Multicity study of air pollution and mortality in Latin America (the ESCALA study).

The ESCALA* project (Estudio de Salud y Contaminación del Aire en Latinoamérica) is an HEI-funded study that aims to examine the association between exposure to outdoor air pollution and mortality in nine Latin American cities, using a common analytic framework to obtain comparable and updated information on the effects of air pollution on several causes of death in different age groups. This report summarizes the work conducted between 2006 and 2009, describes the methodologic issues addressed during project development, and presents city-specific results of meta-analyses and meta-regression analyses. The ESCALA project involved three teams of investigators responsible for collection and analysis of city-specific air pollution and mortality data from three different countries. The teams designed five different protocols to standardize the methods of data collection and analysis that would be used to evaluate the effects of air pollution on mortality (see Appendices B-F). By following the same protocols, the investigators could directly compare the results among cities. The analysis was conducted in two stages. The first stage included analyses of all-natural-cause and cause-specific mortality related to particulate matter < or = 10 pm in aerodynamic diameter (PM10) and to ozone (O3) in cities of Brazil, Chile, and México. Analyses for PM10 and O3 were also stratified by age group and O3 analyses were stratified by season. Generalized linear models (GLM) in Poisson regression were used to fit the time-series data. Time trends and seasonality were modeled using natural splines with 3, 6, 9, or 12 degrees of freedom (df) per year. Temperature and humidity were also modeled using natural splines, initially with 3 or 6 df, and then with degrees of freedom chosen on the basis of residual diagnostics (i.e., partial autocorrelation function [PACF], periodograms, and a Q-Q plot) (Appendix H, available on the HEI Web site). Indicator variables for day-of-week and holidays were used to account for short-term cyclic fluctuations. To assess the association between exposure to air pollution and risk of death, the PM10 and O3 data were fit using distributed lag models (DLMs). These models are based on findings indicating that the health effects associated with air pollutant concentrations on a given day may accumulate over several subsequent days. Each DLM measured the cumulative effect of a pollutant concentration on a given day (day 0) and that day's contribution to the effect of that pollutant on multiple subsequent (lagged) days. For this study, exposure lags of up to 3, 5, and 10 days were explored. However, only the results of the DLMs using a 3-day lag (DLM 0-3) are presented in this report because we found a decreasing association with mortality in various age-cause groups for increasing lag effects from 3 to 5 days for both PM10 and O3. The potential modifying effect of socioeconomic status (SES) on the association of PM10 or O3 concentration and mortality was also explored in four cities: Mexico City, Rio de Janeiro, São Paulo, and Santiago. The methodology for developing a common SES index is presented in the report. The second stage included meta-analyses and metaregression. During this stage, the associations between mortality and air pollution were compared among cities to evaluate the presence of heterogeneity and to explore city-level variables that might explain this heterogeneity. Meta-analyses were conducted to combine mortality effect estimates across cities and to evaluate the presence of heterogeneity among city results, whereas meta-regression models were used to explore variables that might explain the heterogeneity among cities in mortality risks associated with exposures to PM10 (but not to O3). The results of the mortality analyses are presented as risk percent changes (RPC) with a 95% confidence interval (CI). RPC is the increase in mortality risk associated with an increase of 10 microg/m3 in the 24-hour average concentration of PM10 or in the daily maximum 8-hour moving average concentration of O3. Most of the results for PM10 were positive and statistically significant, showing an increased risk of mortality with increased ambient concentrations. Results for O3 also showed a statistically significant increase in mortality in the cities with available data. With the distributed lag model, DLM 0-3, PM10 ambient concentrations were associated with an increased risk of mortality in all cities except Concepci6n and Temuco. In Mexico City and Santiago the RPC and 95% CIs were 1.02% (0.87 to 1.17) and 0.48% (0.35 to 0.61), respectively. PM10 was also significantly associated with increased mortality from cardiopulmonary, respiratory, cardiovascular, cerebrovascular-stroke, and chronic obstructive lung diseases (COPD) in most cities. The few nonsignificant effects generally were observed in the smallest cities (Concepción, Temuco, and Toluca). The percentage increases in mortality associated with ambient O3 