Exposure to air pollution during commuting can be an important contribution to total air pollution exposure, because most commuting takes place during rush hours when pollution concentrations can reach high levels and subjects are close to traffic emissions.1 Few studies have examined acute (within hours) effects of air pollution exposure in traffic on health.2–6 Numerous studies have shown that long-term exposure to traffic-related air pollution as well as short-term (daily) changes in traffic-related pollutants are related to cardiopulmonary mortality and morbidity.7 Hence the hypothesis arose that short exposures in traffic may also cause health effects.
Acute respiratory effects of exposure to air pollution from motor vehicles have been studied in exposure chambers. Studies looking at effects of diesel exhaust exposure in healthy people as well as in asthmatics did not find effects on lung function parameters (forced vital capacity [FVC], forced expiratory flow in the first second [FEV1], peak expiratory flow [PEF], maximum midexpiratory flow).8–12 Exposure to diesel exhaust has been associated with increased airway resistance9,10,12 and airway inflammation.8,10,11,13 Larsson et al3 found increased airway inflammation after air pollution exposure in a busy road tunnel, but no effects on lung function.
These studies give insights into possible acute health effects and mechanisms of concentrated air pollution exposure, but results are difficult to translate to real-world exposures where concentrations are lower and mixtures more complex. Only a limited number of studies have examined acute respiratory effects of short-term in-traffic air pollution exposure in real-world settings. McCreanor et al2 found increased airway inflammation and reduced lung function in asthmatic pedestrians walking in Oxford Street, London, associated with increased exposures to elemental carbon, particle number (PN), and particulate matter (PM2.5). Strak et al6 found a suggestion of an increase in airway inflammation in healthy cyclists, associated with exposure to PN, and a positive association between PN and PEF.
The aim of our study was to investigate exposure to and respiratory health effects of air pollutants encountered in various modes of commuting.
Our study (Transport Related Air pollution, Variance in commuting, Exposure and Lung function, or TRAVEL) examined commuters' exposure to air pollution and associated health effects. We studied 34 healthy, nonsmoking volunteers. PN, PM2.5, PM10, and soot exposures were measured during predefined 2-hour morning commutes of the volunteers in cars, buses, and by bicycle on 47 days. Details on air pollution exposures encountered during commuting have been published previously.14 Inhaled doses were estimated using heart-rate monitoring.15 Before and after exposure in traffic, respiratory function was assessed by spirometry and interrupter airway resistance, and airway inflammation was measured by nitric oxide in exhaled air. The study has been approved by the Medical Ethical Committee of Utrecht University Medical Centre.
The design of the study was to compare the day-to-day variability in the difference in health status before and after exposure in traffic with the day-to-day variability in air pollution concentrations in traffic. The difference between the post- and preexposure can be affected by circadian variation and by exercise during the experiment. In the design, we took care that measurements were always performed in the same periods and that exercise levels were constant (per mode of transport). Circadian variation and exercise therefore could lead to a mean difference between post- and preexposure values, but not in day-to-day variation.
We thus compare the change in health status from 8 am to 4 pm for observations with low air pollution concentrations and observations with high air pollution concentrations. The days with low concentrations serve as our control exposure. We did not include a filtered air control exposure in a laboratory setting because it was not feasible to implement control exposures using clean air supply or filtration devices during bus and cycle trips.
Volunteers were recruited through Intranet websites of their employers. All volunteers were civil servants working in Arnhem, the Netherlands, employed by the local or regional government, or the regional public health service. The inclusion criteria were age between 18 and 56 years and not smoking. People were excluded if they had chronic obstructive pulmonary disease or if they had asthmatic symptoms and used asthma medication. To minimize exposures outside the experiment, participants exposed to fumes or dust at work were excluded, as were participants living more than 20 minutes' commuting time from the site where health measurements were performed and from where commuting started in the experiment.
