Exposure to traffic and traffic-related air pollution is associated with a wide array of health effects. Time spent in a vehicle, in active transportation, along roadsides, and in close proximity to traffic can substantially contribute to daily exposure to air pollutants. For this study, we evaluated daily time spent in transportation and traffic-influenced microenvironments by urban Canadians using the Canadian Human Activity Pattern Survey (CHAPS) 2 results. Approximately 4–7% of daily time was spent in on- or near-road locations, mainly associated with being in a vehicle and smaller contributions from active transportation. Indoor microenvironments can be impacted by traffic emissions, especially when located near major roadways. Over 60% of the target population reported living within one block of a roadway with moderate to heavy traffic, which was variable with income level and city, and confirmed based on elevated NO2 exposure estimated using land use regression. Furthermore, over 55% of the target population ≤ 18 years reported attending a school or daycare in close proximity to moderate to heavy traffic, and little variation was observed based on income or city. The results underline the importance of traffic emissions as a major source of exposure in Canadian urban centers, given the time spent in traffic-influenced microenvironments.
Traffic-related air pollution (TRAP) is associated with adverse cardiorespiratory health effects, including exacerbation of asthma, incident asthma, reduced lung function, myocardial infarction, progression of atherosclerosis, and cardiovascular mortality (Health Effects Institute 2010). Recent evidence also links TRAP exposure to a wide array of other adverse health impacts throughout the life course ranging from adverse birth outcomes to dementia (Stieb et al. 2016; Oudin et al. 2016; Pedersen et al., 2013; Chen et al. 2017a). TRAP is a complex mixture consisting of particle and gaseous components and includes both primary air pollutants, which are directly emitted from vehicle exhaust, brake, and tire wear, as well as secondary air pollutants that form from reactions of primary pollutants in the atmosphere. Pollutants commonly measured as surrogates for traffic exposure include carbon monoxide (CO), nitrogen dioxide (NO2), elemental or black carbon (EC or BC), benzene, and ultrafine particles (UFP). Traffic-related emissions impact ambient air quality and can be a major source of exposure to air pollutants in urban areas. In Canada, it has been estimated that about ten million people (approximately 32% of the population) live within 500 m of highways or 100 m of major urban roads (Brauer et al. 2013). In urban areas, traffic is also the principal source of exposure to noise (Allen et al. 2009; Davies et al. 2009), which has been associated with a variety of adverse impacts from annoyance to ischemic heart disease (World Health Organization 1999, 2011).
For many epidemiological studies, traditional exposure assessment methods have relied on use of central site monitors or predicted ambient concentrations at residence to assign exposures to the study population. However, these approaches can lead to exposure misclassification and/or bias by not incorporating a person’s daily movements and activities into the exposure estimate (Setton et al. 2011; Baxter et al. 2013; Ozkaynak et al. 2013). Exposure estimates have been improved when considering daily time spent away from the home and daily time in transit (Dons et al. 2011; de Nazelle et al. 2013; Dias and Tchepel 2014; Ragettli et al. 2015; Shekarrizfard et al. 2016). By characterizing both time spent and pollutant concentrations in individual microenvironments, it is possible to identify locations that contribute substantially to overall exposure and to guide exposure mitigation strategies (Almeida-Silva et al. 2014; Tchepel et al. 2014; Williams and Knibbs 2016).
The Canadian Human Activity Pattern Survey (CHAPS) 2 was conducted by Health Canada to provide information on daily time-activity patterns, potential exposure to environmental contaminants, occupational activities, and housing characteristics (Matz et al. 2014). The CHAPS 2 results can be used to inform exposure assessment, risk assessment, and risk management activities related to environmental health. In the CHAPS 2 survey, several questions were focussed on daily time spent in activities and locations associated with TRAP exposure. Time spent on or near roadways, especially roadways with heavy traffic, can be a substantial contributor to daily exposure to TRAP and to air pollution in general. Roadways and transit locations (e.g., locations associated with commuting activities) represent exposure hot spots, with pollutant concentrations exceeding those measured at regional outdoor monitors (Weichenthal et al. 2015).
Initial findings from CHAPS 2 indicated that age was a significant predictor of daily time spent in a vehicle, with adults spending the most time (1.5 h per day) in this microenvironment (Matz et al. 2014). In the present study, our objectives were to (1) generate population-representative estimates of daily time spent in transportation and traffic-influenced microenvironments, as possible sources of TRAP exposure for urban Canadians; (2) identify factors influencing these time-activity patterns; (3) evaluate internal consistency between 24-h recall diary and survey responses pertaining to time spent in traffic-influenced microenvironments; and (4) confirm self-reported residential proximity to roadways with moderate to heavy traffic based on estimated ambient NO2 concentration at the respondent’s residence from a land-use regression (LUR) model.
