Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Air pollution exposure and subclinical health impacts in commuter cyclists Cole, Christie 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_november_cole_christie.pdf [ 6MB ]
Metadata
JSON: 24-1.0107223.json
JSON-LD: 24-1.0107223-ld.json
RDF/XML (Pretty): 24-1.0107223-rdf.xml
RDF/JSON: 24-1.0107223-rdf.json
Turtle: 24-1.0107223-turtle.txt
N-Triples: 24-1.0107223-rdf-ntriples.txt
Original Record: 24-1.0107223-source.json
Full Text
24-1.0107223-fulltext.txt
Citation
24-1.0107223.ris

Full Text

  AIR POLLUTION EXPOSURE AND SUBCLINICAL HEALTH IMPACTS IN COMMUTER CYCLISTS  by Christie Cole  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in The Faculty of Graduate and Postdoctoral Studies (Occupational and Environmental Hygiene)  THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)  August 2014   © Christie Anne Cole, 2014   ii Abstract  Background:  Cycling is a form of active transportation, resulting in health benefits via increased physical activity. Less is known of traffic-related air pollution exposures and the resulting physiological responses experienced by urban commuter cyclists. The aim of this study was to measure systemic inflammation and lung function changes amongst cyclists by comparing responses between high and low- air pollution routes.  Methods:  Male and female participants (n = 38) rode an instrumented bicycle for approximately 1-hour along a Residential and a Downtown designated bicycle route in a randomized crossover trial during the summer and fall of 2010 and 2011. Heart rate, power output, location and particulate matter air pollution (PM10, 2.5, and 1 and particle number concentration [PNC]) were measured at 6-second intervals during trials. Endothelial function [RHI], lung function, and blood measurements of C-reactive protein [CRP], Interleukin-6 [IL-6], and 8-hydroxy-2’-deoxyguanosine [8-OHdG] were assessed within one hour pre- and post-trial. A subset of 23 participants each completed a post-ride cycle ergometer minute ventilation ( ) measurement to estimate air pollution intake, based on heart rate measurements.  Results: Geometric mean (GM) PNC exposures and intakes were higher along the Downtown (GM exposure = 16 226 particles/cm3; intake = 4.54 x 1010 particles) compared to the Residential route (GM exposure = 10 011 particles/cm3; intake = 3.13 x 1010 particles). The mean  cycling: rest  iii ratio was 3.0. In linear mixed-effect regression models, post-cycling RHI was 22% lower following the Downtown route compared to the Residential route (RHI of -0.38, 95% CI of -0.75 to -0.02), but this was not associated with exposure or intake of measured air pollutants. IL-6 and 8-OHdG levels increased after cycling trials along the Downtown route, but no significant association was found with PNC exposure or intake in mixed effect models.   Conclusions: Although air pollution exposures and intakes were higher along the Downtown route and RHI was significantly decreased following trials on this route, this decrease was not associated with air pollution exposure or intake. This suggests other drivers of systemic inflammation related to cycling on the Downtown route may have been responsible for the observed association.     iv Preface The implication of increased minute ventilation resulting in increased pollution exposure to commuter cyclists and possible physiological impacts, was a result of discussions between myself and Dr. Michael Brauer, with suggestions from Dr. Christopher Carlsten and Dr. Michael Koehle. Data collection was completed with the support of Dr. Carlsten’s lab facility for the start and end of the cycling trials, as well as for blood sample storage and processing of two blood biomarkers.   I had the assistance of Alistair Scott and Angela White, who helped with participant recruitment and data collection during all of the trials. C-reactive protein serum samples were analyzed by the Vancouver General Hospital Department of Pathology and Laboratory Medicine, while Catherine Steer analyzed the interleukin-6 and 8-hydroxy-2’-deoxyguanosine samples. I completed all statistical calculations, with advice from my committee, a Statistics department student consultant (Dongxu Wang), and Dr. Sarah Henderson. Nima Hazar created a simple computer program so that heart rate, pollution instrument, and location data could be matched at 1-second time points.  Assistance was obtained from Amy Thai for using the air pollution instruments on the bicycles, and Dr. Luc Int Panis contributed prior knowledge of the P-trak tilt electronics for modifying our P-trak to function on the bumpy bicycle environment. Dr. Winnie Chu kindly did the electronic modification of the P-trak, and Alistair Scott was of great help to implementing the design of the P-trak cradle for use on the bicycle. Dr. Ian McKendry kindly lent the GRIMM particle monitor for two summers of use.   v Barbara Karlen, Tracy Kirkham, Rebecca Abernethy, Vincent Zenarosa, Mandy Pui, Luisa Giles, and Dr. Meaghan MacNutt have all contributed their knowledge of laboratory skills and equipment, offered additional periodic assistance with data collection, and shared their experience in dealing with research participants. I am grateful to all those mentioned for their invaluable knowledge and support. My sincere thanks also goes to all participants that volunteered their enthusiasm and time.  This study was made possible by funding from Health Canada.  The University of British Columbia Research Ethics Board certificate number for this study is H10-00902, the Vancouver Coastal Health REB certificate number is V10-00902, and Health Canada’s REB certificate number is 2011- 0009.    vi Table of Contents Abstract ...................................................................................................................................................... ii Preface ....................................................................................................................................................... iv Table of Contents ................................................................................................................................... vi List of Tables ......................................................................................................................................... xii List of Figures ....................................................................................................................................... xvi List of Abbreviations ....................................................................................................................... xxiii Acknowledgements ............................................................................................................................xxv Chapter 1: Introduction ................................................................................................................... 1 1.1 Low levels of physical activity in Canada and worldwide ........................................................... 1 1.2 Cycling as a solution to inactivity ......................................................................................................... 2 1.3 Cycling is accessible to many users ..................................................................................................... 4 Cycling is efficient for short distances ................................................................................................................... 6 Many may benefit from increased participation in cycling ........................................................................... 8 1.4 Motivators and deterrents of cycling .................................................................................................. 8 Cycling risks associated with air pollution exposure ................................................................................... 10 1.5 Minute ventilation in cyclists and other commuters .................................................................. 16 The product of exposure and minute ventilation produces the total intake of a pollutant .......... 18 1.6 Acute health effects in cyclists related to PM exposures ........................................................... 19 Air pollution exposure health impacts in cyclists .......................................................................................... 19 1.7 Particulate matter and traffic-related air pollution .................................................................... 26 What is air pollution? ................................................................................................................................................. 26  vii Particulate matter ....................................................................................................................................................... 28 PM fate upon inhalation ............................................................................................................................................ 29 PM Mechanisms of health impacts from PM air pollution exposures .................................................... 30 i) Oxidative stress and release of pro-inflammatory mediators ............................................................................. 31 ii) ANS imbalance ......................................................................................................................................................................... 33 iii) Translocation of PM or constituents into systemic circulation ........................................................................ 35 Measures with inputs from multiple pathways .............................................................................................. 35 1.8 Rationale ..................................................................................................................................................... 38 1.9 Objectives and hypotheses ................................................................................................................... 39 Chapter 2:  Methods......................................................................................................................... 43 2.1 Participant recruitment ......................................................................................................................... 43 Recruitment results .................................................................................................................................................... 45 2.2 Individual trial method .......................................................................................................................... 46 Pre-test forms ............................................................................................................................................................... 46 Pre-cycling clinical measurements ...................................................................................................................... 48 EndoPAT™ testing ........................................................................................................................................................................ 48 Spirometry testing ....................................................................................................................................................................... 49 Blood draw ...................................................................................................................................................................................... 50 Bicycle trial details ...................................................................................................................................................... 51 Bicycle equipment setup ........................................................................................................................................................... 51 Air pollution monitors ............................................................................................................................................................... 51 Post-cycling measurements .................................................................................................................................... 57 Velotron HR - minute ventilation test protocol .............................................................................................. 57 2.3 Cycling route data processing and preparation for analysis ................................................... 59 2.4 Pre and post clinical measures data processing and preparation for analysis ................. 60 2.5 Three methods of intake estimates ................................................................................................... 60  viii 2.6 Statistical analyses .................................................................................................................................. 62 Participant descriptive statistics ........................................................................................................................... 62 Analysis of variables ................................................................................................................................................... 62 “Intake 2” as the intake estimate variable ......................................................................................................... 63 Chapter 3: Results and Analysis ................................................................................................. 64 3.1 Descriptive data and baseline measurements of study participants .................................... 64 Smoke exposure ........................................................................................................................................................... 66 Allergy and cold symptom score ........................................................................................................................... 67 Medication use by participants .............................................................................................................................. 67 Transportation to the study location .................................................................................................................. 67 3.2 Air pollution exposures ......................................................................................................................... 68 Comparison by route ................................................................................................................................................. 68 3.3 Pollution exposures and subclinical health end points ............................................................. 72 Pearson product-moment correlation coefficients of different air particle size measurements 79 A trial during an air quality event ......................................................................................................................... 79 3.4 Physiological baseline measurements and comparisons of outcomes by route ............... 82 Reactive Hyperemia Index ....................................................................................................................................... 83 Spirometry measurements ...................................................................................................................................... 85 Blood indicators of inflammation or oxidation ............................................................................................... 87 CRP...................................................................................................................................................................................................... 87 IL-6...................................................................................................................................................................................................... 88 8-OHdG ............................................................................................................................................................................................. 88 Correlations of blood components when compared by Route ................................................................. 89 3.5 Estimating intake ..................................................................................................................................... 90 3.6 Exposure and intake estimates ........................................................................................................... 95  ix Correlation of intake estimate data ................................................................................................................... 105 Paired t-tests of intake estimates ....................................................................................................................... 105 3.7 Other route differences ...................................................................................................................... 109 Power output ............................................................................................................................................................... 110 Heart rate ...................................................................................................................................................................... 111 Cadence .......................................................................................................................................................................... 112 3.8 Mixed effects models............................................................................................................................ 113 Using intakes to make mixed effects models for clinical health measurements ............................. 113 3.9 Testing effect modification of significant models using stratified analysis..................... 122 Effect modification of RHI by sex, BMI, and age with Route .................................................................... 122 Effect modification of FEV1/FVC by sex, BMI, and age with PNC intake ............................................. 125 Chapter 4: Discussion .................................................................................................................. 127 Results summary .......................................................................................................................................... 127 4.1 PNC and PM exposures along the two routes .............................................................................. 128 4.2 Minute ventilation and intake parameters .................................................................................. 130 4.3 Endothelial function using RHI, and biomarkers to compare two routes ........................ 131 4.4 Evaluation of blood biomarkers and relationship with RHI ................................................. 134 Sympathetic nervous system activation may explain the results ......................................................... 137 4.5 Strengths of this study ......................................................................................................................... 140 4.6 Limitations .............................................................................................................................................. 141 4.7 Generalizability of these results ...................................................................................................... 144 4.8 Implications of this research, with public health consequences ......................................... 144 4.9 Recommendations for future research ......................................................................................... 146 4.10 Conclusion ............................................................................................................................................. 147 References ............................................................................................................................................ 149  x Appendix A – CAPaH Protocol ....................................................................................................... 196 Appendix B – CAPaH Advertisement ........................................................................................... 204 Appendix C – CAPaH Introductory Letter V3 ........................................................................... 205 Appendix D – CAPaH Letter of Consent ...................................................................................... 206 Appendix E – CAPaH Screening Questionnaire ....................................................................... 211 Appendix F – CAPaH Pre-Test Questionnaire .......................................................................... 215 Appendix G – CAPaH Common Cold Questionnaire ............................................................... 219 Appendix H – SOP 4 P-trak Procedures ..................................................................................... 220 Appendix I – SOP 3 GRIMM Dust Monitor Procedures .......................................................... 223 Appendix J – SOP 2 – GPS Procedures ......................................................................................... 226 Appendix K – SOP 1 – PowerTap Procedures .......................................................................... 228 Appendix L – Form 6 Velotron test data form ......................................................................... 230 Appendix M – Nimacizer file preparation and analysis ....................................................... 231 Appendix N – PM2.5 and PNC time series plots from all trials ............................................ 234 Appendix O – NASA Satellite August 4, 2010 image of British Columbia forest fires 269 Appendix P – Press Release from British Columbia Wildfire Management Branch “Weather Produce Smoky Skies on the Coast” ........................................................................ 271 Appendix Q – Air Quality Health Index Categories ................................................................ 272  xi Appendix R – The HR - minute ventilation relationship curve for each 2011 participant. .......................................................................................................................................... 273 Appendix S – Additional comparisons of Intakes 1, 2, and 3 .............................................. 296 Appendix T – Paired t-test tables for other route comparisons........................................ 301 Appendix U – Spirometry results by trial. ................................................................................. 302 Appendix V – Testing RHI data for normality .......................................................................... 304 Appendix W – Correlation of PMs, with removal of trial 122C .......................................... 306    xii List of Tables Table 1- Energy expenditure for two sample cases, comparing calories expended during 30 minutes of walking, cycling, and driving. ................................................................................ 4 Table 2- Ratio PNC and PM2.5 personal exposure of cyclists compared to car passengers. ........ 13 Table 3- Comparison of studies measuring minute ventilation ratios in cyclists and other commuters .............................................................................................................................. 17 Table 4- Comparison of previous studies measuring health responses of air pollution exposure to cyclists .................................................................................................................................... 20 Table 5- List of variables used for this analysis. ............................................................................ 63 Table 6- Descriptive data of participants and summary of physiological baseline measurements 64 Table 7- Air pollution exposure measurements calculated from means of trials, including all trials except 122C. ........................................................................................................................... 71 Table 8- Exposure and clinical data differences by route for each participant. ............................. 73 Table 9- GM pollutant exposures along the Downtown and Residential routes. ........................... 74 Table 10- The IQR calculated from all included trials. .................................................................. 75 Table 11- The Pearson product-moment correlation of GMs of all air pollution data from each trial. ........................................................................................................................................ 79 Table 12- Baseline measurements and table of mean differences comparing the change in measurement for time points before and after each ride along the indicated route. .............. 82 Table 13- Clinical Measurement Summary by Route (Downtown and Residential) of post- and pre- cycling clinical measurements, with Paired t-test data. .................................................. 84 Table 14- Pearson product-moment correlation coefficients for the ∆Post-Pre for Biomarkers and RHI for each bicycling route. ................................................................................................. 90  xiii Table 15- Measured HR and minute ventilation during the Velotron stepwise ergometer test for participant 101, a female. ....................................................................................................... 92 Table 16- Predictive minute ventilation equations from the Velotron test for 23 participants. ..... 94 Table 17- All trial mean PowerTap summary data, minute ventilation, and GM pollutant exposures with one trial per row. ........................................................................................... 96 Table 18- PNC Intake estimates 1, 2, and 3 for each trial. ........................................................... 100 Table 19- Summary statistics for PNC estimated intakes (particles), for Downtown and Residential routes. ................................................................................................................ 105 Table 20- Paired t-tests of intakes by route type for each pollutant. ............................................ 107 Table 21- Paired t-tests between intake levels. ............................................................................ 108 Table 22- Mixed effects model coefficients of subclinical health measures for the Route variable, scaled to the Residential route. ............................................................................................. 116 Table 23- Mixed effects model coefficients of subclinical health measure, modeled using the GM concentration of PNC for each trial. ..................................................................................... 116 Table 24- Mixed effects model coefficients of subclinical health measure, modeled using the GM concentration of PM2.5 for each trial. .................................................................................... 117 Table 25- Beta-coefficients for PNC, PM2.5 and Route variables modeled for clinical measures. .............................................................................................................................................. 120 Table 26- RHI modeled with Downtown Route and possible effect modifiers. .......................... 123 Table 27- FEV1/FVC modeled with Downtown Route and possible effect modifiers. ............... 125 Table 28- Subject 101 Velotron test data points and predicted minute ventilation values .......... 273 Table 29- Subject 154 Velotron test data points and predicted minute ventilation values .......... 274 Table 30- Subject 157 Velotron test data points and predicted minute ventilation values .......... 275  xiv Table 31- Subject 158 Velotron test data points and predicted minute ventilation values. ......... 276 Table 32- Subject 159 Velotron test data points and predicted minute ventilation values. ......... 277 Table 33- Subject 160 Velotron test data points and predicted minute ventilation values. ......... 278 Table 34- Subject 161 Velotron test data points and predicted minute ventilation values. ......... 279 Table 35- Subject 162 Velotron test data points and predicted minute ventilation values. ......... 280 Table 36- Subject 163 Velotron test data points and predicted minute ventilation values. ......... 281 Table 37- Subject 164 Velotron test data points and predicted minute ventilation values. ......... 282 Table 38- Subject 165 Velotron test data points and predicted minute ventilation values. ......... 283 Table 39- Subject 166 Velotron test data points and predicted minute ventilation values. ......... 284 Table 40- Subject 169 Velotron test data points and predicted minute ventilation values. ......... 285 Table 41- Subject 170 Velotron test data points and predicted minute ventilation values. ......... 286 Table 42- Subject 172 Velotron test data points and predicted minute ventilation values. ......... 287 Table 43- Subject 178 Velotron test data points and predicted minute ventilation values. ......... 288 Table 44- Subject 181 Velotron test data points and predicted minute ventilation values. ......... 289 Table 45- Subject 184 Velotron test data points and predicted minute ventilation values. ......... 290 Table 46- Subject 186 Velotron test data points and predicted minute ventilation values. ......... 291 Table 47- Subject 187 Velotron test data points and predicted minute ventilation values. ......... 292 Table 48- Subject 190 Velotron test data points and predicted minute ventilation values. ......... 293 Table 49- Subject 193 Velotron test data points and predicted minute ventilation values. ......... 294 Table 50- Subject 197 Velotron test data points and predicted minute ventilation values. ......... 295 Table 51- "Intake 2" of PMs by trial ............................................................................................ 296 Table 52- Summary statistics for estimated mass of PM2.5 intakes for Downtown and Residential routes. ................................................................................................................................... 299  xv Table 53- Paired t-test of Power Output, comparing the Residential and Downtown routes. ..... 301 Table 54- Paired t-test of HR (in bpm), comparing the Residential and Downtown routes. ....... 301 Table 55- Paired t-test of cadences along the Residential and Downtown Routes ...................... 301 Table 56- Spirometry results by trial, showing Post minus Pre differences for each parameter. 302   xvi List of Figures Figure 1. Steps of cycling trial measurements. As two trials were required, these steps were repeated on a second day using the alternate bicycle route from the previous test. ............... 48 Figure 2. Endo-PAT test set up in lab. ........................................................................................... 49 Figure 3. Side view of bicycle outfitted with all test equipment, and top view of rear panniers containing air the P-trak, GPS, and GRIMM monitors. ......................................................... 54 Figure 4. The Downtown cycling route. From Google Maps (2014). ........................................... 55 Figure 5. The Residential (top) and Previous Residential (bottom) cycling routes. From Google Maps (2014). .......................................................................................................................... 56 Figure 6. Age distribution of participants at the start of the study. ................................................ 65 Figure 7. Boxplot of BMI distribution of all 38 participants. ........................................................ 66 Figure 8. Box plots of GM values for each size class of PM. ........................................................ 69 Figure 9. Histograms showing counts of the GM of the PNC measurements for each trial along the Residential and Downtown routes. ................................................................................... 76 Figure 10. Histograms showing counts of the GM concentrations of the PM2.5 measurements from each trial along the Residential and Downtown routes. ......................................................... 77 Figure 11. PM 2.5 and PNC time series plot along Residential route for participant 157. ............ 78 Figure 12. PM 2.5 and PNC time series plot along Downtown route for participant 157. ............ 78 Figure 13. Hourly PM2.5 measurements from the Metro Vancouver Kitsilano air monitor station, through the summer of 2010. ................................................................................................. 80 Figure 14. Box plots showing results of endothelial function and blood measures. ...................... 85 Figure 15. Individual trajectory boxplots of spirometry results by route. ...................................... 87 Figure 16. HR and minute ventilation plot for participant 101 ...................................................... 92  xvii Figure 17. Scatter plot of PNC Intake 1, 2 and 3 (particles) along each route. ............................ 104 Figure 18. Scatter plots, and correlation coefficients of each intake estimate for PNC. .............. 109 Figure 19. Box plot demonstrating the mean power output (in Watts) of participants for 74 (of 76 possible) trials where this data was available. ...................................................................... 110 Figure 20. Box plot demonstrating the mean HR for each of the 74 (of 76 possible) trials. ....... 111 Figure 21. Box plot of the mean cadences achieved along the Downtown and Residential routes. .............................................................................................................................................. 112 Figure 22. Boxplots showing effect modification of RHI route values by sex, BMI, and age. ... 124 Figure 23. Time series plot for subject 100 along the Residential route. ..................................... 234 Figure 24. Time series plot for subject 100 along the Downtown route. ..................................... 234 Figure 25. Time series plot for subject 113 along the Residential route. ..................................... 235 Figure 26. Time series plot for subject 113 along the Downtown route. ..................................... 235 Figure 27. Time series plot for subject 116 along the Residential route. ..................................... 236 Figure 28. Time series plot for subject 116 along the Downtown route. ..................................... 236 Figure 29. Time series plot for subject 119 along the Residential route. ..................................... 237 Figure 30. Time series plot for subject 119 along the Downtown route. ..................................... 237 Figure 31. Time series plot for subject 122 along the Residential route. ..................................... 238 Figure 32. Time series plot for subject 122 along the Downtown route. ..................................... 238 Figure 33. Time series plot for subject 129 along the Residential route. ..................................... 239 Figure 34. Time series plot for subject 129 along the Downtown route. ..................................... 239 Figure 35. Time series plot for subject 130 along the Residential route. ..................................... 240 Figure 36. Time series plot for subject 130 along the Downtown route. ..................................... 240 Figure 37. Time series plot for subject 133 along the Residential route. ..................................... 241  xviii Figure 38. Time series plot for subject 133 along the Downtown route. ..................................... 241 Figure 39. Time series plot for subject 134 along the Residential route. ..................................... 242 Figure 40. Time series plot for subject 134 along the Downtown route. ..................................... 242 Figure 41. Time series plot for subject 141 along the Residential route. ..................................... 243 Figure 42. Time series plot for subject 141 along the Downtown route. ..................................... 243 Figure 43. Time series plot for subject 142 along the Residential route. ..................................... 244 Figure 44. Time series plot for subject 142 along the Downtown route. ..................................... 244 Figure 45. Time series plot for subject 145 along the Residential route. ..................................... 245 Figure 46. Time series plot for subject 145 along the Downtown route. ..................................... 245 Figure 47. Time series plot for subject 147 along the Residential route. ..................................... 246 Figure 48. Time series plot for subject 147 along the Downtown route. ..................................... 246 Figure 49. Time series plot for subject 150 along the Residential route. ..................................... 247 Figure 50. Time series plot for subject 150 along the Downtown route. ..................................... 247 Figure 51. Time series plot for subject 152 along the Residential route. ..................................... 248 Figure 52. Time series plot for subject 152 along the Downtown route. ..................................... 248 Figure 53. Time series plot for subject 154 along the Residential route. ..................................... 249 Figure 54. Time series plot for subject 154 along the Downtown route. ..................................... 249 Figure 55. Time series plot for subject 157 along the Residential route. ..................................... 250 Figure 56. Time series plot for subject 157 along the Downtown route. ..................................... 250 Figure 57. Time series plot for subject 158 along the Residential route. ..................................... 251 Figure 58. Time series plot for subject 158 along the Downtown route. ..................................... 251 Figure 59. Time series plot for subject 159 along the Residential route. ..................................... 252 Figure 60. Time series plot for subject 159 along the Downtown route. ..................................... 252  xix Figure 61. Time series plot for subject 160 along the Residential route. ..................................... 253 Figure 62. Time series plot for subject 160 along the Downtown route. ..................................... 253 Figure 63. Time series plot for subject 161 along the Residential route. ..................................... 254 Figure 64. Time series plot for subject 161 along the Downtown route. ..................................... 254 Figure 65. Time series plot for subject 162 along the Residential route. ..................................... 255 Figure 66. Time series plot for subject 162 along the Downtown route. ..................................... 255 Figure 67. Time series plot for subject 163 along the Residential route. ..................................... 256 Figure 68. Time series plot for subject 163 along the Downtown route. ..................................... 256 Figure 69. Time series plot for subject 165 along the Residential route. ..................................... 257 Figure 70. Time series plot for subject 165 along the Downtown route. ..................................... 257 Figure 71. Time series plot for subject 166 along the Residential route. ..................................... 258 Figure 72. Time series plot for subject 166 along the Downtown route. ..................................... 258 Figure 73. Time series plot for subject 169 along the Residential route. ..................................... 259 Figure 74. Time series plot for subject 169 along the Downtown route. ..................................... 259 Figure 75. Time series plot for subject 170 along the Residential route. ..................................... 260 Figure 76. Time series plot for subject 170 along the Downtown route. ..................................... 260 Figure 77. Time series plot for subject 172 along the Residential route. ..................................... 261 Figure 78. Time series plot for subject 172 along the Downtown route. ..................................... 261 Figure 79. Time series plot for subject 178 along the Residential route. ..................................... 262 Figure 80. Time series plot for subject 178 along the Downtown route. ..................................... 262 Figure 81. Time series plot for subject 184 along the Residential route. ..................................... 263 Figure 82. Time series plot for subject 184 along the Downtown route. ..................................... 263 Figure 83. Time series plot for subject 186 along the Residential route. ..................................... 264  xx Figure 84. Time series plot for subject 186 along the Downtown route. ..................................... 264 Figure 85. Time series plot for subject 187 along the Residential route. ..................................... 265 Figure 86. Time series plot for subject 187 along the Downtown route. ..................................... 265 Figure 87. Time series plot for subject 190 along the Residential route. ..................................... 266 Figure 88. Time series plot for subject 190 along the Downtown route. ..................................... 266 Figure 89. Time series plot for subject 193 along the Residential route. ..................................... 267 Figure 90. Time series plot for subject 193 along the Downtown route. ..................................... 267 Figure 91. Time series plot for subject 197 along the Residential route. ..................................... 268 Figure 92. Time series plot for subject 197 along the Downtown route. ..................................... 268 Figure 93. Subject 101 curve and predictive equation. ................................................................ 273 Figure 94. Subject 154 curve and predictive equation. ................................................................ 274 Figure 95. Subject 157 curve and predictive equation. ................................................................ 275 Figure 96. Subject 158 curve and predictive equation. ................................................................ 276 Figure 97. Subject 159 curve and predictive equation. ................................................................ 277 Figure 98. Subject 160 curve and predictive equation. ................................................................ 278 Figure 99. Subject 161 curve and predictive equation. ................................................................ 279 Figure 100. Subject 162 curve and predictive equation. .............................................................. 280 Figure 101. Subject 163 curve and predictive equation. .............................................................. 281 Figure 102. Subject 164 curve and predictive equation. .............................................................. 282 Figure 103. Subject 165 curve and predictive equation. .............................................................. 283 Figure 104. Subject 166 curve and predictive equation. .............................................................. 284 Figure 105. Subject 169 curve and predictive equation. .............................................................. 285 Figure 106. Subject 170 curve and predictive equation. .............................................................. 286  xxi Figure 107. Subject 172 curve and predictive equation. .............................................................. 287 Figure 108. Subject 178 curve and predictive equation. .............................................................. 288 Figure 109. Subject 181 curve and predictive equation. .............................................................. 289 Figure 110. Subject 184 curve and predictive equation. .............................................................. 290 Figure 111. Subject 186 curve and predictive equation. .............................................................. 291 Figure 112. Subject 187 curve and predictive equation. .............................................................. 292 Figure 113. Subject 190 curve and predictive equation. .............................................................. 293 Figure 114. Subject 193 curve and predictive equation. .............................................................. 294 Figure 115. Subject 197 curve and predictive equation. .............................................................. 295 Figure 116. Scatter plot of PM2.5 Intake estimates one (1), two (2) and three (3) along each route. .............................................................................................................................................. 299 Figure 117. Scatter plots, and correlation coefficients of each intake estimate of PM2.5 shown in Table 10. Trial 122C was removed from this analysis completely, while 119D and 193D (missing HR data) could not be correlated as matched pairs due to incomplete data. ......... 300 Figure 118. Scatter plots, and correlation coefficients of each intake estimate of PM2.5 shown in Table 10. Trial 122C was removed from this analysis completely, while 119D and 193D (missing HR data) could not be correlated as matched pairs due to incomplete data. ......... 300 Figure 119. RHI measurements plotted in histograms using the raw data values (left pane), and log-transformed values (right pane) ..................................................................................... 304 Figure 120. Raw (left pane) and log-transformed (right pane) RHI data, in quantile-quantile plots .............................................................................................................................................. 305  xxii Figure 121. Correlation of GMs of exposure data, including (left) and then excluding (right) the Residential trial by subject 122. PM1, PM2.5 and PM10 are measured in the unit of µg/m3, while PNC is measured in the unit of pt/cc. ......................................................................... 306      xxiii List of Abbreviations ANS - autonomic nervous system BDNF - brain-derived neurotropic factor ß - beta value, referring to the beta- coefficient BMI - body mass index, calculated by dividing mass (in kilograms) by height in metres squared bpm - beats per minute (unit of heart rate) CI - confidence interval (95% unless otherwise stated) CMAs - Canadian census metropolitan areas  CRP - C-reactive protein EndoPAT™ - endothelial peripheral arterial tonometry, medical device produced by Itamar Medical, Caesarea, Israel. FEV1 - forced expiratory volume in 1 second, in Litres FEF25—75 - mid expiratory flow rate, the average forced expiratory flow rate between the 25% and 75% intervals of the FVC FVC - forced vital capacity, the total volume of air forcibly exhaled in Litres GM - geometric mean GSD - geometric standard deviation HR - heart rate, in beats per minute IL-6 - interleukin-6 8-OHdG - 8-hydroxy-2’-deoxyguanosine P-Trak- in reference to the TSI P-Trak model 8525 PA - physical activity PM - particulate matter, used in reference to all size classifications PM10, PM2.5, PM1 and UFP  xxiv PM1 - particulate matter having an aerodynamic diameter of <1 µm PM2.5 - fine particulate matter, having an aerodynamic diameter of <2.5 µm PM10 - coarse particulate matter, having an aerodynamic diameter of <10 µm PNC - particle number concentration, usually in the units of particles per cubic centimetre (pt/cc)  pt/cc - particles per cubic centimetre RHI - reactive hyperemia index, the unit of measure of endothelial function produced by EndoPAT™ SD - arithmetic standard deviation TRAP - traffic-related air pollution UFP - ultrafine particles, having an aerodynamic diameter of ≤ 0.1 µm  - minute ventilation, a measure of the amount of air breathed in Litres per minute    xxv Acknowledgements I greatly appreciate the members of my thesis committee for their encouragement and guidance, and feel very lucky that this group of people were all available, providing enthusiastic support to help bring this project to fruition. I must acknowledge the encouragement, cycling enthusiasm, and patience I received from my supervisor Michael Brauer. I have had a number of challenges preventing me from completing this project but he nonetheless made himself available to help me see this through. A big thank you goes to Chris Carlsten for lending lab space and resources as well as his helpful advice. Mike Koehle has been a fantastic resource for all matters cycling and exercise physiology. I attribute the success of this project largely to the inspirational people on this team- each person has taught me the importance of thinking critically and using good scientific methodology, from their own unique backgrounds. I have learned so much during my time in this program, and realize there is so much more left to learn.  