concentrations were smaller than for those associated with PM10. All-natural-cause mortality was significantly related to O3 in Mexico City, Monterrey, São Paulo and Rio de Janeiro. Increased mortality risks for some specific causes were also observed in these cities and in Santiago. In the analyses stratified by season, different patterns in mortality and O3 were observed for cold and warm seasons. Risk estimates for the warm season were larger and significant for several causes of death in São Paulo and Rio de Janeiro. Risk estimates for the cold season were larger and significant for some causes of death in Mexico City, Monterrey, and Toluca. In an analysis stratified by SES, the all-natural-cause mortality risk in Mexico City was larger for people with a medium SES; however we observed that the risk of mortality related to respiratory causes was larger among people with a low SES, while the risk of mortality related to cardiovascular and cerebrovascular-stroke causes was larger among people with medium or high SES. In São Paulo, the all-natural-cause mortality risk was larger in people with a high SES, while in Rio de Janeiro the all-natural-cause mortality risk was larger in people with a low SES. In both Brazilian cities, the risks of mortality were larger for respiratory causes, especially for the low- and high-SES groups. In Santiago, all-natural-cause mortality risk did not vary with level of SES; however, people with a low SES had a higher respiratory mortality risk, particularly for COPD. People with a medium SES had larger risks of mortality from cardiovascular and cerebrovascular-stroke disease. The effect of ambient PM10 concentrations on infant and child mortality from respiratory causes and lower respiratory infection (LRI) was studied only for Mexico City, Santiago, and São Paulo. Significant increased mortality risk from these causes was observed in both Santiago (in infants and older children) and Mexico City (only in infants). For O3, an increased mortality risk was observed in Mexico City (in infants and older children) and in São Paulo (only in infants during the warm season). The results of the meta-analyses confirmed the positive and statistically significant association between PM10 and all-natural-cause mortality (RPC = 0.77% [95% CI: 0.60 to 1.00]) using the random-effects model. For mortality from specific causes, the percentage increase in mortality ranged from 0.72% (0.54 to 0.89) for cardiovascular disease to 2.44% (1.36 to 3.59) for COPD, also using the random-effects model. For O3, significant positive associations were observed using the random-effects model for some causes, but not for all natural causes or for respiratory diseases in people 65 years or older (> or = 65 years), and not for COPD and cerebrovascular-stroke in the all-age and the > or = 65 age groups. The percentage increase in all-natural-cause mortality was 0.16% (-0.02 to 0.33). In the meta-regression analyses, variables that best explained heterogeneity in mortality risks among cities were the mean average of temperature in the warm season, population percentage of infants (< 1 year), population percentage of children at least 1 year old but < 5 years (i.e., 1-4 years), population percentage of people > or = 65 years, geographic density of PM10 monitors, annual average concentrations of PM10, and mortality rates for lung cancer. The ESCALA project was undertaken to obtain information for assessing the effects of air pollutants on mortality in Latin America, where large populations are exposed to relatively high levels of ambient air pollution. An important goal was to provide evidence that could inform policies for controlling air pollution in Latin America. This project included the development of standardized protocols for data collection and for statistical analyses as well as statistical analytic programs (routines developed in R by the ESCALA team) to insure comparability of results. The analytic approach and statistical programming developed within this project should be of value for researchers carrying out single-city analyses and should facilitate the inclusion of additional Latin American cities within the ESCALA multicity project. Our analyses confirm what has been observed in other parts of the world regarding the effects of ambient PM10 and 03 concentrations on daily mortality. They also suggest that SES plays a role in the susceptibility of a population to air pollution; people with a lower SES appeared to have an increased risk of death from respiratory causes, particularly COPD. Compared with the general population, infants and young children appeared to be more susceptible to both PM10 and O3, although an increased risk of mortality was not observed in these age groups in all cities. (ABSTRACT TRUNCATED)

Romieu I ，Gouveia N ，Cifuentes LA ，de Leon AP ，Junger W ，Vera J ，Strappa V ，Hurtado-Díaz M ，Miranda-Soberanis V ，Rojas-Bracho L ，Carbajal-Arroyo L ，Tzintzun-Cervantes G ，HEI Health Review Committee ... -  《-》

[Meta-analysis of the Italian studies on short-term effects of air pollution].

