Learning Objectives
Become familiar with current knowledge regarding the cardiorespiratory effects of air pollution during outdoor exercise , including the authors’ previous study performed during the summer months.
Summarize the methods and findings of the new study, including changes in cardiorespiratory parameters related to air pollution measures during wintertime physical activity.
Discuss the implications for protective advice to address the balance between short-term risks from air pollution and the longer-term benefits of outdoor physical activity.
We recently reported that air pollution was associated with subclinical changes in cardiorespiratory physiological parameters in older adults exercising outdoors in a rural area during summer months.1 1 winter outdoor physical activity in older adults. Winter activity may have different cardiorespiratory effects owing to factors such as altered autonomic control of cardiovascular function in response to cold,2 3 winter thermal inversions trapping pollutants at ground level may particularly influence air quality in rural and smaller urban areas in winter .4 air pollution concentrations are elevated, older adults and those with cardiac or respiratory disease are advised to reduce the duration and/or intensity of outdoor physical activity.5,6 air pollution in smaller communities. In the U.S. and many other countries, advice is provided through an air quality index (AQI), which in most cases reflects a single pollutant, which is highest relative to its standard7 2 ), ozone (O3 ), and particulate matter of median aerodynamic diameter of 2.5 μm or less (PM2.5 ), where weights were derived from an analysis of air pollution and mortality in Canada's largest cities.8 8 winter wood smoke events or summer wildfires, and thus not being sufficiently protective against adverse health effects (Personal communication, Mr. Eric Taylor, British Columbia Ministry of Environment and Climate Change Strategy).
The primary objective of this study was to address these possible shortcomings by examining the cardiorespiratory effects of air pollution in a panel study of winter Outdoor Physical Activity and Health (OPAH) in older adults, employing the same design as in our earlier summer study.1 air pollution exposures (including as measured by AQHI values) would be associated with subclinical effects on blood pressure, heart rate, urinary oxidative stress markers, fraction of exhaled nitric oxide, oxygen saturation, pulmonary function, endothelial function, and heart rate variability (HRV). Unlike our summer study site, which was characterized by moderate levels of ozone and fine particles similar to those observed in North American cities, and low levels of other pollutants, the winter site experiences moderate levels of multiple pollutants, which affords an opportunity to evaluate the performance of multi-pollutant models and alternative aggregate pollutant indices.
METHODS
Methods were described in detail in our previous paper reporting findings from our summer study.1
Study Location and Participant Recruitment
The study was conducted in Prince George, British Columbia, a city of approximately 74,000 in central British Columbia. Local pollution sources include pulp mills, an oil refinery, chemical and metal processing, sawmills, rail and on-road vehicle traffic, residential wood burning, and road dust.9,10 Winter Games.) To avoid biasing study participants in relation to perceived air quality, the study hypothesis was not disclosed. Participants were told that the study pertained to winter OPAH. Inclusion criteria were age at least 55 years, nonsmokers, nonexposed at home to environmental tobacco smoke and without seasonal allergies. Exclusion criteria were unstable angina, atrial flutter, atrial fibrillation, paced rhythm, left bundle branch block, an implanted cardioverter-defibrillator, or allergy to latex or adhesives. The study was approved by Health Canada and University of Northern British Columbia Research Ethics Boards and written consent was obtained from all study participants.
