corresponding indoor regression models had limited predictive power, reinforcing the complexity of characterizing the indoor environment when only limited information about key predictors is available. This finding also underscores the likelihood that these regression models might characterize indoor concentrations of pollutants with ambient origins better than they can the indoor concentrations from all sources. Our findings provide direction for future studies characterizing indoor exposure sources and patterns, and our epidemiologic simulation reinforced the importance of reducing measurement error in a context where many traffic-related air pollutants are influenced by both indoor and outdoor sources. The combination of analytical techniques used in our study could ultimately allow for more refined exposure characterization and evaluation of the relative contributions of various sources to health outcomes in epidemiologic studies.

Levy JI ，Clougherty JE ，Baxter LK ，Houseman EA ，Paciorek CJ ，HEI Health Review Committee ... -  《-》

Source apportionment of indoor residential fine particulate matter using land use regression and constrained factor analysis.

Source contributions to urban fine particulate matter (PM(2.5) ) have been modelled using land use regression (LUR) and factor analysis (FA). However, people spend more time indoors, where these methods are less explored. We collected 3-4- day samples of nitrogen dioxide and PM(2.5) inside and outside of 43 homes in summer and winter, 2003-2005, in and around Boston, Massachusetts. Particle filters were analysed for black carbon and trace element concentrations using reflectometry, X-ray fluorescence (XRF), and high-resolution inductively coupled mass spectrometry (ICP-MS). We regressed indoor against outdoor concentrations modified by ventilation, isolating the indoor-attributable fraction, and then applied constrained FA to identify source factors in indoor concentrations and residuals. Finally, we developed LUR predictive models using GIS-based outdoor source indicators and questionnaire data on indoor sources. FA using concentrations and residuals reasonably separated outdoor (long-range transport/meteorology, fuel oil/diesel, road dust) from indoor sources (combustion, smoking, cleaning). Multivariate LUR regression models for factors from concentrations and indoor residuals showed limited predictive power, but corroborated some indoor and outdoor factor interpretations. Our approach to validating source interpretations using LUR methods provides direction for studies characterizing indoor and outdoor source contributions to indoor cocentrations. By merging indoor-outdoor modeling, factor analysis, and LUR-style predictive regression modeling, we have added to previous source apportionment studies by attempting to corroborate factor interpretations. Our methods and results support the possibility that indoor exposures may be modeled for epidemiologic studies, provided adequate sample size and variability to identify indoor and outdoor source contributions. Using these techniques, epidemiologic studies can more clearly examine exposures to indoor sources and indoor penetration of source-specific components, reduce exposure misclassification, and improve the characterization of the relationship between particle constituents and health effects.

Clougherty JE ，Houseman EA ，Levy JI 《-》

Assessment and statistical modeling of the relationship between remotely sensed aerosol optical depth and PM2.5 in the eastern United States.