Two-hour measurements were taken between 8 am and 10 am on 47 weekdays (Tuesdays and Thursdays) between June 2007 and June 2008. Exposures were measured simultaneously in either diesel and petrol cars, or in electric trolley buses and diesel buses (during their regular service, while other passengers got on and off the bus), and along 2 cycling routes of different traffic intensity. On cycling sampling days, technicians rode 3-wheeled cargo bicycles to transport the equipment; on car sampling days the cars were driven by the technicians. On each trip, 3 or 4 volunteers rode in the cars or buses or cycled close to the cargo bicycle.
Exposure assessment methodology and quality control have been reported in detail previously.14 PN concentrations were measured using condensation particle counters (Model 3007, TSI, MN). PM2.5 was measured with active sampling personal DataRAMs (Model 1200, MIE, Bedford, MA), and readings were corrected for relative humidity. Both condensation particle counters and DataRAMs measured real-time concentrations, recording every second. PM10 was collected on Teflon filters, using Harvard Impactors (Air Diagnostics and Engineering Inc, Naples, ME) and pumps (model SP-280E, Air Diagnostics and Engineering Inc., Harrison, ME) with a flow of 10 L/min.16 Soot content of the filters was determined using a smoke stain reflectometer (model M43D, Diffusion Systems Ltd, London, United Kingdom) and converted into an absorption coefficient. For estimations of elemental carbon (EC) inhaled doses, absorption (soot) levels were converted into exposure concentrations using the equation EC (μg/m3) = 1.6053 × absorption (10−5 m−1) − 0.2620.17
To estimate doses of inhaled air pollutants, minute ventilation levels for all individuals during all trips were estimated using continuous heart rate monitoring data. Heart rates of the participants were recorded during commuting using Polar RS400 heart rate monitors (Polar Electro, Kempele, Finland). All participants performed a submaximal bicycle ergometer test during which heart rate and minute ventilation were measured simultaneously at increasing cycling intensity, to calibrate heart rate data for each subject.15 The trip mean ventilation rates were used to calculate inhaled air pollution doses during commuting. Inhaled doses were divided by body surface area to correct for differences in airway epithelial surface area. Body surface area was calculated as the square root of height times weight divided by 3600.
For 76 of the 352 individual trips, we did not have heart rate data due to equipment failure or because participants forgot to turn on the heart rate monitors. For 69 of these trips we used the mean heart rate of the other trips of the same individual in the same mode of transport (bus, car or bicycle) to estimate minute ventilation. Excluding these doses from the dataset did not change the associations between inhaled doses and health effects. For 7 trips we did not have heart rate data for the same individual in the same mode of transport, and so for these, minute ventilation levels remained missing.
Hourly ambient nitrogen dioxide (NO2) and ozone data were obtained from the nearest urban background station (16 km distance) of the National Air Quality Monitoring Network of the Dutch National Institute for Public Health and the Environment. We obtained daily ambient PM10 data from the nearest regional background station (20 km distance), as there were no urban background PM10 data from nearby monitoring stations available.
FVC, PEF, FEV1, and maximum midexpiratory flow were measured using a pneumotachometer (Jaeger, Viasys Healthcare, Hoechberg, Germany). Daily calibration of the pneumotachometer was performed. Temperature, relative humidity and air pressure were recorded for corrections of spirometry parameters. Tests were performed according to European Respiratory Society criteria.18 All tests were carried out by the same technician.
The fractional concentration of exhaled nitric oxide (NO) was measured using NioxMino (Aerocrine AB, Solna, Sweden) as a marker of airway inflammation. Ambient NO levels were measured before the test. In accordance with the American Thoracic Society protocol,19 exhaled NO tests were performed before the spirometry tests, as forced expiration may influence exhaled NO levels. Two NioxMino devices were used in the study.
Airway resistance was measured using a MicroRint (Micro Medical Ltd, Kent, United Kingdom), which uses the interrupter technique.20 Maneuvers were discarded when expiration was hampered or when interruption of expiration was not timed correctly (visual check). Median values of 5–10 approved maneuvers were used in analyses. One device was used for all measurements.