In recent years, much attention has been given to review reports on the early effects of air pollution on health, measured through daily series of deaths and/or hospital admissions. A number of large planned meta-analyses (in which methods for data retrieval and processing are commonly planned a priori for all participating centers) are on going both in the US and in Europe. The National Mortality, Morbidity and Air Pollution Study included data from 90 US cities, whereas APHEA (Air Pollution and Health, a European Approach) considers data from about 30 european cities. The present paper summarizes methods and findings of MISA, a meta-analysis of data from 8 Italian cities. It belongs to an ad hoc supplement of Epidemiologia & Prevenzione (Epidemiol Prev 2001; 25 (2) Suppl: 1-72), the official Journal of the Italian Association of Epidemiology, which contains a full description of the study. MISA was launched on March 2000, within the project "Statistics, Environment and Health" (GRASPA), funded by the Italian Ministry of Education. Additional support was given by the Authorities of the 8 participating cities (from North to South: Turin, Milan, Verona, Ravenna, Bologna, Florence, Rome and Palermo). DAILY HEALTH DATA: Deaths certificate and hospital admission data have been collected respectively from the Local Health Authority and regional files. The same programme for retrieval of data on selected hospital admissions for acute conditions was used in the 8 cities. Main data are summarized in Table 1. DAILY CONCENTRATION OF POLLUTANTS: Most data were obtained from Regional Environmental Protection Agencies, which are responsible for environmental monitoring since 1993. Verona, Palermo and Milan (1990-94) data were obtained from local sources. Monitors with more than 25% of missing data were excluded. Meteorological data were collected by the same monitors and completed with data from monitors situated in the suburbs or (in Milan and Bologna) in the airport. The monitors were selected by a group of experts to ensure comparability. For SO2 and NO2 daily averages of hourly measurements were used, whereas concentrations of ozone and CO were estimated as the maximum 8 hours moving average. Total suspended particulate or PM10 were measured as 24 hours deposition. All analyses used the whole range of observed values (Table 2). Daily data were considered as missing when more than 25% of hourly data were not available. Missing data in one monitor were imputed as average of data from the remaining monitors weighted by the ratio between the specific monitor's year average and the general year average of all the selected city monitors. Missing data in one day were imputed as average of four days (preceding and following day, the same day of the previous and following weeks). In the city of Florence and Palermo PM10 concentrations were available. For the other cities we applied a conversion factor from PTS to PM10 (0.6 for Turin and 0.8 for all the others) estimated through validation studies. Ozone concentrations were used only where background monitors were available (Turin, Verona, Bologna and Florence) and limited to the warm season (May through September). A common protocol for the city-specific analyses was defined on the basis of a structured exploratory analysis. The adopted basic model was a Generalized Additive Model for Poisson data. Effect estimates were age-adjusted (0-64, 65-74, 75+) and formal tests of interaction pollutant-age were conducted. In the first two age groups, indicator variables for seasonality were specified, and cubic splines with fixed number of degree of freedom were specified for the last age group and for all age groups for the morbidity data. Model adequacy was checked by residual analysis and inspection of the partial autocorrelation function. In a sensitivity analysis non linear pollutant effects were considered and overdispersed [table: see text] transitional models were fitted; the analysis was conducted for all lags 0-3 and some distributed lags (0-1, 1-2, 0-3); no multipollutant models were fitted. The same model was fitted to the city data. No model selection was done: Table 3 describes the steps in model building. In the meta-analysis, for each outcome, the estimates for each pollutant and for each city were combined using fixed and random effects models. Heterogeneity of effects was tested according to DerSimonian and Laird. Results were checked using a hierarchical bayesian model, which was used to investigate heterogeneity across cities in a meta-regression phase. Non informative priors were used. Posterior distributions of parameters of interest have been obtained with WinBUGS. 