Canadian Human Activity Pattern Survey 2 survey data
A detailed description of survey methodology and study population for CHAPS 2 has previously been published (Matz et al. 2014). Briefly, a random digit dialing survey was conducted, in 2010–2011, using computer-assisted telephone interview (CATI) technology to collect time-activity data and questionnaire responses. The target population was Canadian residents of all ages with a telephone residing in five urban areas (Vancouver, Edmonton, Toronto, Montreal, and Halifax) and two rural regions (Haldimand-Norfolk, Ontario, and Annapolis Valley-Kings County, Nova Scotia). For the present analysis, only data from urban respondents were considered as traffic-related air pollution is principally associated with urban centers. The CHAPS 2 survey instrument was based on the original CHAPS survey (Leech et al. 1996) and consisted of three main components: questions regarding respondent characteristics and household composition, a 24-h recall diary, and a supplemental questionnaire covering activities related to exposures to specific contaminants, dwelling characteristics, socio-economic status (SES), and health status. The 24-h recall diary was used to collect time-activity information as respondents described their activities starting from midnight of the previous day. The complete survey instrument is available in Supplementary Materials of the methodology publication (Matz et al. 2014). This study was approved by Health Canada’s Research Ethics Board.
Sampling was conducted in summer 2010 and winter 2011. A total of 3551 urban respondents participated in the survey. Details of the response rates and representativeness of the survey are presented in Matz et al. (2014). Survey weights were calculated to account for oversampling of certain age groups, adjustments for non-response, and to allow for generalization of survey results to the entire target area population.
The main purpose of this analysis was to evaluate daily time spent in transportation and traffic-influenced microenvironments. Analysis of time spent in vehicles, engaged in active transportation, or otherwise on or near a roadway was based on data collected in the 24-h recall diary. Time spent in parking lots or gas stations was based on data collected in the questionnaire. Information on time spent in moderate to heavy traffic was collected in both the 24-h recall diary and supplemental questionnaire. For the recall diary of the previous day, any time a respondent indicated that he/she was in a vehicle (e.g., car, truck), in transit (e.g. bus, rapid transit), engaged in active transport (e.g., walking, running, cycling), or waiting at a transit stop, the respondent was asked if this activity was conducted on or near a roadway with moderate to heavy traffic, and if so, for how long. For the supplemental questionnaire, respondents were asked whether in the previous day they had spent time in a car, van, truck, or bus, in moderate to heavy traffic or running, walking, or standing along a road with moderate to heavy traffic, and if so, for how long. These data were collected in both the diary and supplemental questionnaire as respondents may not remember to report all activities during the recall process (Klepeis et al. 2001). In the present study, the 24-h recall diary was considered the primary data source and the supplemental questionnaire questions on traffic-related exposures were used to assess internal consistency. For the recall diary and supplemental questionnaire, a roadway with moderate to heavy traffic was defined as one that has a substantial amount of traffic for several hours of the day, such as a main thoroughfare, a busy boulevard, or a highway. Survey details, including specific questions evaluated in this publication, are provided in the Supplemental Materials.
CHAPS 2 has a complex design, involving stratification by location and season and clustering by households. The reported statistical analyses accounted for this structure by using the respective parameters in all software procedures (Heeringa et al. 2010). Sampling weights were used to make the estimates generalizable to the CHAPS 2 target population. The sampling errors were estimated using Taylor linearization method. Summary statistics (means, percentiles, and percentages) were estimated using SAS SURVEYMEANS and SURVEYFREQ procedures and SUDAAN DESCRIPT procedure. Regression analysis was performed using SAS SURVEYREG procedure. Odds ratios were estimated by fitting logistic models using SAS SURVEYLOGISTIC procedure. Significance of the model parameters was tested using Wald chi-square test statistics. Differences in means were tested using Wald F statistic. Significance level used in the analysis was 0.05. The analysis was performed with SAS Enterprise Guide 4.2 (SAS Institute Inc., Cary, NC) and SUDAAN 10.0.1. The estimated sampling variability was evaluated using Statistics Canada’s guidelines for reliability of household survey data (Statistics Canada 2014): estimates with high sampling variability (coefficient of variation > 16.5 and ≤ 33.3%) were interpreted with caution, and estimates with very high variability (coefficient of variation > 33.3%) were suppressed.
For both the 24-h recall diary and supplemental questionnaire, time spent in each microenvironment was estimated for the entire target population (i.e., including both those who did and did not spend time in the microenvironment) and restricted to “doers” (i.e., only those who spent at least 1 min of time in the microenvironment or activity). The doers provide specific information about that portion of the total population that was in a given microenvironment. Where sample size was sufficient (> 10 respondents), analysis was also conducted for four age groups, 0–4 years, 5–18 years, 19–64 years, and 65+ years. As age is a predictor of time-activity patterns (Matz et al. 2014), it was anticipated the daily activity patterns would be similar within these age groups.
To evaluate internal consistency of the survey responses, Spearman correlation analysis and t tests were used to evaluate the relationship between time spent in traffic reported in the 24-h recall diary and supplemental questionnaire.
Income, age, education level, employment status, and city of residence were chosen a priori as covariates for analysis. Income was evaluated as above or below low-income cut-offs (LICO), defined by Statistics Canada as thresholds below which a household will likely devote a larger share of its income to food, shelter, and clothing compared to an average family, adjusted for household size and community size (Statistics Canada 2013).