Barb Karlen, Luisa Giles, Alistair Scott, Angie White, and Catherine Steer - you have all provided me with support during experiment design, and during data collection. I am lucky to have had a great team in the lab as well. Additionally, I thank my fellow OEH students for the technical and peer support as many have contributed to my learning.   Finally, big thank yous go to supportive friends like Rebecca Abernethy who is always up for any adventure. To Bev, Colin, and family, who were the exact surrogate family I needed while I was on my Masters adventure. To my own family, thank you for giving me the freedom and encouragement to choose my own path. And to Sébastien-- I would not be this far without you.   1 Chapter 1: Introduction 1.1 Low levels of physical activity in Canada and worldwide  There is an abundance of known benefits of physical activity (PA). (1–19) The advantages of physical fitness associated with regular PA include: a reduced risk of mortality by as much as one half in the most physically active compared to lean but sedentary individuals, (5,8,9) increased muscle strength, (2,4) prevention of the initial development or the further progress of cardiovascular diseases, (1,3,4,6,10) a greater control of body weight, (2,4,6,11) improved blood lipid levels, (2,6,12) reduced blood pressures or decreased risk of developing hypertension , (2,6,13) improved insulin sensitivity or glycemic control, (1,2,13,14) lower risk of or improved management of type 2 diabetes mellitus, (1,2,4,6,15) increased bone mineral density in youth, (2) decreased risk of hip fracture in osteoporosis, (2,4) reduced incidence of some site-specific cancers (colon and breast), (1,2,6) possible benefits to quality of life and mental health, (2,4) possible improved academic performance in children and cognitive function in older adults, (7) and possible decreased re-occurrence and mortality rates along with improved quality of life in cancer patients. (1,2,4,6,16–19)   The World Health Organization (WHO) identifies physical inactivity “as the fourth leading risk factor for global mortality”, with approximately one quarter of the cases of breast cancers, colon cancers, and diabetes, and nearly one third of cases of ischemic heart disease attributable to physical inactivity. (20) Despite these known health benefits of PA and risks associated with sedentary lifestyle habits, the majority of the Canadian population struggles to attain the Canadian Physical Activity Guidelines, which suggest a minimum recommended level of 150 minutes of moderate to vigorous PA each week. (21) At present, most adults are not meeting this goal, with  2 approximately 18% of adults aged 18- 39, and only 13% of 40-59 year olds reaching this level of PA during the 2007-2011 Canadian Health Measures Survey data collection periods. (22)  In close connection with low PA levels is the existence of a widespread energy imbalance, with the increasing prevalence of obesity in Canadian adults. (23) The 2009-2011 Canadian Health Measures Survey found that 27% of adult Canadians aged 18 to 79 had a body mass index (BMI = mass(kg)/height(m)2 ) of 30 or higher (classified as obese), while 40% of adults fell in the range of being overweight with a BMI between 25 and 30. (23) Further, this survey found that more than 30% of Canadian children were classified as either overweight or obese. (24) Unhealthy body composition has become a costly issue for Canada, as an estimated $4.6 to $7.1 billion dollars were spent in 2008, on costs related directly (such as hospital care) and indirectly (productivity and associated long-term disability) to obesity-associated health conditions. (25)   With the high prevalence of excess weight and physical inactivity in the population, and with the resultant health care costs, there is a need to encourage healthy lifestyle habits that include regular daily PA.   1.2 Cycling as a solution to inactivity   One potential solution to the problems of physical inactivity and obesity is to incorporate PA as part of a daily routine, by commuting to school or work using active transportation such as walking or bicycling. The Public Health Agency of Canada cites numerous individual and societal benefits to using active transportation, including those from PA, increased social interactions, reduced road congestion, reduced greenhouse gas emissions and saving money on automobile fuel.  3 (26) Bicycling specifically is one activity that can accommodate a large spectrum of ages, physical abilities and fitness levels. Present recommendations suggest that sedentary or obese individuals participate in low impact activities such as cycling, in order to reduce the likelihood of joint stress and injury. (27) As well, bicycles are available in many different frame geometries, in upright as well as tricycle and recumbent styles for people with balance or other physical impairments. Electronic battery assist devices are also available for those individuals who wish to cycle despite larger distances or particularly hilly terrain.   PA through active transportation or through other forms of intentional exercise helps to maintain physical fitness, muscle mass, and flexibility. (28) Sample cases indicated below in Table 1 show the energy expended while cycling at a relaxed pace is higher than light intensity walking, and driving estimates for the male and female examples. When cycling at the low intensity examples, cyclists would have expended approximately twice as much energy compared to when walking over the same 30 minute time period; cyclists would utilize about three times as many calories as they would when driving a car. (29)   4 Table 1- Energy expenditure for two sample cases, comparing calories expended during 30 minutes of walking, cycling, and driving.  Sample cases Caloric expenditure for 30 minutes of different activities (gross estimate) Sex Age Height Weight Walking 4 km/h Cycling 16-19 km/h Driving a car Female 30 162.6 cm (64 in) 59.0 kg (130 lb) 86 kcal 173 kcal 58 kcal Male 30 175.3 cm (69 in) 74.8 kg (165 lb) 110 kcal 221 kcal 74 kcal Note: Estimates based on “The Compendium of Physical Activities Tracking Guide”. (30,31) Data calculated by using activity based caloric expenditure calculator shapesense.com. (29)   1.3 Cycling is accessible to many users Bicycle ownership is common in developed Western countries, but is influenced by a number of factors including household income, traffic, land use mix and bicycle theft. (32) In the Netherlands, 90% of the population aged 15-65 years owns a bicycle. (33) A 2010 survey of the residents of six small US cities shows the range in bike ownership to be between 55.3 and 80.5% , (32) while a 2001 survey found that overall about 45% of US households owned a bicycle. (34) Ownership rates in Canada were unavailable in the same format, but a small 2012 survey in Toronto, ON focusing multi-unit residential buildings indicated bicycle ownership rates of 0.7-0.9 and 0.3-0.5 bicycles per unit in the Downtown and suburban areas, respectively; unfortunately residents per unit was not provided. (35) In 2010 the Bicycle Trade Association of Canada reported annual sales of bicycles amounted to 357,591 bicycles at independent bicycle dealerships alone, and that estimated retail market value amount to approximate $250 million in Canada. (36) While not offering a full picture of per capita bicycle ownership in Canada, bicycles are certainly  5 accessible to a substantial proportion of the population, making regular participation in cycling a realistic option to many Canadian residents.  Despite the wide availability of bicycles in Canada, limited Canadian data from 2001 for seven metropolitan areas show that the bike transportation modal share ranging from 0.8 to 1.9%. (37) This is consistent with the rate of utilitarian (non-recreational) cycling measured in the 2006 Canadian Census, which reported that only 1.4% of workers in the Canadian Census Metropolitan Areas (CMAs) bicycled to their work on a regular basis. (38) CMAs represent data for 69.1% of the Canadian population (2011 data). (39) This rate of participation is very low compared to many parts of the world where motor vehicles are also readily available. Many Northern and Western European countries have a relatively high bicycle mode share; for example, overall rates of utilitarian cycling are much higher in Germany (2008) at 10%, and the Netherlands at 26% (2006), two countries with living standards similar to those in Canada. (40) Mode share values for most European countries represent the values for all trips by bicycle. Additionally, low participation in active transportation is also being passed onto the next generation. Canadian parents indicated that only 24% of 5-17 year olds regularly use active modes of transportation to and from school, while 62% of these children use only inactive modes; this includes 41% that arrive in personal vehicles. (41) Active commuting as a child may contribute to higher PA levels later in life. (42) The only Canadian example of active school commute habits found that 20.2% of 21,000 Ontario high school student self-reported their usual mode of transportation to school as active. (43) In 1969 nearly 48% of American students were active commuters, whereas in 2009 this rate had declined to about 13% in kindergarten to grade 8 students. (44) This compares to much higher rates of active school commuting in Europe, with one recent Dutch study finding 79 children from different cities used  6 active transport for 79% of their school trips, (45) although it was pointed out that the distance from home to school in the Netherlands was quite short (mean distance travelled = 357m walking, 624m cycling), and increased distance to school has been identified as a deterrent to the use of active transportation to school in a recent systematic review. (45,46) With such low participation rates, there is considerable potential to increase the number of people who may collectively benefit from using active transportation.   Cycling is efficient for short distances  Commuting by bicycle provides an efficient form of transportation due to the distances that can be covered within a typical ~ 30 minute commute time. The sample cases in Table 1 show that the walkers would cover two kilometers over 30 minutes of the activity, while the cyclists would cover 8- 9.75 kilometers in the same amount of time, five times the walking distance at a leisurely speed for either activity. Noticeably, the car driver may potentially cover as much as 50 kilometers in 30 minutes, however this mode of transportation is often impacted by traffic conditions, particularly in urban settings. From the most recent census data available in 2006, the median commuting distance in Canada for over thirteen million commuters was 7.6 kilometers, and approximately 59% of commuters commute a distance of 9.9 kilometers or less. (47) Moreover, the 2011 Canadian Census reports that approximately 81% of Canadians presently reside in urban areas, (48) suggesting active transportation in the form walking, bicycling, and other self-powered modes could be a realistic option for many people during shorter trips. (48) Based on the Table 1 examples, half of the commuters mentioned in the 2006 census could be cycling up to 10 kilometres while traveling 16 km/hour, taking no more than 40 minutes’ time.   7 Choosing cycling as transportation provides mobility and an opportunity to integrate PA into one’s daily routine, allowing participants to benefit from improved fitness levels, and frequent cyclists reap the benefits of PA described in Section 1.1. (49–51) In one prospective study in Denmark, those adults who did not cycle to work had a rate of mortality that was 39% higher than those who regularly commuted by bicycle. (5) A meta-analysis of eight studies showed those active commuters who reported walking or cycling to work had an 11% reduction in cardiovascular disease risk (risk ratio 0.89, 95% confidence interval [CI] = 0.81- 0.98), with more robust protective effects calculated for the female participants, (52) however all of the studies from this review looked at self-reported measurements of PA, which may be less reliable measures for statistical use. (53)   A 2009 review of active school transportation children of different ages showed that in most cases (11/13 studies), children who used active transportation (walking, cycling, and in some cases skateboarding) had higher levels of PA than those who were driven or bussed to school, (54) however most studies reported steps per day using an accelerometer, which is known to underestimate PA when cycling. (55) Overall these studies did not find a difference in body weight or BMI in the active commuters compared to those using passive transportation, however comparisons were limited by heterogeneous study design, with varied definitions of active commuters such as “usual”, “minimum 3 days per week”, and “4-5 days per week”. (54) The Danish cohort of 384 third grade children in the European Youth Heart Study found that PA levels were higher in girls that walked, and for girls and boys that cycled to school. (56) When anthropometric and cardiorespiratory measures were re-assessed six years later, cycling to school at the six year or at both time points was a significant predictor of cardiorespiratory fitness  8 measured using a maximal oxygen cycle ergometry test. (49) Risk factors for cardiovascular diseases were also analyzed for this cohort, which showed that current cyclists had favourable risk factor profiles for measures, which included cholesterol/HDL ratio and glucose metabolism for 334 of these participants. (57) In all, active transportation by children and youth may launch lifelong habits of regular PA, establishing superior baseline health and inspiring a continued active lifestyle into adulthood.  Many may benefit from increased participation in cycling Further, a number of recent studies have considered how introducing a drastic increase in cycling may positively impact regions on a population basis. (58–60) In one simulated US Midwest intervention study, 50% of car trips of ≤4km in distance (one-way) were replaced with bicycle trips. Along with regional air pollutant reductions, it was calculated that the health of 31.3 million people would benefit from the reduced mortality rate of approximately 1 295 deaths per year, fewer days lost due to illness from school or work, and lower health care costs from improved air quality and increased fitness. These benefits were calculated to be least $8 billion US per year in the experimental area, or about $2 300 per cyclist each year, unfortunately no information was provided to estimate any increase in injuries or deaths to cyclists per year. (58) An expanded examination of the costs and benefits of cycling is provided in Section 4.8.  1.4 Motivators and deterrents of cycling  In 2008, Pucher and Buehler described how cycling conditions in many countries “are anything but safe, convenient and attractive”, with specific reference to the United States and United Kingdom where utilitarian cycling rates are low compared to countries such as The  9 Netherlands, Denmark and Germany where rates are much higher. (61) Regions with low cycling rates often demonstrate higher rates of fatalities amongst the small cycling population, and overall evidence shows there is improved cycling safety as the participation rate in cycling increases, (62,63) but other variables relating to cycling safety, such as identifying the safest of infrastructure or understanding route choices, are in need of more research to facilitate safer cycling. (64–66) Related concerns and barriers are described in a 2006 survey administered in Vancouver, which identified deterrents to cycling that included injury risk from car traffic, streets with high traffic levels or fast moving traffic (above 50 km/h), ice, snow or debris along the route, precipitation, freezing temperatures, and needing to carry items not easily transported on a bicycle. (67,68) Motivators for cycling included routes away from traffic noise and air pollution, being separated from car traffic, pleasant scenery, ease of ride (with a distance less than five kilometers) and no hills. (67) Responses of 60 000 Canadian Community Health Survey participants to questions about cycling were in agreement with the deterrents listed in the Vancouver survey, which reported that students cycled more frequently for utilitarian purposes, and increased age, female gender, lower education and participants with higher incomes were less likely to cycle on a regular basis. (68)  In North America, cycling on roads is frequently viewed as a high-risk activity, which many attribute to the increased likelihood of injury compared to the risk when using a car (69). In British Columbia in 2005-2007, among car drivers and passengers, there were 72 injuries and 0.97 fatalities per 100 million vehicle kilometres travelled, whereas among bicyclists, there were 264 injuries and 2.60 fatalities per 100 million kilometres traveled. (70) The higher fatality rate for cyclists is echoed elsewhere. For example, during the 2002 to 2005 time period, annual fatalities were 5.8 and 3.6 per 100 million kilometres cycled in the United States and in the UK, (40), while  10 the most recent and reliable fatality rate for car occupants was from 2011, at a rate of 0.69 and 0.39 deaths per 100 million motor vehicle kilometers traveled in the US and UK, respectively. (71) While the fatality risk in the Netherlands was 1.1 for every 100 million kilometres cycled, (40) the risk of death per 100 million kilometers traveled in a motor vehicle was 0.41. (71) Canadian data from 2010 revealed there were a total of 50 cyclist deaths, (71) however there is no total cycling distance available to adequately determine the nationwide fatality rate along a common denominator scale. (72) Overall, Canada has a rate of 0.65 deaths per 100 million kilometers traveled in a motor vehicle. (71) Irregularities exist in the reporting of national cycling injury rates, making comparison between countries inappropriate. One review suggested that commuters “judge injury risk and respond accordingly”, to say that as traffic volumes and speeds increase, the number of people choosing to walk or cycle are shown to decrease due to fears for their own safety. (73) This is in agreement with the corresponding cycling participation rates that are higher in those countries that are also safer for cyclists. (73)  Cycling risks associated with air pollution exposure While cycling has many health benefits, there are also disadvantages to consider in addition to injury risks. Dense traffic provides higher air pollution exposures compared to cycling along lower traffic routes. (74–78) The resulting increased breathing (due to exercising) in combination with the close proximity of traffic-related air pollution (TRAP) may intensify the dose of pollution, and therefore the potential health risk. (79–85)  Cyclists experience variable air pollutant exposures compared to other transport modes. A number of authors have evaluated cycling routes in parallel with exposures measured inside the  11 cabin of automobiles. (77,78,83,86–92) In some cases, cyclists traveled along the same on-street route as the tested automobile. (83,89,91–93) In other cases, comparisons were made between a dedicated cycling route and a separate road used by the car, with shared starting points and destinations for both modes of transport. (77,78,87,89,90,93) These studies are heterogeneous in their sampling methods, which offer some explanation of differences in the ratio of pollutant exposures. Additionally, a number of studies evaluated the impact variables such as open or closed car windows, (90) pollution differences between summer and winter seasons, (94) and changes to exposure levels at different times of the day. (90)   Table 2 demonstrates the particle number concentration (PNC) ratios of cyclists compared to car passengers, showing mean PNC were generally similar between the two modes, (83,87,90,91) but with exposures decreasing as distance from traffic increased. (77,78) PNC describes the quantity of particles in the smallest size classification of particles, approximately in the UFP range (0.1 μm and smaller), depending on the instrument being used to count the particles. Three studies however reported higher geometric mean (GM) PNC exposures to cyclists, (89,90,95) with closed car windows being associated with lower PNC exposure to car occupants, (90) while one review study found higher exposures in cars. (88) The UFP size class appears to be most directly affected by the acute conditions surrounding the PNC measurement location, (96,97) and car window and ventilation setting was recently demonstrated to greatly impact the PNC level inside the car relative to exposures outside on the road (range of ratio inside:outside of 0.08 to 1.0 ). (98) PM2.5 (particulate matter [PM] with an aerodynamic diameter of 2.5 μm and smaller) ratios are more consistently higher for car passengers (compared to cyclists), (84,93–95) with only one  12 study finding higher exposures to cyclists. (90) In-car PM2.5 exposures were higher when the windows were open during measurement. (90)    13 Table 2- Ratio PNC and PM2.5 personal exposure of cyclists compared to car passengers. Details of car fuel type and route specifics also indicated when available. Authors Location Details of condition  ex. car fuel type, route type Pollutant Ratio of PNC or PM2.5 car : bike  Adams et al. (2001) (94) London, UK Gasoline and diesel fuelled Summer Winter    PM2.5   1.2 : 1 1.1 : 1 Kaur et al. (2005 and 2006) (95)  London, UK Gasoline PNC  PM2.5 0.45 : 1  1.1 : 1 McNabola et al. (2008) (84) Dublin, IE Route 1 Route 2 PM2.5 1.49 : 1 2.85 : 1 Boogaard et al. (2009) (87) 11 cities, NL Gasoline  PNC PM2.5 1 : 1 1.1 : 1 Int Panis et al. (2010) (83)  3 cities, BE Diesel  PNC 1 : 1 Zuurbier et al. (2010) (89) Arnham, NL Gasoline  PNC PM2.5 0.9 : 1 1.3 : 1 Knibbs et al. (2011) (88) Review Paper  PNC 1.3 : 1 Quiros et al. (2012) (90) Santa Monica, US (car fuel not indicated) Car windows closed Car windows open  Car windows closed Car windows open   PNC   PM2.5    0.26  : 1 1 :1  0.54 : 1 0.89 : 1 Huang et al. (2012) (99) Beijing, CN Gasoline Taxis PM2.5 0.6 : 1 Pattinson (2009) (77) & Kingham et al. (2013) (78) Christchurch, NZ Gasoline   Cycling on road  Cycling off road  Cycling on road  Cycling off road  PNC    PM2.5    1.1 : 1  2.1 : 1  1.4 : 1  1.3 : 1 Ragettli et al. (2013) (91)  Basel area, CH Gasoline  PNC ~1 : 1  14  A number of factors appear to contribute to the personal exposures of cyclists, however few studies have identified the specific determinants of exposure in the context of cycling. (86) An exposure study of 11 Dutch cities provided the most detail associated with these exposure determinants, finding that higher exposures of PNC to cyclists were associated with passing mopeds in the bicycle lane (+58.4% change in 1 minute average) and to a lesser degree, cars (+3.7%), cycling on a route with connection to the city centre (+4.6%), traffic intensity (+2.3%), cycling on a path shared with bikes and mopeds (+8.5%), at intersections with a right of way road (+12%), intersections with a side road (+8.4%), roundabouts (+10.9%), bridges (+10.4%), and railway crossings (+5.4%). (87) The only predictor variable associated with decreased PNC was bicycles passing the test bike (-0.5%). (87) Other studies have observed increased PNC exposures to walkers and cyclists on city centre routes where taller surrounding buildings formed a street canyon, (89,96,100) lower PNC measures due to meteorological factors such as higher air temperature and higher wind speed, (83,100) heavy vehicle traffic (those in the truck and bus category) or construction work, (93,96,100,101) peaks attributed to passing mopeds and buses, (87) trucks, and busy roads; (89,93,101) higher PNC measurements were also found during morning test sessions versus other times of the day. (101,102)   Infrastructure such as the separation, configuration, and proximity to traffic of bicycle lanes and multi-use pathways may also influence exposures to cyclists. (86,100,103–105) A 2011 study in Portland compared the exposures of cyclists to PNC by setting up particle counters on the driver’s side mirror of a parallel-parked car (adjacent to moving vehicles), the adjacent cycle track (passenger side mirror), and the sidewalk on the other side of a cycle track. (103) Simultaneous measurements showed significantly lower PNC on the cycle track that likely resulted from  15 increased horizontal distance from traffic (the pollutant source). (103) Differences between the high measurements on the traffic side compared to the lower cycle track side ranged from 8-35% for particle counts, (103) which is consistent with other examples that found decreased PNC as traffic separation increased. (93,102) An additional Portland study found barriers designed to reduce noise between roadways and residential areas also reduced PNC levels on the multi-use path on the residential side of the barrier by 12 to 84% compared to the roadside exposures. (104)  In the study of 11 Dutch cities, route types and obstructions were not as predictive of increases in PM2.5 exposures, with only some small increased exposures attributed to passing mopeds (+5.9%), passing pedestrians (+2.4%), and “other” intersections not otherwise described (+2.0%) and traffic intensity (+1.0%), while decreases in PM2.5 were only found at railway crossings (-3.6%), along city centre connected routes (-3.3%), and on bridges (-1.7%). (87) High speeds of adjacent vehicle traffic, highways and major roads, and increased traffic intensities were found to be associated with higher black carbon (which belongs to the PM2.5 size classification) exposures. (106) For each ten additional diesel vehicles (buses and trucks) per hour, a 15% increase in black carbon exposure was observed to cyclists traveling on the adjacent road in a recent study in Montreal. (102) The 2008 Vancouver study found PM3 to be positively correlated with air temperature and larger particle-size fractions to be negatively correlated with precipitation, (100) in agreement with other study groups. (77,83,94,96,107)   As demonstrated above, air pollution exposure for cyclists may be similar to other commuters, and infrastructure such as proximity to traffic and traffic density are some variables that influence these exposures. However, the PA level that is experienced during different commute  16 activities is also a factor in the consequences of air pollutant exposures to individual travelers. This topic will be explored in the following section that discusses how breathing factors may change the exposures experienced by different commuters.  1.5 Minute ventilation in cyclists and other commuters  Minute ventilation ( ) is described as the number of litres of air expired by an individual in one minute (L/min). This can be measured using a respirometer or an instrument such as a MetaMax© (CORTEX Biophysik GmbH, Leipzig, Germany), which is a portable automated device that measures  and other physiological parameters using a mask attached to a digital turbine. (108)  Of all commonly measured commuter groups, cyclists are found to have the highest  of the active transportation modes, due to the higher level of physical effort expended compared to other mode types. (80,81,83,89)  is variable between individual cyclists as it is related to the sex, lung capacity, and body size of the rider, as well as the level of effort (power output) expended, (80–83,109) with bicycle couriers and competitive cyclists often achieving higher  flow rates by cycling at or near their maximal effort level. (110) Qualitative and size differences exist in the respiratory structures and respiration rates of individuals at different ages, resulting in variation in . (111) Health conditions such as asthma or chronic obstructive pulmonary disease may also reduce efficient air exchange in the lungs. (112) Studies have even considered how an unfamiliar cycling position may lead to altered respiration patterns such as reduced tidal volume in some male cyclists, (113) however differences were not significant in a separate study. (114)   17  Higher  translates into larger intakes of pollutants present in the total inhaled air mass. (83) A small number of commuter studies have looked beyond personal exposure measurements, either estimating or actually measuring the  of study participants during different activities, while simultaneously gathering information about exposures. (80–83) Table 3 shows the known studies that have measured air pollution exposures and  while cycling and using at least one other mode of transportation.  Table 3- Comparison of studies measuring minute ventilation ratios in cyclists and other commuters Author (year) Number of participants measured for  measurement technique Transportation mode compared Cycling : Other mode ratio Van Wijnen et al. (1995) (80) n = 8 gasmeter Car passenger 2.3 : 1 O’Donoghue et al. (2007) (81) n = 2 MetaMax© Bus passenger 2.6 : 1 Zuurbier et al. (2009) (82) n = 34; 24 males 10 females HR and VE curve from exercise test M  Car passenger M  Bus passenger F   Car passenger F   Bus passenger 1.9 : 1 1.8 : 1 2.6 : 1 2.6 : 1 Int Panis et al. (2010) (83) n = 55 38 males 17 females MetaMax© Car passenger M   4.5 : 1  F    4.1 : 1  Note: When  was provided by sex, M = males and F = females.  The 2010 study by Int Panis et al. compared not only the  of the participants while cycling and as car passengers, but they also multiplied the  by the total time spent on the commute. (83) The cyclists had much higher  compared to the car passengers, which resulted in a total volume of air inhaled to be 5.8 (standard deviation, SD= 2.3) and 5.9 (SD= 2.0) times  18 larger over the total duration of cycling compared riding in the car, for males and females, respectively. (83) The study by Zuurbier et al. found that cyclists had an overall mean  ratio that was 2.1 times that of the  when the same participants were traveling as car passengers, with a range of between 1.3 to 5.3, however this study did not account for the trip length or total volume of air inhaled. (82) The 2009 study by Zuurbier et al. showed participants had an average cycling speed of approximately 12 km/h, which may reflect why this study found the smallest  ratio compared to the other studies listed. (89) Larger  ratios were associated with increased average speeds in the studies listed, with average speed between 12 - 15 km/h for Van Wijnen et al., (80) approximately 16km/h for O’Donoghue et al., (81) and 16 - 22 km/h for Int Panis et al. . (83) As MetaMax© systems have been validated to measure  with a very low degree of error, those studies using this measurement device are likely the best approximations of the true  during cycling and other commuting activities. (108) Based on the studies using a MetaMax© device, it is most likely that the  ratio of cyclists: sedentary passengers are in the range of 2.6 to 4.5, within the range of cycling speeds (16-22 km/h) experienced in the two applicable studies. (81,83)  The product of exposure and minute ventilation produces the total intake of a pollutant Pollutant exposure concentrations in mass or particles per volume may be converted to a mass or particle number per Litre (ex. 15 µg/m3 * 1 m3/ 1000L = 0.015 µg/L), which may then be multiplied by the  of the individual (ex. 0.015 µg/L * 20L/min = 0.3 µg/min). This enables us to multiply a mass or particle count by the amount of time a person breathes to calculate an approximate amount of pollutant that may be inhaled by the individual at that  over a specific amount of time. This value would not be the same as the quantity that is deposited inside the lungs, but provides the estimated total amount of pollutant that may interact with the airways; many of  19 these particles will leave the lungs upon exhalation depending on variables such as breathing patterns and whether the an individual is male or female. (115)  1.6 Acute health effects in cyclists related to PM exposures  There is evidence that cyclists may be exposed to higher pollution levels upon choosing routes in close proximity to traffic, and that these pollution levels may in some cases exceed those exposed by occupants of other transport modes. Further, the physically active nature of cycling results in increased air intake due to increased . In combination, these factors may result in air pollution intakes that are not experienced by other street users, and have the potential to offset the health benefits of the exercise component of this activity to the individual cyclist. For this reason we will examine previous studies that have measured the acute health effects of pollution experienced by cyclists.  Air pollution exposure health impacts in cyclists Ten recent publications have examined different acute health impacts of air pollution exposure to individuals cycling in real life scenarios outdoors. Shown in Table 4, each of these studies measured a variety of pollutants, however the table is focused on PNC as these values were available from most of the publications; as well, PNC is the particle measure that is among the most responsive to differences in proximity to traffic, allowing larger differences to be seen when contrasting high and low traffic areas. (100) Physiological measurements of lung function and a number of different blood biomarkers were the most common acute health measures, however studies by Weichenthal et al. (2011 and 2012) in Ottawa, and Nyhan et al. (2014) in Ireland also assessed heart rate variability (HRV). (79,116,117) 20 Table 4- Comparison of previous studies measuring health responses of air pollution exposure to cyclists Author, city (year) # participants Pollutant means Health measures Significant results Strak et al. in Utrecht, NL  (2010) (118) 12 PNC High traffic 41 097 pt/cc Low traffic 27 028 pt/cc  Exhaled nitric oxide (exhaled NO), Lung function (forced vital capacity [FVC], forced expiratory volume in 1 second [FEV1], FEV1/FVC ratio, peak expiratory flow  [PEF]) Respiratory symptoms Post-pre PNC positive association with PEF immediately following ride, but no other significant results Jacobs et al. in Antwerp, BE  (2010) (119) 38 PNC Road 28 867 pt/cc Clean room 496 pt/cc (other PMs as well) Exhaled NO, interleukin- 6 [IL-6], platelet function, Clara cell protein, blood cell counts % blood neutrophils increased after road cycling compared to clean room (p=0.003), no other significant changes Bos et al. in Antwerp, BE  (2011) (120) 38 Same as Jacobs et al. (2010) Serum brain-derived neurotropic factor (BDNF) BDNF increased after clean air ride but not after road ride Zuurbier et al. in Arnhem, NL  (2011) (75)  * only cycling data was used 34 PNC (mean of means) High traffic 48 939 pt/cc Low traffic 39 576 pt/cc Urban background 23 798 pt/cc Lung function (FVC, FEV1, FEV1/FVC, PEF, exhaled NO)  PNC associated with decreased PEF and with increased airway resistance only immediately after trial, no change to exhaled NO in cyclists Zuurbier et al. in Arnhem, NL  (2011) (74)  * only cycling data was used 34 PNC (mean of means) High traffic 48 939 pt/cc Low traffic 39 576 pt/cc Urban background High sensitivity CRP, IL-6/8/10, tumor necrosis factor  (TNF-α), Clara cell protein 16, blood cell counts, blood No significant associations between exposures and health end points, however CRP showed a trend for a positive association with PNC dose (p=0.07), and TNF-α showed a trend for  21 Author, city (year) # participants Pollutant means Health measures Significant results 23 798 pt/cc coagulation markers a negative association with PNC dose (p=0.10). Weichenthal et al. in Ottawa, CA (2011) (116) 42 PNC High traffic 19 747 pt/cc Low traffic 10 882 pt/cc Indoors  1162 pt/cc other PMs measured HRV, lung function including the mean forced expiratory flow between the 25th and 75th percent points of the FVC (FEF25-75%), Exhaled NO Mixed effect models show IQR increase in PNC (18200 pt/cc) resulted in decrease of high frequency power 4 hours after start of cycling. PM2.5 increased exhaled NO by a small amount (1.1 ppb 2 hrs post), PNC was associated with increased FEF25-75 1-hr post (β =191mL). Weichenthal et al. in Ottawa, CA (2012) (117) 42 Volatile organic compounds (VOCs), most were higher along high traffic route. Ex. benzene: High traffic = 0.9 μg/m3 Low traffic = 0.3 μg/m3 HRV, lung function, Exhaled NO Mixed effect models show IQR increase in propane/butane, benzene, and 3-methylhexane were associated with changes to HRV, airway inflammation, and lung function, respectively. Nwokoro et al. in London, UK (2012) (85) 26 with blood samples BC higher fraction of commute in 24 hrs Cyclists: 41% Non-cyclists: 19% Airway macrophage carbon, granulocyte-macrophage colony-stimulating factor, IL-1β, IL-2, IL-6, IL-8, TNF-α Cyclists had higher TNF-α compared with non-cyclists; Jarjour et al. in Berkeley, US  (2013) (76) 15 PNC High traffic 19 945 pt/cc Low traffic 13 517 pt/cc (fine PM as well) Lung function (FVC, FEV1, FEV1/FVC, FEF25-75%) No significant differences in lung function measurements Nyhan et al. in Dublin, IE (2014) (79) 10 cyclists (of 32 total commuters) (no PNC); “inhaled” values PM2.5= 52.6 μg PM10= 65.0 μg HRV Significant mixed effect models for HRV metrics PM2.5 : root mean square standard deviation (ΔRMSSD )  22 Author, city (year) # participants Pollutant means Health measures Significant results PM10 :  Δ  SD of N- N intervals and ΔRMSSD Compared to other commuter mode types  In those studies that used real life examples of high and low traffic routes as the comparison, the concentration of PNC (from high: low) ranged from a ratio 1.24 to 1.81: 1. The study by Jacobs et al. was able to produce a PNC exposure contrast of 58.2: 1 for the high traffic: low exposure ratio as they used filtered air in a lab environment for the low exposure, (119) however this extreme ratio of mean exposures was not reproduced in real life cycling settings where PNC ratios ranged from 1.24 to 1.81 on the higher traffic route compared to the low traffic route. (74–76,116–118) Other pollutant exposures that were quantified for high: low exposures found the volatile organic compound benzene to result in exposures that were approximately 3:1, (117) and black carbon exposure when cycling vs. not cycling resulted in approximately 2.6 times as much black carbon exposure associated with their commute compared to the non-cyclists. (85)  Each of the studies measuring health effects after cycling used a different participant base which varied by factors which included age range, sex, BMI, smoking status, atopic status, exposure (cycling) time, post-exposure data collection times, and study location. Therefore, conditions relating to the population as well as the cycling environment may not be easily reproducible by other study groups.   Acute changes from air pollution exposures to the lung function and exhaled nitric oxide (indicating respiratory inflammation) of healthy cyclists has produced mixed results in cycling  23 examples, (75,76,116–119) whereas other examples of PM exposures in asthmatic participants showed short term decreases or non-significant but inverse associations with lung function measures upon exposure. (121–123) Studies in Berkeley and Antwerp showed no changes in lung function after cycling on higher pollution routes. (76,119) There were opposite results for the studies in Utrecht and Arnhem, with the former finding a positive association with PEF and PNC measures immediately after the cycling trial but then non-significant negative associations six-hours post-exposure, while the latter study finding an inverse association between PEF and PNC immediately, but then no association six hours post-cycling. (75,118) The Utrecht study had one hour cycling times trials, while the Arnhem study was a 2 hour cycling time, (75) which may have had different results attributed to the extended exposure time or possibly the effects of exercising on undiagnosed asthmatics. (124)  Those studies that found significant associations between cycling in higher traffic settings with exhaled NO include the Ottawa study, which found that higher benzene levels were positively associated with exhaled NO (percent change of 0.92 to 1.7%) in the one to four hour period following approximately one hour of ride time, (116,117) while the study by Zuurbier et al. found non-significant increases in exhaled NO associated with different pollutants, PNC (2.5%, with 95% CI of -4.5 to 10.1%) and soot (1.3%, with 95% CI of -1.0 to 3.7%). (118) Weichenthal et al. measured exhaled NO immediately following the ride and every hour for three hours after each cycling trial, (116,117) while Zuurbier et al. measured exhaled NO six hours after cycling only. (118) Each study found different pollutants to be associated with the changes to exhaled NO using mixed effects models, but used different time points for test measurements, so the differences could be related to the time points. The Weichenthal et al. study was done in Ottawa, Canada, where 97%  24 of the fuel used in vehicles 4.5 tons and less is gasoline, whereas in many countries in Western Europe, 50% of passenger vehicles are diesel-fueled, (125,126) This may result in different air pollutant ratios such in urban ambient air, as with soot. (89) Further, the time length of the cycling exposures (Weichenthal et al.= 1 hour, Zuurbier et al.= 2 hours) may have led to different pollutant intakes between these two studies, potentially altering the type and magnitude of health responses. (116–118)  Three studies found significant changes to blood biomarkers and blood cell counts in relation to air pollutant exposures. (119,120) Jacobs et al. found significant changes, in this case finding an increased number of blood neutrophils after cycling on a road compared to filtered air indoors. (119) Bos et al. found a 14% increase in serum BDNF which is understood to enhance brain plasticity in animals, (7) after participants cycled indoors for 20 minutes, but no increase was observed after 20 minutes of cycling outdoors near a major roadway. (120) Tumour-necrosis factor- αwas the only biomarker in the study by Nwokoro et al. that significantly increased in cycling commuters, but not those that commuted by walking or using public transport. (85)  Three of the mentioned studies completed measures of interleukin-6 (IL-6) (Jacobs et al., Zuurbier et al., Nwokoro et al.), however the measurements between cycling exposures did no differ significantly in any of these studies. (74,85,119) The exposure by Jacobs et al. was 20 minutes long, followed by 30 minutes of waiting time prior to the blood draw, (119) which may not have been sufficient time to develop a significant IL-6 response. Other studies that included some asthmatic participants did find significant changes, but had longer exposure times, finding significant changes three hours or more after exposure, (127,128) though significant responses to  25 air pollution exposures have not been consistent throughout the literature for acute-phase response biomarkers. (129,130) Zuurbier et al. did use two-hour cycling exposure times, but did not see significant changes to most biomarkers six hours after the end of the cycling trial, (74) which may be related to the healthy status of participants as those studies showing changes to biomarkers included asthmatic participants. (127,128) Only the study by Zuurbier et al. measured C-reactive protein (CRP), finding positive associations with PNC dose six hours after the end of the exposures (effect estimate = 1.8%, 95% CI = -0.1 to 3.8). (74)  The Ottawa study that assessed HRV measures found that increased PNC exposures resulted in changes to HRV within four hours of cycling along routes with higher levels of PNC; this group also associated other pollutants such as ozone with further changes to HRV measurements. (116) This Ottawa study was the only cycling-specific study of HRV and air pollutant exposures, however a recent study of personal exposures did also find delayed changes to HRV that were inversely associated with PNC exposure in a small group of middle aged participants during foot travel and cycling activities. (131)  Most of the above cycling studies have a female participation rate of approximately 25%, however one study by Strak et al. managed to achieve a 2: 1 ratio of females to males. (118) Some of the above studies included participants that were known to have asthma or allergies, (116) while most other studies excluded potential participants with these health conditions. Lastly, the exposure times generally were approximately one hour in length, however the study by Jacobs et al. (2010), from which Bos et al. (2011) obtained their blood samples, totaled only 20 minutes. (119,120) Due  26 to the unique conditions surrounding each of these air pollution exposure studies, it is reasonable to expect inconsistent results in the health measures for these cycling participants.  Overall, few studies of healthy cyclists have reported significant acute changes in measures of either lung function or inflammatory biomarkers, attributable to different air pollutants. Those studies that measured higher and lower exposure routes showed some small negative health impacts to HRV, (79,117) PEF , (75,118) lung function measures, (117) and exhaled NO, (116) associated with higher exposures of PNC, (75,116–118) total PM, (79) and VOCs, (117) but this means the lower air pollution routes did not show these negative health impacts. All studies described using participants that were either non-asthmatic or had no known acute or chronic medical conditions, and all participants in this group of studies were healthy non-smoking adults aged 18-65. Therefore, the results of the exposures described can be applied to healthy, non-smoking adults with no known chronic health conditions.  1.7 Particulate matter and traffic-related air pollution  As shown above, activities in different settings can result in dissimilar exposures to air pollutants. To better appreciate the potential impacts of these exposures to cyclists, the following section describes air pollution in more detail, with a focus on the known health impacts related to PM exposures.   What is air pollution?  Air pollution is described by the World Health Organization as the “contamination of the indoor or outdoor environment by any chemical, physical, or biological agent that modifies the  27 natural characteristics of the atmosphere”. (132) Air pollutants may originate from a number of sources, influenced by both natural and anthropogenic processes including windblown dust and volcanic eruptions, sea spray, biomass burning, fossil fuel combustion for cooking, power generation, industrial activities, and for use in the transportation of commercial goods, food, and people. (133) Pollutants include gases such as carbon monoxide, sulfur oxides and nitrogen oxides, as well as heavy metals such as lead or mercury, and PM of different sizes. (134)   Exposure to air pollutants is associated with significant adverse health effects, including premature death. (135) These impacts are observable even in Canada, where air quality is typically superior to many other countries in the world.  