Exposure Assessment
A dedicated Airpointer® (Recordum Messtechnik GmbH, Vienna, Austria) monitor was deployed approximately 0.3 km from the site used for weekly health measures in 2014 and at the site in 2015, recording continuous hourly measures of carbon monoxide (CO), nitrogen dioxide (NO2 ) ozone (O3 ), particulate matter of median aerodynamic diameter less than 2.5 μm (PM2.5 ), sulphur dioxide (SO2 ), and temperature. Missing values were filled using data from a nearby (0.5 km in 2014 and 2 km in 2015) National Air Pollution Surveillance program monitor. We classified days as “smoky” as a result of residential wood burning using the following criteria developed by Hong et al11 2.5 less than 0.5, standard deviation of hourly PM2.5 at least 3, and mean temperature less than 13°C. Three aggregate air pollution measures were calculated using the continuous data. The AQHI was calculated as follows:
where all pollutants are entered as 3-hour moving average concentrations in ppb (NO2 , O3 ) or μg/m3 (PM2.5 ).8 7 I p 2 , O3 , PM2.5 , and SO2 , where:
C p
BC Hi C p
BC Lo C p
I Hi BC Hi
I Lo BC Lo
Values of BC Hi BC Lo I Hi I Lo 7 2.5 was based on a “nowcast” employing hourly values for a trailing 12-hour period, with greater weight given to more recent values when there is greater hour to hour variability.12
Finally, the combined oxidant capacity (Ox) of O3 and NO2 was calculated as a weighted average with weights equivalent to their respective redox potentials:
Ox = [(1.07×NO2 )+(2.075×O3 )]/3.145.13
In addition to continuous data, 24-hour PM2.5 filter-based samples were collected using the Harvard Impactor (HI) loaded with a 37 mm Teflon filter. HI samples were connected to OMNI 400 personal sampling pumps (BGI by Mesa Labs, Butler, New Jersey), which supplied the 10-L per minute target flow. Samples were collected every 6 days over the course of each sampling campaign. Filters were analyzed gravimetrically following EPA guidelines to provide mass concentration. Inductively coupled plasma-mass spectrometry (ICP-MS) was used to quantify the composition of each PM2.5 sample for 36 elements. Blanks were randomly deployed throughout the study. The median blank level was used as a blank correction value if 50% or more of the blanks were found to be above the limit of detection.
Collection of Health Data
Personal characteristics, health history data, and housing characteristics were determined at study enrollment using a baseline health questionnaire, and daily and weekly questionnaires documented recent medication use, outdoor activity, symptoms, and indoor exposures. Participants completed daily measurements of blood pressure, peak expiratory flow rate (PEFR) and oximetry, and weekly measurements of HRV, reactive hyperemia index as a measure of endothelial function, spirometry, fraction of exhaled nitric oxide (FeNO), and urinary oxidative stress markers before and after 30 minutes of prescribed outdoor activity. Our objective was not to examine the effect of air pollution on exercise performance or change in physiological measures pre- to post exercise. Rather, prescribed outdoor activity was intended to ensure that participants experienced a consistent minimum exposure to outdoor air pollution and that study findings would be relevant to informing advice provided to older adults exercising outdoors. Pre-exercise measures for daily (at home) exercise were carried out immediately before exercise after sitting for 15 minutes, and post exercise measures were carried out after sitting for 15 minutes following exercise. For weekly measures, pre-exercise measures were carried out up to 1.5 hours before and 2 hours after exercise, due to the longer duration of testing procedures for several weekly measures. Details of instrumentation and measurement protocols are described elsewhere.1 outdoor exercise (eg, walking), at the same time, location, and level of effort each day. We advised them not to exercise outdoors when there was a severe weather warning (eg, ice storm) or an air quality advisory. On days when participants underwent weekly measurements, prescribed walking routes were employed. On other days, exercise was unsupervised. We monitored each participant for up to 10 weeks with weekly measurements carried out at the same time of day and day of the week.