Research in scientific, public health, and policy disciplines relating to the environment increasingly makes use of high-dimensional remote sensing and the output of numerical models in conjunction with traditional observations. Given the public health and resultant public policy implications of the potential health effects of particulate matter (PM*) air pollution, specifically fine PM with an aerodynamic diameter < or = 2.5 pm (PM2.5), there has been substantial recent interest in the use of remote-sensing information, in particular aerosol optical depth (AOD) retrieved from satellites, to help characterize variability in ground-level PM2.5 concentrations in space and time. While the United States and some other developed countries have extensive PM monitoring networks, gaps in data across space and time necessarily occur; the hope is that remote sensing can help fill these gaps. In this report, we are particularly interested in using remote-sensing data to inform estimates of spatial patterns in ambient PM2.5 concentrations at monthly and longer time scales for use in epidemiologic analyses. However, we also analyzed daily data to better disentangle spatial and temporal relationships. For AOD to be helpful, it needs to add information beyond that available from the monitoring network. For analyses of chronic health effects, it needs to add information about the concentrations of long-term average PM2.5; therefore, filling the spatial gaps is key. Much recent evidence has shown that AOD is correlated with PM2.5 in the eastern United States, but the use of AOD in exposure analysis for epidemiologic work has been rare, in part because discrepancies necessarily exist between satellite-retrieved estimates of AOD, which is an atmospheric-column average, and ground-level PM2.5. In this report, we summarize the results of a number of empirical analyses and of the development of statistical models for the use of proxy information, in particular satellite AOD, in predicting PM2.5 concentrations in the eastern United States. We analyzed the spatiotemporal structure of the relationship between PM2.5 and AOD, first using simple correlations both before and after calibration based on meteorology, as well as large-scale spatial and temporal calibration to account for discrepancies between AOD and PM2.5. We then used both raw and calibrated AOD retrievals in statistical models to predict PM2.5 concentrations, accounting for AOD in two ways: primarily as a separate data source contributing a second likelihood to a Bayesian statistical model, as well as a data source on which we could directly regress. Previous consideration of satellite AOD has largely focused on the National Aeronautics and Space Administration (NASA) moderate resolution imaging spectroradiometer (MODIS) and multiangle imaging spectroradiometer (MISR) instruments. One contribution of our work is more extensive consideration of AOD derived from the Geostationary Operational Environmental Satellite East Aerosol/Smoke Product (GOES GASP) AOD and its relationship with PM2.5. In addition to empirically assessing the spatiotemporal relationship between GASP AOD and PM2.5, we considered new statistical techniques to screen anomalous GOES reflectance measurements and account for background surface reflectance. In our statistical work, we developed a new model structure that allowed for more flexible modeling of the proxy discrepancy than previous statistical efforts have had, with a computationally efficient implementation. We also suggested a diagnostic for assessing the scales of the spatial relationship between the proxy and the spatial process of interest (e.g., PM2.5). In brief, we had little success in improving predictions in our eastern-United States domain for use in epidemiologic applications. We found positive correlations of AOD with PM2.5 over time, but less correlation for long-term averages over space, unless we used calibration that adjusted for large-scale discrepancy between AOD and PM2.5 (see sections 3, 4, and 5). Statistical models that combined AOD, PM2.5 observations, and land-use and meteorologic variables were highly predictive of PM2.5 observations held out of the modeling, but AOD added little information beyond that provided by the other sources (see sections 5 and 6). When we used PM2.5 data estimates from the Community Multiscale Air Quality model (CMAQ) as the proxy instead of using AOD, we similarly found little improvement in predicting held-out observations of PM2.5, but when we regressed on CMAQ PM2.5 estimates, the predictions improved moderately in some cases. These results appeared to be caused in part by the fact that large-scale spatial patterns in PM2.5 could be predicted well by smoothing the monitor values, while small-scale