For example, a study of 11 Canadian cities from 1980 to 1991 demonstrated that an average of 17.9 deaths per day during that period could be attributed to gaseous air pollutants such as carbon monoxide (CO), nitrogen dioxide (NO2), sulfur dioxide (SO2), and ozone (O3) air pollutants. (136) In 2004 Health Canada estimated that the sum of short-term and long-term air pollution exposure effects (by CO, NO2, SO2, O3, and fine PM) led to approximately 5 900 2 100 excess deaths per year in Canada, based on data available for 8.9 million people in major Canadian cities. (137) More recently, the Global Burden of Disease 2010 estimated that long-term exposure to ambient (PM) air pollution to was responsible for approximately 7 200 deaths in Canada. (138)  In addition to being pervasive in almost all environments, there is no known level of PM air pollution that has been deemed completely safe or risk-free for everyone. Children and the elderly as well as those with pre-existing pulmonary or cardiovascular conditions and diabetes are shown in epidemiological studies to be especially sensitive to health impacts related to PM  28 exposure. (139,140) Further, there is strong mechanistic support that corroborates increased the increase in cardiovascular morbidity and mortality which is associated with both short and long-term exposures to PM, especially PM in the smaller size categories. (141) With this evidence in mind, this study will focus on better understanding the health implications of cycling in the presence of PM.  Particulate matter  PM is measured based on size, determined by the aerodynamic diameter of each particle. PM10 are classified as those particles with an aerodynamic diameter of less than 10µm, while PM2.5 are those particles having an aerodynamic diameter of less than 2.5 µm. (142) PM10 and PM2.5 are measured by providing a mass concentration, typically in micrograms per metre cubed (µg/m3). PM that is 0.1µm in diameter or smaller are identified as ultrafine particles (UFPs), and a count of particles per cubic centimeter (number concentration) rather than a mass value may be provided to quantify this size fraction. Particle number concentration (PNC) is a description of the quantity of particles in the size range of approximately 0.02- 0.1 µm per cubic centimetre of air, depending on the instrument used for measurement. Limitations of portable instrument technology prevent smaller UFP particles from easily being counted outside of a laboratory setting.  Primary (directly emitted) sources of PM10 include sea spray, pollen, dust from mining and extraction industries, and dust from roads and construction sites as well as agricultural sources, but this size classification also encompasses the particles in the PM2.5 size range. (143) Major sources of PM2.5 include wildfires, residential combustion of wood for home heating and fossil fuel combustion. PM2.5 may also arise from secondary formation due to the reaction of gaseous  29 pollutants emitted from vegetation, solvent evaporation and from combustion by motor vehicles, power plants and industrial sources. (143)  UFPs may originate from both anthropogenic and naturally influenced sources. UFPs vary in composition; some have a carbonaceous or metallic core, and transition metals or water-soluble components may exist on the surface of each particle. (144) Combustion processes are a common source of particles in the UFP size range, along with diesel emissions, and cigarette smoke; UFP –sized particles also produced while cooking food. (97) With the diverse sizes and sources of PM, it is important to keep in mind how the sources and resulting composition of these pollutants may be associated with toxicity and therefore impact health.  PM fate upon inhalation   Upon inhalation, larger sizes of PM typically are deposited in the upper respiratory tract. (112) The mucociliary escalator system assists in the process of clearance of these particles. Goblet cells produce mucous, which traps foreign particles along with microorganisms. (112) In the non-gas exchanging regions of the lungs, cells that line respiratory tract have cilia, which beat in a coordinated fashion to move the mucous and trapped materials upwards to the throat, where it can then be expectorated, or the mucous may be swallowed and move through the digestive tract (112). Particles that impact inside the nose or within the throat may also be expelled through sneezing or coughing.   The majority of particles in the range of approximately PM10 - PM2.5 move further into the bronchi, while particles between approximately 2 µm to 0.1 µm (<PM2) in aerodynamic diameter  30 are able to reach and deposit in the alveoli. (145) A substantial proportion of UFPs deposit in upper airways, however there is evidence that some UFPs may pass beyond the lungs into the circulatory system, as evidenced by radioisotope labeled carbon UFPs found in the systemic circulation of adult male nonsmokers. (146,147) The proportion of UFPs that deposit in the respiratory tract is shown to increase with exercise due to increased mouth breathing, , and tidal volume, with the particles in the smallest size ranges (count median diameter) having the highest fractional deposition rate. (109,148) Overall, human experiments suggest that less than one percent of the inhaled UFPs travel beyond lung tissue into the blood and other organs, with animal data showing translocation rates as high as 10% in certain tissues beyond the alveoli. (146,147)   The deposition of PM differs by particle size, but also by activity level and age due to airway size. (149,150) Breathing while exercising results in increased air exchanged through the mouth resulting in lower nasal deposition, however the size of the particles and the shape of the individual’s airways are also factors in the destination and proportion of particles left in the pulmonary system. (150)  PM Mechanisms of health impacts from PM air pollution exposures Brook et al. (2010) have proposed a schematic showing the hypothesized mechanistic pathways of the acute and chronic PM exposure to the cardiovascular system. (151) The three major pathways suggested by Brook et al. are: i) the release of pro-inflammatory mediators or other molecules from lung cells that act on blood vessels, ii) disruption to the heart rhythm and autonomic nervous system (ANS) balance, due to the interaction of particles with the lung and nerve receptors, and iii) potential translocation of smaller sizes of PM or components into systemic circulation.  31 (151) The possible mechanisms for these pathways are discussed below, with a focus on those processes relevant to this research project.  i) Oxidative stress and release of pro-inflammatory mediators  The pro-inflammatory mediators described by Brook et al. in pathway i) includes the interactions between reactive oxygen species and lung cells, cells that are part of the immune response, and the resulting cytokines which are secreted by the liver following the initial local reaction in the lung tissue. (151) Biomarkers such as CRP and IL-6 are potential mediators of inflammation and may, along with 8-hydroxy-2’-deoxyguanosine (8-OHdG), be related to pathways of systemic oxidative stress.   IL-6 and CRP When stressed, the immune system releases IL-6 from alveolar macrophages and monocytes, which transfers into the systemic circulation, stimulating downstream immune responses. (151,152) IL-6 has the ability to function as both a pro-inflammatory and an anti-inflammatory molecule, depending on its association with different cytokines. (153–155) Upon exposure to foreign material such as PM, the respiratory tract, lung IL-6 promotes the production of acute phase proteins by the liver, such as CRP. (156) CRP is synthesized in hepatocytes in the liver, most often binding to molecules containing phosphocholine but also autologous ligands which mark damaged cell membranes and apoptotic cells for activation of the complement system in order to clear antigens, and foreign or damaged cells. (157,158) Under acute-phase response conditions, CRP will rise within 6 hours of the exposure. (157) Human coronary artery smooth muscle cells have also been demonstrated to synthesize CRP in vitro, (159) and CRP has also been  32 found in human atherosclerotic plaques within smooth muscle-like cells and the thickened intima of arteries and veins. (160)  Airway epithelial cells exposed in vitro to residual oil fly ash (161) and ozone (162) have been shown to increase production of inflammatory cytokines including IL-6. In an animal experiment with rabbits exposed to PM10 over the course of 4 weeks, measurements of lung (such as macrophage-containing particles) and systemic inflammation (including serum IL-6) were associated with impairments in endothelial function. (163) Exposure to ambient PM air pollution has been associated with increased CRP in a young adults aged 18-25 (164) and increased CRP and IL-6 in elderly participants with a history of Coronary Artery Disease, (165) however no associations were seen in male patients with chronic pulmonary disease and CRP, (139) nor were there consistent significant associations between IL-6, CRP, and air pollution exposures to adult commuters on bus, bicycle, or in car. (74) Participants tested in filtered air compared to busy streets in Antwerp, BE did not experience significant difference in IL-6 changes, (119) however in a variety of real-world locations, CRP was associated with PM levels in single-pollutant models for young adults. (166,167) Other investigations have considered associations of CRP concentration measurements in conjunction with PM pollution, (168,169) finding positive, (165) near-significant, (170) or non-significant associations in measurements following short term exposures, which may be related to incomplete PM exposure information such as in home air filtration studies where only home air PM values were measured. (129,130,171,172)  One 2010 meta-analysis of 54 long term prospective studies linked high CRP levels with increased risk ratios of between 1.23 to 1.71 for a number cardiovascular events, (169) while  33 another recent 15 year follow up of patients with coronary heart disease identified an adjusted hazard ratio of 1.4 for CRP (95% CI: 1.2 – 1.6), in patients whose CRP measured above the 3.0 mg/L median level. (173) Prospective studies have linked elevated IL-6 with increased risk of myocardial infarction in apparently healthy males, with those in the highest quartile of IL-6 measurements having a relative risk 2.3 time higher than those with measures in the lowest quartile, (174) and older women diagnosed with cardiovascular disease demonstrated an increased risk of all-cause mortality, with a relative risk of 4.6 for those in the highest tertile of IL-6 measurements compared to those in the lowest tertile. (175)  Overall, Brook et al. concluded that data from these and other studies of CRP that this biomarker did not consistently increase after short-term exposures, and was not consistent enough to be used as a measure of systemic inflammation. (151) On the other hand, Brook et al. credited IL-6 with more consistent increases in pulmonary inflammation given its origin in alveolar macrophages and lung epithelial cells, (163) and with the appearance of this biomarker in both bronchoalveolar lavage fluid as well as in the general circulation. (151)  ii) ANS imbalance The ANS regulates multiple functions in the body including heart rate (HR), blood pressure, and respiration. (176) Altered autonomic control of cardiac function associated with poor air quality has been demonstrated, with studies describing reduced HRV in (most often) older adults upon exposure to pollutants that include PM10, PM2.5, UFP, and black carbon. (131,168,177–179) One example of evidence pointing to this ANS imbalance was that a measure of HRV was significantly increased 20 minutes following the instillation of urban air particles a murine  34 experiment. (180) Further, this experiment administered ANS suppressing agents just prior to particle instillation, resulting in no significant changes to HR or HRV compared to the control animals. (180)   As another example, older adults in Boston experienced reduced HRV (measured as SD of normal R-R intervals), with the strongest associations with the mean 24-hour PM2.5 measurements, with individuals suffering from ischemic heart disease, hypertension, and diabetes showing the greatest associations. (181) In another study, patients with implanted cardioverter defibrillators showed increased arrhythmias associated with air pollution episodes that included measurements of fine particles. (182) One study compared concentrated ambient particle exposures of elderly participants to exposures of young adults, finding reduced HRV in the elderly group immediately following the 2 hour exposure, and persisting 24 hours after the exposure, but with no observed changes in HRV to the younger participant group. (183) Other research groups have found changes to HRV measures in apparently healthy young adults. (116,164)  Alterations to ANS function from PM air pollution have also been observed through changes to blood pressure (BP) measurements, although the results are mixed for healthy and young adult populations, possibly related to the different methods used by each study. (184–189) Healthy adult participants experienced significantly higher diastolic BP (mean increase = 6 mmHg) after short exposures to higher levels of PM2.5 and ozone together, (186) however a similar exposure of concentrated ambient particles to healthy adults did not produce changes to blood pressures, though significant impacts to another measure of vasoconstriction (flow mediated dilatation) were seen. (187) Cardiac rehabilitation patients were found to have increased blood pressures when  35 ambient PM2.5 measurements were elevated in the previous five days, but only those patients whose resting HR was 70 beats per minute (bpm) or higher showed this significant increase in mean diastolic BP (6.95 mmHg), which could be the result of being in healthier condition, or could mean their HR let to a lower pollution intake. (188) Elderly participants with a history of coronary heart disease experienced increases to both systolic (mean increase = 8.3 mmHg) and diastolic (mean increase = 5.8 mmHg) BP associated with organic carbon. (185)  iii) Translocation of PM or constituents into systemic circulation Brook et al. describe pathway iii) as the potential translocation of the smallest sizes of PM (in the UFP range) or pollutant components into systemic circulation. (151) While there is little known of the extent or the exact mechanisms by which translocation happens in humans, there have been multiple experiments done on rats that have demonstrated the translocation of water-soluble non-essential metals, small increases of carbon-13, or water soluble essential metals and acid-soluble metals (respectively), all of which were found to increase in extrapulmonary tissues and organs within 24 hours of whole body (190) or intratracheal instillation. (191) As well, different metal components such as zinc and nickel have been found to increase serum cholesterol and some blood cell types after exposure to mice of the sulfate form each respective metal as PM. (192)  Measures with inputs from multiple pathways Endothelial function and 8-OHdG are both downstream clinical measures that may be related to numerous potential sources of strain in the body. These may be impacted by systemic oxidative stress, which may be the direct (through pulmonary inflammation and oxidative stress)  36 or indirect (through burdens of PM translocation into the circulatory system, or imbalance in ANS effects) result of stressors on the other pathways described in Brook et al. (151)  Translocation of PM or constituents, pulmonary oxidative stress and inflammation, and ANS imbalance may therefore all contribute to Endothelial dysfunction, with measurable influences on 8-OHdG. These two measures are discussed below.   Endothelial Function  The endothelium is the single layer of cells that covers the inner lining of blood vessels, playing multiple roles which include sensing changes in hemodynamic forces, responding to chemicals and preventing platelet adhesion. (193) These cells act to contract, relax and maintain the blood vessels, regulating vascular tone (including blood pressure) and the oxygen supplied to tissues downstream. (194,195) The maintenance of homeostasis in the endothelium is of particular importance as endothelial dysfunction is shown to be predictive of cardiovascular events in both healthy patients and people with coronary atherosclerosis. (196) When released from vascular endothelial cells, nitric oxide (NO) functions as a biological mediator for blood vessels and platelet activation, causing effects that include relaxation in vascular smooth muscle cells. (197,198) One of the systemic biomarkers mentioned above, CRP, may adversely affect endothelial cells by reducing the production of NO. (199) Blood vessels that have been damaged by atherosclerotic plaque may also have an impaired ability to release NO. (200) When superoxide anions (a reactive oxygen species, ROS) are present, a rapid reaction results and the half-life of NO is substantially reduced; (201) this leaves less NO available to act in the vessel. (202) Numerous examples of human and animal studies have shown that blood vessel function is compromised due to PM exposure. (163,187,202,203) Vascular tone and endogenous fibrinolysis (the use of enzymes to  37 prevent the formation of blood clots) were also found to be impaired in healthy non-smoking males exposed to diesel exhaust. (202) Markers of endothelial function are primarily used in a research role to investigate the mechanisms of atherosclerosis and to evaluate the effectiveness of interventions, (204) however a few studies have shown that endothelial dysfunction can predict coronary vascular disease events when tools such as flow-mediated dilation (205) are used. Studies using peripheral arterial tonometry devices, such as the Endo-PAT device, have correlated lower endothelial function scores with increases in other biomarkers or known cardiovascular risk factors (such as ratio of total to high-density lipoprotein) associated with endothelial function. (206–208)  8-OHdG  8-OHdG is an adduct formed as a result of oxidative damage to DNA. (209) ROS can induce oxidative stress and the formation of 8-OHdG, with a number of exposure studies evaluating health risks from by measuring this marker in urine or blood serum. (164,210–212) In a prior study, post-shift 8-OHdG urine samples were found to be significantly higher in a group of boilermakers exposed to residual oil fly ash and metal fumes. (211) Exposure to sulfate particles and ozone to adult students, and increased PM2.5 from diesel exhaust in two types of workers (bus drivers and emissions inspectors) were positively associated with significantly increased levels of serum and urinary 8-OHdG, respectively. (164,213,214) School children experienced significant increases in (urinary and leukocyte) 8-OHdG when their school was located in an urban (compared to rural) area; this level also correlated to their benzene exposure during school hours, attributed to nearby traffic. (215)   38 1.8 Rationale  Given the present low levels of PA (22) and many health risks associated with excess weight and sedentary lifestyles, (20) cycling offers an accessible and simple solution to improving PA levels in Canadians, along with providing the concurrent benefits of reduced air pollutant emissions and reduced traffic congestion if individuals transition from private vehicles to cycling for transport. As more than 805 of the Canadian population resides in urban locales, (48) it is likely that most cycling occurs in areas near motor vehicle traffic, exposing cyclists to TRAP. Because of the physical effort involved when participating in active commuting such as cycling, participants will have a higher  compared to other commuters, (80,82,83) resulting in a higher intake of ambient air pollutants.   Few studies have attempted to quantify the exposure of cyclists and other commuters to air pollutants, and even fewer studies have correlated intake level to acute health measurements. While it is understood that pollution concentrations vary along cycling routes, (100,107,110,118) there is less understanding of this interaction with breathing parameters, intake of air pollution, and the resulting acute health impacts experienced by the rider.   39 1.9 Objectives and hypotheses   The objective of this study was to assess the acute health impacts of cycling along two different cycling routes in a randomized single-blinded crossover study of healthy male and female participants between the ages of 19-39 years. The study included measures PM air pollution (PNC, PM2.5), endothelial function (RHI) by peripheral arterial tonometry, lung function (FEV1, FVC, FEV1/FVC, FEF25-75), and the blood markers of systemic inflammation (CRP, IL-6) and oxidative stress (8-OHdG).  The study included the following specific objectives:  1) To compare PNC and PM2.5 exposure measurements during one-hour trial rides between the higher and lower traffic routes.  2) To compare measurements of RHI, lung function, CRP, IL-6, and 8-OHdG for each participant between the higher and lower traffic routes.  3) To model the relationship of changes in health effect measurements (endothelial function, lung function measures acquired from spirometry, CRP, IL-6, and 8-OHdG) with measures of air pollution exposure (PM2.5 and PNC).  4) To model the relationship between changes in health effect measurements (RHI, lung function measures acquired from spirometry, CRP, IL-6, and 8-OHdG) with the estimated intake of individual air pollutants (PM2.5 and PNC).  40 A number of interactions were considered between the clinical measures collected in the study in order to develop the specific hypotheses that are described below. Because IL-6 is produced from alveolar macrophages and monocytes in the local lung tissue, and the rise in systemic IL-6 promotes the synthesis of CRP in the liver, (156) increases in IL-6 are likely to be detectable during a shorter timeframe before increases in CRP production are evident, as part of an acute phase response (157). 8-OHdG measures oxidative damage to DNA, and this measure appears to increase on the scale of hours, (211,215) so it may be realistic to see increases in 8-OHdG as well, even in the absence of increased CRP.  Decreases in endothelial function could result from the production of a number of cell signaling molecules including biomarkers that result from inflammation or oxidative stress, which may be found in blood following the cycling trial. However, physical activity is also beneficial to the cardiovascular system, and may help promote beneficial levels of many biomarkers within the body. (216) In this healthy and young population, reduced endothelial function may only be measurable if there is a sufficiently high air pollution dose to outweigh the benefits of the bout of exercise. While acute exercise has been demonstrated to increase plasma and urinary nitrate excretion (a metabolite of NO), measurable immediately following and peaking two to three hours after exercise, (217,218) the presence of ROS in the circulatory system can mean the bioavailability of NO is reduced due to reactions of the NO molecule with oxygen species. (219) The presence of IL-6 may also reduce the bioavailability of NO indirectly by reactions with other enzymes that consequently reduce the production of NO. (220) Decreases in  RHI from increased levels of IL-6 or ROS may be a consequence of air pollution exposure, and would be expected on the time scale of within a few hours following cycling.  41 Finally, spirometry measurements of lung function may detect local lung inflammation, however a substantial air pollution exposure may be required in healthy adults in order to elicit a response. Indeed, mixed results have been observed in previous cycling studies of real-life and laboratory exposures to acute air pollution with some studies finding small positive or negative changes to spirometry measures or exhaled NO, attributed to different pollutants and at differing time points, (75,118) while other studies found no correlations to pollution, differences for spirometry, or exhaled NO. (76,119) Additionally, episodes of poor air quality are rare in Vancouver. (100) The relationship between lung function measures and air pollution in this study may also be complicated by small changes that are commonly observed after a bout of exercise, however the submaximal nature of the exercise portion of this protocol may reduce this likelihood. (221,222)   Hypotheses for study measures: 1) PNC exposure measurements are higher while cycling along the higher traffic route. PM2.5 exposure does not differ between the routes.  2) Cycling for one hour along a route with higher mean PNC results in a smaller increase in post vs. pre-testing RHI, compared to cycling along a route with lower mean PNC.   3) Cycling for one hour along a route with higher mean PNC results in a larger increase of post vs. pre-testing levels of biomarkers (CRP, IL-6, 8-OHdG) in peripheral blood, compared to cycling along a route with lower mean PNC.  42 4) Cycling for one hour along a route with a higher PNC intake (from the combined product of exposure level and ) will result in a smaller increase in post vs. pre-testing RHI, compared to the one hour cycle with lower PNC intake.   43 Chapter 2:  Methods  This chapter describes the development of a cycling study that was used to assess the acute health impacts of air pollution exposures during cycling, by comparing exposures and changes in health outcomes following rides along two routes with different levels of traffic. The full 2011 study protocol is provided in Appendix A “CAPaH Protocol”, with amendments from previous versions of the original 2010 protocol described in Section 2.3. Research ethics board approval was obtained from The University of British Columbia (certificate # H10-00902), the Vancouver Coastal Health Research Institute (certificate # V10-00902), and Health Canada (# 2011- 0009). This study was registered with clinicaltrials.gov under study identifier NCT01708356.   2.1 Participant recruitment  Participants were primarily recruited through poster advertisements along popular cycling routes and university advertising boards. Cycling organization websites based in British Columbia were also used as a way to inform individuals of recruitment for this study. Most participants reported hearing of the study through the posted advertisements along cycling routes, or through word of mouth. A copy of the University of British Columbia Research Ethics Board- approved poster is available in Appendix B as “CAPaH Advertisement”. Participants were provided with the introductory study letter “CAPaH Introductory Letter” (Appendix C) and “CAPaH Letter of Consent” form (Appendix D) upon inquiring about the study.   Non-smoking males and females aged 19-39 years were invited to participate. These individuals were required to live in a non-smoking household and receive no other significant exposure to air pollutants in their home or occupational setting. Potential participants were asked  44 not to join the study if they had ever been diagnosed with asthma, had a history of cardiovascular or respiratory problems, or were taking medications that were known to interfere with endothelial function.    Females were invited to participate if they were either not using any oral contraceptives, or were using only monocyclic oral contraceptives. Those that were pregnant or breastfeeding were excluded from participating. Female participants not taking oral contraceptives were asked to participate in test sessions on days 1-8 of their menstrual cycle, while those on oral contraceptives were asked to participate on any day they were taking a hormone-containing pill. Individuals with known seasonal allergies to pollen were asked to delay participation in the study until their allergy symptoms had subsided.   Study advertisements were posted at recreational buildings and near bike facilities on campus at the University of British Columbia, along popular cycling routes surrounding campus on the west side of Vancouver (mainly in the Kitsilano and Dunbar neighbourhoods), on cork boards and lamp posts around VGH campus, and along popular bicycle routes in the Downtown area. Advertisements were also posted at or near bicycle retailers and at a popular outdoor goods store. Additionally, the study was advertised through university department email lists (such as through the School of Population and Public Health, and School of Kinesiology), and regional cycling advocacy or interest groups as well as through word of mouth. Approximately 100 individuals inquired about participating in the study, but approximately half did not respond after receiving the initial study information. Time commitment, no interest in the project, smoking status, presence of allergy symptoms or asthma, irregular menstrual periods in females, and dislike  45 for providing blood samples were among the reasons that inquiring individuals declined to participate.   All participants were administered a “CAPaH Screening Questionnaire” (see Appendix E), which attempted to identify any health conditions or medications that would preclude the individual from participating in this study. A physician on the research committee was consulted in situations where there was any question relating to the appropriateness of participation by any interested individual; this occurred mostly in the circumstance of an unknown medication.   Recruitment results Of those participants that completed both trials, the research staff knew 24% of those individuals prior to the study, implying that word of mouth was one of the most effective recruitment tools. Most other participants reported seeing and responding to one of the advertisements on campus or along some of the higher traffic cycling routes. Seven participants came in for testing but did not complete two full trials. Five were lost in some way to follow up. Four of participant losses were female; two were not able to come in for testing during the appropriate part of their menstrual cycle, one was unable to commit more time, and one was deterred from the blood test. The male participant could not commit further time to the project. Two participants were each unable to complete an entire test session because of fainting after the blood draw.   46 2.2 Individual trial method Test dates were scheduled over the summer and fall months (May through November in 2010 and 2011), at the convenience of each accepted participant, with a wash out period of at least 12 days between the two sessions. All health measurements were done at the Air Pollution Exposure Laboratory in the basement of the Vancouver General Hospital Research Pavilion, located on the Vancouver General Hospital (VGH) campus, which is at a central location just south of Downtown Vancouver, British Columbia. Testing sessions commenced within one hour of the same time on both test dates (ex. if the participant’s first session started at 8 am, the second session must have commenced no earlier than 7 am, or no later than 9 am).  At the beginning of the first test session, participants were introduced to the testing environment, with a small tour of the lab and provided with an introduction to each of the tests that would be conducted (endothelial peripheral arterial tonometry [EndoPAT™], spirometry, blood tests, trialing the outdoor test bike, and receiving an explanation of the Velotron indoor test). Participants were invited to ask questions throughout the entire length of the test session, to ensure their safety and comfort during their participation. The main steps of each trial are summarized in Figure 1.   Pre-test forms  Each participant was asked to sign the consent form only after all of their questions were answered and the individual confirmed they wished to participate. Study participants were first asked to complete the “CAPaH Pre-Test Questionnaire” (Appendix F) in order to assess that each participant had adhered to the test preparation instructions. In addition, participants were asked  47 about their diet in the past 12 hours in order to normalize the pre-trial consumption of foods that may impact endothelial function, such as inorganic nitrate from sources such as processed meats, greens and beets. (223,224) The individual was then asked to complete the “CAPaH Common Cold Questionnaire” (Appendix G) to help identify whether they were suffering from a respiratory infection, which could produce conflicting test results.    The participant was then asked to sit quietly, and have their blood pressure taken prior to doing the EndoPAT™ test; individuals whose blood pressure exceeded 140/90 mmHg were excluded from the study and encouraged to visit their physician. The EndoPAT™ test method was explained to the participant while they were seated.  48  Figure 1. Steps of cycling trial measurements. As two trials were required, these steps were repeated on a second day using the alternate bicycle route from the previous test.  Pre-cycling clinical measurements EndoPAT™ testing  The EndoPAT™ protocol followed the manufacturer’s User Manuel, Section 6 (Itamar Endo-PAT2000 User Manual, Caesarea, Israel © 2002-2009) using a brachial artery occlusion time of five minutes, with probes on the index fingers. (225) This test resulted in the Reactive Hypermia Index (RHI) score, the measure of the change in vascular tone. In this study, the participant was tested while lying still in a supine position on a bed with the option of a pillow for comfort. A blood End of trialVelotron test (once only)measuring VE and HRPost-cycling clinical measurementsEndoPAT Spirometry Blood draw (CRP, IL-6, 8-OHdG)Bicycle ride (1 hour on either Downtown OR Residential route)PNC and PMs measured GPS location HR and power output measuredPre-cycling clinical measurementsEndoPAT Spirometry Blood draw (CRP, IL-6, 8-OHdG) Pre-test formsConsent form if 1st visit Pre-test questionnaire Common cold questionnaire 49 pressure cuff was placed on the upper non-dominant arm while the dominant arm served as a control. Fingertip sensors were placed on the index finger of both hands. After a 5-minute equilibrium period, the blood pressure cuff was inflated to a suprasystolic pressure of the lesser of 200 mmHg or 50 mmHg higher than the BPsys (to induce brachial arterial occlusion) for 5 minutes. The blood pressure cuff was then released, with the EndoPAT™ 2000 recording for a further 5 minutes. The occlusion was usually done on the non-dominant arm, except when a present injury or other concern existed; this was noted on participant test forms. The occlusion was repeated on the same arm for all testing. Figure 2 below shows the setup used to complete the EndoPAT™ test for each participant. Spirometry testing followed.   Figure 2. Endo-PAT test set up in lab. Note 1. The room lights were normally turned off at the participant’s request, with natural light available from a window beyond the foot of the table. Arm rests (with probes seated inside) are shown on top of bed.  Spirometry testing The open circuit protocol for forced vital capacity was used for the spirometry testing (KoKo spirometer, nSpire Health, Longmont, Colorado, USA). The 2005 American Thoracic Society / European Respiratory Society “Standardisation of Spirometry” document was used to  50 guide staff with consistent techniques. (226) A minimum of three forced expiratory maneuvers were performed, with adequate time in between tests to ensure participant comfort. Participants wore a nose clip, and accepted maneuvers ended with the participant maintaining an upright position, as indicated by the American Thoracic Society standards. When three trials were within 5% (for FVC) of each other, the best of three properly executed maneuvers were accepted as the test result.  Blood draw  Upon completion of the spirometry test, five millilitres of blood was drawn from the median cubital vein in the dominant arm using a 21-gauge butterfly needle and a six-milliliter serum-separator gel tube Vacutainer (Becton Dickinson, Mississauga, ON). The dominant arm was used to accommodate the fact that the Endo-PAT test may sometimes cause bruising. Samples were stored in a refrigerator immediately after the blood draw. Upon the termination of the each session, samples were spun according to the specifications for the centrifuge used and Vacutainer tube requirements. The method used for the analysis of CRP was the Dimension Vista® System Flex® reagent cartridge for high sensitivity CRP, (227) with analysis done by the Department of Pathology and Laboratory Medicine at VGH. A research assistant completed the analyses of IL-6 (Quantikine® ELISA Human IL-6 Immunoassay D6050, R&D Systems, Inc. Minneapolis, MN, USA)(228) and 8-OHdG using the Highly Sensitive 8-OHdG Check ELISA method (Japan Institute for the Control of Aging, NIKKEN SEIL Co., Ltd, Fukuroi, Shizuoka, Japan). (229)  51 Bicycle trial details Bicycle equipment setup  Two KHS (Flite 250, Rancho Dominguez, California) hybrid style bicycles of the same model (one each of frames sized 53 cm and 58 cm) were acquired. One PowerTap hub with integrated power meter and bicycle computer (Powertap Comp, Saris Cycling Group, Madison, Wisconsin) was purchased and set up in a wheel that fit onto the rear wheel bracket of both bike frames. A second wiring harness was purchased, so that both frames could be outfitted with the electrical wires required to hold the cycling computer and monitor the power output. This allowed us to quickly switch the rear wheel back and forth between the two frames.  Air pollution monitors  The air pollution monitors were attached to the rear bicycle rack by using a fabric Mountain Equipment Co-op (Vancouver, British Columbia) brand bicycle pannier on the cyclist’s rear right side, and a Swagman Fat Folding (Swagman Racks, Penticton, British Columbia) basket that was modified into the P-trak cradle on the rear left. The air sampling tubes and electrical wires were wrapped around the top or bottom tubes, up to the handlebars. Tubing and wires was secured to the frame using electrical tape, with careful consideration to ensure tubing was not crimped. Details for each instrument are described below, and Figure 3 follows, showing a photo of one of the bicycles with all of the equipment attached as it was for each cycling trial.   TSI P-trak Ultrafine Particle Counter   The P-trak (P-trak Ultrafine Particle Counter 8525, Shoreview, Minnesota) measured the number concentration of particles per cubic centimeter or PNC, in the size range of 0.02 to 1  52 micrometre in aerodynamic diameter, and was set to log PNC every one second. Suspended in a cradle-type cage to reduce errors from tilting or shocks, it balanced horizontally in the rear right pannier to allow the instrument to stay level, even when ascending or descending hills. The standard tilt sensor was de-activated by a minor electrical modification. The sampling hose was connected to the P-trak’s sample inlet, and was coiled around the top tube of the participant’s bicycled and then affixed with tape to the handlebars in order to sample the air closer to the cyclist’s breathing zone. The tube was oriented forward and then secured with electrical tape. Full details for the operation of the P-trak in the context of this study are available in Appendix H as “SOP 4- P-trak Procedures”.  GRIMM Dust Monitor  The GRIMM Dust Monitor (Model 1.108, GRIMM Technologies, Inc. Douglasville, Georgia) is a light-scattering based instrument that estimates the concentrations of different size fractions of PM using a laser to reflect light off of each particle to count and classify the particles by size. For this study, the GRIMM was set up to record particle sizes of PM10, PM2.5, and PM1 once every 6 seconds. It was placed vertically in the rear left pannier of the participant’s bicycle, also with a tube connected to the sampling outlet. This tube (similarly to the P-trak) was coiled around the top tube of the bicycle, with the opening of the tube oriented forward and then secured with tape. Appendix I “SOP 3- GRIMM Dust Monitor Procedures” provides the full operating procedures for this instrument.   53 GPS  The GPS (DG-100 GPS DataLogger, GlobalSat WorldCom Corporation, New Taipei City, Taiwan) had one-button operation. Once powered-on, a solid green light switched to flashing, signaling it had acquired satellite signals and was recording location information. The start and end times of the ride were determined from the GPS data. Appendix J “ SOP 2 – GPS Procedures” can be viewed for details on the use of the GPS.   PowerTap hub and computer The PowerTap (Powertap Comp, Saris Cycling Group, Madison, Wisconsin) hub and computer system recorded the power output of the cyclist. The other accessories of this system (HR monitor strap, cadence magnet) allowed the PowerTap system to record the HR and cadence of the cyclist at the same instant as the ride time and power output. Torque (in Newton-metres), and speed in kilometers per hour were among the measurements recorded on the PowerTap at intervals that were not user-controlled, but which were typically intervals of two to three seconds. The PowerTap computer (a small pager-sized bicycle computer) was mounted to the handlebars of the test bicycle by sliding it into the cradle. When the HR strap was worn while cycling, the screen showed information such as power output and HR. Participants were periodically asked if numbers were showing up on the bike computer screen to ensure data was being collected. Operating procedures for this instrument are available as Appendix K “SOP 1 – PowerTap”.   It is important to note that while power output is best correlated with , (230) it was impractical to use power output to model , as every time a cyclist stopped pedaling (i.e. at a traffic light or stop sign) the power output fell to zero, and the associated  would then also be  54 designated as the resting . Instead,  was correlated to heart rate for the development of  relationship curves and prediction of air intake throughout the trials.   Figure 3. Side view of bicycle outfitted with all test equipment, and top view of rear panniers containing air the P-trak, GPS, and GRIMM monitors.  Route selection  On the day of testing, just prior to the participant’s first bicycle ride, a coin was flipped to determine on which route the cyclist would be traveling, either the Downtown or Residential route. Each route was designed to take approximately 60 minutes, however the technician who rode with the participant could extend or shorten the ride by ensuring each participant arrived at certain time points to minimize the exposure time differences between trials and amongst all of the participants.  55 Figure 4 shows the map of the Downtown cycling route, while Figure 5 shows the Residential cycling routes, including the initial route version, all created with a web application in conjunction with Google Maps. (231) The Residential route was modified partway through the 2010 test year as we discovered that cyclists were likely being exposed to higher exposures by crossing Cambie Street during a point in the trial when the time (frequently) needed to be extended, so this route is identified as the Previous Residential cycling route.   Figure 4. The Downtown cycling route. From Google Maps (2014).  The starting point of the Downtown route was located where the lower black cross marks the VGH campus (identified in Figure 4), and was travelled in a counter-clockwise direction. The upper black pin marks a turnaround point, where the cyclists were frequently instructed to do laps (shown by the black arrow) along Dunsmuir Street until 20 minutes of the cycling trial remained. This is a bicycle lane that permits safe turnaround opportunities along a relatively high traffic area. After this section of the ride, participants would then continue on to Burrard Street and return back  56 to the hospital campus. All trials completed along the Downtown route had “D” assigned to the end of the subject number to easily identify these trials e.g. 100D, while all trials completed along the Residential route had “C” assigned to the end of the subject number to identify these trials, e.g. 100C.   Figure 5. The Residential (top) and Previous Residential (bottom) cycling routes. From Google Maps (2014).   57 The starting point for the Residential route was located above the black cross on the VGH campus (in Figure 5). The final version of this cycling route (traveled by the majority of participants) was cycled in an out-and-back fashion traveling west (left) of the cross, and then south (down), and to the right. After 30 minutes had passed, the cyclist was to turn around and re-trace the same route to return to the VGH campus. The lower map showing the Previous Residential cycling route was completed in a clockwise direction, but the route length was frequently too short for cyclists to complete in 60 minutes, which led to cyclists going past the start point and continuing on for 50% of the remaining time, to then turn around and return to the VGH campus (as the section of tripled red lines shows). This is a higher traffic area crossing Cambie Street, where cyclists typically were required to wait for the light to change, exposing them to higher pollution levels than what would be expected on many other parts of the route. For this reason, the route was modified to avoid this high-traffic intersection.   Post-cycling measurements The same health measures were repeated after each ride in the same order the pre-ride measurements, whenever possible: EndoPAT™ and spirometry, followed by the blood draw.  Velotron HR - minute ventilation test protocol Each participant completed an indoor exercise protocol at the end of one test session, for the purpose of creating a HR–  relationship curve. This allowed us to use the data collected from the HR monitor during cycling trials, to convert HR data to predicted  values for overall means or at each time point. This indoor exercise test (we refer to this as the Velotron test) was done using a Velotron Dynafit Pro cycle ergometer (Racermate Inc, Seattle, WA, USA) and  58 associated RacerMate software. During this test, the participant wore a Polar HR sensor strap (Polar s810i, Polar Electro, Finland), and used a noseplug when measurements were taken with the respirometer (Spirolab II, Medical International Research, Rome, Italy), which was used to measure  at each required time point. The 2011 data for this test was recorded on a form called “Form 6 Velotron test data”, found in Appendix L.    The Velotron was pre-set for either a male or female participant by selecting the program with appropriate interval changes. Each cyclist was fitted to the Velotron ergometer, and then sat on the bike until their HR stabilized for approximately two minutes to a seated resting rate. The participant was then asked to put on a nose plug, and the volume of air breathed through the spirometer was then measured for a 30 second period, and multiplied by two to obtain the . This was the “resting  on the bicycle”. Next, the participant was asked to start pedaling, and then was advised that the workload would increase on the bicycle every two minutes while they continued to pedal. Participants commenced by cycling at a resistance level 20 watts (for female participants) or 30 watts (for males). The first minute was the adaptation period, while during the second minute of each two-minute interval, a 30-second measurement was taken to determine  at that workload. The HR was recorded at the end of the  measurement.   The next interval was at a resistance level of 40 watts and 60 watts (females and males, respectively). Each of the steps increased by 20 watt increments for females, and 30 watts for males through the remainder of the protocol. Participants were encouraged to consume water whenever they were not using the respirometer in order to maintain adequate hydration status.    59 Participants were encouraged to continue until one of the following points was reached:  1- discomfort or tiredness caused them to want to discontinue the test  2- the highest HR we recorded during the ride had been obtained 3- the resistance level on the bicycle was too difficult to continue turning the pedals above a cadence of approximately 50 revolutions per minute  4- the participant otherwise wished to stop for any reason  Upon completion of the cycling test, participants were encouraged to walk around to cool down, and stretch. Participants were then thanked for their time and participation.  2.3 Cycling route data processing and preparation for analysis  All of the data from the bike-mounted instruments was downloaded from each instrument following each participant’s test session. Variations in start times between each instrument were noted on the day of testing on the participant test forms to help adjust device times to the true start time. The GPS device was deemed to be the “true” time in our data analysis, with all other instrument data being lined up with the start points of the GPS time points. Data from all of the instruments for each individual cycling trial was assembled onto a single Excel spreadsheet in a specific order and layout, which was then analyzed using a custom computer program dubbed “the Nimacizer” designed for this purpose to properly align each time point for the duration of each trial (see Appendix M for “Nimacizer file preparation and analysis” as well as spreadsheet samples).    