In order to provide external confirmation of self-reported residential proximity to moderate to heavy traffic, NO2 concentrations at residence were examined. Residential estimates of ambient NO2 concentrations were derived from a national LUR model combined with deterministic gradients to capture regional and local scale variation. The development of the national LUR model has been described in previous publications (Hystad et al. 2011; Crouse et al. 2015). Briefly, a national LUR model was developed to predict regional NO2 concentrations using National Air Pollution Surveillance (NAPS; http://www.ec.gc.ca/rnspa-naps/) monitoring data collected in 2006 (Hystad et al. 2011; Crouse et al. 2015). Final LUR model predictors include road length within 10 km, 2005–2011 satellite NO2 estimates, area of industrial land use within 2 km, and summer rainfall. The model explains 73% of the variation in mean annual NAPS concentrations from 2006, with a root mean square error of 2.9 ppb. Local scale variation due to vehicle emissions was modeled using deterministic gradients from the literature and kernel density measures as described by Crouse et al. (2015).
NO2 concentrations were estimated for each postal code based on representative points. Representative points are unique coordinates within each postal code that reflect the postal code centroid for postal codes capturing a single block, or multiple central points along a line for postal codes larger than one block (Statistics Canada 2016). In urban areas, postal codes typically represent a single city block or a single apartment building. For postal codes with more than one representative point (e.g., postal codes larger than one city block), NO2 estimates for all representative points within the postal code were averaged to generate the postal code level estimate. Postal code level NO2 concentrations were assigned to CHAPS 2 respondents based on self-reported residential postal code. Mean NO2 concentrations were estimated for subpopulations defined by living within one block of a roadway with moderate to heavy traffic, income level, or education. Differences between subpopulation mean NO2 concentrations were tested using t test. The analysis was performed using DESCRIPT procedure in SUDAAN 10.0.1.
Microenvironments influenced by traffic-related air pollution
Time spent in microenvironments which are influenced by TRAP is summarized in Table 1. Among those who spent any time in these microenvironments (“doers”), over 1 h a day on average was spent in a car or on a bus and over 30 min a day on average was spent walking. The time in these microenvironments increases to approximately 3.5, 2.9, and 2.0 h, respectively, when considering the 95th percentile of the doers, representing a subgroup with potentially high TRAP exposure. Overall, 16.7 and 60.2% of the target population spent time in an enclosed or underground parking garage (20.4 min on average) or surface parking (10.0 min on average), respectively. Only 12.5% of the target population went to a gas station and of these, 57.4% pumped fuel and 30.8% were in a vehicle while someone else pumped fuel. A summary of time spent by doers in all microenvironments that may be influenced by vehicle emissions, collected in the recall diary, is provided in Supplemental Material Table S1.
Time spent in transportation and traffic
24-h recall diary
Based on the 24-h recall diary, more people reported, in general, spending time in a vehicle (Table 2) than using active transportation (Table 3) for each age group. Specifically, it was estimated that ≥ 62.3% of the target population, of each age group, reported being in a vehicle on road, while ≤ 41.5% of each age group spent time in active transportation. Additionally, mean daily time in a vehicle was typically greater than time spent in active transportation. It was also estimated that at least 43.6% of the target population for each age group reported being in moderate to heavy traffic while in an on-road location, for an average of 39.3–54.2 min for the four age groups. In comparison, ≤ 21.7% of each age group was in moderate to heavy traffic while engaged in active transportation, for an average of 20.7–38.9 min. For all respondents, mean total daily time in on- or near-road microenvironments (Table 4) ranged from 56.3–101.0 min, with 22.0–41.0 min on average in moderate to heavy traffic. In the doer group, the mean daily time on- or near-road was 72.3–111.4 min, with 28.1–45.2 min in moderate to heavy traffic, and for those who spent any time in the vicinity of moderate to heavy traffic, mean daily time in moderate to heavy traffic ranged from 42.8–58.2 min. Age was a significant predictor for daily time spent in a vehicle on road and associated time in moderate to heavy traffic (p < 0.001), as well as total daily time spent on-or near road and associated time in moderate to heavy traffic (p < 0.001). Age was not a significant predictor for daily time spent in active transportation (p = 0.25), but was a significant predictor for the associated time in moderate to heavy traffic (p = 0.0063). No summer-winter or weekday-weekend differences were noted for time spent in active transportation.
Daily time spent in the vicinity of moderate to heavy traffic, in various microenvironments, was also assessed in the supplemental questionnaire to assess internal consistency of the CHAPS 2 survey. The estimated portions of the target population and time spent in the moderate to heavy traffic while in a vehicle, while running, walking, or standing along a roadside, and total time are available in the Supplemental Materials (Tables S2, S3, and S4, respectively). From the data collected in the supplemental questionnaire, age was a significant predictor for time spent in moderate to heavy traffic while in a vehicle (p = 0.002) and total time spent in traffic-related microenvironments (p = 0.0014). In comparison, age was not a significant predictor for time spent in moderate to heavy traffic while running, walking, or standing along a roadside (p = 0.27).