10,000 iterations (excluding [table: see text] the first 2000) were retained, while for the meta-regression 100,000 iterations (excluding the first 4000) were stored. To approximate the marginal posteriors only one sample out of five were used. Achieved convergence was assessed using the Gelman and Rubin approach. In the meta-regression the models specified were the following: [formula: see text] i denotes city, j calendar period (1990-1994; 1995-1999). The first model includes only period as effect modifier, while the second model other potential variables. The ui terms (which do not vary with j) represent city specific random effects. For each pollutant, the meta-analysis detected a statistically significant association with mortality for natural causes. But for ozone, positive associations were commonly found for death and hospital admissions for both cardiovascular and respiratory diseases. Indeed, the only estimates whose lower 95% confidence limit bore a negative sign regarded the association between PM10 and mortality from respiratory diseases. Ozone in the warm season was positively and significantly associated with daily mortality and mortality for cardiovascular diseases whereas other estimates did not reach statistical significance and some were negative (only lag 0-1 for external comparability are reported in Table 4). Risks were highest (up to 4%) for respiratory conditions (Table 4). They were more pronounced at lag 1-2 for mortality, and at lag 0-3 for hospital admissions. Age was an effect modifier for mortality, the elderly being more susceptible. In the random effect meta-analysis, at lag 1-2, excess risks for unit increase of the pollutants at age 75+ and at age 0-64 were respectively: 4.9% and -0.4% for SO2, 1.7% and 0.6% for NO2; 2.3% and 0.2% for CO. Corresponding figures for PM10 at lag 0-1 were 1.1% and 0.2%. The effect of PM10 on mortality [table: see text] was greater during the warm season (2.8% vs 0.8%). A complete analysis is reported in the Italian text. Here we provide some details on the effects of PM10, about which the residual heterogeneity across cities was highest (Table 4). In addition, the epidemiological evidence on the hazards from this fraction of particulate matter is more controversial. Table 5 reports the excess risk estimated through the meta-analysis in 1995-99 for a 10 micrograms/m3 increase of PM10 for some outcomes. Proper prior distributions (overdispersed normal and inverse gamma) were adopted in the final bayesian analyses. The sensitivity of results to the choice of the priors were investigated (we defined proper and improper uniform, student's t), obtaining comparable results. Total natural mortality was significantly heterogeneous across cities (Q = 18.96, 5 df, p < 0.001). City-specific estimates are represented graphically in Fig. 1. As expected, the confidence (credibility) intervals are widest [table: see text] for bayesian estimates, intermediate for those obtained under a random effects model, and narrowest for those found under a fixed effects model. Nevertheless, differences in point estimates are negligible. A North-South gradient in risk is obvious. Table 6 shows, for the cities for which mortality data were available, the improvement in precision and the shrinkage of effect estimates toward the overall mean introduced by the bayesian modelling. In the meta-regression, total mortality and a deprivation score were associated with greater effects. The excess risks on hospital admission were modified by the deprivation score and by the NO2/PM10 ratio. Overall, the risk estimates were greater in the calendar period 1995-99 and there was a North-South gradient, with larger effects in cities located in Central and Southern Italy (Florence, Rome, Palermo). The meta-analysis of the Italian studies on short-term effects of air pollution in 8 cities, MISA, exhibits the following features: With the exception of Naples, all greatest Italian cities were included; overall a population of 7 million was enrolled. The study protocol was accurate with regard to the selection of hospital admissions for acute conditions. Monitored data of concentration of pollutant were carefully evaluated before their inclusion in the meta-analysis. City specific analyses were carried out according to a common protocol controlling for seasonality, influenza epidemics, age and meterological variables; [table: see text] the protocol derived from a structured exploratory analysis. The meta-analysis was done using fixed and random effects models; a hierarchical bayesian model was fitted in a sensitivity analysis. The heterogeneity of effects across cities was investigated using a hierarchical bayesian model for meta-regression. While mortality data are of good quality, hospital admission data are more problematic. Since the filing criteria for the latter changed around 1995, comparability of results before and after such date is limited. Moreover, hospital admissions rely on availability of beds, the offer of which may be restricted during the warm season. Comparability of pollutant concentration estimates among cities may have been influenced by differences in monitor characteristics. (ABSTRACT TRUNCATED)

Biggeri A ，Bellini P ，Terracini B ，Italian MISA Group ... -  《Epidemiologia & Prevenzione》

Effects of short-term exposure to air pollution on hospital admissions of young children for acute lower respiratory infections in Ho Chi Minh City, Vietnam.