The participants filled in questionnaires on time spent in traffic in the hours before the clinical measurements, and on prevalence of respiratory symptoms on a 4-point scale (no, light, moderate and severe symptoms).
Clinical tests were performed at the same time each test day to avoid effects of circadian variation in health parameters. Tests were taken before exposure (8 am), directly after exposure (10 am), and 6 hours after exposure (4 pm)—except for exhaled NO, which was measured only before and 6 hours after exposure, because airway inflammation is expected to take time to develop.
Relations between exposure to and inhaled doses of air pollutants and health outcomes were analyzed using mixed models. We used repeated measurement analysis to correct for correlation between measurements of the same subject. Random intercepts for subjects were included to account for individual differences.
Changes in health outcomes between post and pre-exposure measurements were the dependent variables. Log transformations of these measurements were used when the distribution was right-skewed. The magnitude of change in health outcomes was calculated for interquartile ranges (IQRs) of exposure and doses. We performed separate analyses for all pollutants, and also for pairs of 2 pollutants. For the real-time measurements of PN we used not only the mean, but also the median and 95th percentile (p95) of the individual readings to study the effects of peaks in exposures. The Spearman rank correlation coefficient between PN mean and median was 0.91, between PN mean and p95, 0.71, and between PN median and p95, 0.53. For PM2.5 we analyzed only mean values, as median, p95 and p99 values were highly correlated (>0.90). Potential confounders included in analyses for all endpoints were relative humidity, temperature, season, time of test, and minutes of traffic participation before baseline measurements and between the 10 am and 4 pm measurements, to correct for exposures of volunteers before or after the controlled traffic exposure. Ambient background NO2 concentrations were included as a potential confounder to take into account the difference in effects of ambient air pollution on pre- and postexposure health measurements; average NO2 concentrations of the 24 hours preceding the baseline health measurements (8 am) were used. To account for possible learning effects in spirometry, we adjusted for whether or not it was the first test-day of the participant. In exhaled NO analyses the instrument identification (ID) number and ambient (room air) NO level were added to the model. In analyses of airway resistance the technician taking the test was included as potential confounder to account for possible differences among the several technicians involved with the study. Effect modification was analyzed for body mass index (BMI, categorized as below or above 25 kg/m2), fruit and vitamin intake (the high category defined as 7 or more fruits per week or taking vitamin supplements on 5 days or more), hay fever, and whether commuting was done by bicycle or by car or bus. To assess the influence of outliers on regression analyses, Cook's D was calculated. We performed additional analyses leaving out 1% (at maximum, 4) observations with the highest Cook's Distance value. Associations were not affected, except for spirometry. Data were analyzed using SAS 9.1 (SAS Institute Inc, Cary, NC).
Measurements were performed on 47 days: 16 bicycle-sampling days, 16 car-sampling days and 15 bus-sampling days. Exposures varied widely between days (Fig. A and eTable 1, http://links.lww.com/EDE/A450). PM2.5 concentrations were likely overestimated, because of the photometrical instrument used,14 yet the differences between measurement days are likely real. Because of their increased physical activity, estimated minute ventilation of cyclists was higher than that of car and bus passengers. Ventilation of cyclists was on average 23.5 L/min, varying from 11.6–47.7 L/min among the 34 men and women, while for car passengers it was 11.8 L/min, (5.1–20.9 L/min) and for bus passengers 12.7 L/min, (5.4–19.5 L/min).15 Body surface area-adjusted inhaled doses of air pollutants (Fig. B) were calculated by multiplying the concentrations of the air pollutants with the minute ventilation and the duration of the trip, and dividing by body surface area.
Correlations between exposure levels and inhaled doses of the various air pollutants are presented in eTable 2 (http://links.lww.com/EDE/A450). The strongest correlation between exposures was the correlation between PN and soot. There was no correlation between PN and either PM2.5 or PM10 exposure. Correlations between doses were stronger than between exposures, which is explained by the constant influence of minute ventilation for all air pollutants. Spearman rank correlations between exposures and inhaled doses of each air pollutant were between 0.61 and 0.66 for PN, PM10 and soot, and 0.90 for PM2.5. Spearman correlations of PN and of soot exposures during commuting with ambient NO2 were 0.39 and 0.21, respectively.
