Using Distributed Lag Non-Linear Models to Estimate Exposure Lag-Response Associations between Long-Term Air Pollution Exposure and Incidence of Cardiovascular Disease.

Long-term air pollution exposure increases the risk for cardiovascular disease, but little is known about the temporal relationships between exposure and health outcomes. This study aims to estimate the exposure-lag response between air pollution exposure and risk for ischemic heart disease (IHD) and stroke incidence by applying distributed lag non-linear models (DLNMs). Annual mean concentrations of particles with aerodynamic diameter less than 2.5 µm (PM2.5) and black carbon (BC) were estimated for participants in five Swedish cohorts using dispersion models. Simultaneous estimates of exposure lags 1-10 years using DLNMs were compared with separate year specific (single lag) estimates and estimates for lag 1-5- and 6-10-years using moving average exposure. The DLNM estimated no exposure lag-response between PM2.5 total, BC, and IHD. However, for PM2.5 from local sources, a 20% risk increase per 1 µg/m3 for 1-year lag was estimated. A risk increase for stroke was suggested in relation to lags 2-4-year PM2.5 and BC, and also lags 8-9-years BC. No associations were shown in single lag models. Increased risk estimates for stroke in relation to lag 1-5- and 6-10-years BC moving averages were observed. Estimates generally supported a greater contribution to increased risk from exposure windows closer in time to incident IHD and incident stroke.

Long-term exposure to particulate air pollution and black carbon in relation to natural and cause-specific mortality: a multicohort study in Sweden.

OBJECTIVES: To estimate concentration-response relationships for particulate matter (PM) and black carbon (BC) in relation to mortality in cohorts from three Swedish cities with comparatively low pollutant levels. SETTING: Cohorts from Gothenburg, Stockholm and Umeå, Sweden. DESIGN: High-resolution dispersion models were used to estimate annual mean concentrations of PM with aerodynamic diameter ≤10 µm (PM10) and ≤2.5 µm (PM2.5), and BC, at individual addresses during each year of follow-up, 1990-2011. Moving averages were calculated for the time windows 1-5 years (lag1-5) and 6-10 years (lag6-10) preceding the outcome. Cause-specific mortality data were obtained from the national cause of death registry. Cohort-specific HRs were estimated using Cox regression models and then meta-analysed including a random effect of cohort. PARTICIPANTS: During the study period, 7 340 cases of natural mortality, 2 755 cases of cardiovascular disease (CVD) mortality and 817 cases of respiratory and lung cancer mortality were observed among in total 68 679 individuals and 689 813 person-years of follow-up. RESULTS: Both PM10 (range: 6.3-41.9 µg/m3) and BC (range: 0.2-6.8 µg/m3) were associated with natural mortality showing 17% (95% CI 6% to 31%) and 9% (95% CI 0% to 18%) increased risks per 10 µg/m3 and 1 µg/m3 of lag1-5 exposure, respectively. For PM2.5 (range: 4.0-22.4 µg/m3), the estimated increase was 13% per 5 µg/m3, but less precise (95% CI -9% to 40%). Estimates for CVD mortality appeared higher for both PM10 and PM2.5. No association was observed with respiratory mortality. CONCLUSION: The results support an effect of long-term air pollution on natural mortality and mortality in CVD with high relative risks also at low exposure levels. These findings are relevant for future decisions concerning air quality policies.

Cyclists' exposure to air pollution under different traffic management strategies.

We characterized the air pollution exposure of cyclists in the city center of Curitiba (Brazil) and then systematically analyzed the influence of several traffic management strategies (bus lanes, bicycle lanes, traffic calming area, traffic lights, and cleaner vehicle technologies) on the exposure. We focused on concentrations of particulates monitored on-board bicycles: PM2.5, black carbon mass (BC) and particle number concentration (PNC), and also reported on total volatile organic compound concentrations (TVOC). Overall, mean (± standard deviation) exposure was moderate compared to other cities around the world (BC: 6.98 ± 11.53 µg m--3, PM2.5: 33.22 ± 25.64 µg m-3, PNC: 3.93 × 104 ± 4.17 × 104 cm-3, TVOC: 361 ± 99 ppb). Concentrations were higher in the morning rush hour than in the afternoon traffic peak, and exhibited a large spatial variability. Bus stops and signalized traffic intersections emerged as hotspots when compared to the rest of the journey, increasing all particulate concentrations. Lower exposure was found on streets with low traffic (particularly, small number of heavy-duty vehicles) and within shallow canyon structures. The impact of traffic calming areas on cyclists' exposure is still inconclusive and further experimental and modelling studies are needed. Simple emission calculations based on traffic activity and real-world emission factors suggested that replacing the diesel bus fleet with hybrid electric buses might largely decrease (64%) the exposure to BC in the city center. Urban planners could use this valuable information to project new cycleways, which would lead to healthier active transportation. Synchronizing traffic signals might further reduce exposure at intersections.