Statistical Analysis
Before modeling associations with air pollution , outcome variables were log transformed if necessary to reduce skew. Associations with both pre- and post-exercise measures were examined. We employed linear mixed effect regression models in order to account for repeated measures among study participants. Participants were treated as random effects and time-invariant individual covariates as well as time-variant environmental variables were treated as fixed effects. Age, sex, body mass index (BMI ≤25, >25), smoking history (never, ever), and dichotomous variables for medication use (statins, other cardiovascular drugs) were included as covariates in all models. Height and use of respiratory medication were also included in spirometry models. Selection of the optimal covariance structure to account for serial correlation was based on minimization of the Akaike Information Criterion (AIC) and examination of plots of the covariance. We accounted for effects of time (trend and temporal cycles) using a linear function of time and day of week variables. Air pollution variables were entered into models at individual lags of 0 to 2 days, each with natural spline functions of temperature with 3 degrees of freedom at individual lags of 0 to 2 days (nine models per pollutant). We analyzed associations separately for 2014 and 2015 and examined heterogeneity between years using Cochran Q statistic.14 P value of Q was more than 0.05, we calculated fixed effects pooled estimates combining the 2 years. Percent change in health measures associated with air pollution was calculated as follows:
for untransformed variables,
and for log transformed variables,
where β is the regression coefficient, Δx is the increment in pollution concentration, and
is the mean value of the health measure. As percent change was calculated differently for log transformed and untransformed variables, they cannot be compared directly.15 2.5 “nowcast,” and total oxidants, using the AIC. Models with AIC exceeding that of the best fitting model by 9 or more were considered to have substantially worse fit to the data.16
RESULTS
Descriptive Results
Participant characteristics based on the baseline health questionnaire are summarized in Table 1 . In both rounds of data collection, most participants were 60 years of age or older, Caucasian, and had at least some post-secondary education. More than half were female (over two-thirds in 2015). There was a larger proportion with BMI of 25 or over in 2014. The most commonly reported medications were statins and other cardiovascular drugs, while less than 10% reported using respiratory medication or oral hypoglycemic agents. Wood burning appliances in the home were more prevalent in 2014 than 2015. The prevalence of each of cardiac and respiratory disease was 10% or less in both years, while the prevalence of allergies was 11.8% in 2015 versus less than half that in 2014 and the prevalence of diabetes was 13.9% in 2014 versus less than half that in 2015.
TABLE 1: Participant Characteristics
Descriptive statistics for air pollution and weather are summarized in Table 2 . Pollutant concentrations were generally similar in 2014 and 2015, with the exception of PM2.5 concentrations, which were somewhat higher in 2015, and SO2 concentrations, which were somewhat higher in 2014. Twelve percent of days were classified as smoky in both 2014 and 2015. The 75th percentile of the AQHI fell at the top of the “low health risk” category both years and the 75th percentile of the AQI was at the top of the “good” category in 2014 and bottom of the “moderate” category in 2015. The maxima of the AQHI and AQI were, respectively, at the top of the “moderate health risk” and bottom of the “unhealthy for sensitive groups” categories in 2014 and middle of the “moderate health risk” and top of the “moderate” category in 2015. Spearman correlations among pollutants and temperature are summarized in Table 3 . CO and NO2 were highly correlated both years, while PM2.5 was more highly correlated with CO, NO2 , and SO2 in 2014 than in 2015. O3 was poorly or negatively correlated with other pollutants but highly correlated with temperature in both years. Ox was more highly correlated with the AQHI than the AQI and was weakly correlated with individual pollutants other than O3 . The AQHI was more highly correlated with NO2 than with PM2.5 , whereas the converse was true for the AQI, and both were moderately correlated with CO in both 2014 and 2015. The AQI and AQHI were also highly correlated with each other.
TABLE 2: Air Pollution and Weather Descriptive Statistics (Daily 3-h Maximum* )
TABLE 3: Spearman Correlations Among AQHI, Pollutants, Temperature (2014 in Lower Off-Diagonals and 2015 in Upper Off-Diagonals)
Results of elemental analysis of integrated 24-hour PM2.5 samples indicated that concentrations of aluminum, copper, iron, manganese, titanium, and zinc as well as potassium, sodium, and phosphorous were elevated compared with the rural site of our earlier summer study1 https://links.lww.com/JOM/A428 2.5 ).
Descriptive statistics for physiological measures are summarized in Table 4 . Distributions of daily and weekly measures were comparable in 2014 and 2015, with the exception of 8-isoprostane and vascular endothelial growth factor values, both of which were higher in 2015. The distribution of 8-isoprostane values in 2015 was skewed higher due in part to several high values belonging to two participants. Although these values were outliers, they were considered valid, as they were consistently observed in these participants and similar results were obtained when samples underwent multiple reanalyses.