60 2.4 Pre and post clinical measures data processing and preparation for analysis  Each trial produced data from the tests of clinical measures (EndoPAT™, spirometry values, CRP, IL-6, 8-OHdG), from before (“Pre”) and after (“Post”) each bicycle ride. This data was entered onto a spreadsheet, with one row dedicated to each trial. Continuous data such as the physiological measurements of height, weight, total ride time, mean HR and mean estimated  were entered on this spreadsheet. GMs for each pollutant were also included. Categorical variables were also provided from each question on the Pre-test Questionnaire, identifying the route cycled (Residential or Downtown), meal and transportation choices information prior to trials, the sex of the participant (male or female), and whether females took an oral contraceptive.   2.5 Three methods of intake estimates  Intake estimates were calculated using three different methods. The first, “Intake 1”, was calculated from the mean HR during the ride. The mean  was measured at the mean HR for each participant using the HR-  curve from the Velotron test. This mean  was multiplied by the GM of each pollutant (PM10, PM2.5, PM1, UFP), and then multiplied by the ride time in minutes and seconds. As mean  was available for all participants, this intake estimate could be made for all participants.  Intake 1 (μg or particles) =  GM concentration (in pt/cc or μg/m3) * mean (L/min) * ride time (min)   The intake designated as the “Intake 2” estimation was calculated by multiplying the predicted  (from the mean HR of the ride) by 1/60 to obtain the “second ventilation”, and then  61 multiplying each “second ventilation” by the concentration of the pollutant at each 1-second time point. All rows from the cycling trial were summed to calculate the total mass of each pollutant. This was done for the 2010 and 2011 participants. For the 2010 participants, the protocol only required the cyclist to cycle at their mean HR while their mean was measured. For the 2011 participants, the HR-  curve from the Velotron test was used to determine the  at the mean HR during each bicycle rides. The total sum of each single column was the value determined to be this intake for each respective pollutant.  Intake 2 (μg or particles) = SUM of [Second by second concentrations (in pt/cc or μg/m3) * mean 1-second ventilation (at mean HR)]   The third intake “Intake 3” estimate was calculated by using a similar method as Intake 2 for the concentration of the pollutants. A much wider range of  measurements were available for the subset of 23 participants who completed testing in the 2011 study period, due to the lengthened multi-step Velotron protocol. This allowed for second-to-second  predictions to be made for each second of the pollution measurements, as we were able to apply HR changes (and therefore  changes) to each of the second-by-second pollutant measurements. This value is likely the most comprehensive estimate of intake, however it may only be applied to the data of the 23 participants from 2011.     62 Intake 3 (μg or particles) = SUM of [Second by second concentrations (in pt/cc or μg/m3) *  (Litres per second) at each second]  2.6 Statistical analyses Participant descriptive statistics  Descriptive statistics of the participants were compiled by assembling pollution data from each trial into excel spreadsheets for each individual, and then the mean or total sum of pollution exposures, HR, , as well as intake estimates for each participant were calculated. Individual ride pollution data in the form of time series plots were created for each participant.  results from the Velotron protocol were also graphed.  Analysis of variables  Analyses were performed using two approaches. The first approach was to use paired t-tests on all of the dependent variables listed in Table 5, comparing for example the EndoPAT™ RHI for the Residential and Downtown trials for each participant. This method excluded the air pollution measurement data from the analysis and assumed the two exposures were different between the two routes. Independent variables were also included in simple paired t-test analysis, to assess comparisons with other dichotomous variables to answer questions about, for example, whether RHI was different for people classified in the upper or lower half of the BMI category. These independent variables are also listed in Table 5.  63 Table 5- List of variables used for this analysis. Variable type Variable name Variable unit Data type  Dependent RHI RHI units Continuous  FEV1 spirometry L or ml Continuous  FVC spirometry L or ml Continuous  CRP mg/dL Continuous  IL-6 pg/mL Continuous  8-OHdG ng/mL Continuous Independent Age years Continuous  Age at median younger/older Dichotomous  Sex female or male Dichotomous  BMI kg/m2 Continuous  BMI at median low or high Dichotomous  Cycling route Downtown or Residential Dichotomous  PM1 µg/m3 Continuous  PM2.5 µg/m3 Continuous  PM10 µg/m3 Continuous  PNC pt/cc Continuous  The second approach used was to create mixed effects models using fixed variables (e.g. bivariate route comparing the Residential and Downtown routes, or continuous air pollution intake values), and random variables (the participants) to predict the change in the clinical measures. An example would be to use the Intake 2 estimates of PM2.5 with a random participant variable to model the changes in CRP that would occur with each interquartile range (IQR) increase of PM2.5.   “Intake 2” as the intake estimate variable   As “Intake 2” was deemed to be the most precise intake estimate, this intake method was used as the intake estimate used for all calculations and modeling for the remainder of this thesis, except when otherwise indicated.   64 Chapter 3: Results and Analysis  3.1 Descriptive data and baseline measurements of study participants Shown in Table 6, a total of 38 participants successfully completed two cycling trials during the summer and fall of 2010 (15 participants) and 2011 (an additional 23 participants). Of the 10 female participants, 4 were taking monocyclic oral contraceptives. Figure 6 shows the age distribution of the all participants who completed two trials.   Table 6- Descriptive data of participants and summary of physiological baseline measurements Variable Baseline Mean [SD] or Total Range Total participants (n = males+ females) 38 (28 males + 10 females) - Age (years) 29 [5.6] 20-39 BMI (kg/m2) 22.8 [1.9] 18.3 - 28.0 # of participants whose first trial was along the Downtown route 17 - Morning test (finished by 12:30pm) 23 - Common Cold Questionnaire score  (≥3 = probable viral infections) 0.4 [0.9] 0 - 3           65  Figure 6. Age distribution of participants at the start of the study.  66   The mean BMI of all participants was 22.8 kg/m2, with an SD of 1.9 (boxplot in Figure 7). With testing times starting in the morning or early afternoon, 23 participants opted to test in the morning.   Figure 7. Boxplot of BMI distribution of all 38 participants. Smoke exposure   Only non-smoking participants were invited to participate however two of the participants described themselves as being “occasional smokers”, smoking no more than once every two to three weeks. These participants were accepted into the study provided they did not smoke within the two-week period before each trial.   67 Allergy and cold symptom score  At least three of the participants reported hay fever symptoms at certain times of the year. These participants were invited to schedule a test time only when they did not have active allergy symptoms, even if otherwise physically active outdoors during allergy season. Existence of these symptoms was partially verified by the “common cold questionnaire” score for cold symptoms. In the case of a cold (infectious rhinitis), a score of three or higher indicated a likely viral infection. Participants with known infectious rhinitis had their trial rescheduled until symptoms had subsided. Most participants were aware of the presence of their own allergy symptoms, and were asked to wait until they were allergy symptom-free prior to participating in the study.   Medication use by participants  Participants were asked about the prescribed medications they were currently taking. Medications associated with cardiovascular or lung function precluded the participation of that individual from the study, however a number of participants stated they were taking medications for problems relating to thyroid (two) and anxiety (one). These participants were asked to take their medications as prescribed by their physician, but to ensure they took it at the same time before starting each test to make the two trials as comparable as possible.  Transportation to the study location  The mode of transportation that participants used to arrive at the study location was predominantly by bicycle (66%), however public transit-- taking the bus, train, or a combination of both (20%), arriving by private car (8%), by walking (3%), arriving or by taxi (1%) were also used. To make the study results reproducible, each participant was asked do their best to choose  68 the same method of transportation to the test location on both days. A total of 33 out of the 38 participants (87%) were able to replicate their transportation to the test on both days.  3.2 Air pollution exposures  Comparison by route  Figure 8 depicts the distributions of GM air pollution levels from all tests along the two cycling routes. These exposures had similar GM concentrations for the larger-sized PM, but showed more pronounced differences between routes for PNC (the particle count per cubic centimeter, in the size range of 0.02 to 1 micrometre for the P-Trak instrument that was used) between the two routes. The distribution of GM concentrations from all trials (by route type) is shown. Only the PNC measurements were significantly different between the two routes (p< 0.01). Data were missing from trial 101C (PM, PNC), 150C (PNC), 190C (PM, PNC). In Figure 8, extreme values from trial 122C are not visible with the y- axis limits set for PM1 (57.6 μg/m3), PM2.5 (63.2 μg/m3) or PM10 (96.1 μg/m3).  69  Figure 8. Box plots of GM values for each size class of PM. Table 7 shows the overall mean, median and GM for each route, calculated from the mean exposure measurements from each cycling trial. Welch two-sample t-test 95% CIs and P-values for each pollutant were compared by Route.  When considering variability along the Residential cycling route compared to the Downtown route, the SDs of the means of each trial and the geometric standard deviations (GSDs) were always higher along the Residential route when comparing the between-participant PNC data, however the SD was higher for all of the other PM sizes along the Downtown route (shown in Table 7). Only the PNC demonstrated a statistically significant difference between the means along the Downtown versus the Residential route, however the mean PM10 values along the two routes  70 approached statistical significance as well, with most trips along the Downtown route resulting in higher exposures.   71 Table 7- Air pollution exposure measurements calculated from means of trials, including all trials except 122C.   Downtown Route  Residential Route Welch’s two sample  t-test Pollutant mean [SD] median; range (min,max) GM GSD mean [SD] median; range (min,max) GM GSD 95% CI of the difference of D vs. C  (low, high) P-value PNC (pt/cc) 16874 [4838] 15740  (9597, 29060) 16226 1.33 11130 [4939] 10390  (2972, 22130) 10011 1.88 3442, 8049 4.6 x 10-6 PM1 (µg/m3) 5.0 [4.2] 3.7 (1.2, 19.7) 3.8 2.0 3.8 [2.9] 2.9 (0.5, 12.1) 2.9 2.1 -0.44, 2.92 0.15 PM2.5  (µg/m3) 7.3 [5.3] 5.5 (2.4, 24.0) 6.0 1.9 5.8 [3.7] 4.4 (1.1, 15.3) 4.7 1.9 -0.58, 3.65 0.15 PM10  (µg/m3) 12.7 [7.3] 10.5 (4.3, 33.4) 10.9 1.8 9.9 [5.6] 8.8 (2.2, 28.5) 8.4 1.8 -0.14, 5.90 0.061 Note 1: Data was missing for the following trial data points - 101C (PNC, PM1, PM2.5, and PM10), 113D (PNC), 119D (PM1, PM2.5, and PM10), 150C (PNC), and 190C (PM1, PM2.5, and PM10). As well, PM1, PM2.5, and PM10 data from trial 122C was excluded due to these values being highly influential.   72 3.3 Pollution exposures and subclinical health end points  Table 8 shows the difference between trials for each of the air pollutants and clinical measures. The positive numbers show there was a higher GM for the Downtown route in comparison to the Residential route. EndoPAT™ RHI and blood biomarkers are in the same format of the difference (“post ride” - “pre ride”) between trials (“Downtown” values -  “Residential” values). A negative RHI value signifies a decrease in endothelial function compared to the Pre measurement. For the blood biomarkers, a positive number for these measurements means there was an increase after the Downtown ride, a decrease after the Residential ride, or both.   73 Table 8- Exposure and clinical data differences by route for each participant.  Participant Difference of GM Concentration of Air Pollutants Difference of RHI Measurements Difference of Blood Biomarker Measurements RouteDT - RouteRes Post - Pre Route DT - Res (DTpost-pre) - (Respost-pre) PM10 (μg/m3) PM2.5 (μg/m3) PM1 (μg/m3) PNC (pt/cc) DT Res CRP (mg/dL) IL-6 (pg/ mL) 8-OHdG (ng/mL) 100 15 -6.2 3.6 -308 -0.51 0.28 -0.79 0.00  -0.13 101 - - - - 1.1 -0.61 1.71 -0.63 -2.2 0.040 113 5.1 -1.0 0.50 12400 -0.47 0.64 -1.11 -0.010 2.5 0.28 116 -7.4 3.8 0.30 8560 -2.2 -0.26 -1.94 -0.030 - 0.020 119 21 -16 15 2950 -0.078 -0.018 -0.06 0.010 13 - 122 -75 49 -46 9610 -0.055 0.019 -0.07 -0.010 -4.9 -0.15 129 6.1 -5.2 3.3 3480 -0.12 1.1 -1.22 0.11 2.8 0.28 130 -14 11 -9.5 6960 1.4 -1.2 2.60 0.23 -2.0 - 133 16 -4.5 3.0 15000 -0.83 0.10 -0.93 0.090 -1.0 0.25 134 -0.60 1.9 -1.5 -10500 0.99 1.1 -0.11 0.00  - 141 1.8 -1.6 0.50 4710 -0.026 0.39 -0.42 0.030 -0.63 - 142 -2.9 -0.4 1.2 10800 -0.46 0.029 -0.49 -0.040  -0.20 145 12 -2.2 1.0 10300 -0.17 - -0.17 0.050 0.60 0.25 147 -5.3 3.1 -1.8 -13900 -0.32 0.78 -1.10 0.00 3.5 -0.090 150 8.6 -2.3 2.5 - 1.2 0.86 0.34 -0.030 0.34 - 152 -7.4 9.2 -8.5 -12400 -0.74 0.68 -1.42 0.030 -1.3 - 154 1.7 -1.4 1.3 8460 1.1 0.64 0.46 0.030 -0.45 -0.090 157 0.90 -0.30 0.30 3690 0.99 -0.48 1.47 -0.13 -1.5 0.10 158 14 -9.5 6.8 9350 -0.28 0.013 -0.29 -0.030 6.9 -0.010 159 -5.0 2.0 -0.50 -5000 -1.7 - -1.70 0.00 1.4 0.060 160 13 -9.2 4.1 -3030 0.35 0.65 -0.30 0.18 15 - 161 6.6 -7.3 7.3 -919 -0.36 1.8 -2.16 0.040 3.4 0.050 162 6.0 -4.3 2.7 1350 0.058 0.078 -0.02 -0.070 -0.24 0.00 163 2.8 -1.0 0.70 796 -0.50 0.10 -0.60 -0.22 -3.7 - 164 1.5 -1.7 2.0 285 -1.4 0.20 -1.60 -0.020 -1.4 0.010 165 -2.6 -0.20 0.0 11000 -0.078 -0.20 0.12 0.060 1.0 -0.13 166 -7.1 5.6 -2.9 8950 0.12 -0.51 0.63 0.00 6.2 -0.11 169 2.0 -0.30 1.3 11200 0.50 0.097 0.40 -0.010 2.1 0.080 170 -6.3 4.8 -3.7 15700 0.25 0.69 -0.44 0.00 -0.63 0.080 172 -6.3 5.4 -3.6 -275 -0.34 -0.39 0.05 0.070 -3.0 0.030 178 -0.2 1.1 -1.0 9740 0.014 -0.085 0.10 -0.14  0.24 181 -0.6 0.50 -0.30 4790 -0.39 0.68 -1.07 0.00 -2.6 -0.050 184 -4.1 4.1 -3.5 6960 - - - 0.14 -0.21 -0.070 186 -6.1 7.1 -6.7 6130 -0.29 -0.098 -0.19 0.00 2.8 0.00 187 -0.90 1.6 -1.1 5240 -0.0060 -0.067 0.06 0.27 -8.1 0.030 190 - - -  -2.6 -0.33 -2.27 0.010 -8.3 0.040 193 -8.5 4.7 -2.2 -1060 -0.40 1.9 -2.30 0.070 15 - 197 19.7 -20 17 -3100 -0.15 0.32 -0.47 0.040 -3.2 -0.14  74 Table 9 provides a directional summary for the clinical measures. Following the hypotheses, if RHI decreased more after the Downtown route (in comparison to the Residential route), then the differences calculated above should produce a negative result. More inflammation or oxidative stress produced after cycling on the Downtown route (compared to the Residential route) would be expected to produce positive results. Missing data points are identified as “Not Available”. Differences are calculated by subtracting the “pre” measurement from the “post” measurement.   Table 9 shows the comparisons of the GM of the exposures along the two routes, using t-tests paired by each participant. The three larger PM size classifications were not significantly different in the paired t-tests comparing the GM exposures, however the GM PNC concentration was significantly higher along the Downtown compared to the Residential route.  Table 9- GM pollutant exposures along the Downtown and Residential routes.  Pollutant Mean Estimate Downtown Mean  Estimate Residential Mean Difference (Downtown - Residential) of paired trials  95% CI of the  Mean Difference  Low, High  PNC (pt/cc) 14 007 9 809 3 969* 1 529, 6 409 PM1 (μg/m3) 4.5 3.7 0.35 -1.3, 2.0 PM2.5 (μg/m3) 6.6 5.6 0.38 -1.7, 2.4 PM10 (μg/m3) 11.3 9.4 1.4 -1.4, 4.2  Note 1: Incomplete pairs are not included; incomplete or missing trials include 101C, 113D, 150C, 190C.   75  Table 10 shows the IQRs of the pollutants, calculated from the one-second averages from all of the included trials. Those trials that were excluded had missing data because of technician or instrument error. Pollutant data was not available from trials 101C (PNC, PM1, PM2.5, and PM10), 113D (PNC), 119D (PM1, PM2.5, and PM10), 150C (PNC), and 190C (PM1, PM2.5, and PM10). As well, PM1, PM2.5, and PM10 data from trial 122C was excluded due to these values being highly influential.  Table 10- The IQR calculated from all included trials. Pollutant IQR PNC 7 600 pt/cc PM1 3.6 ug/m3 PM2.5 4.7 ug/m3 PM10 7.3 ug/m3   76 Figure 9 shows the histograms for the GM PNC for each trial along the Residential and Downtown routes, which was the only pollutant that was significantly different between the two routes (p< 0.01). Pollutant data was not available from trials 101C, and 190C, and data from trial 122C was excluded due to being an extreme data point.   Figure 9. Histograms showing counts of the GM of the PNC measurements for each trial along the Residential and Downtown routes.  Figure 10 shows histograms of the GM concentrations of PM2.5 along both routes, demonstrating that the data from the Downtown route is slightly more skewed to the right. While a number of higher GM concentrations of PM2.5 are found along the Downtown route, no significant differences between the mean concentrations were identified when the two routes were compared using paired t-tests (in Table 7).   77  Figure 10. Histograms showing counts of the GM concentrations of the PM2.5 measurements from each trial along the Residential and Downtown routes.  Figure 11 and Figure 12 show an example of the variability of PNC and PM2.5 exposures experienced by the same participant, on the two routes. The PNC exposures appeared to vary significantly during a one-hour ride compared to PM2.5 concentration over the same one-hour time period. The Downtown route in particular showed many sudden increases in PNC, while the Residential route showed fewer spikes in PNC. Appendix N shows the time series plots for all trials with this available PNC and PM2.5 data.  78  Figure 11. PM 2.5 and PNC time series plot along Residential route for participant 157.   Figure 12. PM 2.5 and PNC time series plot along Downtown route for participant 157.  79 Pearson product-moment correlation coefficients of different air particle size measurements Table 11 shows the Pearson product-moment correlation coefficient of each of the particle sizes measured, when all completed trials are included in the analysis. GMs for each trial were used in this calculation. All the measurements collected with the GRIMM monitor (PM1, PM2.5, and PM10) were all highly correlated with each other, while there was a very low correlation of these measurements with the PNC measurements. For this reason we have focused our analyses on the PNC, and PM2.5 measurements for simplification of the exposure data obtained from the GRIMM monitor.  Table 11- The Pearson product-moment correlation of GMs of all air pollution data from each trial. Pearson product-moment correlation coefficient of GMs   PNC PM1 PM2.5 PM1 -0.0089 - - PM2.5 0.0010 0.99 - PM10 0.067 0.96 0.97  A trial during an air quality event  The trial completed by participant 122 on August 5th, 2010 was along the Residential route. Figure 13 demonstrates how air pollutant measurements at the Metro Vancouver Kitsilano Air Monitor station (with PM2.5 shown) were two to three times higher than measurements taken during the rest of the summer period of 2010. This was attributed to a large amount of forest fire smoke within the province of British Columbia. Notably, other trials surrounding this date were completed on August 9th, 11th, and 17th, however no other extreme concentrations were noted in this data. A satellite image of the fire locations and smoke from NASA verified the presence of widespread fire  80 smoke (see Appendix O). As well, a press release from the British Columbia Wildfire Management Branch (Appendix P) warned that the weather pattern at the time was producing smoky conditions in many coastal communities in the province.    Figure 13. Hourly PM2.5 measurements from the Metro Vancouver Kitsilano air monitor station, through the summer of 2010.  Note 1: Measurements demonstrated very high concentrations surrounding the dates of August 5th and August 15th. Trial 122C was conducted on the Residential route on August 5th, 2010.    The Kitsilano air quality monitor is the nearest fixed monitor available to the study location, measuring PM2.5 approximately three kilometers away from the study testing location and start point of the ride. Air quality warnings from Environment Canada’s Air Quality Health Index (AQHI) sat in the “Moderate” range of the health index for air pollution risks to health. This level corresponded with the advice that those with lung or heart problems should consider reducing their 04-Aug14-Aug0510152025303540455031-May 20-Jun 10-Jul 30-Jul 19-Aug 08-Sep 28-SepPM2.5 in ug/m^3DatePM2.5 Concentration June 1- September 30, 2010PM2.5 81 time spent outdoors if they are experiencing any related symptoms, however members of the general population are not recommended to modify their outdoors activities unless they are experiencing air pollution-related health symptoms (see Appendix Q for full table of the AQHI explanatory messages). The participant wished to go through with testing and completed the trial with no complaints of symptoms, however air quality monitors on the bike found levels of all of the pollutants measured to be substantially higher than the other trial measurements. For example, the GM of PM10 for trial 122C was 96 ug/m3, while the mean of the rest of the GMs recorded for both Residential and Downtown routes was 11 ug/m3. The GM of the PM2.5 for trial 122C was 63 ug/m3, while the mean of the GMs of all other trials was 6.3 ug/m3. The GM PNC for trial 122C was 7 590 pt/cc, while the mean of the GMs of PNC for the rest of the trials was 11 873 pt/cc.   Further information about the highly influential data from trial 122C is available in Appendix W. For these reasons, the influential data from trial 122C has been excluded from the analysis of the rest of the data, unless indicated otherwise.    82 3.4 Physiological baseline measurements and comparisons of outcomes by route  Mean baseline measurements are shown in Table 12, with mean (post-pre ride) differences for the measures also shown. Additionally, Figure 14 shows the individual trajectory plots of the clinical measures for each participant.  Table 12- Baseline measurements and table of mean differences comparing the change in measurement for time points before and after each ride along the indicated route. Variable Baseline Mean [SD]  Range Change (Post-Pre) Downtown Route Mean [SD] Change (Post-Pre)  Residential Route Mean [SD] Post-Pre Range  Downtown Post-Pre Range Residential Systolic BP (mmHg) 117 [9] 90 - 139 -2 [7] -2 [6] -16, 14 -16, 14 Diastolic BP (mmHg) 67 [6] 52 - 83 1 [5] 0 [6] -9, 10 -13, 13 Reactive Hyperemia Index (RHI) 2.0 [0.64] 1.29 - 4.28 -0.18 [0.86] 0.25 [0.63] -2.6, 1.4 -1.2, 1.9 FEV1 (mL) 4280 [677] 2670 - 5860 46 [119] 46 [174] -270, 260 -580, 330 FVC (mL) 5610 [932] 3300 - 7250 46 [151] 21 [309] -280, 270 -960, 1030 FEV1/FVC ratio 0.77 [0.06] 0.63 - 0.92 0.00 [0.02] 0.01 [0.04] -0.03, 0.05 -0.11, 0.06 FEF25-75% 3.70 [0.89] 2.15 - 6.11 0.11 [0.25] 0.082 [0.42] -0.43, 0.62 -0.84, 0.90 CRP (mg/dL) 0.84 [1.2] 0.11 -7.7 0.01 [0.06] 0.01 [0.11] -0.10, 0.23 -0.22, 0.53 IL-6  (pg/mL) 3.6 [4.3] 0.023 - 16 0.55 [3.3] -0.45 [3.5] -5.1, 13 -14, 5.3 8-OHdG (ng/mL) 0.20 [0.12] 0.012 - 0.73 0.00 [0.11] -0.03 [0.11] -0.30, 0.24 -0.39, 0.16 Note 1: Spirometry data was missing from the trials 122D and 190C, with all applicable data removed from these subjects when complete pairs were needed.   83 Reactive Hyperemia Index  The baseline Reactive Hyperemia Index (RHI) measurement for all participants was a mean of 2.0 with an SD of 0.64, shown in Table 12. This pre-testing measurement included both of the “pre-ride” recordings for each participant. After riding along the Downtown route, the mean change was -0.18 for the RHI measure, with an SD of 0.86 (range= -2.6 to 1.4). After cycling along the residential route, the mean change for RHI was 0.25 with an SD of 0.63 (range= -1.2 to 1.9). It is important to note that an improved endothelial function means we expect the RHI would increase after the ride, so a larger positive number is the more favorable change if the cyclist experience an improved RHI after cycling along the Residential route. Figure 14 shows the results from the trials along the Residential route lean more towards the positive side of the scale, whereas trials along the Downtown route are closer to zero.   Shown in Table 13, the results from the paired t-test with for each participant’s RHI demonstrate that the mean change in RHI was -0.38 (lower, or more decreased) after cycling the Downtown route compared to the change observed on the Residential route. The 95% CI for this measure was -0.75 to -0.02.  Note that RHI was treated as a normally distributed data set. Appendix V shows graphs and the results of a test for normality.    84 Table 13- Clinical Measurement Summary by Route (Downtown and Residential) of post- and pre- cycling clinical measurements, with Paired t-test data.  Variable Downtown Route Residential Route ∆ Downtown - ∆ Residential Change (Post-Pre) Mean (95% CI) Change (Post-Pre) Mean (95% CI) Mean Difference (95%CI) Paired  t-test P-value Endothelial Function- EndoPAT™ RHI   -0.18 (-0.46, 0.11)   0.25 (0.03, 0.47)*   -0.38 (-0.75, -0.02)*   0.04*   Spirometry (unit)  FVC (L)   0.046  (-0.0038, 0.097) 0.021  (-0.084, 0.13) 0.023 (-0.085, 0.13) 0.66   FEV1 (L)   0.046  (-0.0066, 0.086) 0.049 (-0.011, 0.11) -0.00056 (-0.065, 0.064) 0.99  FEV1/FVC (ratio)  0.0014 (-0.0046, 0.0073)  0.0043 (-0.0086, 0.017)  0.0027 (-0.016, 0.0099)   0.66    FEF25-75% (L/s)  0.11 (0.023, 0.19)*  0.083 (-0.058, 0.22)  0.026 (-0.13, 0.19)  0.74   Blood Measures (unit)  CRP (mg/dL)  0.0088 (-0.012, 0.029) 0.0066 (-0.030, 0.043) 0.0022 (-0.043, 0.048) 0.92  IL-6 (pg/ml)  0.55 (-0.59, 1.67) -0.61 (-1.8, 0.57) 0.95 (-0.98, 2.89) 0.32  8-OHdG (ng/ml) -0.00045 (-0.040, 0.04) -0.031 (-0.071, 0.0082) 0.029 (-0.023, 0.081) 0.26  85 Note 1: Bolded values with the presence of * indicates statistical significance, where p< 0.05. Note 2: Paired spirometry data was missing for subjects 122 and 190.   Figure 14. Box plots showing results of endothelial function and blood measures.  Note: Lines connect individual ∆ (pre-post ride) measurements for each individual participant, with the box plots indicating the range of ∆ responses.  Spirometry measurements  All spirometry measurements showed little change (calculated by using the “post” measurement, minus the “pre” measurement of the Downtown - Residential route), with a mean increase in FEV1  by 46 mL after cycling either of the routes (Table 12). The SD after cycling the  86 Downtown route was 119 mL (range: -270 to 260 mL), and the Residential route had a SD of 174 mL with a range of -580 to 330 mL. FVC showed a broader range of measurements after cycling the Residential route, with measurements between -960 and 1030 mL (mean= 21 mL, SD= 309 mL), whereas FVC after cycling the Downtown route was mean= 46 mL (SD= 151 mL), and had a range of -280 to 270 mL. A negative value here would suggest reduced lung function. Figure 15 shows individual trajectory boxplots for each spirometry measure, showing that for FEV1 and FVC, the median values were approximately equal. The FEV1/FVC plot shows the Residential route had a much wider variability in ratios compared to the Downtown route, however no results from the paired t-tests in Table 13 demonstrated statistical significance between the Downtown and Residential routes. There was a small statistically significant increase in FEF25-75% after cycling along the Downtown route compared to the baseline measures (0.11 L/s), however no significant difference was found after cycling along the Residential route, though there was a mean increase of 0.083 L/s. Increases in this measure of lung function appear to be typical after exercise in healthy populations. (232) Spirometry results by trial are found in Appendix U.   87  Figure 15. Individual trajectory boxplots of spirometry results by route.  Note: Lines connect individual ∆ (pre-post ride) measurements for each participant, with the box plots indicating the range of ∆ responses.  Blood indicators of inflammation or oxidation CRP  Shown in Table 12, the mean baseline level of CRP for all participants was 0.84 mg/dL, with an SD of 1.2 mg/dL. After cycling in an environment where we expect inflammation to increase in the body, we expect the post-pre change to be a positive number. The paired t-test -1.0-0.50.00.51.0Change in Forced Vital CapacityRoute(n=37)Change in FVC, in LRoute D Route C(FVC) for Two Routes-0.6-0.4-0.20.00.2Change in Forced Expiratory Volume in 1 SecondRoute (n=37)Change in FEV1, in LRoute D Route C(FEV1) for Two Routes-0.10-0.050.000.05Change in FEV1/FVC for Two RoutesRoute (n=37)Change FEV1/FVC in %Route D Route C-0.50.00.5Change in Forced Expiratory FlowRoute (n=37)Change in FEF25-75% in L/sRoute D Route C(FEF25-75%), for Two Routes 88 results shown in Table 13 demonstrate no significant difference in CRP serum concentrations between the two routes, with a mean difference of 0.   IL-6  The mean baseline measurement of IL-6 was 3.6 pg/mL with an SD of 4.3 (Table 12). If higher air pollution exposure or intake were associated with IL-6 production, the post-pre difference would be expected to be larger than zero. After cycling along the Downtown route, the mean post-pre IL-6 measurement was 0.55 pg/mL with an SD of 3.3. The mean change after cycling along the Residential route was -0.45 with an SD of 3.5. Table 13 shows no significant difference for paired t-test results between the two routes, although there was a mean difference of 0.95 pg/mL and the 95% CI distributed more in the positive range in this interval (between -0.98 and 2.89). This demonstrates that approximately 3/4 of individuals had an increase in IL-6 when comparing the differences between the Downtown and Residential routes.  8-OHdG  The mean baseline measurement of 8-OHdG was 0.20 ng/mL (shown in Table 12), with an SD of 0.12. The mean post-pre change would be expected to increase after exposure to conditions that would provoke increased oxidative stress. For this measure as well there were no significant findings, but asymmetry in the results, with 2/3 of individuals having an increase in 8-OHdG after cycling these two routes, with the 95% CI of -0.03 to 0.07  89 Correlations of blood components when compared by Route   Pearson product-moment correlation coefficients shown in Table 14 demonstrate how the changes in the clinical measures compare with each other. Only coefficient values for the CRP measures were significantly correlated with IL-6 measurements on the Downtown route (shown bolded with an asterisk *, p< 0.05), with the other measures being only minimally correlated or negatively correlated with one another. There were no statistically significant Pearson product-moment correlation coefficient values for the Residential route blood or RHI measures.  90 Table 14- Pearson product-moment correlation coefficients for the ∆Post-Pre for Biomarkers and RHI for each bicycling route.  Downtown route correlations  CRP IL-6 8-OHdG IL-6 0.47* - - 8-OHdG 0.18 0.13 - RHI 0.13 0.19 -0.13 Residential route correlations  CRP IL-6 8-OHdG IL-6 -0.0061 - - 8-OHdG 0.15 0.066 - RHI -0.24 -0.19 -0.26 Note 1- Only complete pairs are included in this analysis of Pearson product-moment correlation coefficients. Clinical measures not available for the following trials were: RHI- 145C, 159C, 184C & D CRP- (all available) IL-6- 100D, 116D, 134C, 142C, 178D 8-OHdG- 119C, 130C, 130D, 134C, 141C, 150D, 152C, 160C, 163D, 193D Note 2- Bolded values with the presence of * indicates statistical significance, where p< 0.05.  3.5 Estimating intake  After collecting data in 2010, it was determined that using a step-wise cycle ergometer protocol would ameliorate the air pollution intake estimate. The previous protocol only provided the mean  data at the mean HR during each trial, however this did not provide information about  91 the  at other HRs, limiting the intakes that could be estimated for each 2010 trial to Intake 1 and Intake 2. Upon realizing this limitation, the protocol was modified to capture as much of the range as possible for the HR and  data points for each participant. The Velotron ergometer was set to increase resistance so that each participant was forced to increase their power output at two-minute intervals, with increases seen in the HR and at each step of increased resistance. This allowed for the measurement of  through a wider HR range that most participants experienced during each ride. With this new protocol, the Intake 3 estimation could be made for 23 of the participants that participated in the study during the summer of 2011.   Table 15 shows an example of the data collected for participant 101, a female participant. This data table was used to make a graph such as the one shown in Figure 16, which was then fit with a predictive line and shows the predictive equation used to interpolate all  data points in between. In this case, the predictive equation that best fit the data was an exponential formula, however the equation fit type could be any type (logarithmic, exponential, linear) as long as it was the best fit that individual’s data. The equivalent data was similar for male participants with the exception of the intervals between the power output increases, which were 30-watts for every two-minute step, rather than the 20-watt intervals that were used for females. Appendix R shows the HR - relationship curve for each participant.   92 Table 15- Measured HR and minute ventilation during the Velotron stepwise ergometer test for participant 101, a female.   Figure 16. HR and minute ventilation plot for participant 101   Table 16 shows the predictive formulas calculated from the Velotron test, including the lowest and highest HR data points. The coefficient of determination (R2) values for the formulas ranged from 0.72 to 1 (rounded), however most predictive lines were fit with an R2 that was 0.95 y = 0.0007x2.2253R² = 0.978401020304050600 20 40 60 80 100 120 140 160 180Minute Ventilation (Litres per minute)Heart Rate (beats per minute)Subject 101 Minute Ventilation Relationship"measured minute ventilation" Predictive Equation LineProgrammed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using following "power" regression equation): = 0.0007 (HR)^2.2253 (L/min) 0 73 8.72 9.81 20 98 18.80 18.89 40 101 21.40 20.20 60 106 24.44 22.49 80 112 24.36 25.42 100 120 32.16 29.64 120 130 34.80 35.42 140 140 37.16 41.77 160 150 49.64 48.70 180 156 48.88 53.14  93 or higher. The equation type of the predictive line with the highest R2 was selected in order to best interpolate the values in between the known HR data points.  94 Table 16- Predictive minute ventilation equations from the Velotron test for 23 participants. Participant HR Min, Max  Predictive  equation (L/min) R2 Regression line type 101 73, 156 0.0007 (HR)^2.2253 0.98 power 154 57, 132 0.0169 (HR)^1.6522 0.98 power 157 57, 154 0.0113(HR)^1.7904 0.97 power 158 57, 154 -0.0071(HR)2+ 2.1814(HR)-90.485 0.97 polynomial 159 94,156 -0.0048(HR)2+ 1.6751(HR)- 99.3 0.87 polynomial 160 54, 124 0.0148(HR)2 +1.4824 (HR) + 53.664 0.99 polynomial 161 63, 154 0.0021(HR)2 +0.038(HR) + 0.6164 0.99 polynomial 162 93, 136 -0.0151(HR)2 + 3.9554 (HR) – 219.81 0.72 polynomial 163 73, 152 0.0051(HR)2 – 0.4077 (HR) + 16.297 0.99 polynomial 164 80, 174 0.0005(HR)^2.3064 0.99 power 165 65, 133 2.4183e^0.0249(HR) 1.00 (0.997) exponential 166 65, 160 0.0036(HR)^2 -0.3676(HR) + 31.608 0.95 polynomial 169 71, 136 6.9369e^0.017(HR) 0.98 exponential 170 82, 148 0.0006(HR)^2 + 0.2484(HR) – 9.1433 0.93 polynomial 172 76, 171 0.018 (HR)^1.6229 0.95 power 178 82, 169 0.0071(HR)^2 – 1.1033(HR) + 64.241 0.97 power 181 62, 134 -0.0001(HR)^2 + 0.3594(HR) – 10.881 0.98 polynomial 184 89, 148 0.007(HR)^2 – 0.754 (HR) + 30.883 0.98 polynomial 186 88, 162 57.13ln(HR) – 229.87 0.96 logarithmic 187 49, 148 1.7296 e^ 0.0256(HR) 0.95 exponential 190 86, 179 2.2025 e^ 0.0192(HR) 0.97 exponential 193 69, 158 0.0427 e^ 1.4758(HR) 0.98 exponential 197 67, 156 -0.0006(HR)2 +0.6749(HR) - 24.05 0.95 polynomial  95 3.6 Exposure and intake estimates  Table 17 provides the mean exposure data and information used to calculate intakes for each of the trials. GM pollution measurements of mean HR, mean power output, and  are shown to demonstrate the range in data that was collected. Each row shows the data output from each trial, with Residential route (trials marked with “C”) and the Downtown (trials marked with “D”) cycling routes separated. The PowerTap provided the HR and power output data. The mean  is the  measured by the respirometer at the mean HR recorded during each trial. GM concentration of each pollutant is shown. Missing data points (shown as empty boxes) are due to equipment failure (100D, 193D) or operator error (101C, 150C, 190C).     96 Table 17- All trial mean PowerTap summary data, minute ventilation, and GM pollutant exposures with one trial per row.  Participant Route Mean HR in bpm Mean Power Output (Watts) Mean  L/min GM PNC (pt/cc) PM1 (µg/m3) PM2.5 (µg/m3) PM10 (µg/m3) 100 C 125 106 55.7 15439 2.4 3.9 11 100 D 146 114  12362 6.1 10 26 101 C        101 D 119 93 32.4 8633 7.5 12 17 113 C 112 117 39.9 5594 1.9 2.9 5.9 113 D 112 115 38.2 18021 2.3 3.9 11 116 C 144 127 50.5 9794 4.4 10 19 116 D 119 100 53.7 18353 4.7 6.4 12 119 C 105 78 36.0 9225 1.6 3.4 6.3 119 D 129 110 43.8 12172 17 20 27 122 C 128 122 67.1 7590 58 63 96 122 D 116 107 50.7 17196 11 14 21 129 C 104 124 41.6 16004 3.4 5.3 13 129 D 118 141 50.4 19480 6.6 11 19 130 C 114 98 35.4 11761 12 15 27 130 D 109 78 34.0 18722 2.5 4.5 13 133 C 124 97 40.8 5058 1.6 2.3 4.3 133 D 126 97 30.9 20086 4.5 6.8 20 134 C 138 151 54.8 8091 2.5 3.9 7.5 134 D 146 166 39.3 7038 1.0 2.1 6.9 141 C 108 151 37.7 7765 1.4 2.2 4.8 141 D 108 152 61.2 12478 1.9 3.8 6.6 142 C 165 176 49.3 3136 1.0 2.7 8.4 142 D 148 165 54.9 13943 2.2 3.1 5.6 145 C 114 93 41.1 6926 1.3 1.9 3.5 145 D 113 107 36.8 17191 2.3 4.0 16 147 C 123 180 48.3 21095 3.4 6.4 12 147 D 115 160 58.0 7187 1.6 3.3 6.6 150 C 153 89 44.0  2.0 4.2 7.7 150 D 170 101 48.5 25198 4.6 6.5 16 152 C 144 180 73.6 19935 10 12 16 152 D 142 155 79.4 7540 1.4 3.1 8.2 154 C 112 146 41.0 6883 1.3 2.1 3.1 154 D 100 97 33.1 15342 2.6 3.5 4.8 157 C 121 133 61.0 8034 3.9 5.3 6.9 157 D 103 106 47.2 11724 4.3 5.6 7.7 158 C 140 163 68.2 2927 0.5 1.1 2.1 158 D 123 131 68.2 12280 7.2 11 16 159 C 125 83 35.1 18694 2.2 4.3 8.5 159 D 122 82 33.6 13696 1.6 2.3 3.6 160 C 109 120 67.9 15160 3.3 4.2 5.9 160 D 109 112 67.9 12126 7.4 14 19 161 C 136 116 44.6 14976 5.7 7.8 12 161 D 115 85 32.8 14058 13 15 19 162 C 124 140 38.5 7050 1.1 2.1 3.6 162 D 138 162 38.5 8404 3.8 6.4 9.7  97 Participant Route Mean HR in bpm Mean Power Output (Watts) Mean  L/min GM PNC (pt/cc) PM1 (µg/m3) PM2.5 (µg/m3) PM10 (µg/m3) 163 C 121 110 41.6 11384 0.6 1.2 2.2 163 D 116 103 37.6 12180 1.3 2.1 5.0 164 C 161 141 60.3 15067 3.5 5.7 9.1 164 D 160 115 59.5 15353 5.5 7.4 11 165 C 112 126 40.0 827 1.9 3.4 8.5 165 D 106 119 34.0 11772 1.9 3.5 5.9 166 C 155 182 61.1 2078 4.4 8.1 11 166 D 135 138 47.6 11032 1.5 2.5 4.1 169 C 120 159 54.2 11376 2.7 5.3 8.5 169 D 103 126 39.4 22528 3.9 5.7 11 170 C 138 116 36.6 4510 5.7 7.7 11 170 D 119 92 28.9 20257 2.0 2.8 4.8 172 C 141 117 55.1 9016 4.9 7.9 11 172 D 141 122 55.1 8741 1.3 2.5 4.6 178 C 144 161 52.6 7788 2.9 4.3 5.4 178 D 142 133 50.7 17532 1.9 3.2 5.2 181 C 114 95 28.8 9646 3.9 5.5 8.8 181 D 100 79 24.1 14431 3.5 5.0 8.2 184 C 89 120 21.3 8582 6.8 9.2 13 184 D 86 79 18.3 15540 3.3 5.0 8.4 186 C 130 106 47.6 6876 10 12 15 186 D 129 97 47.2 13008 3.5 5.2 8.8 187 C 159 141 84.6 5020 3.7 5.2 7.7 187 D 145 100 66.5 10262 2.6 3.7 6.7 190 C 89 98 12.7     190 D 87 90 12.7 12785 7.1 9.4 14 193 C 118 117 48.3 13846 9.4 14 22 193 D    12783 7.1 9.5 14 197 C 127 103 52.0 13943 2.9 3.8 6.5 197 D 113 171 44.6 10843 20 24 26    98 GM pollutant data from Table 17 was used to calculate Intake 1 estimates shown in Table 18. The three intake estimates are shown for each trial. In Table 18, only those estimates for PNC are shown, while Intake 2 estimates for PM1, PM2.5, and PM10   are available in Appendix S. Intake estimate values that are more than three SDs from the median of each intake (for each pollutant) have been marked with an asterisk (*). These values have been excluded from the remaining analyses concerning intakes.  Along with ride time,  at rest (while sitting on the ergometer) and mean  are provided for each trial. Mean  was used in the calculation of Intake 1 and 2, while estimates outside of the 95% CI are denoted with an asterisk * and not included in mean calculations. Estimated total volume of air breathed during equivalent time at rest and while riding during each trial is shown, along with the ratio of at rest compared to while cycling for Intake 3 of each trial. The resting  (measured during the indoor cycling test) is shown, when available (for 2011 participants). The ratio of  (ride: rest) is also shown, when available. Partial measurements were used to calculate means during rides, however Intake 2 or 3 were not calculated for those trials missing more than 25% of the second-to-second data HR or pollution data.  Comparing the ratios of different Intakes to the resting  (or resting volume of air breathed in) during the period of time equivalent to cycling brings forward some interesting points. When using the “  in L/min at rest*” group mean data with the “Mean  L/min during ride by HR”, the small group of females had a  ratio (from riding compared to at rest) of 3.6, while the males’ ratio was 3.0. The combined ratio for both sexes was 3.0. These intake volume ratios would apply to Intake 1 and Intake 2 estimates. Conversely, when using Intake 3: rest ratio (looking at the  99 column on the far right), the female intake volume ratio was 2.9, the males’ was 3.0. This highlights the possibility that using different Intake estimation methods may result in significant discrepancies between all participants, or in this particular case, the female sub-group in this study. It appears the female Intake 1 estimates could be overestimating the intake for some participants, though there may be other possible explanations including the fact that Intake 1 relies on mean values, which are easily influenced by extreme data points.   100 Table 18- PNC Intake estimates 1, 2, and 3 for each trial. Participant  and Trial Ride Time (min)  in L/min at rest* Mean  L/min during ride by HR  GM concentration along route of PNC (pt/cc) Intake 1  Intake 2 Intake 3 At rest  Intake 1 and 2 Intake 3 Intake 3: rest PNC (particles) PNC (particles) PNC (particles) Estimated volume of air inhaled (Litres) (resting  x ride time) Estimated volume of air inhaled (Litres) Estimated volume of air inhaled (Litres) Ratio of  riding compared to at rest* 100C 63.98  55.7 15439 5.50 x 1010 6.16 x 1010   3563.7   100D 59.68   12362         101C 66.28 8.72    3.18 x 1010  650.0     101D 65.88 8.72 32.4 8633 1.84 x 1010 2.23 x 1010 2.24 x 1010 646.1 2134.5 1995.2 3.7 113C 64.92  39.9 5594 1.45 x 1010 1.74 x 1010   2590.3    113D 62.48  38.2 18021 4.30 x 1010    2386.7    116C 62.60  50.5 9794 3.10 x 1010 3.60 x 1010   3161.3    116D 66.98  53.7 18353 6.60 x 1010 7.46 x 1010   3596.8    119C 64.68  36.0 9225 2.15 x 1010 2.70 x 1010   2328.5    119D 55.65  43.8 12172 2.97 x 1010 1.91 x 1010   2437.5    122C 72.88  67.1 7590 3.71 x 1010 4.34 x 1010       122D 65.88  50.7 17196 5.74 x 1010 6.38 x 1010   3340.1    129C 64.40  41.6 16004 4.29 x 1010 4.50 x 1010   2679.0    129D 63.68  50.4 19480 6.25 x 1010 7.85 x 1010   3209.5    130C 60.28  35.4 11761 2.51 x 1010 2.64 x 1010   2133.9    130D 63.78  34.0 18722 4.06 x 1010 5.64 x 1010   2168.5    133C 59.00  40.8 5058 1.22 x 1010 1.34 x 1010    2407.2    133D 58.48  30.9 20086 3.63 x 1010 3.93 x 1010   1807.0    134C 56.70  54.8 8091 2.51 x 1010 2.60 x 1010   3107.2    134D 65.10  39.3 7038 1.80 x 1010 2.38 x 1010   2558.4    141C 59.38  37.7 7765 1.74 x 1010  1.93 x 1010   2238.6    141D 65.20  61.2 12478 4.98 x 1010 5.90 x 1010   3990.2    142C 61.20  49.3 3136 9.46 x 109 1.19 x 1010   3017.2    142D 66.43  54.9 13943 5.08 x 1010 5.63 x 1010   3647.0    145C 64.48  41.1 6926 1.84 x 1010 2.02 x 1010   2650.1    145D 61.40  36.8 17191 3.88 x 1010 4.51 x 1010   2259.5    147C 61.82  48.3 21095 6.30 x 1010 6.62 x 1010   2985.9    147D 59.65  58.0 7187 2.49 x 1010 3.36 x 1010   3459.7     101 Participant  and Trial Ride Time (min)  in L/min at rest* Mean  L/min during ride by HR  GM concentration along route of PNC (pt/cc) Intake 1  Intake 2 Intake 3 At rest  Intake 1 and 2 Intake 3 Intake 3: rest PNC (particles) PNC (particles) PNC (particles) Estimated volume of air inhaled (Litres) (resting  x ride time) Estimated volume of air inhaled (Litres) Estimated volume of air inhaled (Litres) Ratio of  riding compared to at rest* 150C 61.78  44.0      2718.3    150D 60.08  48.5 25198 7.34 x 1010 8.47 x 1010   2913.9    152C 62.78  73.6 19935 9.21 x 1010* 9.76 x1010*   4620.6    152D 59.42  79.4 7540 3.56 x 1010 4.57 x 1010   4717.9    154C 59.55 13.8 41.0 6883 1.68 x 1010 2.05 x 1010 1.98 x 1010 801.4 2441.6 2443.5 3.0 154D 64.70 13.8 33.1 15342 3.29 x 1010 4.45 x 1010 4.42 x 1010 870.7 2141.6 2207.8 2.4 157C 66.78 14.1 61.0 8034 3.27 x 1010 3.84 x 1010 3.90 x 1010 1050.6 4073.6 4125.8 4.3 157D 68.58 14.1 47.2 11724 3.79 x 1010 4.64 x 1010 # 1078.9 3237.0 3138.5 3.3 158C 71.25 19.7 68.2 2927 1.42 x 1010 # 1.46 x 1010 1428.1 4859.3 4857.6 3.5 158D 69.38 19.7 68.2 12280 5.81 x 1010 4.96 x 1010 5.47 x 1010 1390.6 4731.7 4342.6 3.5 159C 62.28 8.64 35.1 18694 4.09 x 1010 4.01 x 1010 4.00 x 1010 980.7 2186.0 2079.0 4.1 159D 62.35 8.64 33.6 13696 2.87 x 1010 3.47 x 1010 4.62 x 1010 981.8 2095.0 1974.8 3.9 160C 62.98 17.4 67.9 15160 6.48 x 1010 6.32 x 1010 6.31 x 1010 1056.3 4276.3 3725.5 3.9 160D 60.43 17.4 67.9 12126 4.98 x 1010 2.54 x 1010 2.54 x 1010 1013.5 4103.2 1963.6 3.9 161C 60.78 9.20 44.6 14976 4.06 x 1010 4.31 x 1010 4.27 x 1010 689.6 2710.8 2748.3 4.8 161D 66.08 9.20 32.8 14058 3.05 x 1010 3.49 x 1010 3.51 x 1010 749.7 2167.4 2201.1 3.6 162C 62.48 7.92 38.5 7050 1.70 x 1010 2.59 x 1010 2.67 x 1010 1089.8 2405.5 2236.8 4.9 162D 58.67 7.92 38.5 8404 1.90 x 1010 2.43 x 1010 2.43 x 1010 1023.3 2258.8 2224.2 4.9 163C 66.58 14.6 41.6 11384 3.15 x 1010 3.37 x 1010 3.32 x 1010 913.0 2769.7 2817.9 2.9 163D 65.58 14.6 37.6 12180 3.00 x 1010 3.44 x 1010 3.51 x 1010 899.3 2465.8 2537.8 2.6 164C 61.58 11.5 60.3 15067 5.59 x 1010 5.76 x 1010 5.71 x 1010 754.6 3713.3 3732.5 5.3 164D 64.58 11.5 59.5 15353 5.90 x 1010 7.42 x 1010 7.57 x 1010 791.3 3842.5 3900.2 5.2 165C 60.95 12.5 40.0 827 2.02 x 109 2.44 x 109 2.34 x 109 810.0 2438.0 2484.9 3.2 165D 62.28 12.5 34.0 11772 2.49 x 1010 3.40 x 1010 3.24 x 1010 827.7 2117.5 2209.8 2.7 166C 60.50 20.6 61.1 2078 7.68 x 109 1.06 x 1010 1.02 x 1010 1386.9 3696.6 3554.1 3.0 166D 60.18 20.6 47.6 11032 3.16 x 1010 4.56 x 1010 4.42 x 1010 1379.6 2864.6 2913.8 2.3 169C 59.58 23.0 54.2 11376 3.67 x 1010 3.79 x 1010 3.73 x 1010 1381.8 3229.2 3171.0 2.4 169D 64.78 23.0 39.4 22528 5.75 x 1010 7.52 x 1010 7.60 x 1010 1502.4 2552.3 2681.6 1.7 170C 64.08 13.0 36.6 4510 1.06 x 1010 1.15 x 1010 1.12 x 1010 977.9 2345.3 2248.9 2.8 170D 69.38 13.0 28.9 20257 4.06 x 1010 4.39 x 1010 4.43 x 1010 1058.7 2005.1 1988.2 2.2  102 Participant  and Trial Ride Time (min)  in L/min at rest* Mean  L/min during ride by HR  GM concentration along route of PNC (pt/cc) Intake 1  Intake 2 Intake 3 At rest  Intake 1 and 2 Intake 3 Intake 3: rest PNC (particles) PNC (particles) PNC (particles) Estimated volume of air inhaled (Litres) (resting  x ride time) Estimated volume of air inhaled (Litres) Estimated volume of air inhaled (Litres) Ratio of  riding compared to at rest* 172C 59.08 18.2 55.1 9016 2.94 x 1010 3.39 x 1010 3.37 x 1010 1199.7 3255.3 3287.0 3.0 172D 62.28 18.2 55.1 8741 3.00 x 1010 3.76 x 1010 3.64 x 1010 1264.7 3431.6 3469.4 3.0 178C 63.57 17.9 52.6 7788 2.60 x 1010 3.20 x 1010 3.10 x 1010 1367.4 3343.8 3420.9 2.9 178D 63.75 17.9 50.7 17532 5.67 x 1010 6.30 x 1010 6.17 x 1010 1371.3 3232.1 3335.2 2.8 181C 63.42 10.3 28.8 9646 1.76 x 1010 1.84 x 1010 1.78 x 1010 698.7 1826.5 1757.9 2.8 181D 66.12 10.3 24.1 14431 2.30 x 1010 2.82 x 1010 2.84 x 1010 728.5 1593.5 1559.5 2.3 184C 62.80 18.9 21.3 8582 1.15 x 1010 1.59 x 1010 1.52 x 1010 1202.2 1337.6 1672.9 1.1 184D 68.00 18.9 18.3 15540 1.93 x 1010 2.58 x 1010 2.53 x 1010 1301.7 1244.4 1463.1 1.0 186C 63.08 23.6 47.6 6876 2.06 x 1010 2.17 x 1010 2.16 x 1010 1635.0 3002.6 2845.9 2.0 186D 64.38 23.6 47.2 13008 3.95 x 1010 4.53 x 1010 4.49 x 1010 1668.7 3038.7 2941.6 2.0 187C 61.70 5.46 84.6 5020 2.62 x 1010 2.19 x 1010 2.19 x 1010 374.1 5219.8 4503.6 16 187D 70.22 5.46 66.5 10262 4.79 x 1010 5.13 x 1010 5.02 x 1010 425.8 4669.6 4491.5 12 190C 71.98 10.9 12.7     826.5 914.1 861.9 1.2 190D 64.67 10.9 12.7 12785 1.05 x 1010 1.90 x 1010 1.81 x 1010 742.6 821.3 893.7 1.2 193C 57.08 21.8 48.3 13846 3.82 x 1010 4.08 x 1010 4.07 x 1010 1260.9 2757.0 2794.9 2.2 193D 68.63 21.8  12783    1516.1    197C 62.68 14.7 52.0 13943 4.54 x 1010 4.95 x 1010 4.81 x 1010 1158.0 3259.4 3219.7 3.5 197D 65.05 14.7 44.6 10843 3.15 x 1010 3.85 x 1010 3.78 x 1010 1201.8 2901.2 2869.9 3.0 Mean  Males 63.56 15.79 # 48.07 # - 3.32 x 1010 3.86 x 1010  3.54 x 1010 1107.2 3005.3 2875.9 3.0 Mean Females 63.07 10.43i # 40.69 i # - 3.53 x 1010 3.86 x 1010 i  3.81 x 1010 i 751.7 i 2536.0 2359.6 i 2.9 i Mean Downtown 63.94 - 44.79 14007 3.90 x 1010 4.54 x 1010 4.17 x 1010 - 2918.4 2728.7 2.8 Mean Residential 62.95 - 47.81 9746 2.79 x 1010 3.13 x 1010 2.99 x 1010 - 3036.1 2935.9 3.1 Overall Mean 63.44 14.76 44.84 11964 3.37 x 1010 3.86 x 1010 3.59 x 1010 1046.3 2894.5 2770.3 3.0   103 Notes for Table 18 -The following definitions apply for each level of intake: Intake 1- defined as the mean  multiplied by the GM concentration of the respective pollutant, multiplied by the time of the trial, ex. 38.00 L/min * 15 000pt/cc of PNC * 61.50 mins Intake 2- defined as the sum of [the second-by-second concentrations of the respective pollutant multiplied by the mean  (in L/min) / 60 secs] Intake 3- defined as each second-to-second pollution concentration multiplied by the second-to-second , using HR as the predictor of ). i - Interpret cells with “i” cautiously as they are based on the data of only 5 female participants. # - These means calculated using complete pairs only (using data from 2011 trials) Trials identified as “C” are the Residential trials, while “D” are the Downtown trials. * -  in L/min at rest was only measured for the 2011 participant data, measured from the first step of the Velotron ergometer test. This data was used to calculate the Intake 3: rest ratio for the 2011 participants. Trial 122C is italicized at it is used in a number of analyses, but the data is provided for comparison.  