Characteristics of the 34 participants are given in Table 1 Four persons reported shortness of breath during exercise but not at rest, and 5 subjects reported hay fever. Nobody used respiratory medication. Lung function parameters of the participants were higher than predicted using standard European Respiratory Society equations.18
In total 352 personal lung-function measurements were taken on the 47 sampling days (pre- and postexposure). The volunteers participated at least 5 and at most 12 days. Fewer than 5% of the observations showed changes in scores of respiratory symptoms (coughing, phlegm, wheezing, shortness of breath, and chest tightness) after exposure. Because of the small variation in scores, these differences were not further analyzed.
Spirometry and Air Pollution
There were no effects of exposures or inhaled doses on changes in FVC (Table 2).
PN doses were associated with small, negative effects on maximum midexpiratory flow (MMEF) and FEV1 at 4 pm compared with 8 am , but PN and soot exposures were associated with small, positive effects on maximum midexpiratory flow and FEV1 at 10 am compared with 8 am . PN, PM10 and soot exposures and inhaled doses were associated with decreased PEF at 10 am compared with 8 am . At 4 pm , PEF was still decreased but not by as much. Differences between adjusted and unadjusted estimates were small (eTable 3, http://links.lww.com/EDE/A450).
PN and PM10 exposures were related to higher effects on PEF in subjects with high BMI compared with low BMI, and to higher effects in car/bus trips compared with bicycle trips, but these differences were not significant (eTable 4 and eFigures 1 and 2, http://links.lww.com/EDE/A450). The association between PN dose and PEF was higher in subjects with high BMI and in car/bus trips. There were no differences in effect estimates between high and low fruit intake, and with exclusion of volunteers with hay fever. Excluding ambient NO2, or including ambient PM10 or ozone in the analyses instead of NO2, did not change the results.
Exhaled Nitric Oxide and Air Pollution
Exhaled NO measurements and residuals of regression analyses were skewed; therefore, we applied natural-log transformation to these values.
There was only modest effect of exposure to air pollutants or dose in the complete group of commuters on the difference between post- and pre-exposure exhaled NO measurements (Table 3). PN and soot exposures were associated with increases in exhaled NO at 4 pm compared with 8 am for commuting by bus or car. Effect estimates of PN dose were higher in car/bus trips compared with bicycle trips (eTable 4, http://links.lww.com/EDE/A450). Estimates for subgroups with high and low BMI, high and low fruit intake, and with exclusion of subjects with hay fever did not differ from complete group estimates (eTable 4 and eFigure 3, http://links.lww.com/EDE/A450). Unadjusted and adjusted estimates did not differ (eTable 5, http://links.lww.com/EDE/A450). Including ambient PM10 or ozone in the analyses instead of NO2 did not change the results.
Airway Resistance and Air Pollution
As airway resistance measurements and residuals of regression analyses were skewed, the values were natural-log transformed. There were no associations of the measured air pollutant exposures with changes in airway resistance at 4 pm compared with 8 am (Table 4). However, airway resistance increased with PN inhaled dose. Estimates adjusted for confounders did not differ consistently from unadjusted estimates, with some estimates increasing and others decreasing (eTable 6, http://links.lww.com/EDE/A450).
Median values of 5–10 maneuvers were used as airway resistance score. Sensitivity analyses on the influence of variation between the maneuvers within one test did not show differences in effects of air pollutants. Associations were stronger for commuting by car or bus than for commuting by bicycle (eTable 4 and eFigure 4, http://links.lww.com/EDE/A450). The estimates within the high BMI group were higher than within the low BMI group, although the study lacked power for this comparison (eTable 4). Estimates did not change upon exclusion of volunteers with self-reported hay fever, or when stratifying for fruit intake or cycling (eFigure 4, http://links.lww.com/EDE/A450). Excluding ambient NO2, or including ambient PM10 or ozone in the analyses instead of NO2, did not change the results.