There is emerging evidence, largely from studies in Europe and North America, that economic deprivation increases the magnitude of morbidity and mortality related to air pollution. Two major reasons why this may be true are that the poor experience higher levels of exposure to air pollution, and they are more vulnerable to its effects--in other words, due to poorer nutrition, less access to medical care, and other factors, they experience more health impact per unit of exposure. The relations among health, air pollution, and poverty are likely to have important implications for public health and social policy, especially in areas such as the developing countries of Asia where air pollution levels are high and many live in poverty. The aims of this study were to estimate the effect of exposure to air pollution on hospital admissions of young children for acute lower respiratory infection (ALRI*) and to explore whether such effects differed between poor children and other children. ALRI, which comprises pneumonia and bronchiolitis, is the largest single cause of mortality among young children worldwide and is responsible for a substantial burden of disease among young children in developing countries. To the best of our knowledge, this is the first study of the health effects of air pollution in Ho Chi Minh City (HCMC), Vietnam. For these reasons, the results of this study have the potential to make an important contribution to the growing literature on the health effects of air pollution in Asia. The study focused on the short-term effects of daily average exposure to air pollutants on hospital admissions of children less than 5 years of age for ALRI, defined as pneumonia or bronchiolitis, in HCMC during 2003, 2004, and 2005. Admissions data were obtained from computerized records of Children's Hospital 1 and Children's Hospital 2 (CH1 and CH2) in HCMC. Nearly all children hospitalized for respiratory illnesses in the city are admitted to one of these two pediatric hospitals. Daily citywide 24-hour average concentrations of particulate matter (PM) < or =10 microm in aerodynamic diameter (PM10), nitrogen dioxide (NO2), and sulfur dioxide (SO2) and 8-hour maximum average concentrations of ozone (O3) were estimated from the HCMC Environmental Protection Agency (HEPA) ambient air quality monitoring network. Daily meteorologic information including temperature and relative humidity were collected from KTTV NB, the Southern Regional Hydro-Meteorological Center. An individual-level indicator of socioeconomic position (SEP) was based on the degree to which the patient was exempt from payment according to hospital financial records. A group-level indicator of SEP was based on estimates of poverty prevalence in the districts of HCMC in 2004, obtained from a poverty mapping project of the Institute of Economic Research in HCMC, in collaboration with the General Statistics Office of Vietnam and the World Bank. Poverty prevalence was defined using the poverty line set by the People's Committee of HCMC of 6 million Vietnamese dong (VND) annual income. Quartiles of district-level poverty prevalence were created based on poverty prevalence estimates for each district. Analyses were conducted using both time-series and case-crossover approaches. In the absence of measurement error, confounding, and other sources of bias, the two approaches were expected to provide estimates that differed only with regard to precision. For the time-series analyses, the unit of observation was daily counts of hospital admissions for ALRI. Poisson regression with smoothing functions for meteorologic variables and variables for seasonal and long-term trends was used. Case-crossover analyses were conducted using time-stratified selection of controls. Control days were every 7th day from the date of admission within the same month as admission. Large seasonal differences were observed in pollutant levels and hospital admission patterns during the investigation period for HCMC. Of the 15,717 ALRI admissions occurring within the study period, 60% occurred in the rainy season (May through October), with a peak in these admissions during July and August of each year. Average daily concentrations for PM10, O3, NO2, and SO2 were 73, 75, 