104 Figure 17 shows all three PNC Intake estimate types plotted for the Downtown (DT) and Residential (Res) routes (Intake 1 = DT1 and Res1, Intake 2 = DT2 and Res2, Intake 3= DT3 and Res3). The statistical bar shows the central point (the median), and the 25th and 75th percentile range. The Downtown PNC values have consistently higher median values (shown by the central black point), compared to all three estimates for the Residential trials. All data points were jittered horizontally for better contrast. Intake estimate “3” was not available for a total of 33/76 trials for UFP, and 32/76 trials for PM2.5. Five intake estimates were not available for both PNC and PM2.5 for intake estimate “2”, six estimates were not available for PNC, and five were not available for PM2.5 for Intake 1 (Participant 122C- Residential trial was excluded). All of the Intake estimate values are found in Table 19.  Figure 17. Scatter plot of PNC Intake 1, 2 and 3 (particles) along each route.   105 The Intake 1 estimate along the Downtown route was higher than the Intake 1 estimate from the Residential route; this trend follows for all Intake estimates. Table 19 shows how the mean and median Intake 1, 2, or 3 values for all of the intake estimates along the Downtown route were higher than the intake estimates along the Residential route. A scatter plot and tables for comparison of PM2.5 Intake 1, 2, and 3 estimates are available in Appendix S.  Table 19- Summary statistics for PNC estimated intakes (particles), for Downtown and Residential routes.  PNC count Downtown Residential Intake 1 Intake 2 Intake 3 Intake 1 Intake 2 Intake 3 Mean 3.90 x 1010 4.54 x 1010 4.17 x 1010 2.98 x 1010 3.29 x 1010 2.99 x 1010 Median 3.71 x 1010 4.48 x 1010 4.10 x 1010 2.56 x 1010 2.94 x 1010 3.10 x 1010 SD 1.55 x 1010 1.79 x 1010 1.60 x 1010 1.94 x 1010 1.96 x 1010 1.59 x 1010 Min 1.05 x 1010 1.90 x 1010 1.81 x 1010 2.02 x 109 2.44 x 109 2.34 x 1010 Max 7.34 x 1010 8.47 x 1010 7.60 x 1010 9.21 x 1010 9.76 x 1010 6.31 x 1010  Correlation of intake estimate data   We expected that Intake 3 would be the intake that is closest to the true intake value, as it uses second-by-second data along with the more precise  measured from the longer ergometer test protocol. Intake 2 incorporated the second-to-second changes in pollution concentration, but still uses the mean . It is important to note that 38 participants were included in the analysis of Intakes 1 and 2, but Intake 3 only included the data of 22 participants.   Paired t-tests of intake estimates Table 20 shows the results of paired t-tests for each type of intake estimate for pairs of trials from each participant. For each of the three Intake estimates described, statistically significant  106 differences were found when the PNC was compared between the Downtown and Residential routes, however no other PM size classification had significant differences by intake estimate type.   107 Table 20- Paired t-tests of intakes by route type for each pollutant.  Mean Intake Downtown Mean  Intake Residential Mean Difference of paired trials (Downtown - Residential) 95% CI of the Mean Difference Low, High Intake 1 PNC (particles) 3.90 x 1010 2.98 x 1010 1.01 x 1010 2.15 x 109, 1.81 x 1010 * PM1 (μg) 12.4 11.2 0.73 -4.7, 6.1 PM2.5 (μg) 18.3 17.0 0.90 -6.1, 7.9 PM10 (μg) 30.9 27.7 3.5 -5.2, 12.1 Intake 2 PNC (particles) 4.54 x 1010 3.29 x 1010 1.18 x 1010 3.16 x 109, 2.05 x 1010 * PM1 (μg) 20.4 10.8 8.9 -6.5, 24.2 PM2.5 (μg) 26.0 16.5 8.6 -7.4, 24.7 PM10 (μg) 39.3 27.9 11.0 -5.4, 27.3 Intake 3 PNC (particles) 4.17 x 1010 2.99 x1010 1.44 x 1010 5.23 x 109, 2.37 x 1010 * PM1 (μg) 13.1 10.9 2.3 -3.7, 8.4 PM2.5 (μg) 18.4 16.4 2.0 -5.6, 9.6 PM10 (μg) 27.2 24.6 2.4 -6.0, 10.8  108 Table 21 shows the comparison of Intakes using paired t-tests for the group mean data by intake estimate type. The paired t-tests between PNC Intake 1 and 2, and Intake 1 and 3, show a statistically significant 95% CI. When PM2.5 intakes are compared using paired t-tests, none of the mean differences are significantly different from one another.   Table 21- Paired t-tests between intake levels.  Pollutant Pairs of Intake Numbers Mean Difference of paired estimates 95% CI  low, high PNC 1 vs. 2 -4.47 x 109 -5.94 x 109, -3.00 x 109 * PNC 1 vs. 3 -3.91 x 109 -6.01 x 109, -1.81 x 109 * PNC 2 vs. 3 -4.53 x 107 -7.09 x 108, 6.18 x 108 PM2.5 1 vs. 2 -3.54 -9.4, 2.3 PM2.5 1 vs. 3 0.85 -0.95, 2.7 PM2.5 2 vs. 3 5.94 -3.2, 15 Note 1: Units for PNC are in particles, for PM2.5 are in μg Note 2: Missing data (113D, 119D) and excluded data (122C) are not included in this analysis.  Note 3: Bolded values with the presence of * indicates statistical significance where p< 0.05.  Figure 18 shows the scatter plots and correlation coefficients of the PNC Intake estimates. All of the coefficients are higher than 0.9, with the correlation between Intake 2 and Intake 3 being the highest at 0.992. Intake 2 is very highly correlated with the most precise Intake 3. For this reason, Intake 2 was selected for use as the Intake estimation variable for the analysis of the health response data with the mixed effects models in Section 3.8. Because Intake 2 was available for  109 most participants and the three Intake measures were all well correlated, Intake 2 will simply be described as the Intake for each subject, moving forward.   Figure 18. Scatter plots, and correlation coefficients of each intake estimate for PNC.  Note 1: Trial 122C excluded as were 113D (PNC), 119D (PM2.5), and 193 (incomplete pair)   3.7 Other route differences  Two separate routes were used for each trial, with differences identified that may impact the results of this study. Route characteristics that could not be held constant such as number of red lights or intersections encountered, traffic volumes, or road gradient could have impacted the measures presented below. These measures have been compared using paired t-tests to assist in  110 identifying where differences between routes existed. Paired t-tests for other route comparisons of independent variables are found in Appendix T. Means are shown with the bold line, 25th and 75th are indicated by the box, and the whiskers indicate the 95% CI for each group of data.  Power output  Figure 19 shows the distribution of (within trial) mean power output along the Residential route including 37 qualifying trials), with an overall mean of 118 watts, compared to an overall mean along the Downtown route of 99 watts.  Figure 19. Box plot demonstrating the mean power output (in Watts) of participants for 74 (of 76 possible) trials where this data was available.  A paired t-test comparing the power output along the two cycling route (by participant) found the Residential route was associated with a higher mean power output during the trial. The 050100150200Difference in Mean Power Output of Two Routesby route, p<0.05Mean Power, in WattsDowntown Residential 111 mean difference was -9.9 watts with a 95% CI of -18 to -1.8 watts for the Downtown route, compared to Residential route.  Heart rate  The mean measurements in Figure 20 show a small difference in HR when comparing the trials on the Downtown route compared to the Residential route.    Figure 20. Box plot demonstrating the mean HR for each of the 74 (of 76 possible) trials.  The mean HR along the Residential route was 127 bpm, while the mean HR of all the Downtown trials was 122 bpm. A t-test paired by trial found the mean HR difference between the 50100150200Difference in Mean HR on Two Routesby route, p<0.05Heart Rate in beats per minuteDowntown Residential 112 two routes was small, at -4.5 bpm (with the Downtown route being lower than the Residential route). The CI for the HR was -8.6 to -0.3. HR data was missing for trials 101C and 193D due to equipment malfunction.  Cadence  Mean pedaling cadence (measured by the PowerTap) is visible in the box plot of Figure 21. The mean cadence of the cyclists along the Residential route was 54 rpm, while the mean cadence along the Downtown route was lower, at 48 rpm. Cadence data was not available for trials 101C, 190C, or 197D due to equipment malfunction.   Figure 21. Box plot of the mean cadences achieved along the Downtown and Residential routes.  The mean difference of the two routes was -5.8 rpm, with a higher cadence along the 020406080Difference in Mean Cadence on Two Routesby route, p<0.05Mean Cadence, in revolutions per minuteDowntown Residential 113 Residential compared to the Downtown route. The 95% CI for the paired t-test is -7.5 to -4.0.   3.8 Mixed effects models  In order to better assess the health impacts of the air pollution exposures, the Intake estimates, in conjunction with other available variables, were used to models the impact of the chosen air pollutant (PNC) on each of the clinical end points. The Intake estimates allow us to incorporate the  of the participant, which has be shown to be quite variable, with Intake volume estimates calculated to being between approximately 1800 to 4700L for the lowest and highest volumes measured in the study, respectively.   Using intakes to make mixed effects models for clinical health measurements  The clinical measurement data was modeled using route, the PNC or PM2.5 pollutant Intake estimate, or both route and the pollutant intake estimate together as fixed effects, in a mixed effects model, along with participant-specific random effects, as below:  Mixed effects models:  1) Change in clinical measurement = ß Route + participant   2) Change in clinical measurement = ß GM exposure + participant  3) Change in clinical measurement = ß Pollutant intake + participant    4) Change in clinical measurement = ß Route + ß Pollutant intake  + participant  Ex. Post-Pre RHI  = Post-Pre RHI  = ß Route (Dt. or Res.) + ß UFP  + participant   114 With these four model structures in mind, models with the ß -coefficients for endothelial function, spirometry, and blood biomarker end points are presented in the following tables. Statistically significant values are denoted with an asterisk (*). Note that the ß-coefficient for each air pollutant has been multiplied by the IQR of that respective pollutant, in order to present a realistic effect-estimate for the variable in each model. Table 10 provides the IQRs for each pollutant. The exercise component of the cycling trials was expected to play a substantial part in the pollution intakes to the participants, and significant differences between routes were observed in both the GM exposure concentrations (Table 9) and the Intake estimates (Table 20) for PNC. Because the activity of cycling and the variable exposures that were present along a route were important components of this study, Intake estimates were prioritized for models as they were considered to best capture the experience of the cycling trial with variable exposures combined with variable , rather than the GM exposure alone.  The first mixed effect model set (shown as Table 22) was for the clinical measures scaled to the Residential route along with the participant variable (however no ß-coefficient was shown for the participant variable). The ß-coefficient for the Downtown route was significant for the RHI measure, showing a decreased RHI for the model when cyclists used the Downtown route. Given that the ß-coefficient for the Downtown route was -0.43, and the mean RHI before all trials was 2.0 (see Table 12), the effect of cycling the Downtown route could result in a 22% mean decrease in the RHI measured % (ß of -0.43/ RHI of 2.0 = -0.215), compared to after cycling along the Residential route.   115  No other variable had a statistically significant ß-coefficient or CI. However, the CIs for IL-6 and 8-OHdG both showed a large proportion (approximately 80%) of the interval falling on the positive side of the scale, suggesting there may be some kind of effect of the Route on these two biomarkers.    116 Table 22- Mixed effects model coefficients of subclinical health measures for the Route variable, scaled to the Residential route. Outcome measurement Downtown Route  ß-coefficient 95% CI RHI -0.43* -0.78, -0.077* FEV1  (mL) 0.72 -59, 61 FVC (mL) 26 -75, 130 FEV1/FVC (%) -0.0035 -0.015, 0.0084 CRP (mg/dL) 0.0022 -0.038, 0.043 IL-6 (pg/mL) 0.99 -0.60, 2.6 8-OHdG (ng/mL) 28 -20, 76 Note: Bolded values with the presence of * indicates statistical significance, where p< 0.05.  The second group of models is shown in Table 23 and Table 24. Here the clinical measures have been modeled using the GM concentration of PNC and PM2.5 respectively, for each trial. For PNC (Table 23), the 95% CI shows a small mean increase in IL-6 and 8-OHdG, while there was a mean decrease FVC, though none of the ß-coefficient values were statistically significant.  Table 23- Mixed effects model coefficients of subclinical health measure, modeled using the GM concentration of PNC for each trial. Outcome measurement GM of PNC ß-coefficient 95% CI RHI 7.8 x 10-6 -0.24, 0.35 FEV1  (mL) -0.0010 -59, 43 FVC (mL) -0.0075 -140, 27 FEV1/FVC  9.8 x 10-7 -0.0025, 0.017 CRP (mg/dL) 8.1 x 10-7 -0.018, 0.030 IL-6 (pg/mL) 9.6 x 10-5 -0.56, 2.0 8-OHdG (ng/mL) 4.9 x 10-6 -0.0030, 0.077 Note: All ß-coefficient values were multiplied by the IQR of PNC, which was 7 600 pt/cc (found in Table 8).   117 For models using the GM of PM2.5 (Table 24), the 95% CI shows a small mean increase in IL-6 and 8-OHdG, while there was a mean decrease in FEV1 and FVC, though none of the ß-coefficient values were statistically significant.  Table 24- Mixed effects model coefficients of subclinical health measure, modeled using the GM concentration of PM2.5 for each trial. Outcome measurement GM of PM2.5 ß-coefficient 95% CI RHI -0.011 -0.25, 0.15 FEV1  (mL) -6.7 -66, 3.0 FVC (mL) -8.8 -100, 19 FEV1/FVC -0.00020 -0.0063, 0.0082 CRP (mg/dL) 0.00047 -0.014, 0.019 IL-6 (pg/mL) 0.14 -0.20, 1.5 8-OHdG (ng/mL) 0.0022 -0.019, 0.040 Note: All ß-coefficient values were multiplied by the IQR of PM2.5, which was 4.7 μg/m3 (found in Table 8). The third set of models show the clinical measures with the pollutant intake variable, and the fourth group of models includes both the pollutant intake and the route variable, with both shown in Table 25. The ß-coefficients for the models that include pollutant intake as the independent variable are found in column sections [a] and [c] of Table 25. No statistically significant coefficients were visible, except for the coefficient associated with the PNC intake for FEV1/FVC. The PNC ß-coefficient associated with FEV1/FVC however shows a very small increase (by 2.8 x 10-9 mL), making this a clinically insignificant change. IL-6 and 8-OHdG both have asymmetrically distributed 95% CIs that trend toward a positive change upon increased intakes of PNC, however again this means little on a clinical level at these intakes, due to very small ß- coefficient values (ex. 1.7 x 10-7 for IL-6). The PM2.5   ß-coefficient associated with IL-6 and 8-OHdG were not statistically significant, however the values were larger and more clinically meaningful. There was a trend of these two biomarkers increasing after larger intakes of PM2.5.  118 The ß-coefficients for the fourth model type is found in Table 25, column sections [b] and [d], show the Downtown route was again associated with statistically and clinically significant decreases in the RHI score in both of the pollutant intake models that were combined with the Route variable. In the model made for RHI that used the PNC and Route variables (found in column section [b] of Table 25], the ß-coefficient for PNC was very small and did not show a significant effect, however the Downtown route had a larger ß-coefficient compared to the model with Route alone (from Table 22), with a value of -0.49. Comparing this to the mean RHI before all trials (shown in Table 12), the effect of cycling the Downtown route could result in a 24% mean decrease in the RHI measured (ß of -0.49 divided by RHI of 2.0 = -0.25), compared to the Residential route. In the model made for RHI using the PM2.5 and Route variables (found in column section [d] of Table 25], the ß-coefficient for PM2.5 was very small and also showed little effect of this pollutant on the measured RHI, however the Downtown route also had a larger ß-coefficient compared to the model with Route alone (from Table 22), with a value of -0.44. Comparing this to the mean RHI before all trials (shown in Table 12), the model (alongside PM2.5) of the effect of cycling on the Downtown route resulted in a 22% mean decrease in the RHI measured (ß of -0.44 divided by RHI of 2.0 = -0.22, compared to the Residential route.   The ß-coefficients for the pollutants in column section [b] and [d] were not statistically significant nor clinically significant, nor did they show any near-significant trends. The PNC model containing FEV1/FVC was statistically significant, but again a very small number, making the result clinically insignificant (ß = 3.3 x 10-9).    119 Many of the ß-coefficients for the biomarkers showed a trend of increasing, as evidenced by a CI that was distributed more towards the positive side of the scale. This was found with increased intake of PNC (IL-6), increased PM2.5 (IL-6 and 8-OHdG), and with the Downtown route with both pollutant intakes (IL-6 and 8-OHdG for both pollutants, and CRP modeled with the increased PM2.5 intake). The IL-6 coefficient also shows a trend of positive change in models with the Route combined with the PM2.5 variable and with PM2.5 alone, which may suggest increased inflammation resulted from the PM2.5 intake. Increased 8-OHdG was most evident with the in models that contained the Route variable, with the 95% CI of the ß-coefficient falling predominantly in the positive range, indicating increased oxidative stress measurable by 8-OHdG following trials along the Downtown route.   120 Table 25- Beta-coefficients for PNC, PM2.5 and Route variables modeled for clinical measures.  Section [a] and [c] are in the format: RHI = ß [PNC * IQR of PNC] + participant Section [b] and [d] are in the format: ex. RHI = ß [PNC * IQR of PNC] + ß [Route D] + participant Clinical measure [a] PNC ß-coefficient (95% CI) [b] PNC ß-coefficient (95% CI) Downtown Route ß-coefficient (95% CI) [c] PM2.5 ß-coefficient (95% CI) [d] PM2.5 ß-coefficient (95% CI) Downtown Route ß-coefficient (95% CI) RHI  1.5 x 10-8 (-6.0 x 10-8, 9.1 x 10-8)  4.5 x 10-8 (-3.2 x 10-8, 1.2 x 10-7 )  -0.49 (-0.88,  -0.096)*  -0.00020 (-0.019, 0.019)  -0.00016 (-0.019, 0.018)  -0.44 (-0.80,  -0.080)*  FEV1 (mL)  6.0 x 10-6 (-7.9 x 10-6, 2.0 x 10-5)  6.3 x 10-6 (-8.6 x 10-6, 2.1 x 10-5)  -3.2 (-73, 67)  -0.11 (-3.5, 3.3)  -0.10 (-3.5, 3.3)  -13 (-79, 54)  FVC (mL)  -1.0 x 10-5 (-3.3 x 10-5, 1.3 x 10-5)  -1.4 x 10-5 (-3.8 x 10-5, 1.1 x 10-5)  53 (-61, 170)  0.48 (-5.4, 6.4)  0.47 (-5.5, 6.4)  17 (-91, 130)  FEV1/FVC (%)  2.8 x 10-9 (1.8 x 10-11, 5.5 x 10-9)*  3.3 x 10-9 (4.1 x 10-10, 6.2 x 10-9)*  -0.0078 (-0.021, 0.0057)  2.2 x 10-4 (-4.9 x 10-4, 9.4 x 10-4)  2.3 x 10-4 (-4.9 x 10-4, 9.5 x 10-4)  -0.0049 (-0.018, 0.0080)  CRP (mg/dL)  5.1 x 10-10 (-8.0 x 10-9, 9.0 x 10-9)  3.9 x 10-10 (-8.6 x 10-9, 9.4 x 10-9)  0.0020 (-0.044, 0.048)  8.0 x 10-5 (-1.5 x 10-3, 1.7 x 10-3)  6.5 x 10-5 (-1.5 x 10-3, 1.7 x 10-3)  0.018 (-0.013, 0.048)  IL-6 (pg/mL) 1.7 x 10-7 (-1.6 x 10-7, 5.1 x 10-7)  1.3 x 10-7 (-2.2 x 10-7, 4.7 x 10-7)  0.87 (-0.84, 2.6)  0.072 (-0.0098, 0.15)  0.072 (-0.010, 0.15)  1.1 (-0.58, 2.7)  8-OHdG (ng/mL) 4.1 x 10-6 (-7.9 x 10-6, 1.6 x 10-5) 8.7 x 10-7 (-1.2 x 10-5, 1.4 x 10-5) 29 (-27, 86) 0.84 (-1.8, 3.5) 0.80 (-1.8, 3.4) 34 (-14, 86)  121 Note 1: Bolded values with the presence of * indicates statistical significance, where p< 0.05.  Note 2: All ß-coefficient values associated with a pollutant were multiplied by the IQR of the respective IQR for either PNC (7 600 pt/cc) or PM2.5 (4.7 μg/m3) (found in Table 8).     122 3.9 Testing effect modification of significant models using stratified analysis  Mixed effects models of those clinical measurements that resulted in statistically significant ß - coefficients were tested for effect modification.   Effect modification of RHI by sex, BMI, and age with Route  In Table 26, RHI was modeled with the additional variables of male or female sex, low or high BMI (separated at the median), and younger or older age (separated at the median). These variables were tested as effect modifiers in combination with the Route variable, with the Route variable being scaled to the Residential route.  The ß-coefficient for the females was statistically significant, though both sexes had RHI ß-coefficients that predominantly fell on the negative side. The specific ß-coefficient for the females shows a coefficient that was approximately 2.5 times larger (-0.81) than the coefficient for the males (-0.29), and the low value for the CI (-1.6) for the females was also about 2.5 times larger than the range of the males (-0.66). As the effect estimates for females indicated a larger decrease in comparison to males, suggesting that females experienced a greater impact of the Downtown route on changes in RHI compared to males.  123 Table 26- RHI modeled with Downtown Route and possible effect modifiers.  Route D ß -coefficient with sex, BMI, and age modeled separately.  In the form:  RHI = Route, with the Route variable modeled by separating the variable by sex (by male or female), BMI at the median, or age at the median. Variable ß-coefficient Downtown Route 95% CI Route -0.43* -0.78, -0.077* Route + males -0.29 -0.66, 0.082 Route + females -0.81* -1.6, -0.0048* Route + low BMI -0.36 -0.85, 0.13 Route + high BMI -0.50 -1.0, 0.0072 Route + Younger Age -0.11 -0.50, 0.28 Route + Older Age -0.75* -1.3, -0.18* Note 1: Bolded values with the presence of * indicates statistical significance, where p< 0.05. Note 2: BMI and age variables are split at the median to define low/high and younger/older, respectively.  When BMI was split at the median into a high-BMI or low-BMI group, there were no statistically significant differences in the ß-coefficients. However, those participants in the higher half of the BMI measures appeared to experience a greater negative impact in RHI from the Downtown route exposure.   For the younger and older study participants, the ß -coefficient for the Downtown route was not statistically significant when only the younger half of the participants were included in the model, but the coefficient was significant, at -0.75, when the older half of the participants were included, suggesting the effects of the Downtown route on the RHI of the older participants  124 may be disproportionately influencing this measure. This suggests that the older participants experienced greater negative impacts to their RHI.   Figure 22. Boxplots showing effect modification of RHI route values by sex, BMI, and age. Note 1: Centre black horizontal lines indicate median values, boxes indicate 25th and 75th percentiles (the interquartile range), and whiskers show 95% confidence intervals. Red stars indicate statistically significant ß-coefficient estimates. Exact data values are indicated in Table 26.    125 Effect modification of FEV1/FVC by sex, BMI, and age with PNC intake  Table 27 shows the ß-coefficient for PNC intake was statistically significant when the Route variable was included in the model for FEV1/FVC. While this value was small (3.6 x 10-9), we will nonetheless consider effect modification by the independent variables.   Table 27- FEV1/FVC modeled with Downtown Route and possible effect modifiers. PNC Intake ß -coefficient with sex, BMI, and age modeled separately.  In the form:  FEV1/FVC = Route + PNC, with the Route variable modeled by separating the variable by sex (by male or female), BMI at the median, or age at the median. Variable ß-coefficient Downtown Route 95% CI Route 3.6 x 10-13 -4.4 x 10-15, 7.2 x 10-13 Route + males 2.9 x 10-13 -1.4 x 10-13, 7.2 x 10-13 Route + females 5.2 x 10-13 -1.6 x 10-13, 1.2 x 10-12 Route + low BMI 4.4 x10-13 -1.0 x10-13, 9.8 x10-13 Route + high BMI 2.8 x10-13 -2.0 x10-13, 7.5 x10-13 Route + Younger Age 5.9 x 10-13 1.4 x 10-13, 1.0 x 10-12* Route + Older Age 9.7 x 10-14 -4.6 x 10-13, 6.6 x 10-13 Note 1: Bolded values with the presence of * indicates statistical significance, where p< 0.05. Note 2: BMI and age variables are split at the median to define low/high and younger/older, respectively.   A potential effect modifier includes the sex variable included in a model with PNC intake. When males and females were modeled separately, the male participants appeared to have a greater increase in FEV1/FVC. Participants having a lower BMI also appeared to have a larger impact in FEV1/FVC compared to the higher BMI group, though neither of the groups had a significant CI. When age separated at the median was compared, the younger participants showed a significant  126 increase in FEV1/FVC when cycling on the Downtown route, though by a very small margin in terms of clinical significance.   127 Chapter 4: Discussion Results summary  This study assessed PM air pollution exposures and clinical health measures from 38 healthy young adults along two cycling routes in Vancouver. Detailed information of heart rate combined with bicycle computer information allowed air pollution intake estimation for each cyclist for each 1-hour trial. The mean ratio for  while cycling compared to at rest was 3.0. GM exposure of PNC was significantly higher along the Downtown route (16 226 pt/cc) compared to the Residential route (10 011 pt/cc), with a mean PNC Intake that was also significantly higher along the Downtown compared to the Residential route, however no other PM size classes were significantly different along the two routes. Relative to the Residential route, the Downtown route mixed effect linear models showed a mean 22% decrease (ß = -0.43) in RHI, but no association was seen with pollution intakes.   128 4.1 PNC and PM exposures along the two routes  The hypothesis was that higher PNC would be found along the Downtown route when compared to the Residential route, due to larger vehicle volumes and the closer proximity to cars along the road. An earlier study of cyclist pollution exposures conducted in Vancouver (100) identified high traffic locations such as a highway-classified road and a bridge during rush hour as being associated with peak PNC concentrations. Overall the mean PNC along the single route used for this 2008 study was 56 417 pt/cc. Results from the current study showed pollutant measurements that were significantly different on the Downtown route compared to the Residential route, however the reported mean PNC concentration in the 2008 study was nearly three and a half times higher than the data we collected along the two routes we studied. Higher pollutant exposures may have been avoided in those areas susceptible to peak hour congestion, such as over the Burrard Bridge, as trials were typically conducted at 09:30-10:30, or 14:00-15:00, rather than 07:00-09:00 as what occurred in the 2008 study. Due to modified infrastructure in between the 2008 study and the current study, Downtown trials occurred on some cycling lanes that are now separated by concrete partitions and additional distance from car traffic, compared to the painted lanes experienced by Thai et al. that provided little separation between the cyclist and the traffic in the adjacent lane. (100) Further, improvements in local air quality since that study was completed may have reduced the observed exposures in this new study.   Comparable short-term peaks in PNC were visible in time series plots along both routes, but the PNC difference between the Downtown and Residential route was statistically significant (p = 4.0 x 10-6), with a 95% CI of predicted values between 3 601 pt/cc and 8 355 pt/cc. This amounted to a Downtown: Residential ratio of GMs of 1.7. This ratio is relatively consistent with  129 the finding significant differences between the PNC levels along high- and low-traffic routes in other similar studies in Berkeley (high/low ratio of mean PNC concentration = 1.3), (76)  Arnhem (1.2), (75,89) Utrecht (1.6) and Ottawa (1.8). (116–118)   In contrast to PNC levels, the mean PM2.5 exposures that were measured in this study were not significantly different between the two routes (Downtown GM = 6.0 μg/m3, Residential GM= 4.7 μg/m3) after excluding a single atypical measurement session that occurred when forest fire smoke was present at high concentrations over the City of Vancouver in August 2010. A Berkeley study also found no significant difference in PM2.5 exposures between high- and low-traffic routes, (76) which also agreed with the findings of the homogeneous tendency of PM2.5 concentrations in the previous Vancouver study by Thai et al., (100) however this was contrary to findings in other studies in Ottawa, (116) London (UK) and Dublin, which found a significant difference between the mean PM2.5 exposures along high and low traffic route at ratios of 1.5, 1.6- 2.8, and 2, respectively. (84,92,233) Arithmetic mean PM2.5 concentrations measured while cycling in other geographic locations were substantially higher, including in London (23.5- 34.5 µg/m3), Dublin (75.60 – 110.20 µg/m3), eleven Dutch cities (overall mean = 44.5 µg/m3), and Beijing (49.10 µg/m3), demonstrating the potential for PM2.5 intakes elsewhere to greatly exceed the intakes experienced by the participants in this Vancouver study. (84,87,92,99) Due to these sometimes-large contrasts between the air pollution concentrations in other locales and the exposures observed in this study, one must not extrapolate pollution concentrations along different route types as infrastructure may vary. As well, one should not assume the same the findings with respect to health effects would be observed in other locations.   130 Looking at cycling studies with low and high air pollution exposures, our study’s “high” PNC concentration was quite low compared to the high-exposure conditions of other studies, such as one Dutch study that had a mean PNC of 27 028 and 41 097 on the low-traffic and high-traffic routes, respectively. (118) Mean PNC concentrations in a 2010 lab-based filtered-air condition compared to a road test condition provided an exposure contrast of the mean values 496 pt/cc and 28 867 pt/cc, however despite how there was a 58-fold difference between the low and high exposure conditions, only one of the markers of inflammation (percentage of blood neutrophils) showed a significant change between the two conditions on 28-58 year old adults. (119)  4.2 Minute ventilation and intake parameters  The  our participants experienced while cycling varied widely, but we calculated an overall  ratio of 3.0 while cycling compared to at rest. This is a higher  ratio than what most studies found when measuring commuters in sedentary situations as either car or bus passengers (1.8 – 2.6 : 1). (80–82)  Males from this study had a mean predicted  while cycling of 48.07 (SD= 15.43) L/min (data from both routes), while female cyclists had mean  values of 40.69 (SD= 10.67) L/min over both routes. Cycling studies in other urban environments have recorded mean  measurements of 22.0L/min for males, and 27.6 L/min for females in one Dutch study (82), and  59.1 L/min and 46.2 L/min for males and females respectively, in a Belgian study. (83) These studies demonstrate that  can vary widely, particularly as it is highly associated with the effort level, fitness level, and terrain experienced by the cyclist.    131 The Belgian study took their data a step further, calculating the total volume of air inhaled during the approximately 15 to 21 minute long rides. (83) Their study found that male participants inhaled a mean 924.8 L of air, and females inhaled a mean 801.4 L of air over the duration of the ride. (83) Data from the present study produced an estimated mean total volume inhaled of 3 005.3 L and 2 536.0 L for males and females, respectively, but it must be emphasized this is an estimate as regression equations were correlated with HR and used to calculate the total air intake volume for each trial of the present study. The Belgian study used a Metamax© device, enabling them to take much more precise recordings of air volume breathed, however if their data were to be multiplied by a factor of three or four to obtain the value for cycling for one hour, it seems the estimated volumes for this trial are in a similar range.  4.3 Endothelial function using RHI, and biomarkers to compare two routes A mean increase of 0.25 (95% CI = 0.03, 0.47) was observed for RHI after cycling along the Residential route, when evaluated with paired t-tests, but there was no significant difference when comparing the mean change for the Downtown trials (mean = -0.18, 95%CI = -0.46, 0.11). The mean increase in RHI observed after cycling along the Residential route suggested an acute benefit to endothelial function following a one-hour bout of cycling in a recreational or commuter cycling setting. Further, because there was no significant mean change following the cycling trials along the Downtown route, there was no indication that these trials produced any acute negative health effects at the pollution concentrations and conditions experienced in this study. In fact improvements in RHI following rides along the Downtown route were within the 95% CI. The Route type was the only variable found to be associated with RHI using mixed effects models (Downtown or Residential route, mean difference in RHI of -0.38, 95% CI of -0.75 to -0.02). This  132 could suggest that the benefits to RHI are more so on the Residential route, however it is also plausible that these results were related to a higher level of exercise experienced on the Residential route, implied by the higher recorded mean cadence, power output, and mean heart rates on this route. The changes observed to RHI between the two routes are therefore not explained by the air pollution Intake or exposure alone.   Linear mixed-effects models found ß- coefficients for the Downtown route in the range of -0.43 to -0.49 when modeled alone or with PNC or PM2.5 intakes, however no significant associations were seen with the ß- coefficients for pollution intakes. In this group of participants, the mean pre-trial RHI was 2.0, which was a normal RHI value according to Itamar Medical’s EndoScore, the EndoPAT™ risk score calculator which states that normal RHI values are > 1.67. (234) A lower RHI (by a mean of 22 to 25%) was found to be associated with cycling along the Downtown route (from the pre-cycling measurements), and this may lower the RHI to EndoScores between 1.51 and 1.57. These values fall in the abnormal range of EndoScore values that are designated by Itamar Medical as being associated with increased risk of a cardiovascular event (≤1.67). (234) Keeping in mind that these ß- coefficients were estimates, the 95% CI for these route ß- coefficients were between -0.077 and -0.88; subtracted from the mean pre-trial RHI of 2.0, this results in an RHI of between 1.12 and 1.92 after cycling along the Downtown route. The mean falls in the abnormal range, however the wide 95% CI falls within both the normal and abnormal range of EndoScores, making it inappropriate to conclude that the mean RHI would necessarily fall to abnormal levels immediately following a cycling trial along the Downtown route. If the mean was a good approximation of the true acute change that is experienced in risk of coronary heart disease risk, then the higher risk is similar to adding the diabetic condition (ß = 0.428) to total cholesterol  133 equations for the Framingham 10-year risk for coronary heart disease in a male, according to the underlying predictive values acquired from the Framingham Heart Study. (235) The Framingham 10-year risk characterizes one’s risk of an acute cardiovascular event over the next 10 years using information about cholesterol, blood pressure, age, sex, and smoking status. (235) A person in the lowest risk category for each predictor type may have a one percent 10 year risk of developing an acute cardiovascular event, while someone in the highest risk category for most predictors may have upwards of a 50% risk. (235)   One needs to be extremely cautious in interpreting the data in this way however, as the protocol of this study did not measure endothelial function at multiple time points following the conclusion of the bicycle ride, meaning that there is no indication of the duration or reversibility of the measured changes in endothelial function.  Few other air pollution exposure studies have measured RHI using the Endo-PAT, particularly in young healthy individuals aged 18-40 years, (171,236,237) and those that have did not observe associations with ambient levels of PNC. (171) In a lab-based study of young adults by Bräuner et al., there was a nearly 10-fold difference between the exposure of the clean (filtered, 1 270 pt/cc) air condition and the polluted air (unfiltered exposure, 10 067 pt/cc), but no significant RHI change observed. This may be related to the relatively low concentration of the polluted condition, as Bräuner et al.’s high exposure condition was only half the concentration measured in another European study, located at a nearby outdoor monitoring station in Copenhagen adjacent to a busy street (median = 22 809 pt/cc). (171) While one study in young healthy individuals Endo-PAT demonstrated significant decreased endothelial function 12 hours after second-hand smoke  134 exposure, (236)  it is possible that other components of cigarette smoke (due to the unique chemicals and properties associated with cigarette smoke compared to TRAP), may produce divergent health effects in populations of different ages or other health conditions. Healthy middle-aged and elderly participants in their homes experienced increased endothelial function upon improved air quality (mean reduction of 2 000 pt/cc) in their bedroom when HEPA filtration was used; (172) this was consistent with RHI results from an elderly population, with improved endothelial function (increased 8.1%)  in filtered conditions compared to unfiltered air in their homes. (129)  4.4 Evaluation of blood biomarkers and relationship with RHI No statistically significant results were seen in the changes in biomarkers, either depending on route or PNC intake. Levels of IL-6 and 8-OHdG generally changed in the positive direction after cycling trials along the Downtown route (but not on the Residential route), however this difference could not be attributed to PNC intake. There were numerous cases of improvements in clinical measures from the trials along the Downtown route, where the measures showed a change in the beneficial direction. RHI increased in 11/37 trials, and CRP, IL-6, and 8-OHdG decreased after the cycling trial in 13/38, 18/33, and 11/29 instances, respectively.  Though few studies have evaluated air pollution effects by comparing endothelial function along with these biomarkers, those that have done so have produced mixed results. Most studies have shown non-significant changes to CRP, (170,171) IL-6, (85,119,129,170,171) and 8-OHdG (164) when compared at different PM exposures, and only a few reported significant RHI differences in young to middle-aged healthy participants (p = 0.03) (170) and healthy older participants (p = 0.04), (129) while others reported no RHI difference in the young and healthy  135 (238) or elderly groups. (172) In many of these studies, the findings were nearly significant, including an experiment by Bräuner et al. where the healthy young participants were exposed to outdoor air (mean PNC = 12 200) and filtered air conditions (mean PNC = 1270), experiencing a decrease of nearly 10% in CRP when exposed to the filtered condition. (171) For this measure the percent change was -9.97, with a 95% CI of (-19.7, 1.01), showing a large proportion of results are likely to fall in the negative range; the RHI value showed only a comparatively non-significant change (percent change of 1.11%, 95% CI -3.92, 7.25), nor was there anything notable with IL-6 measures (-0.20%, 95% CI -12.2, 13.9). (171) In a wood-smoke exposure study by Allen et al., significant RHI results were found alongside borderline significant CRP results (p = 0.06). (170)  It was expected that the blood biomarkers that would be most likely to show a change were those that would respond quickly (within one hour, given the protocol length) to the presence of inflammatory substances and reactive oxygen species. While no blood biomarker was significantly increased after the cycling trials on either route, both IL-6 and 8-OHdG increased at a non-significant level, which may suggest that a second blood test a few hours later may have provided further insight into the direction and extent of any inflammatory response. Because CRP is produced in the liver only after the IL-6 response reaches a systemic level, (156) any increases in CRP would not likely be seen within one hour of the end of the cycling trial. Increased levels of 8-OHdG should be measurable on a systemic level shortly after increases in oxidative stress, however previous studies finding significant increases in this biomarker had post-measurements collected at ‘end of shift’ and ‘end of day’ scenarios, (164,211,215) leaving substantially more than one hour to  for a response to develop.   136 Changes to endothelial function were expected to result from the contributions of increases in inflammatory mediators (which include IL-6 and CRP), as well as increases in reactive oxygen species within the systemic circulation, among other pathways hypothesized by indicated by Brook et al. (151) As there are some indications that the health measures obtained post-trial are not attributable to air pollution concentration, it is difficult to conclude that the increases in the blood biomarkers contributed to any non-significant reductions to endothelial function, particularly as the cycling trial and exposure included exercise, which may be acutely improved by a bout of physical activity. (239).  Finally, decreases in endothelial function could result from increases in inflammatory biomarkers present in the blood after the cycling trial, however physical activity also shows benefits to the cardiovascular system. (216) In this healthy and young population, reduced endothelial function may be measurable only if there is a sufficiently high air pollution dose to outweigh the benefit of the bout of exercise.   Common effect modifiers identified in these studies include age, sex, and BMI. Older males are shown to have lower RHI measurement compared to older women, and individuals with metabolic syndrome (which is related to body composition and BMI) are also characterized with having decreased endothelial function as measured by RHI. (206) Most studies measuring endothelial function are aware of the sensitive measurement, and appear to adjust for these variables in their analysis. (129,170,171,237,238) The present study observed a greater decrease in RHI measurements in females, higher BMI participants, and those that were older. Larger increases in FEV1/FVC ratio were also observed in males, lower BMI participants, and those participants  137 that were younger, though FEV1/FVC ratio has been shown to be dependent on age due to a mathematical relationship. (240)   Sympathetic nervous system activation may explain the results With the mixed results related to exposures and changes in endothelial function, it is difficult to confidently attribute changes seen in the present results to the PNC exposures of the cyclists. Given that Route was the only variable that was associated with RHI in the mixed effects models, other variables that may be associated with the Route variable should be considered. Noticeable differences between the Downtown and Residential routes included the volume of traffic, which may increase stress or anxiety in those participants who are less comfortable cycling around higher traffic volumes.   On the Downtown route there are taller buildings surrounded by mostly pavement, with fewer green spaces where trees and grass are found. The presence of green space, experiences in nature, or even pleasant photographs of rural scenes has been found to reduce both psychological and physical indicators of stress, (241) with increased levels of self-esteem, (242,243) improved mood, (242–244) reduced blood pressures, (242,244) reduced HR, (245) and decreased levels of salivary cortisol (a hormone present at higher levels under stressful conditions). (242,244,245) In one study, those with the highest BP and cortisol levels received the most benefit by participating in a forest walking activity. (244) Additionally, health measurements while forest viewing or walking compared to walking and viewing in an urban area found increased high frequency components of HRV (a signal that the participant was relaxed), and increased parasympathetic nerve activity— the “rest and digest” activity which counteracts the “fight or flight” of the  138 sympathetic nervous system, (246) both while viewing and walking in a forest environment. (245) While our study did not test for these physical indicators of stress during the cycling activity, nor was it within the scope of the present study to evaluate differences in green space, it is plausible that psychological stressors may have been present due to the quantity of green space along the bicycle routes. Therefore the presence or lack or green space may have resulted in physiological impacts to the cyclists, including changes to the biomarkers associated with stress.   