Measures of time spent in moderate to heavy traffic while being in a vehicle and of total time in moderate to heavy traffic were highly correlated (ρ = 0.75–0.83) between the 24-h recall diary and the supplemental questionnaire, for all respondents and by age group. Similar degrees of correlation were also observed by household income level, employment status, and education level (ρ ≥ 0.71) (Supplemental Material Table S5). Although highly correlated, respondents reported spending more time in moderate to heavy traffic in the supplemental questionnaire compared to the 24-h recall diary; on average, a difference of 9.1 min more was reported while in a vehicle (p < 0.05) and 15.0 min more overall (p < 0.05). Mean differences for reported times in moderate to heavy traffic between the 24-h recall diary and supplemental questionnaire were variable by age group, household income level, employment status, and education level (Supplemental Material Table S6). In some instances, wide 95% confidence intervals were estimated, reflecting the smaller sample sizes in the subgroups being compared.
Proximity to roadway
Overall, an estimated 60.6% (95% CI 57.9–63.3%) of the target population reported living within one block of a roadway with moderate to heavy traffic, and this was reported more commonly by those with household income ≤ LICO, but not by those with less than university education (Table 5). It was also reported more commonly in Toronto, Halifax, Vancouver, and Edmonton than in Montreal. Additionally, an estimated 55.7% (95% CI 50.3–61.1%) of the target population up to 18 years reported attending a school or daycare within one block of a roadway with moderate to heavy traffic. This was not significantly associated with income, but the odds were significantly greater for Toronto compared to the reference city, Montreal (Table 6). Estimates based on LUR modeling indicated that ambient NO2 concentrations were greater at the residences of those who reported living within one block of a roadway with moderate to heavy traffic compared to those that did not (19.2 ppb vs. 16.7 ppb, p < 0.001), providing a confirmation of the supplemental questionnaire responses. A significantly greater NO2 exposure was also noted for those with household income ≤ LICO (19.6 ppb vs. 17.7 ppb, p = 0.002). With respect to differences by education level, NO2 exposure was significantly lower for those who had completed secondary education compared to those that had completed university (18.0 ppb vs. 19.0 ppb, p = 0.027).
Those who reported living within one block of a roadway with moderate to heavy traffic spent more time in the vicinity of moderate to heavy traffic compared to those that did not report living in close proximity to traffic (Table 7). From the diary, daily time spent in moderate to heavy traffic while in active transportation was greater among those living in proximity to traffic for those 0–4 years (Δ = 20.4 min, p = 0.0427), 19–64 years (Δ = 9.5 min, p = 0.0015), and 65+ years (Δ = 8.7 min, p = 0.0043), though total time in active transportation was not different between the groups. From the supplemental questionnaire, time spent in moderate to heavy traffic while walking, running, or standing along a roadside was greater for those 0–4 years (Δ = 19.9 min, p = 0.0159) and 19–64 years (Δ = 8.7 min, p < 0.0001). No differences were noted for time spent in moderate to heavy traffic associated with being in a vehicle or total time in moderate to heavy traffic between those who reported living within one block of a major roadway and those who did not.
The main purpose of this study was to evaluate daily time spent by urban Canadians in transportation and traffic-influenced microenvironments. The CHAPS 2 survey results indicated that urban Canadians spend approximately 4–7% of daily time (56.3–101.0 min) in on- or near-road locations by age group, and this increased to 5–8% (72.8–111.4 min) among individuals who spent any time in those locations. Children (0–18 years), considered a sensitive subpopulation to the adverse effects of air pollution, spend on average approximately 1 h a day on or near roadways. These results are consistent with other recent studies, which have reported about 6% of daily time spent in transportation microenvironments (Dons et al. 2011, 2012; de Nazelle et al. 2013). Dons et al. (2011) further identified that for full-time workers, daily time spent in transportation increases to 7.8% compared to 6.2% for the entire study population. Additionally, motorized vehicles were the most commonly used mode of transportation and represented larger portions of daily time compared to active transportation (Dons et al. 2011; de Nazelle et al. 2013).
Results from the present study highlight the important role of traffic as a major source of air pollution exposure for Canadians living in urban centers. For each of the age groups considered in this study, a substantial amount of daily time was spent in transportation-influenced microenvironments with moderate to heavy traffic. Importantly, a significant proportion of the target population engaged in active transportation near roadways (approximately 30–40% across the age groups) and approximately 11–22% of the target population did so in the proximity of moderate to heavy traffic. For some age groups, living within one block of a roadway with moderate to heavy traffic was associated with increased daily time spent in traffic while engaged in active transportation. This implies that there may be age-related differences in the degree of exposure misclassification when exposure is assigned based on place of residence (Setton et al. 2011; Gurram et al. 2015; Ragettli et al. 2015). Overall, the results highlight the potential impact of traffic emissions on the population and that access to active transportation options in areas or along routes less impacted by vehicle emissions may have public health benefits.