22, and 22 microg/m3, respectively, with higher pollutant concentrations observed in the dry season (November through April) compared with the rainy season. As the time between onset of illness and hospital admission was thought to range from 1 to 6 days, it was not possible to specify a priori a single-day lag. We assessed results for single-day lags from lag 0 to lag 10, but emphasize results for an average of lag 1-6, since this best reflects the case reference period. Results were robust to differences in temperature lags with lag 0 and the average lag (1-6 days); results for lag 0 for temperature are presented. Results differed markedly when analyses were stratified by season, rather than simply adjusted for season. ALRI admissions were generally positively associated with ambient levels of PM10, NO2, and SO2 during the dry season (November-April), but not the rainy season (May-October). Positive associations between O3 and ALRI admissions were not observed in either season. We do not believe that exposure to air pollution could reduce the risk of ALRI in the rainy season and infer that these results could be driven by residual confounding present within the rainy season. The much lower correlation between NO2 and PM10 levels during the rainy season provides further evidence that these pollutants may not be accurate indicators of exposure to air pollution from combustion processes in the rainy season. Results were generally consistent across time-series and case-crossover analyses. In the dry season, risks for ALRI hospital admissions with average pollutant lag (1-6 days) were highest for NO2 and SO2 in the single-pollutant case-crossover analyses, with excess risks of 8.50% (95% CI, 0.80-16.79) and 5.85% (95% CI, 0.44-11.55) observed, respectively. NO2 and SO2 effects remained higher than PM10 effects in both the single-pollutant and two-pollutant models. The two-pollutant model indicated that NO2 confounded the PM10 and SO2 effects. For example, PM10 was weakly associated with an excess risk in the dry season of 1.25% (95% CI, -0.55 to 3.09); after adjusting for SO2 and O3, the risk estimate was reduced but remained elevated, with much wider confidence intervals; after adjusting for NO2, an excess risk was no longer observed. Though the effects seem to be driven by NO2, the statistical limitations of adequately addressing collinearity, given the high correlation between PM10 and NO2 (r = 0.78), limited our ability to clearly distinguish between PM10 and NO2 effects. In the rainy season, negative associations between PM10 and ALRI admissions were observed. No association with O3 was observed in the single-pollutant model, but O3 exposure was negatively associated with ALRI admissions in the two-pollutant model. There was little evidence of an association between NO2 and ALRI admissions. The single-pollutant estimate from the case-crossover analysis suggested a negative association between NO2 and ALRI admissions, but this effect was no longer apparent after adjustment for other pollutants. Although associations between SO2 and ALRI admissions were not observed in the rainy season, point estimates for the case-crossover analyses suggested negative associations, while time-series (Poisson regression) analyses suggested positive associations--an exception to the general consistency between case-crossover and time-series results. Results were robust to differences in seasonal classification. Inclusion of rainfall as a continuous variable and the seasonal reclassification of selected series of data did not influence results. No clear evidence of station-specific effects could be observed, since results for the different monitoring stations had overlapping confidence intervals. In the dry season, increased concentrations of NO2 and SO2 were associated with increased hospital admissions of young children for ALRI in HCMC. PM10 could also be associated with increased hospital admissions in the dry season, but the high correlation of 0.78 between PM10 and NO2 levels limits our ability to distinguish between PM10 and NO2 effects. Nevertheless, the results support the presence of an association between combustion-source pollution and increased ALRI admissions. There also appears to be evidence of