spatial patterns in AOD appeared to weakly reflect the variation in PM2.5 inferred from the observations. Using a statistical model that allowed for potential proxy discrepancy at both large and small spatial scales was an important component of our modeling. In particular, when our models did not include a component to account for small-scale discrepancy, predictive performance decreased substantially. Even long-term averages of MISR AOD, considered the best, albeit most sparse, of the AOD products, were only weakly correlated with measured PM2.5 (see section 4). This might have been partly related to the fact that our analysis did not account for spatial variation in the vertical profile of the aerosol. Furthermore, we found evidence that some of the correlation between raw AOD and PM2.5 might have been a function of surface brightness related to land use, rather than having been driven by the detection of aerosol in the AOD retrieval algorithms (see sections 4 and 7). Difficulties in estimating the background surface reflectance in the retrieval algorithms likely explain this finding. With regard to GOES, we found moderate correlations of GASP AOD and PM2.5. The higher correlations of monthly and yearly averages after calibration reflected primarily the improved large-scale correlation, a necessary result of the calibration procedure (see section 3). While the results of this study's GOES reflectance screening and surface reflection correction appeared sensible, correlations of our proposed reflectance-based proxy with PM2.5 were no better than GASP AOD correlations with PM2.5 (see section 7). We had difficulty improving spatial prediction of monthly and yearly average PM2.5 using AOD in the eastern United States, which we attribute to the spatial discrepancy between AOD and measured PM2.5, particularly at smaller scales. This points to the importance of paying attention to the discrepancy structure of proxy information, both from remote-sensing and deterministic models. In particular, important statistical challenges arise in accounting for the discrepancy, given the difficulty in the face of sparse observations of distinguishing the discrepancy from the component of the proxy that is informative about the process of interest. Associations between adverse health outcomes and large-scale variation in PM2.5 (e.g., across regions) may be confounded by unmeasured spatial variation in factors such as diet. Therefore, one important goal was to use AOD to improve predictions of PM2.5 for use in epidemiologic analyses at small-to-moderate spatial scales (within urban areas and within regions). In addition, large-scale PM2.5 variation is well estimated from the monitoring data, at least in the United States. We found little evidence that current AOD products are helpful for improving prediction at small-to-moderate scales in the eastern United States and believe more evidence for the reliability of AOD as a proxy at such scales is needed before making use of AOD for PM2.5 prediction in epidemiologic contexts. While our results relied in part on relatively complicated statistical models, which may be sensitive to modeling assumptions, our exploratory correlation analyses (see sections 3 and 5) and relatively simple regression-style modeling of MISR AOD (see section 4) were consistent with the more complicated modeling results. When assessing the usefulness of AOD in the context of studying chronic health effects, we believe efforts need to focus on disentangling the temporal from the spatial correlations of AOD and PM2.5 and on understanding the spatial scale of correlation and of the discrepancy structure. While our results are discouraging, it is important to note that we attempted to make use of smaller-scale spatial variation in AOD to distinguish spatial variations of relatively small magnitude in long-term concentrations of ambient PM2.5. Our efforts pushed the limits of current technology in a spatial domain with relatively low PM2.5 levels and limited spatial variability. AOD may hold more promise in areas with higher aerosol levels, as the AOD signal would be stronger there relative to the background surface reflectance. Furthermore, for developing countries with high aerosol levels, it is difficult to build statistical models based on PM2.5 measurements and land-use covariates, so AOD may add more incremental information in those contexts. More generally, researchers in remote sensing are involved in ongoing efforts to improve AOD products and develop new approaches to using AOD, such as calibration with model-estimated vertical profiles and the use of speciation information in MISR AOD; these efforts warrant continued investigation of the usefulness of remotely sensed AOD for public health research.