Compared to time spent at home, time spent in transportation locations accounts for a small portion of the day but has been demonstrated to account for a greater portion of total daily exposure to air pollution (Dons et al. 2011, 2012; de Nazelle et al. 2013; Dias and Tchepel et al. 2014; Lane et al. 2015). In each of these studies, the percent contribution of transportation microenvironments to time-weighted exposure for black carbon, PM2.5, ultrafine particles, or NO2 was ~ 2–4 times greater than the percent of daily time associated with transportation, since transportation locations typically have greater concentrations of these pollutants compared to home and work locations. Additionally, greater exposure error and bias in risk estimates have been reported when study subjects spend more time away from home, travel greater distances, or spend more time in travel, compared to estimates based on residence location only (Setton et al. 2011; Gurram et al. 2015; Ragettli et al. 2015). People that commute by car are exposed to higher levels of air pollutants than active commuters, including both pedestrians and cyclists (Cepeda et al. 2017). However, modes of active transportation can result in greater personal exposure per trip compared to traveling in vehicle for a given route, due to the increased travel time in areas with higher levels of traffic (Good et al. 2016), as well as a larger inhaled dose due to the increased breathing rate (Dons et al. 2012; Ragettli et al. 2015; Cepeda et al. 2017). Nonetheless, there is evidence that outdoor physical activity results in a net health benefit, despite the increased exposure to air pollutants (Andersen et al. 2015, Cepeda et al. 2017).
Indoor air quality at home, work, or school, can also be influenced by traffic emissions, especially when these indoor locations are situated near a major roadway. Given that Canadians spend approximately 21 h per day indoors (Matz et al. 2014), this represents an additional and potentially significant contribution of traffic emissions to daily exposure to air pollution. It has been estimated that approximately 32% of the Canadian population lives within 100 m of a major urban road or 500 m of a highway (Brauer et al. 2013). In the present study, 60.6% of the target population reported living within one block of a roadway with moderate to heavy traffic. This difference may be explained, at least in part, by differences in the population under consideration: in the present analysis, the sample was restricted to the urban population of five major Canadian cities, in contrast to the Brauer et al. (2013) report which included both urban and rural populations. In addition, 55.7% of the target population up to 18 years, in the present study, reported attending a school or daycare within one block of a roadway with moderate to heavy traffic. In a study of ten major Canadian cities, 16.3 and 36.1% of public elementary schools, based on geocoding, were within 75 and 200 m of a major roadway, respectively (Amram et al. 2011). For Montreal and Vancouver, two of the cities included in CHAPS 2, over 50% of the public elementary schools were within 200 m of a major roadway. Moreover, schools in neighborhoods with higher dwelling density and lower median income were closer to a major roadway. In the present study, household income did not have a significant association with attending a school or daycare within one block of a roadway with moderate to heavy traffic, and little variation was observed between cities. Concerns over potential health effects in children and youth have led to recent investigations of possible interventions to reduce exposure to TRAP in schools (MacNeill et al. 2016, van der Zee et al. 2017).
Living within one block of a roadway with moderate to heavy traffic and elevated NO2 exposure (an indicator of traffic exposure) were both associated with lower household income in this study. Similar results have been reported in previous Canadian evaluations. Lower SES residential areas were more prevalent within 200 m of a major highway, compared to high SES areas, in Vancouver, Toronto, and Montreal (Canadian Institute for Health Information 2011). An opposite trend was noted for Edmonton, which was attributed to the major highway passing through suburban areas and not the city center. Additionally, several studies have evaluated NO2 levels and social geography in major Canadian cities, identifying predictors of potential susceptibility or environmental injustice. Higher NO2 levels have been associated with various neighborhood SES characteristics in Toronto, Montreal, and/or Vancouver, including higher unemployment (Crouse et al. 2009), lower household income (Buzzelli and Jerrett 2007; Crouse et al. 2009; Su et al. 2010; Pinault et al. 2016a, b), people living alone (Crouse et al. 2009; Pinault et al. 2016a), low education (Buzzelli and Jerrett 2007), single-parent homes (Buzzelli and Jerrett 2007; Pinault et al. 2016b), visible minorities (Crouse et al. 2009; Pinault et al. 2016b), and linguistic isolation (Pinault et al. 2016a). However, for both Toronto (Buzzelli and Jerrett 2007) and Montreal (Crouse et al. 2009), elevated levels of NO2 were also associated with indicators of higher SES, including greater dwelling value, high-status occupation, higher education, and/or higher income, in some city center neighborhoods. These results were attributed to gentrification of downtown areas in Toronto and, for Montreal, historically affluent enclaves and presence of universities in the city center. Although these results demonstrate that the relationship between measures of SES and TRAP exposure is not entirely straightforward and may differ within and between cities, there is growing evidence that lower SES residential areas in Canada may be more impacted by the negative effects of traffic.
Strengths and limitations
A key strength of CHAPS 2 is the large sample size, including over 3500 respondents from five major urban centers, which provided detailed time-activity data that can be evaluated by age group. Twenty-four-hour recall diaries are a standard method for collecting time-activity information (Leech et al. 1996; Klepeis et al. 2001) and the reproducibility of the data has been established (Freeman et al. 1999; Wu et al. 2011). Use of global positioning system (GPS) devices has been proposed to improve the accuracy of location data and reduce participant burden (Breen et al. 2014; Nethery et al. 2014); however, these methods are not feasible for studies with a large number of participants.