TABLE 4: Physiological Measures Descriptive Statistics (Pre-Exercise)
Associations With Air Pollution
Associations of the AQHI with daily cardiorespiratory measures are shown in Fig. 1 by lag of air pollution and temperature, expressed as percent change per interquartile range increment in AQHI. Associations with post-exercise measures are shown, although results for pre- and post-exercise were generally similar. Data for all measures were untransformed. Pooled estimates combining 2014 and 2015 are shown, except where there was significant heterogeneity between the 2 years, in which case estimates are shown separately for each year. The AQHI was associated with a significant increase in heart rate at lag 1 and 2 days as well as decreases in diastolic and systolic blood pressure at lag 1 day, and in forced expiratory volume in 1 (FEV1 ) second at lag 2 days. Associations with peak flow in 2015 were sensitive to specification of the autocorrelation structure. Results were somewhat weaker and less consistent using AR1 structure (shown in Figure) compared with compound symmetry structure. In general, associations were not sensitive to temperature lag. Associations with oxygen saturation were inconsistent between the 2 years at every pollutant and temperature lag. Individual air pollutants also exhibited significant associations with several outcomes (See Figures, Supplemental Digital Content 2, https://links.lww.com/JOM/A429 2 (lag 1), O3 (lag 0), PM2.5 (lags 1 and 2), and SO2 (lag 2). Diastolic and systolic blood pressure exhibited significant negative associations at lag 1 day with CO, NO2 , and PM2.5 , while decreased FEV1 was only significantly associated with NO2 (lag 1). All of these significant associations were insensitive to temperature lag. Associations with individual pollutants were consistently of greatest magnitude for NO2 .
FIGURE 1: Percent change (point estimates and 95% confidence intervals) in daily cardiorespiratory measures per interquartile range AQHI by lag of pollutant (P) and temperature (T). Pooled estimates by lag combining 2014 and 2015 are shown, except where there was significant heterogeneity between years, in which case results are shown separately for 2014 and 2015. BP dias, diastolic blood pressure; BP syst, systolic blood pressure; FEV1 , forced expiratory volume in 1 s; O2 satn, blood oxygen saturation; PEFR, peak expiratory flow rate.
Weekly spirometry, HRV measures, endothelial function, FeNO, and urinary oxidative stress markers did not exhibit consistent associations with the AQHI or individual pollutants measured over the previous 2 days (See Figures, Supplemental Digital Content 3, https://links.lww.com/JOM/A430
Multi-Pollutant Models
As heart rate exhibited the most consistent associations with multiple pollutants, we used this outcome to examine the goodness of fit of multi-pollutant models. Of multi-pollutant indices, Ox exhibited the largest magnitude association, while the AQHI and AQI exhibited similar magnitude associations (Fig. 2 ). With respect to individual pollutants, O3 and PM2.5 exhibited the least sensitivity to inclusion of other pollutants as covariates, while the magnitude of the association with NO2 was reduced considerably in models with other pollutants. Model AIC summed over 2014 and 2015 (results were similar in both years) was lowest, indicating best fit, for the model with CO alone followed by AQHI and Ox (Table 5 ). The AIC for the AQI model was higher, indicating worse fit.
FIGURE 2: Percent change (point estimates and 95% confidence intervals) in heart rate per interquartile range increase in pollutant concentration for multi-pollutant indices and single versus multi-pollutant models. “None” indicates single pollutant model. Results for 2014 and 2015 are pooled.
TABLE 5: Multi-Pollutant Model Fit in Descending Order of Model AIC
Changes Over Study Duration
Associations of daily measures of blood pressure, heart rate, FEV1 , and PEFR with day of study as a measure of time trend were generally insensitive to lag of air pollution and temperature. Figure 3 presents median values across lags of air pollution and temperature of percent change of each outcome measure per 70 days (the study duration) relative to the mean observed value of each outcome measure. Significant changes were observed over the course of the study for several measures. Pooled estimates of effects in 2014 and 2015 indicated a significant reduction in heart rate [−1.4%, 95% confidence interval (95% CI) −0.4, to −2.3], diastolic (−1.0%, 95% CI −0.3, −1.8%) and systolic blood pressure (−1.2%, 95% CI −0.5%, −1.9%), and FEV1 (−8.3%, 95% CI −7.0%, −9.6%) over the course of the study. No consistent associations were observed between day of study and other cardiorespiratory measures or urinary oxidative stress markers (not shown).
FIGURE 3: Percent change in daily cardiorespiratory measures over study duration (point estimates and 95% confidence intervals). Pooled estimates combining 2014 and 2015 are shown. BP dias, diastolic blood pressure; BP syst, systolic blood pressure; FEV1 , forced expiratory volume in 1 s; O2 satn, blood oxygen saturation; PEFR, peak expiratory flow rate.