Paciorek CJ ，Liu Y ，HEI Health Review Committee 《-》

Effects of long-term exposure to traffic-related air pollution on respiratory and cardiovascular mortality in the Netherlands: the NLCS-AIR study.

Evidence is increasing that long-term exposure to ambient air pollution is associated with deaths from cardiopulmonary diseases. In a 2002 pilot study, we reported clear indications that traffic-related air pollution, especially at the local scale, was related to cardiopulmonary mortality in a randomly selected subcohort of 5000 older adults participating in the ongoing Netherlands Cohort Study (NLCS) on diet and cancer. In the current study, referred to as NLCS-AIR, our objective was to obtain more precise estimates of the effects of traffic-related air pollution by analyzing associations with cause-specific mortality, as well as lung cancer incidence, in the full cohort of approximately 120,000 subjects. Cohort members were 55 to 69 years of age at enrollment in 1986. Follow-up was from 1987 through 1996 for mortality (17,674 deaths) and from late 1986 through 1997 for lung cancer incidence (2234 cases). Information about potential confounding variables and effect modifiers was available from the questionnaire that subjects completed at enrollment and from publicly available data (including neighborhood-scale information such as income distributions). The NLCS was designed for a case-cohort approach, which makes use of all the cases in the full cohort, while data for the random subcohort are used to estimate person-time experience in the study. Full information on confounders was available for the subjects in the random subcohort and for the emerging cases of mortality and lung cancer incidence during the follow-up period, and in NLCS-AIR we used the case-cohort approach to examine the relation between exposure to air pollution and cause-specific mortality and lung cancer. We also specified a standard Cox proportional hazards model within the full cohort, for which information on potential confounding variables was much more limited. Exposure to air pollution was estimated for the subjects' home addresses at baseline in 1986. Concentrations were estimated for black smoke (a simple marker for soot) and nitrogen dioxide (NO2) as indicators of traffic-related air pollution, as well as nitric oxide (NO), sulfur dioxide (SO2), and particulate matter with aerodynamic diameter < or = 2.5 microm (PM2.5), as estimated from measurements of particulate matter with aerodynamic diameter < or = 10 microm (PM10). Overall long-term exposure concentrations were considered to be a function of air pollution contributions at regional, urban, and local scales. We used interpolation from data obtained routinely at regional stations of the National Air Quality Monitoring Network (NAQMN) to estimate the regional component of exposure at the home address. Average pollutant concentrations were estimated from NAQMN measurements for the period 1976 through 1996. Land-use regression methods were used to estimate the urban exposure component. For the local exposure component, geographic information systems (GISs) were used to generate indicators of traffic exposure that included traffic intensity on and distance to nearby roads. A major effort was made to collect traffic intensity data from individual municipalities. The exposure variables were refined considerably from those used in the pilot study, but we also analyzed the data for the full cohort in the current study using the exposure indicators of the pilot study. We analyzed the data in models with the estimated overall pollutant concentration as a single variable and with the background concentration (the sum of regional and urban components) and the local exposure estimate from traffic indicators as separate variables. In the full-cohort analyses adjusted for the limited set of confounders, estimated overall exposure concentrations of black smoke, NO2, NO, and PM2.5 were associated with mortality. For a 10-microg/m3 increase in the black smoke concentration, the relative risk (RR) (95% confidence interval [CI]) was 1.05 (1.00-1.11) for natural-cause (nonaccidental) mortality, 1.04 (0.95-1.13) for cardiovascular mortality, 1.22 (0.99-1.50) for respiratory mortality, 1.03 (0.88-1.20) for lung cancer mortality, and 1.04 (0.97-1.12) for noncardiopulmonary, non-lung cancer mortality. Results were similar for NO2, NO, and PM2.5. For a 10-microg/m3 increase in PM2.5 concentration, the RR for natural-cause mortality was 1.06 (95% CI, 0.97-1.16), the same as in the results of the American Cancer Society Study reported by Pope and colleagues in 2002. The highest relative risks were found for respiratory mortality, though confidence intervals were wider for this less-frequent cause of death. No associations with mortality were found for SO2. Some of the associations between the traffic indicator variables used to assess traffic intensity near the home and mortality reached statistical significance in the full cohort. For an