In this study, internal consistency (high correlation) was observed between time spent in moderate to heavy traffic based on the 24-h recall diary and supplemental questionnaire responses, though a greater amount of time was reported in the supplemental questionnaire portion of the survey. This could arise if respondents failed to recall some travel activities in the diary portion, but included the associated time in traffic with the supplemental questionnaire. Alternatively, respondents may overestimate time in traffic when not prompted by a sequential recall dairy. Similarly, a dietary survey study found greater reported intake of fruits and vegetables when assessed using questionnaires compared to 24-h dietary recall interviews (Eaton et al. 2013).
Confirmation of questionnaire responses regarding living near a roadway with moderate to heavy traffic was provided by LUR-based NO2 concentration estimates; mean NO2 exposure was greater for those who reported living within one block of a roadway with moderate to heavy traffic than those who did not and was similarly associated with household income. There are several potential limitations in the use of the national model developed by Hystad et al. (2011). First, the model was developed to predict mean annual concentrations from 2006. However, while the regional background values were based on 2006 monitoring data, the final estimates produced by the model were adjusted to account for local variation based on road network information, which did not vary significantly over time. Assignment of ambient concentrations using postal codes may introduce additional error due to the differences in the postal code size between different geographic areas. This source of error is likely minor in the reported analyses, because they were limited to urban areas, where postal codes typically reflect a single city block. Despite their potential limitations, these NO2 estimates have been applied widely in previous studies to elucidate associations between air quality and health (e.g., Ashley-Martin et al. 2016; Stieb et al. 2016; Chen et al. 2017b).
Response rates were low (Matz et al. 2014), in keeping with a downward trend for telephone surveys which has steepened in recent years (Pew Research Center 2012). Reduced response rates may introduce bias if respondents differ from non-respondents for variable(s) of interest. However, proper weighting and adjustment for non-response can allow for generalization of survey results to the entire target population despite low response rates (Pew Research Center 2012).
Overall, the CHAPS 2 study provides quantitative estimates of time spent in traffic-influenced microenvironments by urban Canadians, both in vehicle and when engaged in active transportation. Previous studies have reported that total daily personal exposure was disproportionately impacted by time spent in transportation microenvironments. In this population representative study, we found that the proportion of daily time spent in transportation microenvironments was consistent with the previous exposure studies, suggesting that traffic emissions may be a major contributor to air pollution exposure for Canadians living in urban centers.
Allen RW, Davies H, Cohen MA, Mallach G, Kaufman JD, Adar SD (2009) The spatial relationship between traffic-generated air pollution and noise in 2 US cities. Environ Res 109:334–342
Almeida-Silva M, Wolterbeek HT, Almeida SM (2014) Elderly exposure to indoor air pollutants. Atmos Environ 85:54–63
Amram O, Abernethy R, Brauer M, Davies H, Allen RW (2011) Proximity of public elementary schools to major roads in Canadian urban areas. Int J Health Geogr 10:68
Andersen ZJ, de Nazelle A, Mendez MA, Garcia-Aymerich J, Hertel O, Tjonneland A et al (2015) A study of the combined effects of physical activity and air pollution on mortality in elderly urban residents: the Danish diet, cancer, and health cohort. Environ Health Perspect 123:557–563
Ashley-Marin J, Lavigne E, Arbuckle TE, Johnson M, Hystad P, Crouse DL et al (2016) Air pollution during pregnancy and cord blood immune system biomarkers. J Occup Enivron Med 58:979–986
Baxter LK, Burke J, Lunden M, Turpin BJ, Rich DQ, Thevenet-Morrison K et al (2013) Influence of human activity patterns, particle composition, and residential air exchange rates on modeled distributions of PM2.5 exposure compared with central-site monitoring data. J Expo Sci Environ Epidemiol 23:241–247
Brauer M, Reynolds C, Hystad P (2013) Traffic-related air pollution and health in Canada. CMAJ 185:1557–1558
Breen MS, Long TC, Schultz BD, Crooks J, Breen M, Langstaff JE (2014) GPS-based microenvironment tracker (MicroTrac) model to estimate time-location of individuals for air pollution exposure assessments: model evaluation in central North Carolina. J Expo Sci Environ Epidemiol 24:412–420
Buzzelli M, Jerrett M (2007) Geographies of susceptibility and exposure in the city: environmental inequality of traffic-related air pollution in Toronto. Can J Reg Sci 30:195–210