Long-Term Exposure to Particulate Air Pollution, Black Carbon, and Their Source Components in Relation to Ischemic Heart Disease and Stroke.

BACKGROUND: Long-term exposure to particulate matter (PM) in ambient air has been associated with cardiovascular mortality, but few studies have considered incident disease in relation to PM from different sources. OBJECTIVES: We aimed to study associations between long-term exposure to different types of PM and sources, and incident ischemic heart disease (IHD) and stroke in three Swedish cities. METHODS: Based on detailed emission databases, monitoring data, and high-resolution dispersion models, we calculated source contributions to PM with aerodynamic diameter ≤10µm (PM10), PM with aerodynamic diameter ≤2.5µm (PM2.5), and black carbon (BC) from road wear, traffic exhaust, residential heating, and other sources in Gothenburg, Stockholm, and Umeå. Registry data for participants from four cohorts were used to obtain incidence of IHD and stroke for first hospitalization or death. We constructed time windows of exposure for same-year, 1- to 5-y, and 6- to 10-y averages preceding incidence from annual averages at residential addresses. Risk estimates were based on random effects meta-analyses of cohort-specific Cox proportional hazard models. RESULTS: We observed 5,166 and 3,119 incident IHD and stroke cases, respectively, in 114,758 participants. Overall, few consistent associations were observed between the different air pollution measures and IHD or stroke incidence. However, same-year levels of ambient locally emitted BC (range: 0.01-4.6 µg/m3) were associated with a 4.0% higher risk of incident stroke per interquartile range (IQR), 0.30 µg/m3 [95% confidence interval (CI): 0.04, 7.8]. This association was primarily related to BC from traffic exhaust. PM10 (range: 4.4-52 µg/m3) and PM2.5 (range: 2.9-22 µg/m3) were not associated with stroke. Associations with incident IHD were observed only for PM2.5 exposure from residential heating. DISCUSSION: Few consistent associations were observed between different particulate components and IHD or stroke. However, long-term residential exposure to locally emitted BC from traffic exhaust was associated with stroke incidence. The comparatively low exposure levels may have contributed to the paucity of associations. https://doi.org/10.1289/EHP4757.

Modelling urban cyclists' exposure to black carbon particles using high spatiotemporal data: A statistical approach.

This is a pioneering work in South America to model the exposure of cyclists to black carbon (BC) while riding in an urban area with high spatiotemporal variability of BC concentrations. We report on mobile BC concentrations sampled on 10 biking sessions in the city of Curitiba (Brazil), during rush hours of weekdays, covering four routes and totaling 178 km. Moreover, simultaneous BC measurements were conducted within a street canyon (street and rooftop levels) and at a site located 13 km from the city center. We used two statistical approaches to model the BC concentrations: multiple linear regression (MLR) and a machine-learning technique called random forests (RF). A pool of 25 candidate variables was created, including pollution measurements, traffic characteristics, street geometry and meteorology. The aggregated mean BC concentration within 30-m buffers along the four routes was 7.09 µg m-3, with large spatial variability (5th and 95th percentiles of 1.75 and 16.83 µg m-3, respectively). On average, the concentrations at the street canyon façade (5 m height) were lower than the mobile data but higher than the urban background levels. The MLR model explained a low percentage of variance (24%), but was within the values found in the literature for on-road BC mobile data. RF explained a larger variance (54%) with the additional advantage of having lower requirements for the target and predictor variables. The most impactful predictor for both models was the traffic rate of heavy-duty vehicles. Thus, to reduce the BC exposure of cyclists and residents living close to busy streets, we emphasize the importance of renewing and/or retrofitting the diesel-powered fleet, particularly public buses with old vehicle technologies. Urban planners could also use this valuable information to project bicycle lanes with greater separation from the circulation of heavy-duty diesel vehicles.