Subgroup Analyses
We did not observe significant variation in associations with air pollution by sex or use of statins (not shown), although the small number of statin users reduced our ability to detect effect modification by statin use.
DISCUSSION
We observed significant associations between air pollution and subclinical changes in daily measures of heart rate, blood pressure, and FEV1 among older adults exercising outdoors in winter in a small city characterized by moderate concentrations of both regional pollutants (ozone and PM2.5 ) and traffic-related and industrial air pollutants (CO, NO2 , and SO2 ). Heart rate exhibited the most consistent associations both with individual pollutants and the AQHI. In contrast to our study employing the same design in a rural area during summer,1 air pollution and weekly measures of spirometry, HRV, FeNO, or endothelial function. If associations with air pollution were mediated at least partly by total outdoor activity, rather than just prescribed activity for our study, power to detect effects based on weekly measures may have been reduced if winter participants spent less time outdoors than summer participants, in keeping with documented seasonality of time activity patterns.17 winter participants’ homes may also have been lower relative to summer participants,18 19 2 air pollution in our winter versus summer studies.
In our earlier paper, we summarized findings from numerous previous panel studies that have examined links between air pollution and cardiorespiratory measures among older adults.1 20–22 23,24 winter air pollution episode in Augsburg, Germany.25 air pollution .26,27 air pollution ,23,28–32 26 27 elderly and those with lung disease. Studies of O2 saturation have been inconsistent.20,22,33,34 air pollution have been detected, they have been small in magnitude, at lags of 0 or 1 day.20,33 1 in association with air pollution exposure is consistent with other panel studies of older adults, which also reported decrements in spirometry parameters,35 36,37 winter panel study among asthmatic adults in Eastern Europe at higher pollutant concentrations, in which associations with PEFR were weaker and less consistent than in a parallel panel of children,38 1 .39
Our examination of multi-pollutant indices and models revealed that Ox exhibited the largest magnitude association and the AQHI and AQI exhibited similar magnitude associations with heart rate. Model fit was best for a model with CO alone. This may not be particularly surprising given that the adverse cardiovascular effects of CO are well established.40 2 concentrations, unlike the AQI (owing to a high threshold for NO2 in the AQI formula), suggesting that NO2 is an important element of a composite exposure measure in an environment characterized by at least moderate levels of multiple gaseous and particulate pollutants. NO2 was also highly correlated with CO.
As an incidental finding, we also observed a significant decrease in heart rate over the duration of the study, similar to our findings in our earlier summer study,1 air pollution ), which we interpreted as possible cardiorespiratory benefits of prescribed daily outdoor activity, consistent with the literature.41–45 air pollution and decreased blood pressure, which in frail populations could be considered a possible adverse effect, while the reduction we observed presumably gradually over the course of the study would be considered desirable. These findings over different time scales require replication in additional studies. We also observed a significant decrease in FEV1 over the duration of the study, the interpretation of which is unclear, although it is possible that fatigue or reduced participant motivation over the 10-week duration of the study could have played a role. This finding was not driven by outlying observations.
Strengths and Limitations
Strengths and limitations are essentially the same as we reported in our earlier summer study.1 46
Study limitations include the lack of personal monitoring data, although we deployed a dedicated monitor close to the site where weekly health measure data were collected. This would tend to reduce exposure measurement error relative to weekly health measures. Indeed, we observed high correlations between pollutant concentrations measured at a government monitor 0.5 to 2 km away from the study site and those measured at the dedicated study monitor. While the study site experienced a moderate proportion of smoky days attributable to residential wood smoke, our findings may not be representative of communities where wood smoke is the primary source of winter air pollution . Results of elemental analysis of PM2.5 compared with the rural summer site and large urban settings (Montreal, Toronto, and Vancouver) confirmed the presence of diverse sources, including metal processing, pulp and paper, sawmills, biomass burning, and road dust.47–49 air pollution exposure. However, we cannot conclusively attribute the incidental finding of improvements in several measures over the duration of the study to a cardiorespiratory training effect, as we did not have a control group that did not participate in prescribed daily outdoor activity. We did not collect objective data on the actual duration and intensity of daily outdoor activity (daily physical activity and health measures at home were unsupervised and self-reported), thus there may be additional random error in outcome measures. Nonetheless, health measure data exhibited plausible distributions consistent with our and others’ previous studies. In particular, mean post-exercise heart rate was greater than mean pre-exercise heart rate. While the numerous hypothesis tests we conducted increase the probability of false positives, coherence among multiple measures and with our earlier findings employing the same design at a different site in summer strengthens the likelihood that our findings in fact reflect true associations.