uncontrolled negative confounding within the rainy season, with higher incidence of ALRI and lower pollutant concentrations overall. Exploratory analyses made using limited historical and regional data on monthly prevalence of respiratory syncytial virus (RSV) suggest that an unmeasured, time-varying confounder (RSV, in this case) could have, in an observational study like this one, created enough bias to reverse the observed effect estimates of pollutants in the rainy season. In addition, with virtually no RSV incidence in the dry season, these findings also lend some credibility to the notion that RSV could influence results primarily in the rainy season. Analyses were not able to identify differential effects by individual-level indicators of SEP, mainly due to the small number of children classified as poor based on information in the hospitals' financial records. Analyses assessing differences in effect by district-level indicator of SEP did not indicate a clear trend in risk across SEP quartiles, but there did appear to be a slightly higher risk among the residents of districts with the highest quartile of SEP. As these are the districts within the urban center of HCMC, results could be indicative of increased exposures for residents living within the city center. (ABSTRACT TRUNCATED)

HEI Collaborative Working Group on Air Pollution, Poverty, and Health in Ho Chi Minh City ，Le TG ，Ngo L ，Mehta S ，Do VD ，Thach TQ ，Vu XD ，Nguyen DT ，Cohen A ... -  《-》

Part 2. Association of daily mortality with ambient air pollution, and effect modification by extremely high temperature in Wuhan, China.

Fewer studies have been published on the association between daily mortality and ambient air pollution in Asia than in the United States and Europe. This study was undertaken in Wuhan, China, to investigate the acute effects of air pollution on mortality with an emphasis on particulate matter (PM*). There were three primary aims: (1) to examine the associations of daily mortality due to all natural causes and daily cause-specific mortality (cardiovascular [CVD], stroke, cardiac [CARD], respiratory [RD], cardiopulmonary [CP], and non-cardiopulmonary [non-CP] causes) with daily mean concentrations (microg/m3) of PM with an aerodynamic diameter--10 pm (PM10), sulfur dioxide (SO2), nitrogen dioxide (NO2), or ozone (O3); (2) to investigate the effect modification of extremely high temperature on the association between air pollution and daily mortality due to all natural causes and daily cause-specific mortality; and (3) to assess the uncertainty of effect estimates caused by the change in International Classification of Disease (ICD) coding of mortality data from Revision 9 (ICD-9) to Revision 10 (ICD-10) code. Wuhan is called an "oven city" in China because of its extremely hot summers (the average daily temperature in July is 37.2 degrees C and maximum daily temperature often exceeds 40 degrees C). Approximately 4.5 million residents live in the core city area of 201 km2, where air pollution levels are higher and ranges are wider than the levels in most cities studied in the published literature. We obtained daily mean levels of PM10, SO2, and NO2 concentrations from five fixed-site air monitoring stations operated by the Wuhan Environmental Monitoring Center (WEMC). O3 data were obtained from two stations, and 8-hour averages, from 10:00 to 18:00, were used. Daily mortality data were obtained from the Wuhan Centres for Disease Prevention and Control (WCDC) during the study period of July 1, 2000, to June 30, 2004. To achieve the first aim, we used a regression of the logarithm of daily counts of mortality due to all natural causes and cause-specific mortality on the daily mean concentrations of the four pollutants while controlling for weather, temporal factors, and other important covariates with generalized additive models (GAMs). We derived pollutant effect estimations for 0-day, 1-day, 2-day, 3-day, and 4-day lagged exposure levels, and the averages of 0-day and 1-day lags (lag 0-1 day) and of 0-day, 1-day, 2-day, and 3-day lags (lag 0-3 days) before the event of death. In addition, we used individual-level data (e.g., age and sex) to classify subgroups in stratified analyses. Furthermore, we explored the