Health Impact of PM10, PM2.5 and Black Carbon Exposure Due to Different Source Sectors in Stockholm, Gothenburg and Umea, Sweden.

The most important anthropogenic sources of primary particulate matter (PM) in ambient air in Europe are exhaust and non-exhaust emissions from road traffic and combustion of solid biomass. There is convincing evidence that PM, almost regardless of source, has detrimental health effects. An important issue in health impact assessments is what metric, indicator and exposure-response function to use for different types of PM. The aim of this study is to describe sectorial contributions to PM exposure and related premature mortality for three Swedish cities: Gothenburg, Stockholm and Umea. Exposure is calculated with high spatial resolution using atmospheric dispersion models. Attributed premature mortality is calculated separately for the main local sources and the contribution from long-range transport (LRT), applying different relative risks. In general, the main part of the exposure is due to LRT, while for black carbon, the local sources are equally or more important. The major part of the premature deaths is in our assessment related to local emissions, with road traffic and residential wood combustion having the largest impact. This emphasizes the importance to resolve within-city concentration gradients when assessing exposure. It also implies that control actions on local PM emissions have a strong potential in abatement strategies.

Long-term effects of total and source-specific particulate air pollution on incident cardiovascular disease in Gothenburg, Sweden.

BACKGROUND AND AIMS: Long-term exposure to air pollution increases cardiopulmonary morbidity and mortality, but it is not clear which components of air pollution are the most harmful, nor which time window of exposure is most relevant. Further studies at low exposure levels have also been called for. We analyzed two Swedish cohorts to investigate the effects of total and source-specific particulate matter (PM) on incident cardiovascular disease for different time windows of exposure. METHODS: Two cohorts initially recruited to study predictors of cardiovascular disease (the PPS cohort and the GOT-MONICA cohort) were followed from 1990 to 2011. We collected data on residential addresses and assigned each individual yearly total and source-specific PM and Nitrogen Oxides (NOx) exposures based on dispersion models. Using multivariable Cox regression models with time-dependent exposure, we studied the association between three different time windows (lag 0, lag 1-5, and exposure at study start) of residential PM and NOx exposure, and incidence of ischemic heart disease, stroke, heart failure and atrial fibrillation. RESULTS AND DISCUSSION: During the study period, there were 2266 new-onset cases of ischemic heart disease, 1391 of stroke, 925 of heart failure and 1712 of atrial fibrillation. The majority of cases were in the PPS cohort, where participants were older. Exposure levels during the study period were moderate (median: 13µg/m3 for PM10 and 9µg/m3 for PM2.5), and similar in both cohorts. Road traffic and residential heating were the largest local sources of PM air pollution, and long distance transportation the largest PM source in total. In the PPS cohort, there were positive associations between PM in the last five years and both ischemic heart disease (HR: 1.24 [95% CI: 0.98-1.59] per 10µg/m3 of PM10, and HR: 1.38 [95% CI: 1.08-1.77] per 5µg/m3 of PM2.5) and heart failure. In the GOT-MONICA cohort, there were positive but generally non-significant associations between PM and stroke (HR: 1.48 [95% CI: 0.88-2.49] per 10µg/m3 of PM10, and HR: 1.50 [95% CI: 0.90-2.51] per 5µg/m3 of PM2.5, in the last five years). Effect estimates were stronger for women, non-smokers, and higher socioeconomic classes. Exposure in the last five years seemed to be more strongly associated with outcomes than other exposure time windows. Associations between source-specific PM air pollution and outcomes were mixed and generally weak. High correlations between the main pollutants limited the use of multi-pollutant models. CONCLUSIONS: The main PM air pollutants were associated with ischemic heart disease and stroke (in women) at the relatively low exposure levels in Gothenburg, Sweden. The associations tended to be stronger for women than for men, for non-smokers than for smokers, and for higher socioeconomic classes than for lower. The associations could not be attributed to a specific PM source or type, and differed somewhat between the two cohorts. The results of this study confirm that further efforts to reduce air pollution exposure should be undertaken in Sweden to reduce the negative health effects in the general population.

High-resolution modeling of residential outdoor particulate levels in Sweden.