increase in traffic intensity of 10,000 motor vehicles in 24 hours (motor vehicles/day) on the road nearest a subject's residence, the RR was 1.03 (95% CI, 1.00-1.08) for natural-cause mortality, 1.05 (0.99-1.12) for cardiovascular mortality, 1.10 (0.95-1.26) for respiratory mortality, 1.07 (0.96-1.19) for lung cancer mortality, and 1.00 (0.94-1.06) for noncardiopulmonary, non-lung cancer mortality. Results were similar for traffic intensity in a 100-m buffer around the subject's residence and living near a major road (a road with more than 10,000 motor vehicles/day). Distance in meters to the nearest major road and traffic intensity on the nearest major road were not associated with any of the mortality outcomes. We did not find an association between cardiopulmonary mortality and living near a major road as defined using the methods of the pilot study. In the case-cohort analyses adjusted for all potential confounders, we found no associations between background air pollution and mortality. The associations between traffic intensity and mortality were weaker than in the full cohort, and confidence intervals were wider, consistent with the smaller number of subjects. The lower relative risks of mortality associated with traffic variables in the case-cohort study population could be related to the particular subcohort that was randomly selected from the full cohort, as the risks estimated with the actual subcohort were well below the average estimates obtained for 100 new case-cohort analyses with 100 alternative subcohorts of 5000 subjects each that we randomly selected from the full cohort. Differences in adjusted relative risks between the full-cohort and the case-cohort analyses could be explained by random error introduced by sampling from the full cohort and by a selection effect resulting from the relatively large number of missing data for variables in the extensive confounder model used in the case-cohort analyses. More complete control for confounding probably did not contribute much to the lower relative risks in the case-cohort analyses, especially for the traffic variables, as results were similar when the limited confounder model for the full cohort was used in analyses of the subjects in the case-cohort study population. In additional analyses using black smoke concentrations as the exposure variables, we found that the association between overall black smoke and cardiopulmonary mortality was somewhat stronger for case-cohort subjects who did not change residence during follow-up, and in the full cohort, there was a tendency for relative risks to be higher for subjects living in the three major cities included in the study. Adjustment for estimated exposure to traffic noise did not affect the associations of background black smoke and traffic intensity with cardiovascular mortality. There was some indication of an association between traffic noise and cardiovascular mortality only for the 1.6% of the subjects in the full cohort who were exposed to traffic noise in the highest category of > 65 A-weighted decibels (dB(A); decibels with the sound pressure scale adjusted to conform with the frequency response of the human ear). Examination of sex, smoking status, educational level, and vegetable and fruit intake as possible effect modifiers showed that for overall black smoke concentrations, associations with mortality tended to be stronger in case-cohort subjects with lower levels of education and those with low fruit intake, but differences between strata were not statistically significant. For lung cancer incidence, we found essentially no relation to exposure to NO2, black smoke, PM2.5, SO2, or several traffic indicators. Associations of overall air pollution concentrations and traffic indicator variables with lung cancer incidence were, however, found in subjects who had never smoked, with an RR of 1.47 (95% CI, 1.01-2.16) for a 10-microg/m3 increase in overall black smoke concentration. In the current study, the mortality risks associated with both background air pollution and traffic exposure variables were much smaller than the estimate previously reported in the pilot study for risk of cardiopulmonary mortality associated with living near a major road (RR, 1.95; 95% CI, 1.09-3.51). The differences are most likely due to the extension of the follow-up period in the current study and to random error in the pilot study related to sampling from the full cohort. Though relative risks were generally small in the current study, long-term average concentrations of black smoke, NO2, and PM2.5 were related to mortality, and associations of black smoke and NO2 exposure with natural-cause and respiratory mortality were statistically significant. Traffic intensity near the home was also related to natural-cause mortality. The highest relative risks associated with background air pollution and traffic variables were for respiratory mortality, though the number of deaths was smaller than for the other mortality categories. (ABSTRACT TRUNCATED)