For PN, not only mean but also median and p95 values were analyzed to account for short peaks during the 2-hour exposure. Effect estimates were similar for the 3 variables, eg, the associations between PEF at 10 am and the median of PN was −0.94% and with the 95th percentile was −0.56%.
We found associations of in-traffic exposure to particle number and soot with changes in lung function, airway resistance and exhaled NO. PN, PM10 and soot were associated with peak expiratory flow directly after exposure, but not 6 hours afterward. PN and soot were associated with increased exhaled NO after car and bus trips but not after bicycle trips. PN inhaled dose was associated with an increase in airway resistance directly following exposure, but not 6 hours later. PN doses were associated with small, negative effects on maximum midexpiratory flow and FEV1 at 4 pm compared with 8 am ; PN and soot exposures were associated with small, positive effects on maximum midexpiratory flow and FEV1 at 10 am compared with 8 am . This study adds to the small number of studies suggesting that high exposures to traffic related air pollution during commuting have measurable health effects.
Controlled diesel exhaust exposure studies generally have not found effects on lung function parameters in relation to short-term air pollution exposure.8–12 In our study, in-traffic exposure to ambient PN, PM10 or soot exposure was not associated with effects in healthy cyclists, except for an increase in PEF directly following exposure (Table 5).6 McCreanor et al2 found slightly larger adverse effects of air pollution exposure on FVC, FEV1 and maximum midexpiratory flow in asthmatics walking in central London; PEF was not reported. Effects were strongest directly after the end of the 2-hour exposure. We found a decline in PEF associated with increments in PN and soot exposure, no effects in FVC, and inconsistent effects in FEV1 and maximum midexpiratory flow. We did not study asthmatic volunteers, and air pollutant concentrations in our study were lower than in central London, possibly explaining the differences. All spirometry measurements were taken by the same technician, using the same instrument, ruling out possible effects of the technician on the tests.
We found only negative associations with PEF and no consistent associations with FEV1, the least error-prone spirometric variable. Although PEF is more effort-dependent than FEV1, it seems unlikely that the observed associations between air pollution and PEF are due to bias, as subjects and investigators were unaware of the actual air pollution concentration.
The results of our study agree closely with a recent study of road-tunnel exposures in mild asthmatics: only PEF was affected, with no effect on FEV1.21
Exhaled Nitric Oxide
Strak et al6 found associations between PN exposure of healthy cyclists and exhaled NO comparable to the associations we found in car and bus trips (Table 5). Strak et al found no associations of exhaled NO with PM10 and soot. McCreanor et al2 found slightly higher estimates for the effect on exhaled NO of exposure to EC compared with our study (Table 5), possibly because they measured effects in asthmatics, and exposure levels were higher in their study. They found no association between exhaled NO and PN or PM2.5. McCreanor et al found strongest effects on exhaled NO 3 hours after exposure, while we looked only at exhaled NO 6 hours after exposure.
Direct effects of air pollution on airway inflammation have also been observed in several controlled exposure-chamber studies. Exposure to diesel exhaust was associated with increases in neutrophils in bronchoalveolar lavage8 and bronchial wash.13 Similarly, McCreanor et al2 found increased neutrophil numbers in sputum in asthmatics.
Rudell et al9 found a smaller effect of PN on airway resistance than the effect we found for PN dose and PN exposure (Table 5). This may be explained by their shorter (1 hour) exposure time or their high, nonvariable exposure levels. Stenfors et al10 and Nordenhall et al12 found effects of PM10 from diesel exhaust on airway resistance in healthy adults and asthmatics, respectively. We did not find an effect of PM10 exposure, possibly because we did not measure effects in people with asthma, and possibly because we used the interrupter technique, while the exposure chamber studies measured airway resistance using body plethysmography. Measurements using the interrupter technique may be less sensitive in detecting bronchoconstriction.22 Airway resistance tests were taken by several technicians in our study, but including technician in the model did not change estimates.