Canadian Institute for Health Information (2011) Urban physical environments and health inequalities. Ottawa, Canada, 2011. https://secure.cihi.ca/free_products/cphi_urban_physical_environments_en.pdf. Accessed 4 Jan 2017
Cepeda M, Schoufour J, Freak-Poli R, Koolhaas CM, Dhana K, Bramer WM et al (2017) Levels of ambient air pollution according to mode of transport: a systematic review. Lancet Public Health 2:e23–e34
Chen H, Kwong JC, Copes R, Tu K, Villeneuve PJ, van Donkelaar A et al (2017a) Living near major roads and the incidence of dementia, Parkinson’s disease, and multiple sclerosis: a population-based cohort study. Lancet 389:718–726
Chen H, Kwong JC, Copes R, Hystad P, van Donkelaar A, Tu L et al (2017b) Exposure to ambient air pollution and the incidence of dementia: a population-based cohort study. Environ Int 108:271–277
Crouse DL, Ross NA, Goldberg MS (2009) Double burden of deprivation and high concentrations of ambient air pollution at the neighbourhood scale in Montreal, Canada. Soc Sci Med 69:971–981
Crouse DL, Peters PA, Hystad P, Brook JR, van Donkelaar A, Martin RV et al (2015) Ambient PM2.5, O3, and NO2 exposures and associations with mortality over 16 years of follow-up in the Canadian census health and environment cohort (CanCHEC). Environ Health Perspect 11:1180–1186
Davies HW, Vlaanderen JJ, Henderson SB, Brauer M (2009) Correlation between co-exposures to noise and air pollution from traffic sources. Occup Environ Med 66:347–350
de Nazelle A, Seto E, Donaire-Gonzalez D, Mendez M, Matamala J, Nieuwenhuijsen MJ et al (2013) Improving estimates of air pollution exposure through ubiquitous sensing technologies. Environ Pollut 176:92–99
Dias D, Tchepel O (2014) Modelling of human exposure to air pollution in the urban environment: a GPS-based approach. Environ Sci Pollut Res 21:3558–3571
Dons E, Panis LI, Van Poppel M, Theunis J, Willems H, Torfs R et al (2011) Impact of time-activity patterns on personal exposure to black carbon. Atmos Environ 25:3594–3602
Dons E, Panis LO, Van Poppel M, Theunis J, Wets G (2012) Personal exposure to black carbon in transport microenvironments. Atmos Environ 55:392–398
Eaton DK, O’Malley E, Brener ND, Scanlon KS, Kim SA, Dmissie Z (2013) A comparison of fruit and vegetable intake estimates from three survey questions to estimates from 24-hour dietary recall interviews. J Acad Nutr Diet 113:1165–1174
Freeman NC, Lioy PJ, Pellizzari E, Zelon H, Thomas K, Clayton A et al (1999) Responses to the region 5 NHEXAS time/activity diary. National Human Exposure Assessment Survey. J Expo Anal Environ Epidemiol 9:414–426
Good N, Molter A, Ackerson C, Bachand A, Carpenter T, Clark ML et al (2016) The Fort Collins commuter study: impacts of route type and transport mode on personal exposure to multiple air pollutants. J Expo Sci Environ Epidemiol 26:397–404
Gurram S, Stuart AL, Pinjari AR (2015) Impacts of travel activity and urbanicity on exposure to ambient oxides of nitrogen and exposure disparities. Air Qual Atmos Health 8:97–114
Health Effects Institute (2010) Traffic-related air pollution: a critical review of the literature on emissions, exposure, and health effects. HEI Special Report 17. Health Effects Institute, Boston https://www.healtheffects.org/publication/traffic-related-air-pollution-critical-review-literature-emissions-exposure-and-health. Accessed 4 Nov 2016
Heeringa SG, West BT, Berglund PA (2010) Applied survey data analysis. CRC Press / Taylor & Francis, Boca Raton, FL
Hystad P, Setton E, Cervantes A, Poplawski K, Deschenes S, Brauer M (2011) Creating national air pollution models for population exposure assessment in Canada. Environ Health Perspect 119:1123–1129
Klepeis NE, Nelson WC, Ott WR, Robinson JP, Tsang AM, Switzer P et al (2001) The National Human Activity Pattern Survey (NHAPS): a resource for assessing exposure to environmental pollutants. J Expo Anal Environ Epidemiol 11:231–252
Lane KJ, Levy JI, Scammell MK, Patton AP, Durant JL, Mwamburi M et al (2015) Effect of time-activity adjustments on exposure assessment for traffic-related ultrafine particles. J Expo Sci Environ Epidemiol 25:506–516
Leech JA, Wilby K, McMullen E, Laporte K (1996) Canadian human time-activity pattern survey report of methods and population surveyed. Chronic Dis Can 17:118–123
MacNeill M, Dobbin N, St-Jean M, Wallace L, Marro L, Shin T et al (2016) Can changing the timing of outdoor air intake reduce indoor concentrations of traffic-related pollutants in schools? Indoor Air 26:687–701
Matz CJ, Stieb DM, Davis K, Egyed M, Rose A, Chou B et al (2014) Effects of age, season, gender and urban–rural status on time-activity: Canadian human activity pattern survey 2 (CHAPS 2). Int J Environ Res Public Health 11:2108–2124
Nethery E, Mallach G, Rainham D, Goldberg MS, Wheeler AJ (2014) Using global position systems (GPS) and temperature data to generate time-activity classifications for estimating personal exposure in air monitoring studies: an automated method. Environ Health 13:33