Brunekreef B ，Beelen R ，Hoek G ，Schouten L ，Bausch-Goldbohm S ，Fischer P ，Armstrong B ，Hughes E ，Jerrett M ，van den Brandt P ... -  《-》

The London low emission zone baseline study.

On February 4, 2008, the world's largest low emission zone (LEZ) was established. At 2644 km2, the zone encompasses most of Greater London. It restricts the entry of the oldest and most polluting diesel vehicles, including heavy-goods vehicles (haulage trucks), buses and coaches, larger vans, and minibuses. It does not apply to cars or motorcycles. The LEZ scheme will introduce increasingly stringent Euro emissions standards over time. The creation of this zone presented a unique opportunity to estimate the effects of a stepwise reduction in vehicle emissions on air quality and health. Before undertaking such an investigation, robust baseline data were gathered on air quality and the oxidative activity and metal content of particulate matter (PM) from air pollution monitors located in Greater London. In addition, methods were developed for using databases of electronic primary-care records in order to evaluate the zone's health effects. Our study began in 2007, using information about the planned restrictions in an agreed-upon LEZ scenario and year-on-year changes in the vehicle fleet in models to predict air pollution concentrations in London for the years 2005, 2008, and 2010. Based on this detailed emissions and air pollution modeling, the areas in London were then identified that were expected to show the greatest changes in air pollution concentrations and population exposures after the implementation of the LEZ. Using these predictions, the best placement of a pollution monitoring network was determined and the feasibility of evaluating the health effects using electronic primary-care records was assessed. To measure baseline pollutant concentrations before the implementation of the LEZ, a comprehensive monitoring network was established close to major roadways and intersections. Output-difference plots from statistical modeling for 2010 indicated seven key areas likely to experience the greatest change in concentrations of nitrogen dioxide (NO2) (at least 3 microg/m3) and of PM with an aerodynamic diameter < or = 10 microm (PM10) (at least 0.75 microg/m3) as a result of the LEZ; these suggested that the clearest signals of change were most likely to be measured near roadsides. The seven key areas were also likely to be of importance in carrying out a study to assess the health outcomes of an air quality intervention like the LEZ. Of the seven key areas, two already had monitoring sites with a full complement of equipment, four had monitoring sites that required upgrades of existing equipment, and one required a completely new installation. With the upgrades and new installations in place, fully ratified (verified) pollutant data (for PM10, PM with an aerodynamic diameter < or = 2.5 microm [PM2.5], nitrogen oxides [NOx], and ozone [O3] at all sites as well as for particle number, black smoke [BS], carbon monoxide [CO], and sulfur dioxide [SO2] at selected sites) were then collected for analysis. In addition, the seven key monitoring sites were supported by other sites in the London Air Quality Network (LAQN). From these, a robust set of baseline air quality data was produced. Data from automatic and manual traffic counters as well as automatic license-plate recognition cameras were used to compile detailed vehicle profiles. This enabled us to establish more precise associations between ambient pollutant concentrations and vehicle emissions. An additional goal of the study was to collect baseline PM data in order to test the hypothesis that changes in traffic densities and vehicle mixes caused by the LEZ would affect the oxidative potential and metal content of ambient PM10 and PM2.5. The resulting baseline PM data set was the first to describe, in detail, the oxidative potential and metal content of the PM10 and PM2.5 of a major city's airshed. PM in London has considerable oxidative potential; clear differences in this measure were found from site to site, with evidence that the oxidative potential of both PM10 and PM2.5 at roadside monitoring sites was higher than at urban background locations. In the PM10 samples this increased oxidative activity appeared to be associated with increased concentrations of copper (Cu), barium (Ba), and bathophenanthroline disulfonate-mobilized iron (BPS Fe) in the roadside samples. In the PM2.5 samples, no simple association could be seen, suggesting that other unmeasured components were driving the increased oxidative potential in this fraction of the roadside samples. These data suggest that two components were contributing to the oxidative potential of roadside PM, namely Cu and BPS Fe in the coarse fraction of PM (PM with an aerodynamic diameter of 2.5 microm to 10 microm; PM(2.5-10)) and an unidentified redox catalyst in PM2.5. The data derived for this baseline study confirmed key observations from a more limited spatial mapping exercise published in our earlier HEI report on the introduction of the London's Congestion Charging Scheme (CCS) in 2003 (Kelly et al. 2011a,b). In addition, the data set in the current report provided robust baseline information on the oxidative potential and metal content of PM found in the London airshed in the period before implementation of the LEZ; the finding that a proportion of the oxidative potential appears in the PM coarse mode and is apparently related to brake wear raises important issues regarding the nature of traffic management schemes. The final goal of this baseline study was to establish the feasibility, in ethical and operational terms, of using the U.K.'s electronic primary-care records to evaluate the effects of the LEZ on human health outcomes. Data on consultations and prescriptions were compiled from a pilot group of general practices (13 distributed across London, with 100,000 patients; 29 situated in the inner London Borough of Lambeth, with 200,000 patients). Ethics approvals were obtained to link individual primary-care records to modeled NOx concentrations by means of post-codes. (To preserve anonymity, the postcodes were removed before delivery to the research team.) A wide range of NOx exposures was found across London as well as within and between the practices examined. Although we observed little association between NOx exposure and smoking status, a positive relationship was found between exposure and increased socioeconomic deprivation. The health outcomes we chose to study were asthma, chronic obstructive pulmonary disease, wheeze, hay fever, upper and lower respiratory tract infections, ischemic heart disease, heart failure, and atrial fibrillation. These outcomes were measured as prevalence or incidence. Their distributions by age, sex, socioeconomic deprivation, ethnicity, and smoking were found to accord with those reported in the epidemiology literature. No cross-sectional positive associations were found between exposure to NOx and any of the studied health outcomes; some associations were significantly negative. After the pilot study, a suitable primary-care database of London patients was identified, the General Practice Research Database responsible for giving us access to these data agreed to collaborate in the evaluation of the LEZ, and an acceptable method of ensuring privacy of the records was agreed upon. The database included about 350,000 patients who had remained at the same address over the four-year period of the study. Power calculations for a controlled longitudinal analysis were then performed, indicating that for outcomes such as consultations for respiratory illnesses or prescriptions for asthma there was sufficient power to identify a 5% to 10% reduction in consultations for patients most exposed to the intervention compared with patients presumed to not be exposed to it. In conclusion, the work undertaken in this study provides a good foundation for future LEZ evaluations. Our extensive monitoring network, measuring a comprehensive set of pollutants (and a range of particle metrics), will continue to provide a valuable tool both for assessing the impact of LEZ regulations on air quality in London and for furthering understanding of the link between PM's composition and toxicity. Finally, we believe that in combination with our modeling of the predicted population-based changes in pollution exposure in London, the use of primary-care databases forms a sound basis and has sufficient statistical power for the evaluation of the potential impact of the LEZ on human health.

Kelly F ，Armstrong B ，Atkinson R ，Anderson HR ，Barratt B ，Beevers S ，Cook D ，Green D ，Derwent D ，Mudway I ，Wilkinson P ，HEI Health Review Committee ... -  《-》