Other biomarkers that we evaluated may be impacted be stress as well. In one study, participants exposed to stressful sedentary behavioral tasks displayed a 56% increase in IL-6 measurements two hours after completing the tasks compared to controls; CRP in this case was not statistically significant. (247) This study also found significant mean changes in BP and HR during the tasks, but these returned to normal once completing the stressful tasks. (247) While we did not measure BP during the ride, the IL-6 measurements between the two routes apparently showed a trend, but were not statistically significant, nor were the changes we saw equal in magnitude after cycling the Downtown route (mean IL-6 increase of approximately 15%). The changes we observed for IL-6 were possibly influenced by the stressful aspect of riding Downtown, with a differing magnitude related to the fact that the results from the aforementioned study had a small group of participants with a mean age of 40.9 years.   Our participants would have had a different biomarker response due to the one-hour cycling exercise activity, as exercise is shown to increase blood levels of IL-6. (154,155) Further, there is evidence that stress-associated biomarkers may respond in dissimilar ways due to effect modification by fitness level. (248,249) In one British study, participants classified as being the  139 most fit had less pronounced reductions in HRV during the stressful tasks, while the participants in the lowest tertile for fitness had higher IL-6 responses to stress (p = 0.002). (248) High fitness levels benefitted individuals in exposures to psychological stressors, where a number of examples shows participants experienced decreased systolic BP and HR reactivity, and a faster recovery in HR. (250,251) Respiratory rate, HR, and  measurements were also significantly elevated when cycling ergometer exercise was combined with a concurrent mental challenge, compared to the cycling trial only. (252)  Finally, less green space and large volumes of automobile traffic may also signal the possibility of more noise on the Downtown route. People exposed to high levels of noise exhibit similar increases in sympathetic nervous system activity as those undergoing increased stress, resulting in elevated stress hormones including cortisol and increases in BP and HR along with blood components such as blood clotting factors and blood lipids. (253) These physiological responses to stress are also associated with cardiovascular disease risks and long-term chronic health disorders. (253–255) It is difficult to ascertain the amount of noise that each cyclist could have been exposed to without having a personal monitoring device present for each test, however one 2009 cycling experiment by Boogaard et al. did monitor the noise exposures to cyclists in eleven Dutch cities, finding one-minute mean noise levels ranged between 63-66 dB(A) during the short 10-20 minute rides. (87) A 2011 World Health Organization report discussing the burden of disease of environmental noise in Europe estimated that mean daytime sound pressure levels in excess of 60 db(A) produced a slightly increase risk of ischemic heart diseases (256), which is within the range of measurement acquire in the study by Boogaard et al.. The noise literature suggests that cardiovascular and stress effects related to noise are possible in an urban cycling  140 environment, (87,249,254,257) so the possible impacts on our physiological measures (RHI, BP, HR, and blood biomarkers) must not be dismissed. Overall, there are a number of more stressful components along the Downtown cycling route that may add up to a significant difference in RHI, possibly to the level we saw.  4.5 Strengths of this study  This study design included a number of strengths that increase confidence in the interpretation of results, specifically, the single-blind randomized crossover design minimized the impact of between-participant differences as each participant acted as his or her own control. Further, prospective participants were screened in order to exclude those with health conditions known to be associated with unpredictable impacts to measured RHI, such as medications that may change blood circulation, or known vascular conditions such as Raynaud’s phenomenon. The design included a minimum of two weeks (max. 8 weeks) between each of the trials, therefore reducing the likelihood of any carryover in exposure and subsequent impacts from the previous trial, and also limiting any effects that could have resulted from improved physical fitness training that could have occurred between the two trials. Participants experienced real life air pollution exposures on actual cycling routes in an urban environment.  The participants had varied backgrounds as cyclists, both in their frequency of participation as well as their fitness level, making this study a good example of authentic participants in this age group.    141 The individual participants had varying diets and activity levels. Participants were asked to continue with their typical diets and habits during the periods in between the two cycling trial pre-test preparation periods. While it is possible there may have been exposures, supplements, or other factors that were not divulged to research staff at the time of the test, the majority of participants appeared dedicated to this project, and seemed to wish to contribute as accurate of data as possible.   4.6 Limitations  Despite these strengths, as in all studies with an observational component this study had a number of important limitations. As we measured acute changes to clinical measures, it was important to limit the impacts of known sources of exposure that may have short-term effects. Each individual was therefore asked to use the same transportation to arrive to the testing session on both testing occasions to limit these potential variations. However, day-to-day living would have exposed participants to air pollutants, and these exposures would have varied for each individual, even when using the same mode of transportation to arrive at the test site.  Participants were also asked to repeat their morning meals prior to that day’s test and to follow other dietary restrictions such as limiting caffeine intake, as these can introduce unpredictable outcomes to the physiological measures.   The two routes that were selected had important differences in the infrastructure and terrain, which likely contributed to differences seen in the mean cadence, power output, and mean heart rates, with each being significantly higher along the Residential route in comparison to the Downtown route. As a result of these factors, cyclists may have experienced differences in exercise levels because of the more continuous nature of the Residential route (with fewer stops due to red  142 lights). It is possible that improvements in RHI were related to higher volumes of physical activity along the Residential route, however a more detailed analysis of heart rate and frequency of stops for each trial would be required to shed light on the causes of these variances.    The testing environment indoors was not monitored to ensure consistency, such as for air pollution levels or noise. The environment during the clinical tests had a very real potential to influence results, in particular those that may respond instantaneously such as BP or HR. These responses could be due to interruptions in the room and sudden loud noises, situations that may cause any level of stress such as anxiety about a test, or a frustrating or stressful experience in traffic before or during the trial. While efforts were made to ensure the participants were as comfortable and calm as possible before conducting the clinical measurements, there were a few interruptions that could not be avoided or controlled.  Specific limitations related to the protocol include that we did not measure each participant using the stepwise cycling test, only introducing this more detailed protocol after a number of participants had completed both trials. The respirometer was also found to be restrictive to the breathing of participants at very high  levels. This kept us from measuring the  values at the highest HR measurements during the indoor Velotron stepwise cycling test for some participants. As HR was measured to indirectly predict , this study data was based on a method not validated to measure  during physical exertion which is unlike the Metamax® method used by other studies. (81,83) Further, the air pollution monitors may have recorded concentrations that are less accurate because of the orientation of the intake hoses that were mounted on the handlebars. Proper sampling of air particles requires that isokinetic sampling conditions be used to ensure that particle  143 concentration representing the actual concentration in air is measured, however this requires that the velocity of the air in the sampling tube is equal to the velocity of the air that is being sampled. Because the sampling tubes were facing forward and the speed of the cyclists varied, it is possible that some particle sizes may not have been captured at their true concentration, leading to a sample that may be biased towards smaller or larger particles at different speeds. (258,259) Despite this potential source of error, the impact of sampling orientation would be present during all trials, making it unlikely that the overall study findings would be impacted.   Given that the cycling trials were outdoors, we were unable to control meteorological variables such as temperature or wind. Unpleasant cycling conditions (such as rain or wind) may have reduced cycling speed, therefore modifying the average  of the participant while riding. Some participants also opted out of cycling in very wet or unpleasant conditions, resulting in a tendency for more tests during fair weather days, with few examples of cycling in adverse conditions. Temperature, humidity, and pollutant characteristics may impact particle measurements from the air pollution monitors; for this reason correction factors are often applied to help reduce error in the instrument data, (260,261) however this study did not collect such information and the pollution data was not altered. As well, environmental noise level was not controlled, nor was it measured. This means temperatures or physical stressors could have potentially influenced the clinical measurements on test days with dissimilar environmental conditions.   144 4.7 Generalizability of these results Due to the specific nature of cycling as an activity, only those individuals interested in cycling with interests in the topic of air pollution chose to participate. This leaves the possibility that the results of this study may not be applicable to people who are more sedentary or unable to cycle, to people outside of the study age group of 19-39 years, or to people who have a chronic health condition that excluded them from participating in this study.    Because only 10 female cyclists completed both cycling trials, the pool of participants provides limited power to assess the effects of air pollution on the female sex alone. Therefore, the study results should be interpreted more cautiously when discussing the possible effects seen in the female population.  4.8 Implications of this research, with public health consequences This study demonstrates how acute improvements in RHI may be seen after cycling along a Residential traffic route, but this was not attributable to any differences in PM air pollution intake, as evaluated by the mixed effects models for RHI. With multiple cases of improvements in the RHI and biomarkers after the Downtown trials, this shows that cycling in general, but also along a route away from busy traffic may provide benefits to fitness, and no known drawbacks to one’s cardiovascular and respiratory health in apparently healthy young adults.   Utilitarian cycling is a way to incorporate regular PA into one’s routine, however this research study shows on a limited scale that there may be negative aspects to cycling along some routes in a city. The benefits of regular PA are numerous and widely accepted, however it is  145 important to determine if the health costs of air pollution to cyclists negates any advantages from participation in the activity. The economic, environmental, and health benefits of participation in cycling have been contrasted in the literature. (58–60,262–268) Larger societal benefits in the form of deaths avoided from air pollution (58–60) and population health benefits related to PA gains (58,59,265,266) are observed by most groups, however many acknowledge there to be a small increase in mortality impacts related to increased air pollution or accidents. (269)   One study by Rojas-Rueda et al. analyzed the effects of increased participation in cycling by the population of Barcelona, upon the initial commencement period (March 2007 – August 2009) of a bike-sharing initiative called Bicing. (59) The authors included mortality changes from increased PA, air pollution effects to cyclists as well as to city residents, and road traffic accidents. The estimated changes resulted in a calculated net benefit of 12.28 deaths avoided, with 0.13 and 0.03 deaths attributed air pollution exposure and road accidents, respectively, in addition to a reduction in over nine million kilograms of carbon dioxide emissions. (59) Similar calculations quantifying the effects of a modal shift for 500 000 people using bicycles for short trips in the Netherlands found gains from PA of 3-14 months, 0.8- 40 days lost from air pollution, and 5-9 days lost from traffic accidents. (59)  The most local North American example of this type of assessment was done for 11 metropolitan areas in upper Midwestern US states, where the authors simulated the elimination of round-trips in car ≤ 8 km in length for the region of 31.3 million people. (58) This study found this aggressive intervention would result in small decreases in the levels of PM2.5 on a local and regional scale, while mortality would be reduced by approximately 1295 deaths annually (95% CI: 912,  146 1636) and costs from net health benefits would exceed $4.94 billion per year (95% CI: $0.2 billion, $13.5 billion). (58)  4.9 Recommendations for future research One aspect of this study that poses more questions is that the RHI of the cyclists was negatively impacted on the Downtown cycling route, but this was not clearly attributed to the concentration of any of the measured air pollutants. This may be related to an unknown factor along the specific routes selected, related to a stress-provoking variable such as those discussed in section 4.2, or it may be due to an unmeasured air pollutant that was not well correlated with UFP but nonetheless was present at a higher quantity along the Downtown route. Factors that could not be objectively evaluated during the cycling trials include the presence of noise or stressful situations (such as near collisions). A questionnaire could also be administered after the trial to assess stressors that may have been experienced such as any close calls with traffic or other roads users, awareness of loud noises such as horns and questions relating to perceived or actual danger from situations or infrastructure. Using a camera on the bicycle to measure interactions with other road users, monitoring environmental noise, and evaluating the pollution exposures of cyclists as they travel through the different types of cycling infrastructure are all reasonably simple ways this study could be expanded. Boogaard et al. (2009) measured air pollutants and noise while cycling in 11 Dutch cities, finding moderate correlations between noise and PNC (median r = 0.34). (87) A further recommendation would be to monitor air pollutant concentrations, temperature, and noise for the duration of the time the participant is undergoing the testing to ensure the pre- and post- trial environments are consistent.    147 The effects on the endothelial function (RHI) of participants were seen within one hour of the cycling trials, however no known studies have investigated the long term clinical health impacts to regular bike commuters. It would be valuable to the present cycling population to understand the risks about the possibly cumulative effects of one’s daily commute over a longer period of time than just a single day, if it exists.   Long-term plans for research in this area should focus on making cycling as beneficial as possible to the health of its participants, by ensuring negative health impacts are minimized.   4.10 Conclusion  All of the study objectives outlined in the introduction have been attained, with the successful collection and analysis of PM exposure data and health measures from 38 cyclists along the two specified Downtown and Residential routes. PNC air pollution measurements were on average higher along the Downtown route (GM = 16 226 pt/cc, GSD = 1.33) compared to the Residential route (10 011 pt/cc, GSD =1.88), however 29% of participants experienced higher GM PNC, and more variability was seen along the Residential route. PM2.5 concentrations were not significantly different between the two routes. Intake estimates along the Downtown route were on average 45% higher (4.54 x 1010 particles) than the intake estimates from the Residential route trials (3.13 x 1010 particles).  RHI increased significantly after the Residential trial (mean improvement = 0.25, 95% CI 0.03 to 0.47), but not the Downtown trial (using a paired t-test, p = 0.04). The Route type was the only variable found to be associated with RHI using mixed effects models, with a mean difference  148 in RHI between the Downtown relative to the Residential route of -0.38 RHI units [-22%], 95% CI of -0.75 to -0.02). Exhibiting a possible trend, IL-6 (indicating inflammation) and 8-OHdG (indicating oxidative stress) increased, but not significantly, after the Downtown cycling trials compared to the Residential trials. The changes observed to RHI due to the cycling Route could not be explained by the air pollution exposure or estimated intake alone, suggesting other variables impacting RHI should be considered. On the Residential route, cyclists experienced a higher mean HR, higher mean power output, and higher mean cadence, but other variables such as noise, stress, green space, and fitness level should be considered. Increased exercise intensity along the Residential route could have resulted in benefits that could not be attained along the Downtown route in its current state of infrastructure and configuration.  This study contributes to the area of cycling research by using a crossover study to evaluate acute clinical measurements in a group of cyclists in a real life scenario. With the individual benefits of improving physical fitness and the larger rewards of reducing health care costs associated with improved fitness and air quality, it advisable to support cycling in the conditions described along either of these route types. Both provide the benefit of exercise, and neither conclusively demonstrates harm to subclinical health end points that were measured. With regard to heterogeneity between route types, there may be advantages to cycling along routes that require additional effort or that have lower air pollution levels. To achieve the maximum benefit, we should seek to better understand the variables along the Residential route that may have provided additional benefit to endothelial function in this group of participants, and consider encouraging cyclists to choose routes providing the most beneficial features, all while continuing to support air quality improvements that further the health of the most vulnerable road users.   149 References 1.  Warburton DER, Nicol CW, Bredin SSD. Health benefits of physical activity: the evidence. CMAJ [Internet]. 2006 Mar 14;174(6):801–9. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1402378&tool=pmcentrez&rendertype=abstract 2.  Physical Activity Guidelines Advisory Committee. Physical Activity Guidelines Advisory Committee Report, 2008 [Internet]. Washington, DC.; 2008. Available from: http://www.health.gov/paguidelines/Report/pdf/CommitteeReport.pdf 3.  Kannel W, Sorlie P. Some Health Benefits of Physical Activity: The Framingham Study. Arch Intern Med [Internet]. 1979 [cited 2014 May 23];139(8):857–61. Available from: http://archinte.ama-assn.org/cgi/content/abstract/139/8/857 4.  Haskell WL, Lee I-M, Pate RR, Powell KE, Blair SN, Franklin BA, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc [Internet]. 2007 Aug [cited 2014 Apr 30];39(8):1423–34. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17762377 5.  Andersen LB, Schnohr P, Schroll M, Hein HO. All-cause mortality associated with physical activity during leisure time, work, sports, and cycling to work. Arch Intern Med [Internet]. 2000 Jun 12 [cited 2012 Apr 9];160(11):1621–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/10847255 6.  Fletcher G, Balady G, Blair S, Blumenthal J, Caspersen C, Chaitman B, et al. Statement on Exercise : Benefits and Recommendations for Physical Activity Programs for All Americans. Circulation. 1996;94:857–62.   150 7.  Hillman C, Erickson K, Kramer A. Be smart, exercise your heart: exercise effects on brain and cognition. Nat Rev Neurosci [Internet]. 2008 [cited 2014 May 23];9:58–65. Available from: http://www.nature.com/nrn/journal/v9/n1/abs/nrn2298.html 8.  Hu F, Willett W, Li T, Stampfer M, Colditz G, Manson J. Adiposity as Compared with Physical Activity in Predicting Mortality among Women. N Engl J Med [Internet]. 2004 [cited 2014 Aug 14];(351):2694–703. Available from: http://www.nejm.org/doi/full/10.1056/NEJMoa042135 9.  Blair S, Kohl H, Barlow C, Paffenbarger R, Gibbons L, Macera C. Changes in Physical Fitness and All-Cause Mortality. JAMA. 1995;273(14):1093–8.  10.  Berlin JA, Colditz GA. A Meta-Analysis of Physical Activity in the Prevention of Coronary Heart Disease. Am J Epidemiol [Internet]. 1990 Oct 1;132(4):612–28. Available from: http://aje.oxfordjournals.org/content/132/4/612.abstract 11.  Lee I-M, Djoussé L, Sesso HD, Wang L, Buring JE. Physical Activity and Weight Gain Prevention. JAMA. 2010;303(12):1173–9.  12.  Durstine J, Grandjean P, Davis P, Ferguson M, Alderson N, DuBose K. Blood Lipid and Lipoprotein Adaptations to Exercise. Sport Med. 2001;31(15):1033–62.  13.  Ishikawa-Takata K, Tanaka H, Nanbu K, Ohta T. Beneficial effect of physical activity on blood pressure and blood glucose among Japanese male workers. Diabetes Res Clin Pract [Internet]. Elsevier Ireland Ltd; 2010 Mar [cited 2014 May 25];87(3):394–400. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19879663 14.  Rosenthal M, Haskell WL, Solomon R, Widstrom a, Reaven GM. Demonstration of a relationship between level of physical training and insulin-stimulated glucose utilization in  151 normal humans. Diabetes [Internet]. 1983 May;32(5):408–11. Available from: http://www.ncbi.nlm.nih.gov/pubmed/6341123 15.  Sigal R, Kenny G, Boulé N, Wells G, Prud’homme D, Fortier M, et al. Effects of Aerobic Training, Resistance Training, or Both on Glycemic Control in Type 2 Diabetes. Ann Intern Med [Internet]. 2007 [cited 2014 Jun 20];147:357–69. Available from: http://annals.org/article.aspx?articleid=736439 16.  Adamsen L, Quist M, Andersen C, Møller T. Effect of a multimodal high intensity exercise intervention in cancer patients undergoing chemotherapy: randomised controlled trial. BmJ [Internet]. 2009 [cited 2014 Aug 14];339:b3410. Available from: http://www.bmj.com/content/339/bmj.b3410.long 17.  Holick CN, Newcomb P a, Trentham-Dietz A, Titus-Ernstoff L, Bersch AJ, Stampfer MJ, et al. Physical activity and survival after diagnosis of invasive breast cancer. Cancer Epidemiol Biomarkers Prev [Internet]. 2008 Feb [cited 2014 May 24];17(2):379–86. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18250341 18.  Haydon a MM, Macinnis RJ, English DR, Giles GG. Effect of physical activity and body size on survival after diagnosis with colorectal cancer. Gut [Internet]. 2006 Jan [cited 2014 May 24];55(1):62–7. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1856365&tool=pmcentrez&rendertype=abstract 19.  Craft LL, Vaniterson EH, Helenowski IB, Rademaker AW, Courneya KS. Exercise effects on depressive symptoms in cancer survivors: a systematic review and meta-analysis. Cancer Epidemiol Biomarkers Prev [Internet]. 2012 Jan [cited 2014 May 24];21(1):3–19. Available from:  152 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3253916&tool=pmcentrez&rendertype=abstract 20.  World Health Organization. Health topics: Physical activity [Internet]. Global Strategy on Diet, Physical Activity and Health: Physical Activity. 2013 [cited 2013 Nov 15]. Available from: http://www.who.int/dietphysicalactivity/pa/en/index.html 21.  Canadian Society for Exercise Physiology. Canadian Physical Activity Guidelines [Internet]. 2011 [cited 2013 Sep 5]. Available from: www.csep.ca/guidelines 22.  Statistics Canada. Directly measured physical activity of Canadian adults , 2007 to 2011 [Internet]. Health Fact Sheets. 2013. p. ISSN 192–9118. Available from: http://www.statcan.gc.ca/pub/82-625-x/2013001/article/11807-eng.htm 23.  Statistics Canada. Canadian Health Measures Survey: Body composition of Canadian adults, 2009 to 2011 [Internet]. Health Fact Sheets (82-625-X). Ottawa, ON; 2012. Available from: http://www.statcan.gc.ca/pub/82-625-x/2012001/article/11708-eng.htm 24.  Roberts KC, Shields M, de Groh M, Aziz A, Gilbert J-A. Overweight and obesity in children and adolescents: results from the 2009 to 2011 Canadian Health Measures Survey. [Internet]. Health reports / Statistics Canada, Canadian Centre for Health Information = Rapports sur la santé / Statistique Canada, Centre canadien d’information sur la santé. 2012. p. 37–41. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23061263 25.  Public Health Agency of Canada and Canadian Institute for Health Information. Obesity in Canada: A Joint Report from the Public Health Agency of Canada and the Canadian Institute for Health Information [Internet]. 2011 p. 1–62. Available from: http://www.phac-aspc.gc.ca/hp-ps/hl-mvs/oic-oac/index-eng.php#toc  153 26.  Public Health Agency of Canada. What is Active Transportation? [Internet]. Health Promotion- Healthy Living: Physical Activity. 2010 [cited 2013 Dec 5]. Available from: http://www.phac-aspc.gc.ca/hp-ps/hl-mvs/pa-ap/at-ta-eng.php 27.  Schmidt S. Obesity and Exercise [Internet]. American College of Sports Medicine Access Public Information Articles. 2012 [cited 2013 Dec 5]. Available from: http://www.acsm.org/access-public-information/articles/2012/01/19/obesity-and-exercise 28.  Fentem PH. ABC of sports medicine. Benefits of exercise in health and disease. BMJ [Internet]. 1994 May 14;308(6939):1291–5. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2540212&tool=pmcentrez&rendertype=abstract 29.  Daley J. Activity Based Calorie Burn Calculator [Internet]. 2013 [cited 2013 Nov 25]. Available from: http://www.shapesense.com/fitness-exercise/calculators/activity-based-calorie-burn-calculator.aspx# 30.  Ainsworth BE, Haskell WL, Herrmann SD, Meckes N, Bassett Jr DR, Tudor-Locke C, Greer JL, Vezina J, Whitt-Glover MC LA. The Compendium of Physical Activities Tracking Guide [Internet]. Healthy Lifestyles Research Center, College of Nursing & Health Innovation, Arizona State University. 2003 [cited 2013 Nov 25]. Available from: https://sites.google.com/site/compendiumofphysicalactivities/ 31.  Ainsworth BE, Haskell WL, Herrmann SD, Meckes N, Bassett DR, Tudor-Locke C, et al. 2011 Compendium of Physical Activities: a second update of codes and MET values. Med Sci Sports Exerc [Internet]. 2011 Aug [cited 2013 Nov 9];43(8):1575–81. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21681120  154 32.  Handy SL, Xing Y, Buehler TJ. Factors associated with bicycle ownership and use: a study of six small U.S. cities. Transportation (Amst) [Internet]. 2010 Apr 21 [cited 2013 Nov 19];37(6):967–85. Available from: http://link.springer.com/10.1007/s11116-010-9269-x 33.  CBS Statistics Netherlands. StatLine (translated) Mobility vehicles owned by background characteristics [Internet]. 2013 [cited 2013 Dec 4]. Available from: http://statline.cbs.nl/StatWeb/publication/?DM=SLNL&PA=37856&D1=0&D2=0&D3=1-6,11-22,24-26&D4=0-14,21-22&HDR=G3,T,G1&STB=G2&VW=T 34.  Maness M. Bicycle Ownership in the United States: Empirical Analysis of Regional Differences [Internet]. College Park, Maryland; 2011 p. 1–17. Available from: http://www.academia.edu/1839374/Bicycle_Ownership_in_the_United_States_Empirical_Analysis_of_Regional_Differences# 35.  Riekko H. Bicycle Parking Regulations for Multi-Unit Residential Buildings in Toronto [Internet]. Toronto, ON; 2013 p. 1–18. Available from: http://www.cite7.org/conferences/compendium/2013_Cycling_BicycleParkingRegulationsMultiUnitResidentialBuildingsToronto.pdf 36.  Van den Dool H, Simplicius2wheels, The Bicycle Trade Association of Canada. Cycling in Canada on the Up and Up... [Internet]. I Bike TO. 2011 [cited 2014 May 25]. Available from: http://www.ibiketo.ca/forum/general/cycling-canada-and 37.  Pucher J, Buehler R. Why Canadians cycle more than Americans: A comparative analysis of bicycling trends and policies. Transp Policy [Internet]. 2006 May [cited 2013 Aug 30];13(3):265–79. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0967070X05001381  155 38.  Transport Canada. Urban Bicycle Planning [Internet]. Archived web page. Ottawa, ON; 2010. Available from: http://data.tc.gc.ca/archive/eng/programs/environment-utsp-casestudy-cs77ebikeplanning-1177.htm 39.  Statistics Canada. The Canadian Population in 2011: Population Counts and Growth [Internet]. Ottawa, ON; 2011 p. 1–26. Available from: http://www12.statcan.ca/census-recensement/2011/as-sa/98-310-x/98-310-x2011001-eng.cfm 40.  Pucher J, Buehler R. City Cycling. Pucher J, Buehler R, editors. Cambridge, Massachusetts: The MIT Press; 2012.  41.  Canadian Fitness & Lifestyle Research Institute. Getting Kids Active! 2010 Physical Activity Monitor: Facts and Figures. Ottawa, ON; 2011 p. 1–3.  42.  Yang X, Telama R, Hirvensalo M, Tammelin T, Viikari JS a, Raitakari OT. Active commuting from youth to adulthood and as a predictor of physical activity in early midlife: the young Finns study. Prev Med (Baltim) [Internet]. Elsevier Inc.; 2014 Mar [cited 2014 May 30];59:5–11. Available from: http://www.ncbi.nlm.nih.gov/pubmed/24201092 43.  Robertson-Wilson JE, Leatherdale ST, Wong SL. Social-ecological correlates of active commuting to school among high school students. J Adolesc Health [Internet]. 2008 May [cited 2014 May 30];42(5):486–95. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18407044 44.  McDonald NC, Brown AL, Marchetti LM, Pedroso MS. U.S. school travel, 2009 an assessment of trends. Am J Prev Med [Internet]. Elsevier Inc.; 2011 Aug [cited 2014 Jun 12];41(2):146–51. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21767721 45.  Dessing D, de Vries SI, Graham JM a, Pierik FH. Active transport between home and school assessed with GPS: a cross-sectional study among Dutch elementary school  156 children. BMC Public Health [Internet]. BMC Public Health; 2014 Jan [cited 2014 Jun 20];14(1):227. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3973871&tool=pmcentrez&rendertype=abstract 46.  Wong BY-M, Faulkner G, Buliung R. GIS measured environmental correlates of active school transport: a systematic review of 14 studies. Int J Behav Nutr Phys Act [Internet]. BioMed Central Ltd; 2011 Jan [cited 2014 May 30];8(1):39. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3112372&tool=pmcentrez&rendertype=abstract 47.  Statistics Canada. 2006 Census of Population: Place of work and commuting to work [Internet]. 97-561-XCB2006010. 2006 [cited 2013 Nov 27]. Available from: http://www12.statcan.gc.ca/census-recensement/2006/dp-pd/tbt/Rp-eng.cfm?TABID=1&LANG=E&A=R&APATH=3&DETAIL=0&DIM=0&FL=A&FREE=0&GC=01&GID=837928&GK=1&GRP=1&O=D&PID=90655&PRID=0&PTYPE=88971,97154&S=0&SHOWALL=0&SUB=0&Temporal=2006&THEME=76&VID=0&VNAMEE=&VNAMEF=&D1=0&D2=0&D3=0&D4=0&D5=0&D6=0 48.  Human Resources and Skills Development Canada. Indicators of well-being in Canada: Canadians in Context - Geographic Distribution: [Internet]. 2011 [cited 2013 Sep 13]. Available from: http://www4.hrsdc.gc.ca/.3ndic.1t.4r@-eng.jsp?iid=34 49.  Cooper AR, Wedderkopp N, Jago R, Kristensen PL, Moller NC, Froberg K, et al. Longitudinal associations of cycling to school with adolescent fitness. Prev Med (Baltim) [Internet]. 2008 Sep [cited 2013 Dec 5];47(3):324–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18602943  157 50.  De Geus B, Joncheere J, Meeusen R. Commuter cycling: effect on physical performance in untrained men and women in Flanders: minimum dose to improve indexes of fitness. Scand J Med Sci Sports [Internet]. 2009 Apr [cited 2013 Nov 22];19(2):179–87. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18282219 51.  De Nazelle A, Nieuwenhuijsen MJ, Antó JM, Brauer M, Briggs D, Braun-Fahrlander C, et al. Improving health through policies that promote active travel: a review of evidence to support integrated health impact assessment. Environ Int [Internet]. 2011 May [cited 2012 Mar 29];37(4):766–77. Available from: http://dx.doi.org/10.1016/j.envint.2011.02.003 52.  Hamer M, Chida Y. Active commuting and cardiovascular risk: a meta-analytic review. Prev Med (Baltim) [Internet]. 2008 Jan [cited 2013 Nov 7];46(1):9–13. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17475317 53.  Prince SA, Adamo KB, Hamel ME, Hardt J, Connor Gorber S, Tremblay M. A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act [Internet]. 2008 Jan [cited 2014 May 28];5:56. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2588639&tool=pmcentrez&rendertype=abstract 54.  Faulkner GEJ, Buliung RN, Flora PK, Fusco C. Active school transport, physical activity levels and body weight of children and youth: a systematic review. Prev Med (Baltim) [Internet]. Elsevier Inc.; 2009 Jan [cited 2014 May 30];48(1):3–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19014963  158 55.  Crouter SE, Churilla JR, Bassett DR. Estimating energy expenditure using accelerometers. Eur J Appl Physiol [Internet]. 2006 Dec [cited 2014 May 30];98(6):601–12. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17058102 56.  Cooper AR, Andersen LB, Wedderkopp N, Page AS, Froberg K. Physical activity levels of children who walk, cycle, or are driven to school. Am J Prev Med [Internet]. 2005 Oct [cited 2014 May 30];29(3):179–84. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16168866 57.  Andersen LB, Wedderkopp N, Kristensen P, Moller NC, Froberg K, Cooper AR. Cycling to school and cardiovascular risk factors: a longitudinal study. J Phys Act Health [Internet]. 2011 Nov;8(8):1025–33. Available from: http://www.ncbi.nlm.nih.gov/pubmed/22039135 58.  Grabow ML, Spak SN, Holloway T, Stone B, Mednick AC, Patz JA. Air quality and exercise-related health benefits from reduced car travel in the midwestern United States. Environ Health Perspect [Internet]. US DEPT HEALTH HUMAN SCIENCES PUBLIC HEALTH SCIENCE; 2012 Jan [cited 2012 May 30];120(1):68–76. Available from: http://apps.webofknowledge.com/full_record.do?product=WOS&search_mode=GeneralSearch&qid=4&SID=2D5PLH1FKpOjAG3l1hJ&page=1&doc=1 59.  Rojas-Rueda D, Nazelle A De. The health risks and benefits of cycling in urban environments compared with car use: health impact assessment study. BMJ [Internet]. 2011 [cited 2014 Mar 28];343(d4521):1–8. Available from: http://www.bmj.com/content/343/bmj.d4521  159 60.  Lindsay G, Macmillan A, Woodward A. Moving urban trips from cars to bicycles: impact on health and emissions. Aust N Z J Public Health [Internet]. 2011 Feb [cited 2012 Apr 17];35(1):54–60. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21299701 61.  Pucher J, Buehler R. Making Cycling Irresistible: Lessons from The Netherlands, Denmark and Germany. Transp Rev [Internet]. 2008 Jul [cited 2013 Aug 7];28(4):495–528. Available from: http://www.tandfonline.com/doi/abs/10.1080/01441640701806612 62.  Robinson DL. Safety in numbers in Australia: more walkers and bicyclists, safer walking and bicycling. Health Promot J Austr [Internet]. 2005 Apr;16(1):47–51. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16389930 63.  Jacobsen PL. Safety in numbers: more walkers and bicyclists, safer walking and bicycling. Inj Prev [Internet]. 2003 Sep;9(3):205–9. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1731007&tool=pmcentrez&rendertype=abstract 64.  Wegman F, Zhang F, Dijkstra A. How to make more cycling good for road safety? Accid Anal Prev [Internet]. 2012 Jan [cited 2012 Apr 9];44(1):19–29. Available from: http://dx.doi.org/10.1016/j.aap.2010.11.010 65.  Reynolds CCO, Harris MA, Teschke K, Cripton P a, Winters M. The impact of transportation infrastructure on bicycling injuries and crashes: a review of the literature. Environ Heal A Glob Access Sci Source [Internet]. 2009 Jan [cited 2014 Apr 30];8:47. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2776010&tool=pmcentrez&rendertype=abstract  160 66.  Yiannakoulias N, Bennet SA, Scott DM. Mapping commuter cycling risk in urban areas. Accid Anal Prev [Internet]. 2012 Mar [cited 2012 Apr 9];45:164–72. Available from: http://dx.doi.org/10.1016/j.aap.2011.12.002 67.  Winters M, Davidson G, Kao D, Teschke K. Motivators and deterrents of bicycling: comparing influences on decisions to ride. Transportation (Amst) [Internet]. 2010 Jun 13 [cited 2013 Aug 16];38(1):153–68. Available from: http://link.springer.com/10.1007/s11116-010-9284-y 68.  Winters M, Friesen MC, Koehoorn M, Teschke K. Utilitarian bicycling: a multilevel analysis of climate and personal influences. Am J Prev Med [Internet]. 2007 Jan [cited 2013 Aug 30];32(1):52–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17184961 69.  Pucher J, Dijkstra L. Promoting safe walking and cycling to improve public health: lessons from The Netherlands and Germany. Am J Public Health [Internet]. 2003 Sep;93(9):1509–16. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1448001&tool=pmcentrez&rendertype=abstract 70.  Teschke K, Harris M, Reynolds C. Exposure-based traffic crash injury rates by mode of travel in British Columbia. Can J Public Heal [Internet]. 2013 [cited 2014 Aug 11];104(1):e75–79. Available from: http://journal.cpha.ca/index.php/cjph/article/view/3621 71.  International Traffic Safety Data and Analysis Group. Road Safety Annual Report 2013 [Internet]. Paris; 2013 p. 1–458. Available from: http://www.internationaltransportforum.org/pub/pdf/13IrtadReport.pdf  161 72.  Minister of Transport. Canadian Motor Vehicle Traffic Collision Statistics: 2007 [Internet]. Ottawa; 2011 p. 1–6. Available from: http://scholar.google.com/scholar?hl=en&btnG=Search&q=intitle:Canadian+Motor+Vehicle+Traffic+Collision+Statistics+2011#1 73.  Jacobsen PL, Racioppi F, Rutter H. Who owns the roads? How motorised traffic discourages walking and bicycling. Inj Prev [Internet]. 2009 Dec [cited 2014 Jun 5];15(6):369–73. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19959727 74.  Zuurbier M, Hoek G, Oldenwening M, Meliefste K, Krop E, van den Hazel P, et al. In-traffic air pollution exposure and CC16, blood coagulation, and inflammation markers in healthy adults. Environ Health Perspect [Internet]. United States; 2011 Oct [cited 2012 May 30];119(10):1384–9. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3230432&tool=pmcentrez&rendertype=abstract 75.  Zuurbier M, Hoek G, Oldenwening M, Meliefste K, van den Hazel P, Brunekreef B. Respiratory effects of commuters’ exposure to air pollution in traffic. Epidemiology [Internet]. 2011 Mar [cited 2012 May 30];22(2):219–27. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21228698 76.  Jarjour S, Jerrett M, Westerdahl D, de Nazelle A, Hanning C, Daly L, et al. Cyclist route choice, traffic-related air pollution, and lung function: a scripted exposure study. Environ Heal a Glob access Sci source [Internet]. Environmental Health; 2013 Jan [cited 2013 Sep 20];12(1):14. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3639931&tool=pmcentrez&rendertype=abstract  162 77.  Pattinson WJ. Cyclist exposure to traffic pollution: microscale variance, the impact of route choice and comparisons to other modal choices in two New Zealand cities [Internet]. University of Canterbury; 2009. Available from: http://ir.canterbury.ac.nz/bitstream/10092/3687/1/thesis_fulltext.pdf 78.  Kingham S, Longley I, Salmond J, Pattinson W, Shrestha K. Variations in exposure to traffic pollution while travelling by different modes in a low density, less congested city. Environ Pollut [Internet]. Elsevier Ltd; 2013 Oct [cited 2014 Jan 6];181:211–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23871818 79.  Nyhan M, McNabola A, Misstear B. Comparison of particulate matter dose and acute heart rate variability response in cyclists, pedestrians, bus and train passengers. Sci Total Environ [Internet]. Elsevier B.V.; 2014 Jan 15 [cited 2014 Mar 9];468-469:821–31. Available from: http://www.ncbi.nlm.nih.gov/pubmed/24076503 80.  Van Wijnen JH, Verhoeff a P, Jans HW, van Bruggen M. The exposure of cyclists, car drivers and pedestrians to traffic-related air pollutants. Int Arch Occup Environ Health [Internet]. 1995 Jan;67(3):187–93. Available from: http://www.ncbi.nlm.nih.gov/pubmed/7591177 81.  O’Donoghue RT, Gill LW, McKevitt RJ, Broderick B. Exposure to hydrocarbon concentrations while commuting or exercising in Dublin. Environ Int [Internet]. 2007 Jan [cited 2012 May 27];33(1):1–8. Available from: http://dx.doi.org/10.1016/j.envint.2006.05.005 82.  Zuurbier M, Hoek G, van den Hazel P, Brunekreef B. Minute ventilation of cyclists, car and bus passengers: an experimental study. Environ Health [Internet]. 2009 Jan [cited 2011 Dec 29];8:48. Available from:  163 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2772854&tool=pmcentrez&rendertype=abstract 83.  Int Panis L, de Geus B, Vandenbulcke G, Willems H, Degraeuwe B, Bleux N, et al. Exposure to particulate matter in traffic: A comparison of cyclists and car passengers. Atmos Environ [Internet]. 2010 Jun [cited 2011 Aug 18];44(19):2263–70. Available from: http://linkinghub.elsevier.com/retrieve/pii/S1352231010003225 84.  McNabola A, Broderick BM, Gill LW. Relative exposure to fine particulate matter and VOCs between transport microenvironments in Dublin: Personal exposure and uptake. Atmos Environ [Internet]. 2008 Aug [cited 2012 Mar 15];42(26):6496–512. Available from: http://dx.doi.org/10.1016/j.atmosenv.2008.04.015 85.  Nwokoro C, Ewin C, Harrison C, Ibrahim M, Dundas I, Dickson I, et al. Cycling to work in London and inhaled dose of black carbon. Eur Respir J  Off J Eur Soc Clin Respir Physiol [Internet]. 2012 Nov [cited 2013 Mar 27];40(5):1091–7. Available from: http://www.ncbi.nlm.nih.gov/pubmed/22362851 86.  Kaur S, Nieuwenhuijsen MJ, Colvile RN. Fine particulate matter and carbon monoxide exposure concentrations in urban street transport microenvironments. Atmos Environ [Internet]. 2007 Jul [cited 2012 Mar 19];41(23):4781–810. Available from: http://dx.doi.org/10.1016/j.atmosenv.2007.02.002 87.  Boogaard H, Borgman F, Kamminga J, Hoek G. Exposure to ultrafine and fine particles and noise during cycling and driving in 11 Dutch cities. Atmos Environ [Internet]. 2009 Sep [cited 2012 May 27];43(27):4234–42. Available from: http://dx.doi.org/10.1016/j.atmosenv.2009.05.035  164 88.  Knibbs LD, Cole-Hunter T, Morawska L. A review of commuter exposure to ultrafine particles and its health effects. Atmos Environ [Internet]. PERGAMON-ELSEVIER SCIENCE LTD; 2011 May [cited 2012 Mar 29];45(16):2611–22. Available from: http://apps.webofknowledge.com/full_record.do?product=WOS&search_mode=GeneralSearch&qid=35&SID=2D5PLH1FKpOjAG3l1hJ&page=1&doc=10 89.  Zuurbier M, Hoek G, Oldenwening M, Lenters V, Meliefste K, van den Hazel P, et al. Commuters’ exposure to particulate matter air pollution is affected by mode of transport, fuel type, and route. Environ Health Perspect [Internet]. 2010 Jun [cited 2012 Mar 16];118(6):783–9. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2898854&tool=pmcentrez&rendertype=abstract 90.  Quiros DC, Lee ES, Wang R, Zhu Y. Ultrafine particle exposures while walking, cycling, and driving along an urban residential roadway. Atmos Environ [Internet]. Elsevier Ltd; 2013 Jul [cited 2014 Jan 6];73:185–94. Available from: http://linkinghub.elsevier.com/retrieve/pii/S1352231013002033 91.  Ragettli MS, Corradi E, Braun-Fahrländer C, Schindler C, de Nazelle A, Jerrett M, et al. Commuter exposure to ultrafine particles in different urban locations, transportation modes and routes. Atmos Environ [Internet]. Elsevier Ltd; 2013 Oct [cited 2013 Dec 21];77:376–84. Available from: http://linkinghub.elsevier.com/retrieve/pii/S1352231013003464 92.  Adams H., Nieuwenhuijsen M., Colvile R., McMullen MA., Khandelwal P. Fine particle (PM2.5) personal exposure levels in transport microenvironments, London, UK. Sci Total Environ [Internet]. 2001 Nov [cited 2012 May 27];279(1-3):29–44. Available from: http://dx.doi.org/10.1016/S0048-9697(01)00723-9  165 93.  Kaur S, Clark RDR, Walsh PT, Arnold SJ, Colvile RN, Nieuwenhuijsen MJ. Exposure visualisation of ultrafine particle counts in a transport microenvironment. Atmos Environ [Internet]. 2006 Jan [cited 2013 Sep 20];40(2):386–98. Available from: http://linkinghub.elsevier.com/retrieve/pii/S1352231005008964 94.  Adams H., Nieuwenhuijsen M., Colvile R. Determinants of fine particle (PM2.5) personal exposure levels in transport microenvironments, London, UK. Atmos Environ [Internet]. 2001 Sep [cited 2012 Jun 16];35(27):4557–66. Available from: http://dx.doi.org/10.1016/S1352-2310(01)00194-7 95.  Kaur S, Nieuwenhuijsen M, Colvile R. Personal exposure of street canyon intersection users to PM2.5, ultrafine particle counts and carbon monoxide in Central London, {UK}. Atmos Environ [Internet]. 2005;39(20):3629–41. Available from: http://www.sciencedirect.com/science/article/pii/S1352231005002268 96.  Briggs DJ, de Hoogh K, Morris C, Gulliver J. Effects of travel mode on exposures to particulate air pollution. Environ Int [Internet]. 2008 Jan [cited 2011 Jun 28];34(1):12–22. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17688949 97.  Wallace L, Ott W. Personal exposure to ultrafine particles. J Expo Sci Environ Epidemiol [Internet]. Nature Publishing Group; 2011 Jan [cited 2011 Oct 29];21(1):20–30. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20087407 98.  Knibbs LD, de Dear RJ, Morawska L. Effect of Cabin Ventilation Rate on Ultrafine Particle Exposure Inside Automobiles. Environ Sci Technol [Internet]. 2010;44(9):3546–51. Available from: http://pubs.acs.org/doi/abs/10.1021/es9038209  166 99.  Huang J, Deng F, Wu S, Guo X. Comparisons of personal exposure to PM(2.5) and CO by different commuting modes in Beijing, China. Sci Total Environ [Internet]. 2012 Apr 1 [cited 2012 Apr 6]; Available from: http://dx.doi.org/10.1016/j.scitotenv.2012.03.007 100.  Thai A, McKendry I, Brauer M. Particulate matter exposure along designated bicycle routes in Vancouver, British Columbia. Sci Total Environ [Internet]. 2008 Nov 1 [cited 2011 Sep 19];405(1-3):26–35. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18701140 101.  Berghmans P, Bleux N, Int Panis L, Mishra VK, Torfs R, Van Poppel M. Exposure assessment of a cyclist to PM10 and ultrafine particles. Sci Total Environ [Internet]. 2009 Feb 1 [cited 2012 Mar 4];407(4):1286–98. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19036413 102.  Hatzopoulou M, Weichenthal S, Dugum H, Pickett G, Miranda-Moreno L, Kulka R, et al. The impact of traffic volume, composition, and road geometry on personal air pollution exposures among cyclists in Montreal, Canada. J Expo Sci Environ Epidemiol [Internet]. Nature Publishing Group; 2012 [cited 2014 Jan 6];23(1):46–51. Available from: http://www.ncbi.nlm.nih.gov/pubmed/22910003 103.  Kendrick CM, Moore A, Haire A, Bigazzi A, Figliozzi M, Monsere CM, et al. Impact of Bicycle Lane Characteristics on Exposure of Bicyclists to Traffic-Related Particulate Matter. Transp Res Rec J Transp Res Board [Internet]. 2011 Dec 1 [cited 2012 May 31];2247(-1):24–32. Available from: http://trb.metapress.com/openurl.asp?genre=article&id=doi:10.3141/2247-04 104.  George L, Figliozzi M, Monsere C, Kendrick C, Bigazzi A, Moore A. Evaluation of Transportation Microenvironments Through Assessment of Cyclists’ Exposure to Traffic- 167 related Particulate Matter [Internet]. Portland, OR; 2011 p. 1–34. Available from: http://ntl.bts.gov/lib/41000/41100/41194/Tran.doc 105.  