Oudin A, Forsberg B, Adolfsson AN, Lind N, Modig L, Nordin M et al (2016) Traffic-related air pollution and dementia incidence in northern Sweden: a longitudinal study. Environ Health Perspect 124:306–312
Ozkaynak H, Baxter LK, Dionisio KL, Burke J (2013) Air pollution exposure prediction approaches used in air pollution epidemiology studies. J Expo Sci Environ Epidemiol 23:566–572
Pedersen M, Giorgis-Allemand L, Bernard C, Aguilera I, Andersen AM, Ballester F et al (2013) Ambient air pollution and low birthweight: a European cohort study (ESCAPE). Lancet Respir Med 1:695–704
Pew Research Center (2012) Assessing the representativeness of public opinion surveys. Pew Research Centre, Washington, D.C http://www.people-press.org/2012/05/15/assessing-therepresentativeness-of-public-opinion-surveys/. Accessed 4 Nov 2016
Pinault L, Crouse D, Jerrett M, Brauer M, Tjepkema M (2016a) Spatial associations between socioeconomic groups and NO2 air pollution exposure within three large Canadian cities. Environ Res 147:373–382
Pinault L, Crouse D, Jerrett M, Brauer M, Tjepkema M (2016b) Socioeconomic differences in nitrogen dioxide ambient air pollution exposure among children in the three largest Canadian cities. Health Rep 27:3–9
Ragettli MS, Phuleria HC, Tsai MY, Schindler C, de Nazelle A, Ducret-Stich RE et al (2015) The relevance of commuter and work/school exposure in an epidemiological study on traffic-related air pollution. J Expo Sci Environ Epidemiol 25:474–481
Setton E, Marshall JD, Brauer M, Lundquist KR, Hystad P, Keller P et al (2011) The impact of daily mobility on exposure to traffic-related air pollution and health effect estimates. J Expo Sci Environ Epidemiol 21:42–48
Shekarrizfard M, Faghih-Imani A, Hatzopoulou M (2016) An examination of population exposure to traffic related air pollution: comparing spatially and temporally resolved estimates against long-term average exposures at the home location. Environ Res 147:435–444
Statistics Canada (2013) Low income lines, 2011–2012. Income research series paper. Catalogue no. 75F0002M. http://www5.statcan.gc.ca/olc-cel/olc.action?objId=75F0002M&objType=2&lang=en&limit=0. Accessed 4 Nov 2016
Statistics Canada (2014) Guide to the Labour Force Survey. Catalogue no. 71–543-G. http://www5.statcan.gc.ca/olc-cel/olc.action?objId=71-543-G&objType=2&lang=en&limit=0. Accessed 4 Nov 2016
Statistics Canada (2016) Postal CodeOM Conversion File (PCCF), reference guide. http://www.statcan.gc.ca/pub/92-154-g/92-154-g2016001-eng.htm. Accessed 4 Nov 2016
Stieb DM, Chen L, Hystad P, Beckerman BS, Jerrett M, Tjepkema M et al (2016) A national study of the association between traffic-related air pollution and adverse pregnancy outcomes in Canada, 1999–2008. Environ Res 148:513–526
Su JG, Larson T, Gould T, Cohen M, Buzzelli M (2010) Transboundary air pollution and environmental justice: Vancouver and Seattle compared. GeoJournal 75:595–608
Tchepel O, Dias D, Costa C, Santos BF, Teixeira JP (2014) Modeling of human exposure to benzene in urban environments. J Toxicol Environ Health A 77:777–795
van der Zee SC, Strak M, Dijkema MB, Brunekreef B, Janssen NA (2017) The impact of particle filtration on indoor air quality in a classroom near a highway. Indoor Air 27:291–302
Weichenthal S, Van Ryswyk K, Kulka R, Sun L, Wallace L, Joseph L (2015) In-vehicle exposures to particulate air pollution in Canadian metropolitan areas: the urban transportation exposure study. Environ Sci Technol 49:597–605
Williams RD, Knibbs LD (2016) Daily personal exposure to black carbon: a pilot study. Atmos Environ 132:296–299
World Health Organization (1999) Guidelines for community noise. World Health Organization, Geneva http://www.who.int/docstore/peh/noise/guidelines2.html. Accessed 4 Jan 2017
World Health Organization (2011) Burden of disease from environmental noise—quantification of healthy life years lost in Europe. World Health Organization, Regional Office for Europe, Bonn http://www.who.int/quantifying_ehimpacts/publications/e94888/en/. Accessed 4 Jan 2017
Wu X, Bennett DH, Lee K, Cassidy DL, Ritz B, Hertz-Picciotto I (2011) Longitudinal variability of time-location/activity patterns of population at different ages: a longitudinal study in California. Environ Health 10:80
The authors would like to acknowledge Dr. Perry Hystad (Oregon State University) for providing the NO2 LUR estimates. Funding for this project was provided by Health Canada.
Conflict of interest
The authors declare that they have no conflict of interest.
Electronic supplementary material
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Matz, C.J., Stieb, D.M., Egyed, M. et al. Evaluation of daily time spent in transportation and traffic-influenced microenvironments by urban Canadians. Air Qual Atmos Health 11, 209–220 (2018). https://doi.org/10.1007/s11869-017-0532-6
- Time-activity patterns
- Nitrogen dioxide