Body Mass Index
Associations between exposures and effects in spirometry, exhaled nitric oxide and airway resistance were all stronger in subjects with high BMI (>25 kg/m2) compared with people with low BMI (≤25 kg/m2). Because only one participant had a BMI >30 kg/m2, we were not able to compare effects between obese and nonobese participants. Obesity has been found to enhance effects of air pollution on cardiovascular outcomes,23,24 but effects of body mass on effect estimates of respiratory health have not been reported before.
Our study included 352 observations, more than earlier studies looking at respiratory effects of short-term exposure to traffic emissions (Table 5). The relatively-large size of the study and the large variation in exposures and doses resulted in narrow confidence intervals. Measurements were taken throughout a complete year, including all seasons. The effect of season was included as confounder, and sensitivity analyses were performed excluding people reporting hay fever.
The plausibility of associations with in-traffic exposures is enhanced by associations between 24-hour average concentrations of traffic-related air pollution and various respiratory outcomes, including lung function, acute respiratory symptoms and airway inflammation.25
Estimated pollution doses had a slightly stronger association with health effects (especially for airway resistance) than with exposures. The small difference could be due to the high correlation between doses and exposures, related to the study design in which all volunteers were engaged in very similar activities. Alternatively, correcting for body surface area may not have sufficiently corrected for differences in airway epithelial surface area. The methodology used to estimate minute ventilation based on heart rates was demonstrated to provide reliable estimates of minute ventilation, and thus of doses.15
Inhaled doses were higher in cycling trips than in car and bus trips,14 but we showed that these higher inhaled doses did not lead to stronger health effects. On the contrary, effect estimates of particle-number doses were lower in cycling trips. Effect estimates expressed per particle number exposure were also smaller for cycling trips compared with car and bus trips, but the difference was less pronounced than for particle doses. It is unlikely that the smaller effect estimates in cyclists is explained by errors in the dose calculations, as these errors would need to be correlated with air pollution to produce higher effect estimates.
In conclusion, 2-hour in-traffic exposures to PN, PM10 and soot were associated with effects on peak expiratory flow, exhaled NO and airway resistance in healthy adults. No effects were found on other lung function parameters (FEV1, FVC, maximum midexpiratory flow) or respiratory symptoms.
We thank all volunteers for participating in the study. We thank Virissa Lenters for fieldwork assistance.
1.Kaur S , Nieuwenhuijsen MJ , Colvile RN . Fine particulate matter and carbon monoxide exposure concentrations in urban street transport microenvironments . Atmos Environ
. 2007 ;41 :4781–4810.
2.McCreanor J , Cullinan P , Nieuwenhuijsen MJ , et al. Respiratory effects of exposure to diesel traffic in persons with asthma . N Engl J Med
. 2007 ;357 :2348–2358.
3.Larsson BM , Sehlstedt M , Grunewald J , et al. Road tunnel air pollution induces bronchoalveolar inflammation in healthy subjects . Eur Respir J
. 2007 ;29 :699–705.
4.Adar SD , Adamkiewicz G , Gold DR , Schwartz J , Coull BA , Suh H . Ambient and microenvironmental particles and exhaled nitric oxide before and after a group bus trip . Environ Health Perspect
. 2007 ;115 :507–512.
5.Riediker M , Cascio WE , Griggs TR , et al. Particulate matter exposure in cars is associated with cardiovascular effects in healthy young men . Am J Respir Crit Care Med
. 2004 ;169 :934–940.
6.Strak M , Boogaard H , Meliefste K , et al. Respiratory health effects of ultrafine and fine particle exposure in cyclists . Occup Environ Med
. 2010 ;67 :118–124.
7.Krzyzanowski M, Kuna-Dibbert B, Schneider J, eds; World Health Organization. Health effects of transport-related air pollution. Copenhagen: WHO Regional Office for Europe; 2005. ISBN 92–890–1373–7.