Large-scale exposure assessments that include both between- and within-city differences in air pollution levels are lacking. The objective of this study was to model long-term particle exposure for the whole of Sweden, separating long-range transport from local sources, which were further separated into combustion and road dust. Annual regional, urban and local traffic PM exposure contributions were modeled for 26,000 addresses from a national survey, using a European scale model, an urban model and a local traffic model. Total PM(10) was overall dominated by the regional contribution, ranging from 3.5 µg/m(3) (northernmost) to 13.5 µg/m(3) (southernmost). Local traffic and urban sources contributed nationally on average to 16% of total PM(10), but for urban populations this contribution was larger (for Stockholm around 30%). Generalized to the Swedish adult population, the average residential exposure contributions from regional, urban and local traffic PM(10) were 10.2, 1.3 and 0.2 µg/m(3), respectively. Corresponding exposure to PM(1) was 5.1, 0.5 and 0.03 µg/m(3), respectively. Long-range transport dominates average Swedish residential PM(1) and PM(10) levels, but for urban populations the contributions from urban and local traffic sources are important and may even dominate for residences close to heavily trafficked roads. The study shows the importance of considering both national and city-scale gradients. The approach to exposure modeling at home addresses of a Swedish cohort includes both the regional scale and the urban and local traffic contributions to total PM exposure. With this we can resolve both between- and within-city gradients in national exposure assessments. The within-city exposure is further divided into a submicron (combustion) and a supermicron (road dust generated by studded tires) part. This gives new possibilities to study health impacts of different particles generated in Scandinavian cities.

Fine and coarse particulate air pollution in relation to respiratory health in Sweden.

Health effects have repeatedly been associated with residential levels of air pollution. However, it is difficult to disentangle effects of long-term exposure to locally generated and long-range transported pollutants, as well as to exhaust emissions and wear particles from road traffic. We aimed to investigate effects of exposure to particulate matter fractions on respiratory health in the Swedish adult population, using an integrated assessment of sources at different geographical scales. The study was based on a nationwide environmental health survey performed in 2007, including 25,851 adults aged 18-80 years. Individual exposure to particulate matter at residential addresses was estimated by dispersion modelling of regional, urban and local sources. Associations between different size fractions or source categories and respiratory outcomes were analysed using multiple logistic regression, adjusting for individual and contextual confounding. Exposure to locally generated wear particles showed associations for blocked nose or hay fever, chest tightness or cough, and restricted activity days with odds ratios of 1.5-2 per 10-µg·m(-3) increase. Associations were also seen for locally generated combustion particles, which disappeared following adjustment for exposure to wear particles. In conclusion, our data indicate that long-term exposure to locally generated road wear particles increases the risk of respiratory symptoms in adults.

Traffic noise and cardiovascular health in Sweden: the roadside study.

Long-term exposure to traffic noise has been suggested to increase the risk of cardiovascular diseases (CVD). However, few studies have been performed in the general population and on railway noise. This study aimed to investigate the cardiovascular effects of living near noisy roads and railways. This cross-sectional study comprised 25,851 men and women, aged 18-80 years, who had resided in Sweden for at least 5 years. All subjects participated in a National Environmental Health Survey, performed in 2007, in which they reported on health, annoyance reactions and environmental factors. Questionnaire data on self-reported doctor's diagnosis of hypertension and/or CVD were used as outcomes. Exposure was assessed as Traffic Load (millions of vehicle kilometres per year) within 500 m around each participant's residential address. For a sub-population (n = 2498), we also assessed road traffic and railway noise in L(den) at the dwelling façade. Multiple logistic regression models were used to assess Prevalence Odds Ratios (POR) and 95% Confidence Intervals (CI). No statistically significant associations were found between Traffic Load and self-reported hypertension or CVD. In the sub-population, there was no association between road traffic noise and the outcomes; however, an increased risk of CVD was suggested among subjects exposed to railway noise ≥50 dB(A); POR 1.55 (95% CI 1.00-2.40). Neither Traffic Load nor road traffic noise was, in this study, associated with self-reported cardiovascular outcomes. However, there was a borderline-significant association between railway noise and CVD. The lack of association for road traffic may be due to methodological limitations.

Estimates of black carbon and size-resolved particle number emission factors from residential wood burning based on ambient monitoring and model simulations.