MacNaughton P, Melly S, Vallarino J, Adamkiewicz G, Spengler JD. Impact of bicycle route type on exposure to traffic-related air pollution. Sci Total Environ [Internet]. Elsevier B.V.; 2014 May 16 [cited 2014 Jun 19];490C(2):37–43. Available from: http://www.ncbi.nlm.nih.gov/pubmed/24840278 106.  Dons E, Temmerman P, Van Poppel M, Bellemans T, Wets G, Int Panis L. Street characteristics and traffic factors determining road users’ exposure to black carbon. Sci Total Environ [Internet]. Elsevier B.V.; 2013 Mar 1 [cited 2014 Jun 1];447:72–9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23376518 107.  Bevan MAJ, Proctor CJ, Baker-rogers J, Warren ND. Exposure to Carbon Monoxide , Respirable Suspended Particulates , and Volatile Organic Compounds While Commuting by Bicycle. Environ Sci Technol. 1991;25(4):788–91.  108.  Macfarlane DJ, Wong P. Validity, reliability and stability of the portable Cortex Metamax 3B gas analysis system. Eur J Appl Physiol [Internet]. 2012 Jul [cited 2014 Jun 1];112(7):2539–47. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3371330&tool=pmcentrez&rendertype=abstract 109.  James a. C, Stahlhofen W, Rudolf G, Köbrich R, Briant JK, Egan MJ, et al. Annexe D. deposition of inhaled particles. Ann ICRP [Internet]. 1994 Jan [cited 2014 Apr 25];24(1-3):231–99. Available from: http://ani.sagepub.com/lookup/doi/10.1016/0146-6453(94)90042-6  168 110.  Bernmark E, Wiktorin C, Svartengren M, Lewné M, Aberg S. Bicycle messengers: energy expenditure and exposure to air pollution. Ergonomics [Internet]. 2006 Nov 15 [cited 2013 Apr 11];49(14):1486–95. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17050389 111.  Dietert RR, Etzel ARA, Chen D, Halonen M, Holladay SD, Jarabek AM, et al. Workshop to Identify Critical Windows of Exposure for Children’s Health: Immune and Respiratory Systems Work Group Summary. Environ Health Perspect [Internet]. 2000;108 (suppl(June):483–90. Available from: http.//ehpnet1. niehs. nih.gov/docs/2000/suppl-3/483-490dietert/abstract. html 112.  Lippmann M, Yeates DB, Albert RE. Deposition, retention, and clearance of inhaled particles. Occup Environ Med [Internet]. 1980 Nov 1 [cited 2013 Oct 16];37(4):337–62. Available from: http://oem.bmj.com/cgi/doi/10.1136/oem.37.4.337 113.  Ashe MC, Scroop GC, Frisken PI, Amery C a, Wilkins M a, Khan KM. Body position affects performance in untrained cyclists. Br J Sports Med [Internet]. 2003 Jan;37(5):441–4. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1751358&tool=pmcentrez&rendertype=abstract 114.  Origenes IV MM, Blank SE, Schoene RB. Exercise ventilatory response to upright and aero-posture cycling. Med Sci Sports Exerc. 1993;25(5):608–12.  115.  Jaques P, Kim C. Measurement of total lung deposition of inhales ultrafine particles in healthy men and women. Inhal Toxicol [Internet]. 2000 [cited 2014 Jul 5];715–31. Available from: http://informahealthcare.com/doi/abs/10.1080/08958370050085156 116.  Weichenthal S, Kulka R, Dubeau A, Martin C, Wang D, Dales R. Traffic-related air pollution and acute changes in heart rate variability and respiratory function in urban  169 cyclists. Environ Health Perspect [Internet]. 2011 Oct;119(10):1373–8. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3230442&tool=pmcentrez&rendertype=abstract 117.  Weichenthal S, Kulka R, Bélisle P, Joseph L, Dubeau A, Martin C, et al. Personal exposure to specific volatile organic compounds and acute changes in lung function and heart rate variability among urban cyclists. Environ Res [Internet]. 2012 Oct [cited 2014 Jun 2];118:118–23. Available from: http://www.ncbi.nlm.nih.gov/pubmed/22776327 118.  Strak M, Boogaard H, Meliefste K, Oldenwening M, Zuurbier M, Brunekreef B, et al. Respiratory health effects of ultrafine and fine particle exposure in cyclists. Occup Environ Med [Internet]. England; 2010 Feb [cited 2012 Mar 14];67(2):118–24. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19773283 119.  Jacobs L, Nawrot TS, de Geus B, Meeusen R, Degraeuwe B, Bernard A, et al. Subclinical responses in healthy cyclists briefly exposed to traffic-related air pollution: an intervention study. Environ Health [Internet]. 2010 Jan [cited 2011 Dec 29];9:64. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2984475&tool=pmcentrez&rendertype=abstract 120.  Bos I, Jacobs L, Nawrot TS, de Geus B, Torfs R, Int Panis L, et al. No exercise-induced increase in serum BDNF after cycling near a major traffic road. Neurosci Lett [Internet]. Elsevier Ireland Ltd; 2011 Aug 15 [cited 2011 Nov 7];500(2):129–32. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21708224 121.  McCreanor J, Cullinan P, Nieuwenhuijsen MJ, Stewart-Evans J, Malliarou E, Jarup L, et al. Respiratory effects of exposure to diesel traffic in persons with asthma. N Engl J Med  170 [Internet]. 2007 Dec 6;357(23):2348–58. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18057337 122.  Gong H, Linn WS, Clark KW, Anderson KR, Sioutas C, Alexis NE, et al. Exposures of healthy and asthmatic volunteers to concentrated ambient ultrafine particles in Los Angeles. Inhal Toxicol [Internet]. 2008 Apr [cited 2014 Mar 27];20(6):533–45. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18444007 123.  Penttinen P, Timonen KL, Tiittanen P, Mirme A, Ruuskanen J, Pekkanen J. Number Concentration and Size of Particles in Urban Air : Effects on Spirometric Lung Function in Adult Asthmatic Subjects. 2001;109(4):319–23.  124.  Cerrina J, Denjean A, Alexandre G, Lockhart A, Duroux P. Inhibition of exercise-induced asthma by a calcium antagonist, nifedipine. Am Rev Respir Dis. 1981;123(2):156–60.  125.  Statistics Canada. Canadian Vehicle Survey : Annual 2009 [Internet]. Ottawa, ON; 2009. Available from: http://www5.statcan.gc.ca/access_acces/archive.action?loc=/pub/53-223-x/53-223-x2009000-eng.pdf&archive=1 126.  Campestrini M, Mock P. European Vehicle Market Statistics [Internet]. Washington, DC.; 2011. Available from: http://www.theicct.org/sites/default/files/publications/Pocketbook_LowRes_withNotes-1.pdf 127.  Urch B, Speck M, Corey P. Concentrated ambient fine particles and not ozone induce a systemic interleukin-6 response in humans. Inhal Toxicol [Internet]. 2010 [cited 2014 Jun 9];22(May 2009):210–8. Available from: http://informahealthcare.com/doi/abs/10.3109/08958370903173666  171 128.  Gong H, Sioutas C, Linn WS. Controlled exposures of healthy and asthmatic volunteers to concentrated ambient particles in metropolitan Los Angeles. Res Rep Health Eff Inst [Internet]. 2003 Dec;(118):1—36; discussion 37—47. Available from: http://europepmc.org/abstract/MED/14738210 129.  Bräuner EV, Forchhammer L, Møller P, Barregard L, Gunnarsen L, Afshari A, et al. Indoor particles affect vascular function in the aged: an air filtration-based intervention study. Am J Respir Crit Care Med [Internet]. 2008 Feb 15 [cited 2011 Dec 1];177(4):419–25. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17932377 130.  Carlsten C, Kaufman J, Peretz A. Coagulation markers in healthy human subjects exposed to diesel exhaust. Thromb Res [Internet]. 2007 [cited 2014 Apr 27];120(6):849–55. Available from: http://www.sciencedirect.com/science/article/pii/S0049384807000345 131.  Hampel R, Rückerl R, Yli-Tuomi T, Breitner S, Lanki T, Kraus U, et al. Impact of personally measured pollutants on cardiac function. Int J Hyg Environ Health [Internet]. Elsevier GmbH.; 2013 Oct 19 [cited 2014 Jan 6]; Available from: http://www.ncbi.nlm.nih.gov/pubmed/24231411 132.  World Health Organization. Health Topics: Air Pollution [Internet]. 2013 [cited 2013 Sep 17]. Available from: http://www.who.int/topics/air_pollution/en/index.html 133.  World Health Organization. Air Quality Guidelines Global Update 2005: Particulate matter, ozone, nitrogen dioxide and sulfur dioxide [Internet]. Copenhagen, Denmark: WHO Regional Office for Europe; 2006 [cited 2014 Aug 23]. Available from: http://books.google.com/books?hl=en&lr=&id=7VbxUdlJE8wC&oi=fnd&pg=PR9&dq=Air+Quality+Guidelines+Global+Update+2005:+Particulate+matter,+ozone,+nitrogen+dioxide+and+sulfur+dioxide&ots=w118yLP7sc&sig=8EyEpLO1C9C_3WsBFV-zh7QzzRw  172 134.  United States Environmental Protection Agency. Six Common Pollutants: Particulate Matter (PM) [Internet]. 2013 [cited 2013 Oct 6]. Available from: http://www.epa.gov/air/particlepollution/index.html 135.  Dockery D, Pope C, Xu X, Spengler JD, Ware JH, Fay ME, et al. An Association Between Air Pollution and Mortality in Six U.S. Cities. N Engl J Med [Internet]. 1993 [cited 2013 Sep 18];329(24):1753–9. Available from: http://www.nejm.org/doi/full/10.1056/nejm199312093292401 136.  Burnett RT, Cakmak S, Brook JR. The effect of the urban ambient air pollution mix on daily mortality rates in 11 Canadian cities. Can J Public Heal Rev Can santé publique [Internet]. 1998;89(3):152–6. Available from: http://www.ncbi.nlm.nih.gov/pubmed/9654797 137.  Judek S, Jessiman B, Stieb D, Vet R. Estimated Number of Excess Deaths in Canada Due to Air Pollution [Internet]. unpublished. 2004. p. 1–10. Available from: http://www.metrovancouver.org/about/publications/Publications/AirPollutionDeaths.pdf 138.  Institute for Health Metrics and Evaluation (IHME). Global Burden of Disease Study 2010 (GBD 2010) Results by Risk Factor 1900-2010- Country Level. Seattle, United States; 2013.  139.  Hildebrandt K, Rückerl R, Koenig W, Schneider A, Pitz M, Heinrich J, et al. Short-term effects of air pollution: a panel study of blood markers in patients with chronic pulmonary disease. Part Fibre Toxicol [Internet]. 2009 Jan [cited 2013 Oct 6];6:25. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2762952&tool=pmcentrez&rendertype=abstract  173 140.  Simkhovich BZ, Kleinman MT, Kloner R a. Air pollution and cardiovascular injury epidemiology, toxicology, and mechanisms. J Am Coll Cardiol [Internet]. 2008 Aug 26 [cited 2013 Oct 11];52(9):719–26. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18718418 141.  Martinelli N, Olivieri O, Girelli D. Air particulate matter and cardiovascular disease: a narrative review. Eur J Intern Med [Internet]. European Federation of Internal Medicine.; 2013 Jun [cited 2013 Dec 3];24(4):295–302. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23647842 142.  United States Environmental Protection Agency. Six Common Pollutants: Particulate Matter (PM) Basic Information [Internet]. 2013 [cited 2013 Oct 6]. Available from: http://www.epa.gov/airquality/particlepollution/basic.html 143.  Environment Canada - Health Canada. Priority Substances List Assessment Report for Respirable Particulate Matter [Internet]. Environmental Contaminants. 2000 [cited 2014 Jan 19]. Available from: http://www.hc-sc.gc.ca/ewh-semt/pubs/contaminants/psl2-lsp2/pm10/index-eng.php#a22 144.  Pritchard R, Ghio A, Lehmann J, Winsett D, Tepper J, Park P, et al. Oxidant generation and lung injury after particulate air pollutant exposure increase with the concentrations of associated metals. Inhal Toxicol [Internet]. 1996 [cited 2013 Sep 28];8:457–77. Available from: http://informahealthcare.com/doi/abs/10.3109/08958379609005440 145.  MacNee W, Donaldson K. Mechanism of lung injury caused by PM10 and ultrafine particles with special reference to COPD. Eur Respir J [Internet]. 2003 May 1 [cited 2013 Apr 13];21(Supplement 40):47S–51s. Available from: http://erj.ersjournals.com/cgi/doi/10.1183/09031936.03.00403203  174 146.  Nemmar A, Hoet PHM, Vanquickenborne B, Dinsdale D, Thomeer M, Hoylaerts MF, et al. Passage of Inhaled Particles Into the Blood Circulation in Humans. Circulation [Internet]. 2002 Jan 29 [cited 2013 Sep 25];105(4):411–4. Available from: http://circ.ahajournals.org/cgi/doi/10.1161/hc0402.104118 147.  HEI Review Panel on Ultrafine Particles. HEI Perspectives 3: Understanding the Health Effects of Ambient Ultrafine Particles. Health Effects Institute. 2013 p. 1–122.  148.  Daigle CC, Chalupa DC, Gibb FR, Morrow PE, Oberdörster G, Utell MJ, et al. Ultrafine Particle Deposition In Humans During Rest And Exercise. Inhal Toxicol. 2003;15:539–52.  149.  WHO Regional Office for Europe. Particulate matter. Air Quality Guidelines - Second Edition [Internet]. Second. Copenhagen, Denmark: WHO Regional Office for Europe; 2000. p. 1–40. Available from: http://www.euro.who.int/__data/assets/pdf_file/0019/123085/AQG2ndEd_7_3Particulate-matter.pdf 150.  Becquemin MH, Swift DL, Bouchikhi a, Roy M, Teillac a. Particle deposition and resistance in the noses of adults and children. Eur Respir J [Internet]. 1991 Jun;4(6):694–702. Available from: http://www.ncbi.nlm.nih.gov/pubmed/1889496 151.  Brook RD, Rajagopalan S, Pope CA, Brook JR, Bhatnagar A, Diez-Roux A V, et al. Particulate matter air pollution and cardiovascular disease: An update to the scientific statement from the American Heart Association. Circulation [Internet]. 2010 Jun 1 [cited 2013 Sep 16];121(21):2331–78. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20458016 152.  Hollingsworth JW, Maruoka S, Li Z, Potts EN, Brass DM, Garantziotis S, et al. Ambient Ozone Primes Pulmonary Innate Immunity in Mice. J Immunol [Internet]. 2007 Sep 18  175 [cited 2014 Jun 9];179(7):4367–75. Available from: http://www.jimmunol.org/cgi/doi/10.4049/jimmunol.179.7.4367 153.  Scheller J, Chalaris A, Schmidt-Arras D, Rose-John S. The pro- and anti-inflammatory properties of the cytokine interleukin-6. Biochim Biophys Acta [Internet]. 2011 May [cited 2014 May 27];1813(5):878–88. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21296109 154.  Moldoveanu AI, Shephard ROYJ, Shek PN. Exercise elevates plasma levels but not gene expression of IL-1 beta , IL-6 , and TNF- alpha in blood mononuclear cells. 2000;9:1499–504.  155.  Pedersen B, Febbraio M. Muscle as an endocrine organ: focus on muscle-derived interleukin-6. Physiol Rev [Internet]. 2008 [cited 2014 May 1];88:1379–406. Available from: http://physrev.physiology.org/content/88/4/1379.short 156.  Moshage HJ, Pelt JF Van, Leeuwen V, Limburq PC, Yap SH, Diseases L. The effect of interleukin-1, interleukin-6 and its interrelationship on the synthesis of serum amyloid A and C-reactive protein in primary cultures of adult human hepatocytes. Biochem Biophys Res Commun [Internet]. 1988;155(1):112–7. Available from: http://www.amyloid.nl/Files/Moshage1988.pdf 157.  Pepys MB, Hirschfield GM. C-reactive protein : a critical update. 2003;111(12):1805–12.  158.  Thompson D, Pepys M, Wood S. The physiological structure of human C-reactive protein and its complex with phosphocholine. Structure [Internet]. 1999 [cited 2014 Jul 6];(7):169–77. Available from: http://www.sciencedirect.com/science/article/pii/S0969212699800239  176 159.  Calabró P, Willerson JT, Yeh ETH. Inflammatory cytokines stimulated C-reactive protein production by human coronary artery smooth muscle cells. Circulation [Internet]. 2003 Oct 21 [cited 2013 Oct 11];108(16):1930–2. Available from: http://www.ncbi.nlm.nih.gov/pubmed/14530191 160.  Yasojima K, Schwab C, McGeer EG, McGeer PL. Generation of C-reactive protein and complement components in atherosclerotic plaques. Am J Pathol [Internet]. 2001 Mar;158(3):1039–51. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1850354&tool=pmcentrez&rendertype=abstract 161.  Quay JL, Reed W, Samet J, Devlin RB. Air pollution particles induce IL-6 gene expression in human airway epithelial cells via NF-κ B activation. Am J Respir Cell Mol Biol. Am Thoracic Soc; 1998;19(1):98–106.  162.  Devlin RB, McKinnon KP, Noah T, Becker S, Koren HS. Ozone-induces release of cytokines and fibronectin by alveolar macrophages and airway epithelial cells. Am J Physiol Lung Cell Mol Physiol [Internet]. 1994;266(L612-L619). Available from: http://ajplung.physiology.org/content/266/6/L612 163.  Tamagawa E, Bai N, Morimoto K, Gray C, Mui T, Kazuhiro Y, et al. Particulate matter exposure induces persistent lung inflammation and endothelial dysfunction. Am J Physiol Lung Cell Mol Physiol [Internet]. 2008 [cited 2014 Jan 1];295(1):79–85. Available from: http://ajplung.physiology.org/content/295/1/L79.short 164.  Chuang K-J, Chan C-C, Su T-C, Lee C-T, Tang C-S. The effect of urban air pollution on inflammation, oxidative stress, coagulation, and autonomic dysfunction in young adults.  177 Am J Respir Crit Care Med [Internet]. 2007 Aug 15 [cited 2013 Sep 20];176(4):370–6. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17463411 165.  Delfino RJ, Staimer N, Tjoa T, Polidori A, Arhami M, Gillen DL, et al. Circulating biomarkers of inflammation, antioxidant activity, and platelet activation are associated with primary combustion aerosols in subjects with coronary artery disease. Environ Health Perspect [Internet]. 2008 Jul [cited 2013 Sep 28];116(7):898–906. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2453158&tool=pmcentrez&rendertype=abstract 166.  Strak M, Hoek G, Godri KJ, Gosens I, Mudway IS, van Oerle R, et al. Composition of PM Affects Acute Vascular Inflammatory and Coagulative Markers - The RAPTES Project. PLoS One [Internet]. 2013 Jan [cited 2013 Mar 25];8(3):e58944. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3596332&tool=pmcentrez&rendertype=abstract 167.  Riediker M, Cascio WE, Griggs TR, Herbst MC, Bromberg P a, Neas L, et al. Particulate matter exposure in cars is associated with cardiovascular effects in healthy young men. Am J Respir Crit Care Med [Internet]. 2004 Apr 15 [cited 2014 Jan 9];169(8):934–40. Available from: http://www.ncbi.nlm.nih.gov/pubmed/14962820 168.  Brook RD. Cardiovascular effects of air pollution. Clin Sci (Lond) [Internet]. 2008 Sep [cited 2013 Oct 11];115(6):175–87. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18691154 169.  Kaptoge S, Di Angelantonio E, Lowe G, Pepys MB, Thompson SG, Collins R, et al. C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta-analysis. Lancet [Internet]. 2010 Jan 9 [cited 2014 May  178 30];375(9709):132–40. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3162187&tool=pmcentrez&rendertype=abstract 170.  Allen RW, Carlsten C, Karlen B, Leckie S, van Eeden S, Vedal S, et al. An air filter intervention study of endothelial function among healthy adults in a woodsmoke-impacted community. Am J Respir Crit Care Med [Internet]. 2011 May 1 [cited 2014 Mar 26];183(9):1222–30. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21257787 171.  Bräuner E V, Møller P, Barregard L, Dragsted LO, Glasius M, Wåhlin P, et al. Exposure to ambient concentrations of particulate air pollution does not influence vascular function or inflammatory pathways in young healthy individuals. Part Fibre Toxicol [Internet]. 2008 Jan [cited 2011 Aug 15];5:13. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2579917&tool=pmcentrez&rendertype=abstract 172.  Karottki DG, Spilak M, Frederiksen M, Gunnarsen L, Brauner EV, Kolarik B, et al. An indoor air filtration study in homes of elderly: cardiovascular and respiratory effects of exposure to particulate matter. Environ Health [Internet]. 2013 Jan;12:116. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3893545&tool=pmcentrez&rendertype=abstract 173.  Eldrup N, Kragelund C, Steffensen R, Nordestgaard BG. Prognosis by C-reactive protein and matrix metalloproteinase-9 levels in stable coronary heart disease during 15 years of follow-up. Nutr Metab Cardiovasc Dis [Internet]. 2012 Aug [cited 2014 Jun 6];22(8):677–83. Available from: http://www.sciencedirect.com/science/article/pii/S0939475310002693  179 174.  Ridker PM, Rifai N, Stampfer MJ, Hennekens CH. Plasma Concentration of Interleukin-6 and the Risk of Future Myocardial Infarction Among Apparently Healthy Men. Circulation [Internet]. 2000 Apr 18 [cited 2014 Jun 9];101(15):1767–72. Available from: http://circ.ahajournals.org/cgi/doi/10.1161/01.CIR.101.15.1767 175.  Volpato S, Guralnik JM, Ferrucci L, Balfour J, Chaves P, Fried LP, et al. Cardiovascular Disease, Interleukin-6, and Risk of Mortality in Older Women: The Women’s Health and Aging Study. Circulation [Internet]. 2001 Feb 20 [cited 2014 Jun 9];103(7):947–53. Available from: http://circ.ahajournals.org/cgi/doi/10.1161/01.CIR.103.7.947 176.  Lauer MS. Autonomic function and prognosis. Cleve Clin J Med [Internet]. 2009 Apr [cited 2013 Dec 20];76 Suppl 2:S18–22. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19376976 177.  Franchini M, Mannucci PM. Short-term effects of air pollution on cardiovascular diseases: outcomes and mechanisms. J Thromb Haemost [Internet]. 2007 Nov;5(11):2169–74. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17958737 178.  Gold DR, Litonjua a., Schwartz J, Lovett E, Larson a., Nearing B, et al. Ambient Pollution and Heart Rate Variability. Circulation [Internet]. 2000 Mar 21 [cited 2013 Oct 14];101(11):1267–73. Available from: http://circ.ahajournals.org/cgi/doi/10.1161/01.CIR.101.11.1267 179.  Schwartz J, Litonjua a, Suh H, Verrier M, Zanobetti a, Syring M, et al. Traffic related pollution and heart rate variability in a panel of elderly subjects. Thorax [Internet]. 2005 Jun [cited 2014 May 1];60(6):455–61. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1747419&tool=pmcentrez&rendertype=abstract  180 180.  Rhoden CR, Wellenius G a, Ghelfi E, Lawrence J, González-Flecha B. PM-induced cardiac oxidative stress and dysfunction are mediated by autonomic stimulation. Biochim Biophys Acta [Internet]. 2005 Oct 10 [cited 2014 Apr 24];1725(3):305–13. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16005153 181.  Park SK, O’Neill MS, Vokonas PS, Sparrow D, Schwartz J. Effects of Air Pollution on Heart Rate Variability: The VA Normative Aging Study. Environ Health Perspect [Internet]. 2004 Dec 6 [cited 2013 Apr 29];113(3):304–9. Available from: http://www.ehponline.org/ambra-doi-resolver/10.1289/ehp.7447 182.  Peters A, Liu E, Verrier RL, Schwartz J, Gold DR, Mittleman M, et al. Air pollution and incidence of cardiac arrhythmia. Epidemiology. LWW; 2000;11(1):11–7.  183.  Devlin RB, Ghio a. J, Kehrl H, Sanders G, Cascio W. Elderly humans exposed to concentrated air pollution particles have decreased heart rate variability. Eur Respir J [Internet]. 2003 May 1 [cited 2014 May 1];21(Supplement 40):76S–80s. Available from: http://erj.ersjournals.com/cgi/doi/10.1183/09031936.03.00402403 184.  Brook RD, Urch B, Dvonch JT, Bard RL, Speck M, Keeler G, et al. Insights into the mechanisms and mediators of the effects of air pollution exposure on blood pressure and vascular function in healthy humans. Hypertension [Internet]. 2009 Sep [cited 2014 Apr 29];54(3):659–67. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3706996&tool=pmcentrez&rendertype=abstract 185.  Delfino R, Tjoa T, Gillen D. Traffic-related Air Pollution and Blood Pressure in Elderly Subject With Coronary Artery Disease. Epidemiology [Internet]. 2010 [cited 2014 May 1];21(3). Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3872093/  181 186.  Urch B, Silverman F, Corey P, Brook JR, Lukic KZ, Rajagopalan S, et al. Acute Blood Pressure Responses in Healthy Adults During Controlled Air Pollution Exposures. Environ Health Perspect [Internet]. 2005 May 19 [cited 2013 Sep 25];113(8):1052–5. Available from: http://www.ehponline.org/ambra-doi-resolver/10.1289/ehp.7785 187.  Brook RD. Inhalation of Fine Particulate Air Pollution and Ozone Causes Acute Arterial Vasoconstriction in Healthy Adults. Circulation [Internet]. 2002 Mar 11 [cited 2013 Mar 18];105(13):1534–6. Available from: http://circ.ahajournals.org/cgi/doi/10.1161/01.CIR.0000013838.94747.64 188.  Zanobetti A, Canner MJ, Stone PH, Schwartz J, Sher D, Eagan-Bengston E, et al. Ambient pollution and blood pressure in cardiac rehabilitation patients. Circulation [Internet]. 2004 Oct 12 [cited 2014 Apr 29];110(15):2184–9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15466639 189.  Brook R, Bard R, Morishita M. Hemodynamic, Autonomic, and Vascular Effects of Exposure to Coarse Particulate Matter Air Pollution from a Rural Location. Environ Health Perspect [Internet]. 2014 [cited 2014 Apr 27];(January 2013). Available from: http://ehp.niehs.nih.gov/wp-content/uploads/advpub/2014/3/ehp.1306595.pdf 190.  Oberdörster G, Sharp Z, Atudorei V, Elder A, Gelein R, Lunts A, et al. Extrapulmonary translocation of ultrafine carbon particles following whole-body inhalation exposure of rats. J Toxicol Environ Health [Internet]. 2002 [cited 2014 May 1];64:1531–43. Available from: http://www.tandfonline.com/doi/abs/10.1080/00984100290071658 191.  Wallenborn JG, McGee JK, Schladweiler MC, Ledbetter AD, Kodavanti UP. Systemic translocation of particulate matter-associated metals following a single intratracheal  182 instillation in rats. Toxicol Sci [Internet]. 2007 Jul [cited 2014 Apr 29];98(1):231–9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17434951 192.  Wallenborn JG, Schladweiler MJ, Richards JH, Kodavanti UP. Differential pulmonary and cardiac effects of pulmonary exposure to a panel of particulate matter-associated metals. Toxicol Appl Pharmacol [Internet]. Elsevier Inc.; 2009 Nov 15 [cited 2014 Apr 30];241(1):71–80. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19679144 193.  Münzel T, Sinning C, Post F, Warnholtz A, Schulz E. Pathophysiology, diagnosis and prognostic implications of endothelial dysfunction. Ann Med [Internet]. 2008 Jan [cited 2014 Mar 25];40(3):180–96. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18382884 194.  Verma S. Fundamentals of Endothelial Function for the Clinical Cardiologist. Circulation [Internet]. 2002 Feb 5 [cited 2014 Jan 2];105(5):546–9. Available from: http://circ.ahajournals.org/cgi/doi/10.1161/hc0502.104540 195.  Deanfield JE, Halcox JP, Rabelink TJ. Endothelial function and dysfunction: testing and clinical relevance. Circulation [Internet]. 2007 Mar 13 [cited 2013 Dec 26];115(10):1285–95. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17353456 196.  Halcox JPJ. Prognostic Value of Coronary Vascular Endothelial Dysfunction. Circulation [Internet]. 2002 Jul 22 [cited 2013 Oct 3];106(6):653–8. Available from: http://circ.ahajournals.org/cgi/doi/10.1161/01.CIR.0000025404.78001.D8 197.  Palmer RMJ, Ferrige AG, Moncada S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature [Internet]. 1987 Jun 11;327(6122):524–6. Available from: http://dx.doi.org/10.1038/327524a0  183 198.  Ignarro L, Cirino G, Casini A, Napoli C. Nitric oxide as a signaling molecule in the vascular system: an overview. J Cardiovasc Pharmacol [Internet]. 1999 [cited 2013 Oct 9];34(6):879–86. Available from: http://scholar.google.com/scholar?hl=en&btnG=Search&q=intitle:Nitric+Oxide+as+a+Signaling+Molecule+in+the+Vascular+System:+An+Overview#0 199.  Verma S, Wang C-H, Li S-H, Dumont a. S, Fedak PWM, Badiwala M V., et al. A Self-Fulfilling Prophecy: C-Reactive Protein Attenuates Nitric Oxide Production and Inhibits Angiogenesis. Circulation [Internet]. 2002 Jul 29 [cited 2013 Dec 26];106(8):913–9. Available from: http://circ.ahajournals.org/cgi/doi/10.1161/01.CIR.0000029802.88087.5E 200.  Napoli C, Ignarro LJ. Nitric oxide and atherosclerosis. Nitric Oxide [Internet]. 2001 Apr [cited 2013 Sep 29];5(2):88–97. Available from: http://www.ncbi.nlm.nih.gov/pubmed/11292358 201.  Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci [Internet]. 1990 Feb 1;87 (4 ):1620–4. Available from: http://www.pnas.org/content/87/4/1620.abstract 202.  Mills NL, Törnqvist H, Robinson SD, Gonzalez M, Darnley K, MacNee W, et al. Diesel exhaust inhalation causes vascular dysfunction and impaired endogenous fibrinolysis. Circulation [Internet]. 2005 Dec 20 [cited 2013 Sep 19];112(25):3930–6. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16365212 203.  Frey PF, Ganz P, Hsue PY, Benowitz NL, Glantz S a, Balmes JR, et al. The exposure-dependent effects of aged secondhand smoke on endothelial function. J Am Coll Cardiol  184 [Internet]. 2012 May 22 [cited 2013 Mar 27];59(21):1908–13. Available from: http://www.ncbi.nlm.nih.gov/pubmed/22595411 204.  Widlansky ME, Gokce N, Keaney JF, Vita J a. The clinical implications of endothelial dysfunction. J Am Coll Cardiol [Internet]. 2003 Oct [cited 2013 Sep 22];42(7):1149–60. Available from: http://linkinghub.elsevier.com/retrieve/pii/S073510970300994X 205.  Modena MG, Bonetti L, Coppi F, Bursi F, Rossi R. Prognostic role of reversible endothelial dysfunction in hypertensive postmenopausal women. J Am Coll Cardiol [Internet]. 2002 Aug 7;40(3):505–10. Available from: http://www.ncbi.nlm.nih.gov/pubmed/12142118 206.  Carnovale V, Paradis M-E, Gigleux I, Ramprasath VR, Couture P, Jones PJ, et al. Correlates of reactive hyperemic index in men and postmenopausal women. Vasc Med [Internet]. 2013 Dec 1;18 (6 ):340–6. Available from: http://vmj.sagepub.com/content/18/6/340.abstract 207.  Hamburg NM, Keyes MJ, Larson MG, Vasan RS, Schnabel R, Pryde MM, et al. Cross-sectional relations of digital vascular function to cardiovascular risk factors in the Framingham Heart Study. Circulation [Internet]. 2008 May 13 [cited 2014 May 1];117(19):2467–74. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2734141&tool=pmcentrez&rendertype=abstract 208.  Kuvin JT, Patel AR, Sliney K a, Pandian NG, Sheffy J, Schnall RP, et al. Assessment of peripheral vascular endothelial function with finger arterial pulse wave amplitude. Am Heart J [Internet]. 2003 Jul [cited 2013 Oct 15];146(1):168–74. Available from: http://www.ncbi.nlm.nih.gov/pubmed/12851627  185 209.  Valavanidis A, Vlachogianni T, Fiotakis C. 8-hydroxy-2’ -deoxyguanosine (8-OHdG): A critical biomarker of oxidative stress and carcinogenesis. J Environ Sci Health C Environ Carcinog Ecotoxicol Rev [Internet]. 2009 Apr [cited 2013 Oct 11];27(2):120–39. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19412858 210.  Bräuner EV, Forchhammer L, Møller P, Simonsen J, Glasius M, Wåhlin P, et al. Exposure to ultrafine particles from ambient air and oxidative stress-induced DNA damage. Environ Health Perspect [Internet]. 2007 Aug [cited 2013 Sep 20];115(8):1177–82. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1940068&tool=pmcentrez&rendertype=abstract 211.  Kim JY, Mukherjee S, Ngo L, Christiani DC. Urinary 8-Hydroxy-2´-Deoxyguanosine as a Biomarker of Oxidative DNA Damage in Workers Exposed to Fine Particulates. Environ Health Perspect [Internet]. 2004 Jan 20 [cited 2013 Oct 13];112(6):666–71. Available from: http://www.ehponline.org/ambra-doi-resolver/10.1289/ehp.6827 212.  Han Y-Y, Donovan M, Sung F-C. Increased urinary 8-hydroxy-2’-deoxyguanosine excretion in long-distance bus drivers in Taiwan. Chemosphere [Internet]. 2010 May [cited 2013 Sep 20];79(9):942–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20303570 213.  Wei Y, Han I-K, Hu M, Shao M, Zhang JJ, Tang X. Personal exposure to particulate PAHs and anthraquinone and oxidative DNA damages in humans. Chemosphere [Internet]. 2010 Nov [cited 2012 May 25];81(10):1280–5. Available from: http://dx.doi.org/10.1016/j.chemosphere.2010.08.055 214.  Lee M-W, Chen M-L, Lung S-CC, Tsai C-J, Yin X-J, Mao I-F. Exposure assessment of PM2.5 and urinary 8-OHdG for diesel exhaust emission inspector. Sci Total Environ  186 [Internet]. 2010 Jan 1 [cited 2012 May 30];408(3):505–10. Available from: http://dx.doi.org/10.1016/j.scitotenv.2009.10.012 215.  Buthbumrung N, Mahidol C, Navasumrit P, Promvijit J, Hunsonti P, Autrup H, et al. Oxidative DNA damage and influence of genetic polymorphisms among urban and rural schoolchildren exposed to benzene. Chem Biol Interact [Internet]. 2008 Apr 15 [cited 2012 Apr 18];172(3):185–94. Available from: http://dx.doi.org/10.1016/j.cbi.2008.01.005 216.  Padilla J, Harris RA, Fly AD, Rink LD, Wallace JP. The effect of acute exercise on endothelial function following a high-fat meal. Eur J Appl Physiol [Internet]. 2006;98(3):256–62. Available from: http://link.springer.com/article/10.1007/s00421-006-0272-z# 217.  Bode-Böger SM, Böger RH, Schröder EP, Frölich JC. Exercise increases systemic nitric oxide production in men. J Cardiovasc Risk [Internet]. 1994 Aug;1(2):173–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/7606631 218.  Jungersten L, Ambring A, Wall B, Wennmalm A. Both physical fitness and acute exercise regulate nitric oxide formation in healthy humans. J Appl … [Internet]. 1997 [cited 2014 Aug 20];87:760–4. Available from: http://jap.physiology.org/content/82/3/760.short 219.  Jin RC, Loscalzo J. Vascular Nitric Oxide: Formation and Function. J Blood Med [Internet]. 2010 Aug 1;2010(1):147–62. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3092409&tool=pmcentrez&rendertype=abstract 220.  Hung M-J, Cherng W-J, Hung M-Y, Wu H-T, Pang J-HS. Interleukin-6 inhibits endothelial nitric oxide synthase activation and increases endothelial nitric oxide synthase binding to stabilized caveolin-1 in human vascular endothelial cells. J Hypertens [Internet].  187 2010;28(5). Available from: http://journals.lww.com/jhypertension/Fulltext/2010/05000/Interleukin_6_inhibits_endothelial_nitric_oxide.12.aspx 221.  O’Kroy JA, Loy RA, Coast JR. Pulmonary function changes following exercise. Med Sci Sports Exerc [Internet]. 1992 Dec;24(12):1359—1364. Available from: http://europepmc.org/abstract/MED/1470019 222.  Coast JR, Haverkamp HC, Finkbone CM, Anderson KL, George SO, Herb RA. Alterations in pulmonary function following exercise are not caused by the work of breathing alone. Int J Sports Med. Georg Thieme Verlag Stuttgart{\textperiodcentered} New York; 1999;20(07):470–5.  223.  Machha A, Schechter A. Inorganic nitrate: a major player in cardiovascular health benefits of vegetables? Nutr Rev [Internet]. 2012 [cited 2014 Aug 10];70(6):367–72. Available from: http://onlinelibrary.wiley.com/doi/10.1111/j.1753-4887.2012.00477.x/full 224.  Hord N, Tang Y, Bryan N. Food sources of nitrates and nitrites: the physiologic context for potential health benefits. Am J Clin … [Internet]. 2009 [cited 2014 Aug 18];(6):1–10. Available from: http://ajcn.nutrition.org/content/90/1/1.short 225.  Itamar Medical Limited. Endo-PAT2000 User Manual [Internet]. Caesarea, Israel; 2009. Available from: http://www.itamar-medical.com/images/EndoPAT Brochure March09.pdf 226.  Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, et al. Standardisation of spirometry. Eur Respir J [Internet]. 2005 Aug [cited 2011 Jul 27];26(2):319–38. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16055882 227.  Siemens Healthcare Diagnostics Products GmbH. SIEMENS Dimension Vista System Flex reagent cartridge hsCRP method. Marburg, Germany; 2009. p. 2.   188 228.  R&D Systems Inc. Quantikine ® ELISA Human IL-6 Immunoassay [Internet]. Minneapolis, MN USA; p. 1–14. Available from: http://www.rndsystems.com/pdf/d6050.pdf 229.  Japan Institute For the Control of Aging. “Highly Sensitive 8-OHdG Check” INSTRUCTIONS [Internet]. Fukuroi City, Shizuoka, Japan; p. 1–4. Available from: http://www.jaica.com/e/pdf/8ohdg_kit_hs_instruction.pdf 230.  Coast JR, Welch H. Linear increase in optimal pedal rate with increased power output in cycle ergometry. Eur J Appl Physiol Occup Physiol [Internet]. Springer-Verlag; 1985;53(4):339–42. Available from: http://dx.doi.org/10.1007/BF00422850 231.  Spin Technologies Inc. Vancouver, BC Google Map [Internet]. http://www.runningmap.com/. 2014 [cited 2014 Jun 6]. Available from: http://www.runningmap.com/ 232.  Pichon A, Roulaud M, Denjean A, de Bisschop C. Airway Tone During Exercise in Healthy Subjects: Effects of Salbutamol and Ipratropium Bromide. Int J Sport Med. 26.08.2004 ed. 2005;26(05):321–6.  233.  Weichenthal S, Kulka R, Dubeau A, Martin C, Wang D, Dales R. Supplemental Material. (613):1–13.  234.  Itamar Medical Limited. EndoScore [Internet]. 2014 [cited 2014 Jul 6]. Available from: http://www.itamar-medical.com/EndoPATTM/Patient_Information/EndoPATTM_Basics/EndoScore 235.  Wilson PWF, D’Agostino RB, Levy D, Belanger a. M, Silbershatz H, Kannel WB. Prediction of Coronary Heart Disease Using Risk Factor Categories. Circulation [Internet].  189 1998 May 19 [cited 2014 Jul 10];97(18):1837–47. Available from: http://circ.ahajournals.org/cgi/doi/10.1161/01.CIR.97.18.1837 236.  Bonetti PO, Lardi E, Geissmann C, Kuhn MU, Brüesch H, Reinhart WH. Effect of brief secondhand smoke exposure on endothelial function and circulating markers of inflammation. Atherosclerosis [Internet]. Elsevier Ireland Ltd; 2011 Mar [cited 2014 Apr 19];215(1):218–22. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21215401 237.  Kajbafzadeh M. Assessing the impacts of traffic-related and woodsmoke particulate matter on subclinical measures of cardiovascular health: a HEPA filter intervention study. University of British Columbia (Vancouver); 2014. p. 1–155.  238.  Forchhammer L, Møller P, Riddervold IS, Bønløkke J, Massling A, Sigsgaard T, et al. Controlled human wood smoke exposure: oxidative stress, inflammation and microvascular function. Part Fibre Toxicol [Internet]. BioMed Central Ltd; 2012 Jan [cited 2014 Mar 28];9(1):7. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3369202&tool=pmcentrez&rendertype=abstract 239.  Hallmark R, Patrie JT, Liu Z, Gaesser G a, Barrett EJ, Weltman A. The effect of exercise intensity on endothelial function in physically inactive lean and obese adults. PLoS One [Internet]. 2014 Jan [cited 2014 Jul 22];9(1):e85450. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3896361&tool=pmcentrez&rendertype=abstract 240.  Gólczewski T, Lubiński W, Chciałowski A. A mathematical reason for FEV1/FVC dependence on age. Respir Res [Internet]. 2012 Jan [cited 2014 Mar 4];13:57. Available from:  190 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3461438&tool=pmcentrez&rendertype=abstract 241.  Beyer KMM, Kaltenbach A, Szabo A, Bogar S, Nieto FJ, Malecki KM. Exposure to neighborhood green space and mental health: evidence from the survey of the health of wisconsin. Int J Environ Res Public Health [Internet]. 2014 Jan [cited 2014 Apr 2];11(3):3453–72. Available from: http://www.ncbi.nlm.nih.gov/pubmed/24662966 242.  Pretty J, Peacock J, Sellens M, Griffin M. The mental and physical health outcomes of green exercise. Int J Environ Health Res [Internet]. 2005 Oct [cited 2014 Mar 19];15(5):319–37. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16416750 243.  Barton J, Pretty J. What is the best dose of nature and green exercise for improving mental health? A multi-study analysis. Environ Sci Technol [Internet]. 2010 May 15;44(10):3947–55. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20337470 244.  Horiuchi M, Endo J, Akatsuka S, Uno T, Hasegawa T. Influence of Forest Walking on Blood Pressure , Profile of Mood States and Stress Markers from the Viewpoint of Aging. 2013;9–17.  245.  Park BJ, Tsunetsugu Y, Kasetani T, Kagawa T, Miyazaki Y. The physiological effects of Shinrin-yoku (taking in the forest atmosphere or forest bathing): evidence from field experiments in 24 forests across Japan. Environ Health Prev Med [Internet]. 2010 Jan [cited 2014 Apr 8];15(1):18–26. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2793346&tool=pmcentrez&rendertype=abstract  191 246.  Carney RM, Freedland KE, Veith RC. Depression, the autonomic nervous system, and coronary heart disease. Psychosom Med [Internet]. 2005 [cited 2014 Apr 2];67 Suppl 1(1):S29–33. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15953797 247.  Steptoe A, Willemsen G, Owen N, Flower L, Mohamed-Ali V. Acute mental stress elicits delayed increases in circulating inflammatory cytokine levels. Clin Sci (Lond) [Internet]. 2001 Aug;101(2):185–92. Available from: http://www.ncbi.nlm.nih.gov/pubmed/11473494 248.  Hamer M, Steptoe A. Association between physical fitness, parasympathetic control, and proinflammatory responses to mental stress. Psychosom Med [Internet]. 2007 [cited 2014 Apr 20];69(7):660–6. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17724255 249.  Huang C-J, Webb HE, Zourdos MC, Acevedo EO. Cardiovascular reactivity, stress, and physical activity. Front Physiol [Internet]. 2013 Jan [cited 2014 Mar 27];4(November):314. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3819592&tool=pmcentrez&rendertype=abstract 250.  Forcier K, Stroud LR, Papandonatos GD, Hitsman B, Reiches M, Krishnamoorthy J, et al. Links between physical fitness and cardiovascular reactivity and recovery to psychological stressors: A meta-analysis. Health Psychol [Internet]. 2006 Nov [cited 2014 Apr 20];25(6):723–39. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17100501 251.  McCubbin JA, Cheung R, Montgomery TB, Bulbulian R, Wilson JF. Aerobic Fitness and Opioidergic Inhibition of Cardiovascular Stress Reactivity. Psychophysiology [Internet]. Blackwell Publishing Ltd; 1992;29(6):687–97. Available from: http://dx.doi.org/10.1111/j.1469-8986.1992.tb02047.x  192 252.  Webb HE, Weldy ML, Fabianke-Kadue EC, Orndorff GR, Kamimori GH, Acevedo EO. Psychological stress during exercise: cardiorespiratory and hormonal responses. Eur J Appl Physiol [Internet]. 2008 Dec [cited 2014 Apr 20];104(6):973–81. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18762969 253.  Babisch W. Traffic Noise and Cardiovascular Disease: Epidemiological Review and Synthesis. Noise Heal [Internet]. 2000 [cited 2014 Apr 21];2(8):9–32. Available from: http://www.noiseandhealth.org/article.asp?issn=1463-1741;year=2000;volume=2;issue=8;spage=9;epage=32;aulast=Babisch 254.  Maschke C, Rupp T, Hecht K. a review of present scientific findings with noise. 2000;53:45–53.  255.  Lundberg U. Coping with stress: neuroendocrine reactions and implications for health. 1999;1(4):67–74. Available from: http://www.noiseandhealth.org/text.asp?1999/1/4/67/31722 256.  Fritschi L, Brown L, Kim R. Burden of disease from environmental noise: Quantification of healthy life years lost in Europe [Internet]. Copenhagen; 2011 p. 1–126. Available from: http://www.env-health.org/IMG/pdf/25052011_Conference_on_Noise_-_Rokho_Kim_WHO_Burden_of_disease_Presentation.pdf 257.  Babisch W. Stress hormones in the research on cardiovascular effects of noise. Noise Health [Internet]. 2003;5(18):1–11. Available from: http://www.ncbi.nlm.nih.gov/pubmed/12631430 258.  UK Environment Agency. Technical Guidance Note (Monitoring) M1 Sampling requirements for stack emission monitoring [Internet]. p. 1–33. Available from:  193 https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/296772/geho0110brro-e-e.pdf 259.  Gray L. An Explanation for Particle Sampling in a Moving Gas Stream Within a Duct Using an Unshrouded and Shrouded Probe [Internet]. yosemite.epa.gov. [cited 2014 Aug 12]. p. 1–11. Available from: http://yosemite.epa.gov/rpd/neshaps_pid.nsf/767c54327a568757852573150057f167/774703c9da54d0c186256f1c0067744c/$FILE/Sampling Method_49.pdf 260.  Zhu Y, Yu N, Kuhn T, Hinds WC. Field Comparison of P-Trak and Condensation Particle Counters. Aerosol Sci Technol [Internet]. 2006 Jul [cited 2014 Jul 6];40(6):422–30. Available from: http://www.tandfonline.com/doi/abs/10.1080/02786820600643321 261.  Burkart J, Steiner G, Reischl G, Moshammer H, Neuberger M, Hitzenberger R. Characterizing the performance of two optical particle counters (Grimm OPC1.108 and OPC1.109) under urban aerosol conditions. J Aerosol Sci [Internet]. 2010 Oct [cited 2014 Jul 3];41(10):953–62. Available from: http://www.sciencedirect.com/science/article/pii/S0021850210001679 262.  Woodcock J, Givoni M, Morgan AS. Health impact modelling of active travel visions for England and Wales using an Integrated Transport and Health Impact Modelling Tool (ITHIM). PLoS One [Internet]. 2013 Jan [cited 2014 Jan 14];8(1):e51462. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3541403&tool=pmcentrez&rendertype=abstract 263.  Teschke K, Reynolds CCO, Ries FJ, Gouge B, Winters M. Bicycling: Health Risk or Benefit? Univ Br Columbia Med J. 2012;3(March):6–11.   194 264.  Rabl A, de Nazelle A. Benefits of shift from car to active transport. Transp Policy [Internet]. 2012 Jan;19(1):121–31. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0967070X11001119 265.  Johan de Hartog J, Boogaard H, Nijland H, Hoek G. Do the health benefits of cycling outweigh the risks? Environ Health Perspect [Internet]. 2010 Aug [cited 2013 Sep 20];118(8):1109–16. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2920084&tool=pmcentrez&rendertype=abstract 266.  Maizlish N, Woodcock J, Co S, Ostro B, Fanai A, Fairley D. Health cobenefits and transportation-related reductions in greenhouse gas emissions in the San Francisco Bay area. Am J Public Health [Internet]. 2013 Apr [cited 2014 Mar 28];103(4):703–9. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3673232&tool=pmcentrez&rendertype=abstract 267.  Holm AL, Glümer C, Diderichsen F. Health Impact Assessment of increased cycling to place of work or education in Copenhagen. BMJ Open [Internet]. 2012 Jan [cited 2014 Mar 28];2(4). Available from: http://www.ncbi.nlm.nih.gov/pubmed/22833650 268.  Creutzig F, Mühlhoff R, Römer J. Decarbonizing urban transport in European cities: four cases show possibly high co-benefits. Environ Res Lett [Internet]. 2012 Dec 1 [cited 2014 Mar 28];7(4):044042. Available from: http://stacks.iop.org/1748-9326/7/i=4/a=044042?key=crossref.f9b392df118c03d24e902ff914b851e1 269.  Woodcock J, Tainio M, Cheshire J, O’Brien O, Goodman A. Health effects of the London bicycle sharing system: health impact modelling study. BMJ [Internet]. 2014 Feb 13 [cited  195 2014 Feb 14];348(feb13 1):g425–g425. Available from: http://www.bmj.com/cgi/doi/10.1136/bmj.g425   196 Appendix A – CAPaH Protocol  V3 revised October 25, 2011 Cycling, Air Pollution, and Health Study – Protocol  1. Introduction and Objectives Bicycling as a form of transportation has recently gained the attention of government agencies as a way to increase the level of physical activity in order to improve population health, as well as reduce air pollutant emissions, which are themselves a health concern (Health Canada, 2006).  While cycling is promoted as a form of active transportation as a means to increase physical activity and achieve health benefits, there are risks associated with cycling. In addition to the well-know increased risk of injuries and fatality compared to car drivers (Reynolds et al, 2009), there is also concern regarding potential health risks associated with exposures to air pollutants while cycling in traffic (de Nazelle & Nieuwenhuijsen, 2009) Research has evaluated the exposure levels of bicycle commuters to air pollutants, including ultrafine particles (<0.1µm aerodynamic diameter), fine particles (2.5-10µm aerodynamic diameter; PM2.5), coarse particles (≥10µm aerodynamic diameter; PM10), carbon monoxide, ozone, sulphates, and compounds such as benzene and toluene (Bevan et al., 1991, van Wijnen et al., 1995, Kaur et al., 2007, Thai et al., 2008, Strak et al., 2010). Most exposure research has been done in European centres, and provides limited information about the work rate of the subjects participating in the study. Not knowing the level at which study participants are exercising limits applicability of data to the situation of cyclists, as exposure and actual dose of pollutants to subjects cannot be correlated without information about minute ventilation.   Of the studies that have considered exercise intensity of the cyclists, all results show increased ventilation rates compared to other commuters, mostly automobile drivers, and car and bus passengers. van Wijnen et al. showed that cyclists exhaled an average of 2.3 times the amount of air exhaled by car drivers (1995). The most recent study, by Int Panis et al. (2010), demonstrated that ventilation frequency increased by 1.6 times compared to when the subjects were automobile passengers, and that tidal volume increased by a factor of 2.6 while cycling. In all, the minute ventilation of females increased by a factor of 4.1, while the minute ventilation of males increased by a factor of 4.5 when comparing bicycling to being a car passenger (Int Panis et al., 2010). In total, only five known studies have considered the impact of ventilation rate and dose on their subjects (van Wijnen et al., 1995, Rank et al., 2001, O’Donoghue et al., 2007, Zuurbier et al., 2009, & Int Panis et al., 2010).  Markers of inflammation found in blood have been correlated to exposure to particulate matter, with many of these biomarkers have being associated with risk of cardiovascular morbidity and mortality (Delfino et al., 2008). Endothelial function as a result of indoor air particles has been compared using high-efficiency particle air (HEPA) filters, and microvascular function has been shown to be significantly improved in individuals when exposed to filtered indoor air while at rest (Brauner et al., 2007). Endothelial health can be assessed in part by measures of inflammation found circulating through the blood such as C-reactive protein level, which has been recognized as a strong predictor of cardiovascular disease risk (Verma et al., 2003).    