8.Nightingale JA , Maggs R , Cullinan P , et al. Airway inflammation after controlled exposure to diesel exhaust particulates . Am J Respir Crit Care Med
. 2000 ;162 :161–166.
9.Rudell B , Ledin MC , Hammarstrom U , Stjernberg N , Lundback B , Sandstrom T . Effects on symptoms and lung function in humans experimentally exposed to diesel exhaust . Occup Environ Med
. 1996 ;53 :658–662.
10.Stenfors N , Nordenhall C , Salvi SS , et al. Different airway inflammatory responses in asthmatic and healthy humans exposed to diesel . Eur Respir J
. 2004 ;23 :82–86.
11.Salvi S , Blomberg A , Rudell B , et al. Acute inflammatory responses in the airways and peripheral blood after short-term exposure to diesel exhaust in healthy human volunteers . Am J Respir Crit Care Med
. 1999 ;159 :702–709.
12.Nordenhall C , Pourazar J , Ledin MC , Levin JO , Sandstrom T , Adelroth E . Diesel exhaust enhances airway responsiveness in asthmatic subjects . Eur Respir J
. 2001 ;17 :909–915.
13.Behndig AF , Mudway IS , Brown JL , et al. Airway antioxidant and inflammatory responses to diesel exhaust exposure in healthy humans . Eur Respir J
. 2006 ;27 :359–365.
14.Zuurbier M , Hoek G , Oldenwening M , et al. Commuters' exposure to particulate matter air pollution is affected by mode of transport, fuel type, and route . Environ Health Perspect
. 2010 ;118 :783–789.
15.Zuurbier M , Hoek G , Van den Hazel P , Brunekreef B . Minute ventilation of cyclists, car and bus passengers: an experimental study . Environ Health
. 2009 ;8 :48.
16.Brunekreef B, Janssen NA, de Hartog JJ, et al. Personal, indoor, and outdoor exposures to PM2.5 and its components for groups of cardiovascular patients in Amsterdam and Helsinki. Res Rep Health Eff Inst
17.Cyrys J , Heinrich J , Hoek G , et al. Comparison between different traffic-related particle indicators: elemental carbon (EC), PM2 5 mass, and absorbance . J Expo Anal Environ Epidemiol
. 2003 ;13 :134–143.
18.Quanjer PH , Tammeling GJ , Cotes JE , Pedersen OF , Peslin R , Yernault JC . Lung-volumes and forced ventilatory flows—Report Working Party Standardization of Lung-Function Tests European-Community for Steel and Coal—official statement of the European Respiratory Society . Eur Respir J
. 1993 ;16 :5–40.
19.American Thoracic Society; European Respiratory Society. ATS/ERS recommendations for standardized procedures for the online and offline measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide, 2005. Am J Respir Crit Care Med
20.Pouls KP , Alsema LE , van der Laan H , Evenhuis HM , Penning C . Microrint pulmonary function testing in older adults with an intellectual disability . Respir Med
. 2009 ;103 :1954–1959.
21.Larsson BM , Grunewald J , Skold CM , et al. Limited airway effects in mild asthmatics after exposure to air pollution in a road tunnel . Respir Med
. 2010 ;104 :1912–1918.
22.Phagoo SB , Watson RA , Silverman M , Pride NB . Comparison of four methods of assessing airflow resistance before and after induced airway narrowing in normal subjects . J Appl Physiol
. 1995 ;79 :518–525.
23.Madrigano J , Baccarelli A , Wright R , et al. Air pollution, obesity, genes, and cellular adhesion molecules . Occup Environ Med
. 2010 ;67 :312–317.
24.Dubowsky SD , Suh H , Schwartz J , Coull BA , Gold DR . Diabetes, obesity, and hypertension may enhance associations between air pollution and markers of systemic inflammation . Environ Health Perspect
. 2006 ;114 :992–998.
25.HEI Panel on the Health Effects of Traffic-Related Air Pollution. Traffic-related air pollution: a critical review of the literature on emissions, exposure, and health effects. HEI special report 17. Boston, MA: Health Effects Institute; 2010.