In this paper we derive typical emission factors for coarse particulate matter (PM(10)), oxides of nitrogen (NO(x)), black carbon (BC), and number particle size distributions based on a combination of measurements and air quality dispersion modeling. The advantage of this approach is that the emission factors represent integrated emissions from several vehicle types and different types of wood stoves. Normally it is very difficult to estimate the total emissions in cities on the basis of laboratory measurements on single vehicles or stoves because of the large variability in conditions. The measurements were made in Temuco, Chile, between April 18 and June 15, 2005 at two sites. The first one was located in a residential area relatively far from major roads. The second site was located in a busy street in downtown Temuco where wood consumption is low. The measurements support the assumption that the monitoring sites represent the impact of different emission sources, namely traffic and residential wood combustion (RWC). Fitting model results to the available measurements, emission factors were obtained for PM(10) (RWC = 2160 +/- 100 mg/kg; traffic = 610 +/- 51 mg/veh-km), NO(x) (RWC = 800 +/- 100 mg/kg; traffic = 4400 +/- 100 mg/veh-km), BC (RWC = 74 +/- 6 mg/kg; traffic = 60 +/- 3 mg/veh-km) and particle number (N) with size distribution between 25 and 600 nm (N(25-600)) (RWC = 8.9 +/- 1 x 10(14) pt/kg; traffic = 6.7 +/- 0.5 x 10(14) pt/veh-km). The obtained emission factors are comparable to results reported in the literature. The size distribution of the N emission factors for traffic was shown to be different than for RWC. The main difference is that although traffic emissions show a bimodal size distribution with a main mode below 30 nm and a secondary one around 100 nm, RWC emissions show the main mode slightly below 100 nm and a smaller nucleation mode below 50 nm.

Spatial & temporal variations of PM10 and particle number concentrations in urban air.

The size of particles in urban air varies over four orders of magnitude (from 0.001 microm to 10 microm in diameter). In many cities only particle mass concentrations (PM10, i.e. particles <10 microm diameter) is measured. In this paper we analyze how differences in emissions, background concentrations and meteorology affect the temporal and spatial distribution of PM10 and total particle number concentrations (PNC) based on measurements and dispersion modeling in Stockholm, Sweden. PNC at densely trafficked kerbside locations are dominated by ultrafine particles (<0.1 microm diameter) due to vehicle exhaust emissions as verified by high correlation with NOx. But PNC contribute only marginally to PM10, due to the small size of exhaust particles. Instead wear of the road surface is an important factor for the highest PM10 concentrations observed. In Stockholm, road wear increases drastically due to the use of studded tires and traction sand on streets during winter; up to 90% of the locally emitted PM10 may be due to road abrasion. PM10 emissions and concentrations, but not PNC, at kerbside are controlled by road moisture. Annual mean urban background PM10 levels are relatively uniformly distributed over the city, due to the importance of long range transport. For PNC local sources often dominate the concentrations resulting in large temporal and spatial gradients in the concentrations. Despite these differences in the origin of PM10 and PNC, the spatial gradients of annual mean concentrations due to local sources are of equal magnitude due to the common source, namely traffic. Thus, people in different areas experiencing a factor of 2 different annual PM10 exposure due to local sources will also experience a factor of 2 different exposure in terms of PNC. This implies that health impact studies based solely on spatial differences in annual exposure to PM10 may not separate differences in health effects due to ultrafine and coarse particles. On the other hand, health effect assessments based on time series exposure analysis of PM10 and PNC, should be able to observe differences in health effects of ultrafine particles versus coarse particles.

Design and operation of air quality monitoring networks

Se describe el diseño de una red de monitoreo del aire para grandes áreas urbanas. En primer lugar se señalan las técnicas de monitoreo, sus objetivos y los aspectos a tener en cuenta. Luego se comenta la experiencia realizada en Estocolmo, Suecia; y en Buenos Aires, Argentina, donde se presentan datos de concentraciones de contaminantes a distintas horas, tanto de la Provincia como de la Capital; y se los compara con los estándares de la OMS

Design and operation of air quality monitoring networks

Se describe el diseño de una red de monitoreo del aire para grandes áreas urbanas. En primer lugar se señalan las técnicas de monitoreo, sus objetivos y los aspectos a tener en cuenta. Luego se comenta la experiencia realizada en Estocolmo, Suecia; y en Buenos Aires, Argentina, donde se presentan datos de concentraciones de contaminantes a distintas horas, tanto de la Provincia como de la Capital; y se los compara con los estándares de la OMS