nonlinear shapes ("thresholds") of the exposure-response relations. To achieve the second aim, we tested the hypothesis that extremely high temperature modifies the associations between air pollution and daily mortality. We developed three corresponding weather indicators: "extremely hot," "extremely cold," and "normal temperatures." The estimates were obtained from the models for the main effects and for the pollutant-temperature interaction for each pollutant and each cause of mortality. To achieve the third aim, we conducted an additional analysis. We examined the concordance rates and kappa statistics between the ICD-9-coded mortality data and the ICD-10-coded mortality data for the year 2002. We also compared the magnitudes of the estimated effects resulting from the use of the two types of ICD-coded mortality data. In general, the largest pollutant effects were observed at lag 0-1 day. Therefore, for this report, we focused on the results obtained from the lag 0-1 models. We observed consistent associations between PM10 and mortality: every 10-microg/m3 increase in PM10 daily concentration at lag 0-1 day produced a statistically significant association with an increase in mortality due to all natural causes (0.43%; 95% confidence interval [CI], 0.24 to 0.62), CVD (0.57%; 95% CI, 0.31 to 0.84), stroke (0.57%; 95% CI, 0.25 to 0.88), CARD (0.49%; 95% CI, 0.04 to 0.94), RD (0.87%; 95% CI, 0.34 to 1.41), CP (0.52%; 95% CI, 0.27 to 0.77), and non-CP (0.30%; 95% CI, 0.05 to 0.54). In general, these effects were stronger in females than in males and were also stronger among the elderly (> or = 65 years) than among the young. The results of sensitivity testing over the range of exposures from 24.8 to 477.8 microg/m3 also suggest the appropriateness of assuming a linear relation between daily mortality and PM10. Among the gaseous pollutants, we also observed statistically significant associations of mortality with NO, and SO2, and that the estimated effects of these two pollutants were stronger than the PM10 effects. The patterns of NO2 and SO2 associations were similar to those of PM10 in terms of sex, age, and linearity. O3 was not associated with mortality. In the analysis of the effect modification of extremely high temperature on the association between air pollution and daily mortality, only the interaction of PM10 with temperature was statistically significant. Specifically, the interaction terms were statistically significant for mortality due to all natural (P = 0.014), CVD (P = 0.007), and CP (P = 0.014) causes. Across the three temperature groups, the strongest PM10 effects occurred mainly on days with extremely high temperatures for mortality due to all natural (2.20%; 95% CI, 0.74 to 3.68), CVD (3.28%; 95% CI, 1.24 to 5.37), and CP (3.02%; 95% CI, 1.03 to 5.04) causes. The weakest effects occurred at normal temperature days, with the effects on days with low temperatures in the middle. To assess the uncertainty of the effect estimates caused by the change from ICD-9-coded mortality data to ICD-10-coded mortality data, we compared the two sets of data and found high concordance rates (> 99.3%) and kappa statistics close to 1.0 (> 0.98). All effect estimates showed very little change. All statistically significant levels of the estimated effects remained unchanged. In conclusion, the findings for the aims from the current study are consistent with those in most previous studies of air pollution and mortality. The small differences between mortality effects for deaths coded using ICD-9 and ICD-10 show that the change in coding had a minimal impact on our study. Few published papers have reported synergistic effects of extremely high temperatures and air pollution on mortality, and further studies are needed. Establishing causal links between heat, PM10, and mortality will require further toxicologic and cohort studies.

Qian Z ，He Q ，Lin HM ，Kong L ，Zhou D ，Liang S ，Zhu Z ，Liao D ，Liu W ，Bentley CM ，Dan J ，Wang B ，Yang N ，Xu S ，Gong J ，Wei H ，Sun H ，Qin Z ，HEI Health Review Committee ... -  《-》

Part 1. A time-series study of ambient air pollution and daily mortality in Shanghai, China.

Kan H ，Chen B ，Zhao N ，London SJ ，Song G ，Chen G ，Zhang Y ，Jiang L ，HEI Health Review Committee ... -  《-》