CONCLUSION
We observed associations between the AQHI and subclinical cardiorespiratory effects, as well as superior fit to the data of the AQHI compared with the USEPA AQI. These findings provide support for the utility of the AQHI as a predictor of health effects in smaller urban areas in winter . The consistency of associations of air pollution with increased heart rate in both our winter and previous summer study strengthen the likelihood of a causal association with this outcome. Older adults exercising outdoors in smaller urban areas in winter may benefit from reducing outdoor activity when air pollution levels are particularly high in order to reduce exposure and the risk of acute adverse cardiorespiratory effects. While we evaluated the USEPA AQI and total oxidants as alternative multi-pollutant indices, alternative pollutant weightings within an additive index like the AQHI should also be examined in future studies. We also recommend additional research with the aim of identifying points on the AQHI scale for providing protective advice that optimize the balance between reducing outdoor activity when necessary to reduce short-term health risks from air pollution while preserving longer term benefits of outdoor physical activity.
Acknowledgments
The authors are extremely grateful to the dedicated study participants, who generously gave of their time during the study, and to the conscientious coordinators and field staff who collected the data: Kim Menounos, Melanie Noullett, Emily Williams (Braam), Bobbi Piche, Erica Erasmus, Chelsea Monell, Joshua Power, Kyla Reschny, Kendra Brown, Catherine Blokker, James Connell, Kathryn Nicholls. The authors also thank Messrs. Terry Robert and Jim Vanderwal, Fraser Basin Council, for project management support, Mr. Hongyu You, Health Canada, for assistance in operating the air monitoring equipment during the study, and Dr. Peter Jackson, University of Northern British Columbia for advice and logistical support.
REFERENCES
1. Stieb DM, Shutt R, Kauri L, et al. Cardio-respiratory effects of
air pollution in a panel study of outdoor physical activity and health in rural older adults.
J Occup Environ Med 2017; 59:356–364.
2. Hintsala H, Kenttä TV, Tulppo M, et al. Cardiac repolarization and autonomic regulation during short-term cold exposure in hypertensive men: an experimental study.
PLoS One 2014; 9:e99973.
3. Peng RD1, Dominici F, Pastor-Barriuso R, Zeger SL, Samet JM. Seasonal analyses of
air pollution and mortality in 100 US cities.
Am J Epidemiol 2005; 161:585–594.
4. Brook JR, Dann TF, Galarneau E, Herod D, Charland JP. Taylor E, McMillan A. The state of air quality in Canada: national patterns.
Air Quality Management . Dordrecht: Springer Science+Business Media; 2014. 43–67.
5. Brook RD, Rajagopalan S, Pope CA 3rd, et al. An update to the scientific statement from the American Heart Association.
Circulation 2010; 121:2331–2378.
6. Giorgini P, Rubenfire M, Bard RL, Jackson EA, Ferri C, Brook RD.
Air pollution and exercise: a review of the cardiovascular implications for health care professionals.
J Cardiopulm Rehabil Prev 2016; 36:84–95.
7. United States Environmental Protection Agency. Technical Assistance Document for the Reporting of Daily Air Quality - the Air Quality Index (AQI). Research Triangle Park: United States Environmental Protection Agency; 2016:11–12. Available at:
https://www3.epa.gov/airnow/aqi-technical-assistance-document-may2016.pdf .
8. Stieb DM, Burnett RT, Smith-Doiron M, Brion O, Shin HH, Economou V. A new multipollutant, no-threshold air quality health index based on short-term associations observed in daily time-series analyses.
J Air Waste Manag Assoc 2008; 58:435–450.
9. National Pollutant Release Inventory. Available at:
http://ec.gc.ca/inrp-npri/donnees-data/index.cfm Accessed June 1, 2017.
10. Prince George Air Improvement Roundtable. Air Quality Information. Available at:
http://pgairquality.com/air-quality-information . Accessed June 1, 2017.
11. Hong KY, Weichenthal S, Saraswat A, King GH, Henderson SB, Brauer M. Systematic identification and prioritization of communities impacted by residential woodsmoke in British Columbia, Canada.
Environ Pollut 2017; 220:797–806.
12. Mintz D, Stone S, Dickerson P, Davis A. Transitioning to a New NowCast Method. Research Triangle Park: United States Environmental Protection Agency; 2013:6–7. Available at:
https://www3.epa.gov/airnow/ani/pm25_aqi_reporting_nowcast_overview.pdf .
13. Bratsch SG. Standard electrode potentials and temperature coefficients in water at 298.15 K. J Phys Chem Ref Data 1989; 18:1–21.
14. DerSimonian R, Laird N. Meta-analysis in clinical trials.
Control Clin Trials 1986; 7:177–188.
15. Buteau S, Goldberg MS. A structured review of panel studies used to investigate associations between ambient
air pollution and heart rate variability.
Environ Res 2016; 148:207–247.
16. Burnham KP, Anderson DR, Huyvaert KP. AIC model selection and multimodel inference in behavioral ecology: some background, observations, and comparisons.
Behav Ecol Sociobiol 2011; 65:23–35.
17. Matz CJ, Stieb DM, Davis K, et al. 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 2014; 11:2108–2124.
18. MacNeill M, Kearney J, Wallace L, et al. Quantifying the contribution of ambient and indoor-generated fine particles to indoor air in residential environments.
Indoor Air 2014; 24:362–375.
19. Barnett AG, Sans S, Salomaa V, Kuulasmaa K, Dobson AJ. WHO MONICA Project. The effect of temperature on systolic blood pressure.
Blood Press Monit 2007; 12:195–203.
20. Goldberg MS, Giannetti N, Burnett RT, Mayo NE, Valois MF, Brophy JM. A panel study in congestive heart failure to estimate the short-term effects from personal factors and environmental conditions on oxygen saturation and pulse rate.
Occup Environ Med 2008; 65:659–666.
21. Luttmann-Gibson H, Suh HH, Coull BA, et al. Short-term effects of
air pollution on heart rate variability in senior adults in Steubenville, Ohio.
J Occup Environ Med 2006; 48:780–788.
22. Pope CA 3rd, Dockery DW, Kanner RE, Villegas GM, Schwartz J. Oxygen saturation, pulse rate, and particulate
air pollution : a daily time-series panel study.
Am J Respir Crit Care Med 1999; 159:365–372.
23. Liu L, Kauri LM, Mahmud M, et al. Exposure to
air pollution near a steel plant and effects on
cardiovascular physiology : a randomized crossover study.
Int J Hygiene Environ Health 2014; 217:279–286.
24. Cakmak S, Kauri L, Shutt R, et al. The association between ambient air quality and cardiac rate and rhythm in ambulatory subjects.
Environ Int 2014; 73:365–371.
25. Peters A, Perz S, Döring A, Stieber J, Koenig W, Wichmann HE. Increases in heart rate during an
air pollution episode.
Am J Epidemiol 1999; 150:1094–1098.
26. Brook RD, Rajagopalan S. Particulate matter,
air pollution , and blood pressure.
J Am Soc Hypertens 2009; 3:332–350.
27. Giorgini P, Di Giosia P, Grassi D, Rubenfire M, Brook RD, Ferri C.
Air pollution exposure and blood pressure: an updated review of the literature.
Curr Pharm Des 2016; 22:28–51.
28. Brauer M, Ebelt ST, Fisher TV, Brumm J, Petkau AJ, Vedal S. Exposure of chronic obstructive pulmonary disease patients to particles: respiratory and cardiovascular health effects.
J Expo Anal Environ Epidemiol 2001; 11:490–500.
29. Chen SY, Su TC, Lin YL, Chan CC. Short-term effects of
air pollution on pulse pressure among nonsmoking adults.
Epidemiology 2012; 23:341–348.
30. Harrabi I, Rondeau V, Dartigues JF, Tessier JF, Filleul L. Effects of particulate
air pollution on systolic blood pressure: a population-based approach.
Environ Res 2006; 101:89–93.
31. Hoffmann B, Luttmann-Gibson H, Cohen A, et al. Opposing effects of particle pollution, ozone, and ambient temperature on arterial blood pressure.
Environ Health Perspect 2012; 120:241–246.
32. Ibald-Mulli A, Timonen KL, Peters A, et al. Effects of particulate
air pollution on blood pressure and heart rate in subjects with cardiovascular disease: a multicenter approach.
Environ Health Perspect 2004; 112:369–377.
33. DeMeo DL, Zanobetti A, Litonjua AA, Coull BA, Schwartz J, Gold DR. Ambient
air pollution and oxygen saturation.
Am J Respir Crit Care Med 2004; 170:383–387.
34. Mar TF, Koenig JQ, Jansen K, et al. Fine particulate
air pollution and cardiorespiratory effects in the
elderly .
Epidemiology 2005; 16:681–687.
35. Lee JT, Son JY, Cho YS. The adverse effects of fine particle
air pollution on respiratory function in the
elderly .
Sci Total Environ 2007; 385:28–36.
36. Ni Y, Wu S, Ji W, et al. The exposure metric choices have significant impact on the association between short-term exposure to outdoor particulate matter and changes in lung function: findings from a panel study in chronic obstructive pulmonary disease patients.
Sci Total Environ 2016; 542:264–270.
37. Lagorio S, Forastiere F, Pistelli R, et al.
Air pollution and lung function among susceptible adult subjects: a panel study.
Environ Health 2006; 5:11.
38. Peters A, Goldstein IF, Beyer U, et al. Acute health effects of exposure to high levels of
air pollution in eastern Europe.
Am J Epidemiol 1996; 144:570–581.
39. Sarnat JA, Golan R, Greenwald R, et al. Exposure to traffic pollution, acute inflammation and autonomic response in a panel of car commuters.
Environ Res 2014; 133:66–76.
40. United States Environmental Protection Agency. Integrated Science Assessment for Carbon Monoxide. Research Triangle Park: United States Environmental Protection Agency; 2010:2–6.
41. Huang G, Shi X, Gibson CA, Huang SC, Coudret NA, Ehlman MC. Controlled aerobic exercise training reduces resting blood pressure in sedentary older adults.
Blood Press 2013; 22:386–394.
42. Levy WC, Cerqueira MD, Harp GD, et al. Effect of endurance exercise training on heart rate variability at rest in healthy young and older men.
Am J Cardiol 1998; 82:1236–1241.
43. Seals DR, Chase PB. Influence of physical training on heart rate variability and baroreflex circulatory control.
J Appl Physiol 1989; 66:1886–1895.
44. Thompson Coon J, Boddy K, Stein K, et al. Does participating in physical activity in outdoor natural environments have a greater effect on physical and mental wellbeing than physical activity indoors? A systematic review.
Environ Sci Technol 2011; 45:1761–1772.
45. Warburton DE, Nicol CW, Bredin SS. Health benefits of physical activity: the evidence.
CMAJ 2006; 174:801–809.
46. Bos I, De Boever P, Vanparijs J, Pattyn N, Panis LI, Meeusen R. Subclinical effects of aerobic training in urban environment.
Med Sci Sports Exerc 2013; 45:439–447.
47. Rubin JI, Brown SG, Hafner HR, Roberts PT. Source Apportionment of PM2.5 in Prince George, British Columbia Final Report STI-906052. 06-3268-FR for BC Ministry of the Environment. 2008; Petaluma: Sonoma Technology, Inc, 4–14.
48. Van Ryswyk K, Anastasopolos AT, Evans G, et al. Metro commuter exposures to particulate
air pollution and PM(2.5)-associated elements in three Canadian cities: the Urban Transportation Exposure Study.
Environ Sci Technol 2017; 51:5713–5720.
49. Baumann K, Jayanty RKM, Flanagan JB. Fine particulate matter source apportionment for the chemical speciation trends network site at Birmingham, Alabama, using positive matrix factorization.
J Air Waste Manage Assoc 2008; 58:27–44.