197 Injury risk is a concern for cyclists and those who have considered cycling for recreational and utilitarian purposes. While limited data exists on this subject, it is notable that the fatality risk per distance travelled for pedestrians in the United States is 23 times higher than individuals traveling by car, where for cyclists the risks of fatality are 12 times higher than automobile occupants (Pucher & Dijkstra, 2003). A survey of bicycle commuters in North America, with Canadian respondents representing 3% of the total group surveyed, reported a rate of 37.1 crashes per one million kilometers cycled; alternatively, the cyclists experienced one accident for every 27 000 kilometres cycled (Moritz, 1997).   Understanding how air pollution can impact health of bicycle commuters may allow municipalities to recognize what the needs of cyclists are, to enable cyclists to participate in this activity in the most healthful way, and to help provide long-term guidance for infrastructure development. There is also a need to better appreciate how much air pollution levels affect the health of bicycle commuters. This may help individuals plan ways to reduce their exposure, and more generally, it will help policy makers to better understand how they can advise the population to participate in physical activity in the presence of different concentrations of air pollution.   The purpose of this study is to better understand the relationship between the measurements of air pollution exposure from road traffic, and the resulting health impacts by looking at lung function, endothelial function, and C-reactive protein to identify the presence of systemic inflammation within the body.  Hypotheses  Cycling for 1 hour along a route with higher levels of ultrafine particles results in a smaller increase in post vs pre-testing lung function, compared to cycling along a route with lower air pollution levels.  Cycling for 1 hour along a route with higher levels of ultrafine particles results in a smaller increase in post vs pre-testing endothelial function, compared to cycling along a route with lower air pollution levels.   Methodology  Subject recruitment Subjects will be recruited from the university and recreational cycling community, using posters to notify individuals of the upcoming study (Appendix B). Individuals interested in participating will be invited to contact the researcher by e-mail. The researcher will screen prospective subjects over the phone (using the Screening Questionnaire- Appendix E), and answer any questions or concerns. Those individuals who meet initial inclusion criteria will be provided with an introductory letter (Appendix C) and consent form (Appendix D), and each prospective subject will have a minimum of 48 hours to decide if they would like to participate in the study before they will be scheduled for their first trial.   Inclusion criteria Non-smoking recreationally active individuals between the ages of 19 and 39 years old are invited to participate. Subjects may be male or female; female subjects will be asked to provide information  198 about whether they are currently using contraceptive medications, and those taking oral contraceptives or not taking any contraceptive medications will be allowed to participate. Subjects must be free of all chronic cardiovascular and pulmonary disorders that may interfere with their ability to safely participate in this study, or that may require them to take medications that may interfere with the results of the spirometry, endothelial function, or measurements of inflammatory mediators in blood. Subjects must also feel confident in their ability to safely ride a bicycle along designated city bike routes. All questions relating to these criteria are found in Appendix E.  Exclusion criteria  Cigarette smokers, those who use other nicotine products, or anyone who has smoked any other substance within the last 24 hours will be excluded; subjects must also not work or live in a home where they are regularly exposed to second-hand smoke. In addition, subjects occupationally exposed to high levels of dusts and our fumes will be not be allowed to participate. Individuals outside of the age range of 19 to 39 years old may not participate. Individuals answering “Yes” to any of the questions of the PAR-Q screening questionnaire (Appendix E) will be excluded from the study, as will any individuals presenting with a blood pressure level outside of the ranges of 140 to 90mmHg for systolic blood pressure, and 89 to 60mmHg for diastolic blood pressure. Anyone with cardiovascular or pulmonary conditions, including asthma or a respiratory infection, may not participate in this study. Individuals unable to safely fit on the bicycle will also be excluded from the testing. Females taking contraceptive medications other than oral contraceptives will not be able to participate in this study.   Subjects will be required to avoid analgesics and vitamin supplements for 72 hours before testing. Caffeine and alcohol must also be avoided for 24 hours prior to a trial, with test sessions being rescheduled if necessary.    Method for each trial The subject will schedule the first testing date upon signing the consent form (Appendix D). The evening prior to each test session, the subject will be contacted to ensure diet and medication restrictions allow a valid trial to be carried out.   Subjects who choose to participate will repeat these steps for each bicycle route. The routes will have a pre-determined order for each subject (based on chance), but the route type (better air quality vs. poorer air quality) will not be disclosed to the subject until after the end of the second testing session.   1. The subject will arrive at the lab on the first testing date, and fill out the Pre-test Questionnaire (Appendix F); this includes question about their diet in the last 12 hours. The subject will also be screened for respiratory symptoms using the Common-Cold Questionnaire (Appendix G).   2. The subject will be asked to lie quietly on a bed, and have their blood pressure taken; individuals whose blood pressure exceeds 140/90mmHg will be excluded from the study and encouraged to visit their physician.  The endothelial peripheral arterial tonometry (EndoPAT) test method will be explained to the subject at this point.    199 3. The Endo-Pat measurement will be conducted with subjects lying supine with their forearms comfortably resting on the blue arm supports. A blood pressure cuff will be placed on the upper non-dominant arm while the dominant arm serves as a control. Fingertip sensors are placed on the index finger of both hands. After a 5-minute equilibrium period, the blood pressure cuff is inflated to a suprasystolic pressure (to induce brachial arterial occlusion) for 5 minutes and then released with the EndoPAT 2000 recording for a further 5 minutes.   4. Pre-bicycle spirometry testing will be conducted according to the American Thoracic Society standards. The subject will be seated upright wearing a nose clip, and instructed to inhale rapidly until their lungs are filled as much as possible with air. The subject would then exhale as hard and as quickly as they possibly can into the mouthpiece of the spirometer, until they can no longer expel any air while still remaining upright. This test will be repeated a minimum of three times to ensure repeatability.   5. The pre-trial blood draw will occur, where a maximum of 10mL of blood will be taken.  6. An adjustable bicycle will be sized to fit the subject, and the adjustments will be noted for reproducibility on the next trial. The bicycle will be instrumented with a PowerTap power meter that measures the power output of the rider (see Appendix K), a GlobalSAT GPS data logger (see Appendix J), with a TSI P-Trak ultrafine particle counter (see Appendix H), and a GRIMM optical particle counter (see Appendix I) also mounted on the bicycle. The total weight of the combined instruments is approximately 4 kg. The subject will be given the opportunity to ride the bicycle to ensure they feel comfortable.   The subject will be asked to bring his or her own bicycle helmet for use during the ride, along with a water bottle; for those subjects who do not bring a helmet, we will provide and fit one for their use. The subject will be encouraged to consume water as needed throughout the ride. Bottled water will be provided for those subjects who do not bring their own water bottle. The bicycle will be equipped with all reflectors and a bell as required by law. The subject will be instructed to follow all road safety laws, and told not to compromise their safety at any time. The cycling route will be explained to the subject, and a map of the route will be provided.  7. The subject will be instructed to ride at a work rate that they feel they can sustain for 1 hour. This intensity will be relative; the subject will be told to work at a relative effort level of 12 to 13 on the Borg Scale (American College of Sports Medicine, 2007). Where on the Borg scale from 6-20, 20 is the hardest level of work, one that can only be sustained for approximately 10 seconds (such as the effort of sprinting); an effort level of 6 is equivalent to sitting down, at rest.  8. The subject will exit the research laboratory on the instrumented and fitted bicycle, followed on bicycle by a research assistant who is able to provide support to the subject should the instrumented bicycle malfunction in any way; the research assistant will always cycle behind the subject, and have a cellular phone on their person in the case that any emergency situation arises. The subject will be instructed to cycle for 30 minutes based on the time indicated on the bike computer. The investigator will at this point ensure that all instruments are recording properly, and inform the subject to turn around and follow the same route back to the starting point.   200 9. Once the subject has returned to the laboratory, they will be invited to stretch and cool down, and then sit quietly for 5 minutes. The EndoPAT test will then be repeated.  10. Once the subject’s breathing rate has returned to a resting level, the spirometry test will be repeated according to American Thoracic Society Standards.  11. The post-trial blood sample will be collected in the same manner as the pre-trial collection.  12. The subject will be thanked for their time, and invited to schedule a second testing session (when applicable)  13. When applicable, subjects completing the second test session will have their heart rate and power output information (recorded by the PowerTap Power Meter bicycle computer) assessed by the testing staff. Staff will look at the maximum heart rate attained, and the average power output attained during the bicycle rides. The maximum heart rate and average power output values will be used as guidelines to know when the submaximal stepwise exercise test should stop.  The sub-maximal exercise test is designed to provide information about minute ventilation and heart rate, allowing this relationship to be graphed against the power output values that were measured during the bicycle ride outside.   Wearing a heart rate monitor strap, subjects sit quietly on the bicycle in their most preferred riding position until their heart rate remains the same for 2 minutes. They are then asked to put on the nose plug, and breathe into the respirometer for 30 seconds. At the end of the 30 second period, the minute ventilation is multiplied by 2 (to give Litres per minute), and the heart rate is recorded. This is the “resting minute ventilation on the bicycle”.   Next, the subject is asked to start pedaling, and then is told the workload will increase on the bicycle every two minutes while they ride. Subjects commence by cycling at a resistance level 20 watts (for female subjects) or 30 watts (for males). The first minute is an adaptation period, while during the second minute of each two-minute interval, a 30-second measurement is taken to determine minute ventilation at that work load. The heart rate is recorded at the end of the minute ventilation measurement.  Form 6 is used to record this data.  The next interval is at a resistance level of 40 watts and 60 watts (females and males, respectively). Each of the steps increase by 20 watt increments for females, and 30 watts for males through the remainder of the protocol.  Subjects will be encouraged to consume water whenever they are not using the respirometer in order to maintain adequate hydration status.  Subjects are asked to stop pedaling upon reaching the following points during the test:  1- discomfort or tiredness causes them to want to discontinue the test  2- the highest heart rate we recorded during the ride has been obtained  3- the resistance level on the bicycle is too difficult to continue turning the pedals above  a cadence of approximately 50 revolutions per minute  4- the participant otherwise wishes to stop for any reason   201 Upon completion of the cycling test, participants will be encouraged to walk around to cool down, and stretch. Participants will be then be thanked for their time and participation.   Research Outcomes Air pollution exposure and health impacts are measured as participants cycle along local bicycle routes. This research will allow us to evaluate the dose of air pollution to individuals who use bicycle routes for recreational and utilitarian purposes, and will provide information to cities to enable the design of recreational infrastructure with health effects in mind.  Analysis The primary health endpoints are microvascular function (endothelial dysfunction) assessed using a non-invasive finger-probe test (Endo-PAT2000, Itamar Medical, Israel), a spirometer (KoKo spirometer, nSpire Health, Longmont, Colorado, USA) that measures volume and time for each breath will be used to test lung function, and a blood test to measure C-reactive protein. A standard bicycle with attached PowerTap (Powertap Comp, Saris Cycling Group, Madison WI) will be used to measure power output during the ride along each bicycle route, and then a Velotron Dynafit Pro cycle ergometer (Racermate Inc, Seattle WA) will be used to set the mean power output while the subject rides in the lab while their ventilation rate and tidal volume are measured with a respirometer (Spirolab II, Medical International Research, Rome, Italy).  Paired t-tests will be used to assesses changes in pre-post cycling difference in reactive hyperaemia index (RHI) as measured by the EndoPAT (and change in level of C-reactive protein, CRP) comparing clean vs polluted route, using R Statistics for data analysis (R Development Core Team, Vienna, Austria). In addition we will model the effect on RHI change (and CRP change) of pollutant inhaled dose (concentration x minute ventilation) as a continuous variable in a mixed model with a random effect at the subject level and fixed effects for gender (and possibly also age, depending on the age range of subjects).   Our sample size estimation is based upon the a previous exposure experiment by Brauner et al. (2008) in which endothelial dysfunction was the main outcome measurement and where RHI in elderly (healthy) adults was compared following exposures to traffic particles with and without operation of a HEPA-filter air cleaner in subjects’ homes. Based on the measured difference in RHI from this study (alpha= 0.05, and beta=0.1) the calculated required sample size is 20 people.  and the exposure levels in our proposed study, based on the paper by Thai et al. (2008) in which levels of air pollutants were measured while cycling on different bicycle routes in Vancouver will be approximately 10x greater than in the study of Brauner et al, however for a shorter length of time. Further, since the subjects in our study will be exercising (increased tidal volume and breathing rate during bicycle commuting results in an increase of 4.3 in minute ventilation, meaning a bicycle commuter breathes in 4.3 times the amount of air, and associated pollutants, compared to an individual at rest (Int Panis et al., 2010)) their inhaled doses will be much larger (approximately 40-fold). Balanced against this much larger inhaled dose, compared to Brauner et al, is the fact that our subjects will be young (healthy) adults and the shorter exposure duration. For these reasons we have targeted a somewhat larger sample size of 40 subjects.      202 References  American College of Sports Medicine (2007). Perceived Exertion. ACSM Current Comment. http://www.acsm.org/AM/TemplateRedirect.cfm?Template=/CM/ContentDisplay.cfm&ContentID=8648&Section=Updated_single_page  Bevan, M., Proctor, C., Baker-Rogers, J., & Warren, N. (1991). Exposure to carbon monoxide, respirable suspended particulates, and volatile organic compounds while commuting by bicycle. Environ Sci Technol 25:788-791  Bräuner, E., Moller, P., Forchhammer, L., Barregard, L., Gunnarsen, L., Afshari, A., Wahlin, P., Glasius, M., Dragsted, L., Basu, S., Raaschou-Nielsen, O., & Loft, S. (2008). Exposure to ambient concentration of particulate air pollution does not influence vascular function or inflammatory pathways in young healthy individuals. Am J Respir Crit Care Med 177(4):419-425  Delfino, J., Staimer, N., Tjoa, T., Polidori, A., Arhami, M., Gillen, D., Kleinman, M., Vaziri, N, Longhurst, J., Zaldivar, F., & Sioutas, C. (2008). Circulating biomarkers of inflammation, antioxidant activity, and platelet activation are associated with primary combustion aerosols in subjects with coronary artery disease. Env Health Perspec 116(7):898-906  de Nazelle A, Nieuwenhuijsen M. (2010)  Integrated health impact assessment of cycling. Occup Environ Med. 67(2):76-7. Epub 2009 Oct 9.  Health Canada (2006). Let’s Talk About Health and Air Quality. Accessed February 22, 2010. http://www.hc-sc.gc.ca/ewh-semt/air/out-ext/effe/talk-a_propos-eng.php  Int Panis, L., de Geus, B., Vandenbulcke, G., Willems, H., Degraeuwe, B., Bleux, N., Mishra, V., Thomas, I., & Meeusen, R. (2010). Exposure to particulate matter in traffic: A comparison of cyclists and car passengers. Atmos Environ 44:2263-2270  Kaur, S., Nieuwenhuijsen, M., & Colvile, R. (2007). Fine particulate matter and carbon monoxide exposure concentrations in urban street transport microenvironments. Atmos Environ 41:4781-4810  Moritz, W. (1997). A survey of North American bicycle commuters-design and aggregate results. Transp Res Rec 1578:91-101  O’Donoghue, R., Gill, L., McKevin, R., & Broderick, B. (2007). Exposure to hydrocarbon concentrations while commuting or exercising in Dublin. Environ Int 33: 1-8  Pucher, J., & Dijkstra, L. (2003). Promoting safe walking and cycling to improve public health: lessons from the Netherlands and Germany. Am J Public Health 993:1509-1516  Rank, J., Folke, J., & Jespersen, P. (2001) Difference in cyclists and car drivers exposure to air pollution from traffic in the city of Copenhagen. Sci Total Environ 279:131-136  203 Strak, M., Boogaard, H., Meliefste, K., Oldenwening, M., Zuurbier, M., Brunekreef, B., & Hoel, G. (2010) Respiratory health effects of ultrafine and fine particle exposure in cyclists. Occup Environ Med 67:118-124  Thai, A., McKendry, I., & Brauer, M. (2008). Particulate matter exposure along designated bicycle routes in Vancouver, British Columbia. Sci Total Environ 405:26-35  van Wijnen, J., Verhoeff, A., Jans, H., & van Bruggen, M. (1995). The exposure of cyclists, car drivers and pedestrians to traffic-related air pollutants. Int Arch Occup Environ Health 67:187-193  Verma, S., Buchanon, M., & Anderson, T. (2003) Endothelial Function Testing as a Biomarker of Vascular Disease. Circulation 108: 2054-2059.  Zuurbier, M., Hoek, G., van den Hazel, P., & Brunekreef, B. (2009). Minute ventilation of cyclists, car and bus passengers: an experimental study. Environ Health 8:48  204 Appendix B – CAPaH Advertisement Are you interested in cycling, air quality and impacts on your health?   We invite you to participate in a research study on the health effects of air pollution among cyclists.   We are inviting recreationally active people aged 19-39 years to participate in a research study related to the health impacts of cycling in an urban environment.  You will be asked to bicycle along two routes in Vancouver, while instruments on a bicycle measure the air pollution you are exposed to during the ride. You will be asked to provide a small blood sample and complete a non-invasive test of blood vessel health (similar to a blood pressure measurement) before and after each ride.  Each participant will need to bicycle for one hour on two occasions.  Subjects must be non-asthmatic, and have no other cardiovascular or respiratory diseases.  If you are interested in learning more about the study, please contact: ubc.bike.study@gmail.com   Principal Investigator: Dr. Michael Brauer, Professor School of Environmental Health University of British Columbia            ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com   ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com  ubc.bike.study@gmail.com   205 Appendix C – CAPaH Introductory Letter V3   Cycling, Air Pollution, and Health Study   We are a team of researchers from the University of British Columbia (UBC) who are investigating the health effects of air pollution. We are writing to ask if you would be interested in participating in this study. Recreationally active individuals aged 19-39 may be eligible to participate in the study. People who smoke or have health conditions such as asthma cannot be considered for participation.  We would like to measure the levels of air pollution present while you ride a bicycle along street routes in Vancouver, and compare some measurements of health before and after the ride.  Automobile traffic is one source of air pollution in the Vancouver area that can negatively affect health.  Choosing bike routes in areas where there is less car traffic may decrease the exposure of individuals to air pollution.  The study will take place between July 2, 2010 and November 1, 2011. You will be asked to ride an instrumented bicycle for one hour along two predetermined routes. Instruments on the bicycle will measure the levels of air pollutants in real time during each bike ride. We will measure your lung function using a simple test lasting 5 minutes where you exhale into a tube, and take a small blood sample from your arm. We will also measure the health of your blood vessels in a test lasting approximately 20 minutes that involves having a blood-pressure cuff on one arm, and a sensor on each index finger. Lastly, we will measure your breathing rate while you are cycling on an exercise bike. The tests should cause only minor discomfort. All of these tests will be done before and after cycling on each of the routes. We expect each of the two testing sessions to last approximately 3.5 hours.   All data that we collect will be kept in secure facilities, and all results will remain anonymous.    If you are interested in participating, or would like further details about the study and the procedures to be used, please contact Christie Cole by e-mail at ubc.bike.study@gmail.com. Alternately you can contact myself at the number listed below.  Thank you for your time and attention, Sincerely,    Dr. Michael Brauer, Professor  School of Population and Public Health  Faculty of Medicine, Respiratory Division University of British Columbia  206 Appendix D – CAPaH Letter of Consent   SUBJECT INFORMATION AND CONSENT FORM Cycling, Air Pollution, and Health Study  Principal Investigator: Dr. Michael Brauer ScD School of Population and Public Health Department of Medicine, Respiratory Division The University of British Columbia Phone: (604) 822-9585 Email: michael.brauer@ubc.ca  Co-Investigators: Christie Cole BSc (MSc student) School of Population and Public Health The University of British Columbia Phone: (604) 827-3509 Email: ubc.bike.study@gmail.com  Dr. Michael Koehle, MD PhD School of Human Kinetics, Division of Sports Medicine Faculty of Medicine The University of British Columbia Phone: (604) 822-9331  Dr. Christopher Carlsten, MD MPH Department of Medicine, Respiratory Division The University of British Columbia Phone: (604) 875-4729   1. Introduction  You are being invited to participate in this study because you are a recreationally active individual between the ages of 19 and 39 years old.  2. Your participation is voluntary Your participation in this study is entirely voluntary. If you decide to take part in this study, you are still free to withdraw at any time and without giving any reasons for your decision. Please take time to read the following information before you decide.  3. Who is conducting the study? This study is being conducted by researchers at The University of British Columbia.   4. Background Air pollution is present in urban areas due to a variety of sources, including motor vehicles. Cycling is a common transportation method used around the world, including in urban areas where air pollutants are present. Cycling is also recommended as a method of exercise with beneficial impacts on health. Higher levels  207 of air pollution are known to cause negative health effects, especially in individuals with cardiovascular problems, but we are still learning how air pollution affects healthy people who are exercising. 5. What is the purpose of the study? This study will examine the health impacts of cycling along bicycle routes with known levels of air pollution. The exercise level of each individual will be used to determine his or her dose of air pollutants, and several measures of health will be related to this dose. This study will help provide advice to urban exercisers in choosing where and when to be active, in consideration of any observed health impacts associated with air pollution. In addition, study results will be useful for transportation planners in designing bicycle routes.  6. Who can participate in the study? Recreationally active, non-smoking females and males between the ages of 19 and 39 years old, and who can safely ride the study bicycle.  7. Who should not participate in the study? Anyone who smokes, lives in a smoking household, or works in an environment where they are exposed to high levels of smoke or dusts cannot be included. People with known allergies to pollen, mould or other substances present outdoors may not participate. Individuals who have been told by a doctor that they have asthma or other chronic lung conditions, a cardiovascular disease or a heart condition, or other chronic health condition such as diabetes, may not participate. Individuals taking certain medications (such as for blood pressure, heart conditions, or allergies) may also be excluded. People outside the age range of 19 to 39 years old may not participate in this study. Those individuals reporting symptoms including bone or joint problems limiting activity, chest pain at rest, loss of balance, or dizziness will not be able to participate in this study. Individuals who could be pregnant or that are breast-feeding may not participate. Any individuals who must take medications on a daily basis on the advice of their physician (with the exception of oral contraceptive pills) may not participate.  Lastly, individuals who do not feel they may confidently and safely bicycle along designated bicycle routes within the city, because of lack of cycling experience, fear of traffic, poor hearing or vision, or for any other reason that may not have already been mentioned will not be able to participate in this study.  8. What does the study involve?  Overview of the Study The study will take place during the summer and fall of 2010-11 in Vancouver, British Columbia. Approximately 30 people will be enrolled in this study. On two separate testing days, each person will ride an instrumented bicycle along a bicycle route for 1 hour, and health measurements will be collected before and after each bike ride. Your lung function and a test of the health of your blood vessels will be assessed using non-invasive tests, and you will be asked to provide a small blood sample from your arm. You will wear a heart rate monitor, and a small computer will record your work rate as you cycle. Instruments on the bicycle will record the levels of air pollutants throughout the hour. You will then have your lung and blood vessel function assessed using the same test methods as before. Lastly, we will take a second small sample of blood. At the end of one of the bike rides, you will also ride a stationary bike for about 15 minutes while breathing normally into a mouthpiece connected to a respirometer; this is a device that measures how quickly and how much air you breathe into your lungs while you are exercising, and it does not require any extra effort to breathe.  Here is a list of the specific procedures that will be included if you decide to join this study:  The time for each trial will be approximately 3.5 hours, which includes preparation time, and the pre- and post-testing. Testing will occur on two different weekday mornings, at least two weeks apart. Female participants will be scheduled to be tested during the first week of their menstrual cycle, as variable hormone levels can change test results. You will be asked to consume the same breakfast at the same time on the mornings of both of the testing days. The common cold, the diet that you consume, the time during the  208 menstrual cycle, and medications can impact a number of the tests. For this reason you will be contacted before each testing session, so that information about your diet and these other factors can be recorded. You are not required to answer all questions. You are only required to divulge information to us that you feel comfortable discussing.   Before and after each 1-hour ride, you will have your lung function tested using a spirometer. This test involves breathing into a mouthpiece while it measures the amount of air breathed out, and how rapidly the air is exhaled. This test takes approximately 5 minutes, and will be done according to standard procedures (American Thoracic Society) that are widely used by respiratory physicians when assessing the health of the lungs and the respiratory system.   Next, a non-invasive test of the health of you blood vessels will be performed. This involves having a blood-pressure cuff on one arm and a small sensor on two fingers. The test takes approximately 20 minutes and should not cause any discomfort.   You will ride a bicycle along a predetermined bike route on each test date. You will be asked to ride at a pace that you would consider the normal speed at which you would ride when traveling to work, or for other purposes for which you cycle. The bicycle will be adjusted to fit you comfortably, and if you have your own helmet, we will ask you to bring it with you to wear during the ride. If you do not have a helmet, we will provide one for you to wear. The bike ride itself will be a set route that you will be asked to follow for ½ hour at your chosen speed. Once ½ hour has passed, we will have you turn around and return to the starting location (the lab). You will be accompanied during the ride by a technician who will check your instruments at the turn-around time. You should do your best to ride as you normally would, following traffic regulations.  Air pollution monitoring equipment will record data during the bike rides along each route. Different sizes of particles in the air will be measured using two different instruments that will be mounted on the bike, and they should not hinder your cycling in any way. The bicycle will also be equipped with a bicycle computer that will display your heart rate and work rate.  Once you have returned to the lab, after one of the rides only, we will have you ride a stationary bike for about 15 minutes while you are breathing into a mouthpiece attached to a spirometer. The resistance level will gradually increase until the near-maximum heart rate that you reached during the ride is recorded, or until you can no longer cycle comfortably. This will tell us your breathing rate at different levels of heart rate and power output.   The last test we will do involves taking a small blood sample (10mL, approximately two teaspoons) from your arm, taken before and after the bike ride, which will be used to measure a common marker of inflammation in your blood.  The total amount of direct contact time with study technicians required for you to participate in the study is 3.5 hours for each of the two trials or about 7 hours in total. Note that the two testing days will be spaced at a minimum of two weeks apart.  Samples will be identified by research code only, which will not contain personally identifying information.   9. What are the possible harms of participating? There may be some minor discomfort or bruising on your arm at the site of the blood collection and you may experience light-headedness and/or fainting. As with any blood sample collection, there is a very small risk of infection at the site of the collection. You may become slightly dizzy or lightheaded when doing the lung function testing. There are no risks involved with the measurement of blood vessel health.    209 There are no risks involved with the air pollution sampling. All sampling involves measurement of compounds normally present in outdoor air, and your exposure to air pollutants will be at levels that are normally experienced by residents of Vancouver.  Participating in this study may cause some inconvenience because of the time required for both of the tests. The technician will schedule your testing on dates convenient to you.  There are inherent dangers in using a bicycle, including the possibility that you could fall off or be involved in a collision with objects or other road users. You should not take any risks while on the bicycle, and yield to other road users when it is safer to do so. The routes that are chosen will be along roads that are designated as cycling routes. The bicycle will be kept in good working order, however if you have any concerns about the route choice or the functioning of the equipment, we ask that you bring it to the attention of the technician following from behind as soon as safely possible.  10. What are the possible benefits of participating in this study? You will receive a summary of your health measurements and the measured level of air pollution during each of your two rides at the end of the study. You may also benefit from learning more about air quality in the Vancouver area. You may not directly benefit from participating in this study.   11. What happens if you decide to withdraw your consent to participate? Your participation in this research is entirely voluntary.  You may withdraw from this study at any time. If you decide to participate and then decide to withdraw at any time in the future, your samples will have their ID number removed and they will be discarded according to standard laboratory procedures. There will be no penalty.  If you wish to withdraw from this study, please contact the study coordinator, at the number listed below.    If you choose to enter the study and then decide to withdraw at a later time, all data collected about you during your enrolment in the study will be retained for analysis. Under UBC Policy #85, “Original data for any given study must be retained in the unit of origin for at least five years after the work is published or otherwise presented (if the form of the data permits this, and if assurances have not been given that data would be destroyed to assure anonymity”).   12. After the study is finished At the end of the study, a summary of your personal results will be sent to you.  13. Possible cost of participating in the study The only cost will be of your time. Individuals who either drive or choose to take public transit to the study location (the Air Pollution Exposure Lab at the Vancouver General Hospital Research Pavilion) will be asked to provide receipts or bus transfers so that the cost of parking or transit tickets (the cost of traveling to and from the lab on that day) can be reimbursed. A transit ticket will be provided for the return trip home at the same time as travel to the site is reimbursed to transit riders.  14. Will my taking part in this study be kept confidential?  Your confidentiality will be respected.  No information that discloses your identity will be released or published without your specific consent to the disclosure.  However, research records identifying you may be inspected in the presence of the Investigator and the UBC Research Ethics Board for the purpose of monitoring the research.  However, no records which identify you by name or initials will be allowed to leave the Investigators' offices.  15. Who do I contact if I have questions about the study during my participation?   210 If you have any questions or desire further information about this study before or during participation, you can contact the study coordinator, Christie Cole at 604-827-3509 or the principal investigator Michael Brauer at 604-822-9585.  16. Who do I contact if I have any questions or concerns about my rights as a subject during the study? If you have any concerns about your rights as a research subject and/or your experiences while participating in this study, contact the Research Subject Information Line at the University of British Columbia Office of Research Services at 604-822-8598.  17. Conflict of Interest There are no known conflicts of interest on the part of the study investigators.  Subject consent to participate     I have read and understood this consent form. I have had sufficient time to consider the information provided and to ask for advice if necessary.  I understand that my participation in this study is entirely voluntary and that I may refuse to participate or withdraw from the study at any time without any consequences.  I understand that all of the information collected will be kept confidential and will only be used for scientific objectives.  I understand that I am not waiving any of my legal rights by signing this consent form.  I have been told that I will receive a dated and signed copy of this consent form.  My signature below indicates that I consent to participate in the sections of this study checked below:  Participation in study procedures as outlined above.  I consent to provide blood samples.  I do not consent to provide blood samples.  Study personnel may contact me about future follow-up research studies.    ______________________________________________________________________________ Printed Name of Subject                          Signature Date     ______________________________________________________________________________ Printed Name of Witness                          Signature Date                                     ______________  Printed Name of                                            Signature         Date Principal Investigator   211 Appendix E – CAPaH Screening Questionnaire Prospective Subject Screening Questionnaire- Cycling, Air Pollution, and Health Study Subject #________  Today’s Date (dd/mm/yyyy): ____/____/________ Completed by (interviewer):________________ Month and year of subject’s birth (mm/yyyy): ____/________ Sex (circle):    male  female A. Par Q & You- Please read the questions carefully, and answer honestly Yes No  1. Has your doctor ever told you that you have a heart condition and that you should only do physical activity recommended by a doctor?  Yes No   2. Do you feel pain in your chest when you do physical activity?  Yes No  3. In the past month, have you had chest pain when you were not doing physical activity? Yes No   4. Do you lose your balance because of dizziness or do you ever lose consciousness?  Yes No  5. Do you have a bone or joint problem (for example, back, knee or hip) that could be made worse by a change in your physical activity?      212 Yes No 6. Is your doctor currently prescribing drugs (for example, water pills) for your blood pressure or heart condition?  Yes No 7. Do you know of any other reason why you should not do physical activity? Yes No  8. Could you be pregnant, or are you breast-feeding?  If there is any reason you feel you cannot bicycle for 1 hour without rest, please stop now.   213 B. Other medical conditions- Please indicate if any of the following applies to you:  Yes  No  1. Are you taking any medications for a heart condition, heart disease, blood pressure, or respiratory conditions?  Yes No   2. Do you take any pain medications on a daily basis including aspirin, acetaminophen (ex. Tylenol), or ibuprofen (ex. Advil)?   Yes No 3. Are you regularly taking any supplements, including vitamins, minerals, or nutritional supplements including energy drinks? Please list.    Yes No  4. Do you smoke cigarettes, or use any nicotine products? If you have smoked in the past, but do not currently smoke, please indicate the most recent date:______________________________________________________  Yes  No  5. Do you currently live in a place where you are exposed to second-hand smoke indoors?    Yes No 6. Does your work involve exposure to large amounts of dust, chemicals, or smoke? Please indicate the nature of your occupation, if yes.___________________________ _________________________________________________________________________   214 Yes No 7. Has a physician ever told you that you have asthma? If yes, stop here.  Yes No  8. Do you have any allergies to mould, dust, or pollen? If yes, stop here.   Yes No 9. Is there any other reason that you may not be able to safely ride a bicycle on the street, such as fear of traffic, or poor hearing or vision? If yes, stop here. C. For female subjects only:  Yes    No    1. Do you know the date of your last menstrual period?_____________                                                      Yes    No   2. Are you using any contraceptive medications? Please circle if yes.            Oral contraceptives (“the pill”) – mono-phasic or tri-phasic (if known)           Please indicate which:____________________________________                   Long acting reversible contraceptives (ex. intrauterine device or implant                        such as Implanon), or Depo Provera            NuvaRing              Other (please provide name of medication):________________________        215 Appendix F – CAPaH Pre-Test Questionnaire  Subject #________  Date (dd/mm/yyyy): ____/____/________  Completed by (interviewer):________________  Emergency Contact Information- who can we contact in the event of an emergency?   Contact 1            Contact 2 Name  Name  Phone #  Phone #  Alternate Phone #  Alternate Phone #  Relationship  Relationship    216 Yes No   1. Have you taken any pain-relieving medications over the last 72 hours, including aspirin, acetaminophen (ex. Tylenol), or ibuprofen (ex. Advil)?  Yes No 2. Have you taken any supplements, including vitamins, minerals, or nutritional supplements including energy drinks in the last 24 hours? Please list, and if possible provide time of last dose and quantity.     Yes  No  3. Have you consumed any alcohol in the last 24 hours?  Yes No  4. Have you smoked cigarettes, any other substances, or used any nicotine products in the past 24 hours? Please note that if you have smoked any substance in the past 24 hours, we cannot complete testing at this time. _________________________________________________________  Yes  No  5. Have you consumed any caffeine-containing products in the past 24 hours, including coffee, tea, chocolate, or energy drinks?   Male subjects, please skip to section D. Female subjects, please complete section C.  217 C. For female subjects only: 7. Please provide the date of your last menstrual period_____________    8. Are you using any contraceptive medications?    Yes (provide details below)   No            Oral contraceptives (“the pill”) – mono-phasic or tri-phasic (if known)                 Please indicate which:____________________________________                   Long acting reversible contraceptives (ex. intrauterine device or implant                        such as Implanon or Norplant), or Depo Provera            NuvaRing                                                                  Other              D. Dietary Information Please indicate what you ate for breakfast this morning. You will be asked to eat this same breakfast again on your 2nd bicycle trial---we will contact you the day before to remind you what you ate. Please be as detailed as possible, and include the amount you consumed. ____________________________________________________________________________________________________________________________________________________________ 218 ____________________________________________________________________________________________________________________________________________________________  E. Transportation Information Please indicate how you arrived at the testing location this morning. You will be asked to travel by using the same method on your 2nd bicycle trial. Examples include: public transit (bus, skytrain), automobile, bicycle, rollerblading…. ______________________________________________________________________________   219 Appendix G – CAPaH Common Cold Questionnaire   220 Appendix H – SOP 4 P-trak Procedures  CAPaH P-Trak 8525 Ultrafine Particle Counter procedures – rev-1.1, December 13, 2010  NOTE: The P-Trak used in this study has had the manufacturer’s TILT sensor disabled, due to shock effects inherent of a typical bike ride preventing reliable sampling. A specially designed cage has been built for this study to reduce shock and prevent tilting of the device when the subject ascends/descends hills, and details of the design are included at the end of this procedural manual.  In lab - Before sampling: 1. PTrak uses six (6) regular AA batteries for operation and they can either be single use disposables or rechargeable. Recharge batteries using a standard wall charger for rechargeable batteries. While charging the charger LEDs should flash green, when charged should be solid green. Single use batteries should last ~4 hours. Rechargeables last ~1.5 hours. 2. Connect the serial cable to the unit and the PC. Power on the P-Trak Ultrafine Particle Counter by pressing the black button on the front unit, which is found near the sample inlet port. The unit will power on for 60 seconds. 3. Once powered on the unit will begin to display approximately 0. These are NOT accurate counts as the alcohol wick has not been inserted. 4. Launch the TrakPRO software. Select the drop menu “Instrument setup” and select “Parameters”. Synchronize the unit’s date and time with the PC under the “Clock menu. Set the sampling parameters under “Logging intervals”. Set the logging interval to every 6 seconds (this is the interval at which the machine samples, and is associated with the preset “Log mode 1”). 5. Disconnect the PTrak from the computer. Clear the memory log manually on the unit by pressing the return button (a pink arrow) and selecting “Clear Memory” then “Logged Test Statistics”.  In Field - at sampling site (on the ride): 1. Attach the PTrak cage to the pannier rack of the bicycle. Secure using three (3) reusable cable (zip) ties. 2. Tilt the P-trak, and insert the alcohol cartridge, making sure the porous wick is saturated in pure isopropyl alcohol. Shake off any droplets before inserting the cartridge as these may flood the  221 optics of the device. 3. Power ON the device and wait for 60 seconds as the device powers up. Once the device has been powered on, it will begin to display live readouts of particle counts (in pt/cc). 4. Navigate using the arrow buttons and scroll down to the “Log mode 1” option, and press “Enter”. The device will now begin to log data at the pre-set interval (as set on the PC, every 6 seconds). 5. Sampling is typically 1 hour. When the bicycle ride is completed stop the logging session by simply pressing the “return” arrow (pink) on the PTrak, and then switch off the unit by pressing the button on the front panel. 6. Remove the alcohol cartridge and replace it in its plastic container.  Back at lab - downloading data after sampling: 1. Power ON the device, connect it to the PC using the serial cable, and launch the “TrakPRO” software. 2. After the 60 second power-up, the unit will display ~0 pt/cc (as the alcohol cartridge has been taken out). This can be ignored. 3. In the software, select “File” then “Receive (or press Ctrl-R). The P-Trak must be switched on for this to function properly. 4. Select the test(s) you wish to import from the sampling and click “receive”. The P-trak can now be disconnected and switched off. Save the data by clicking “File” then “Save” (the file will save as a .tkp file) and use the XXXR file name system where XXX is the subject ID and R is the route ID (ie 123D, for subject 123 route D). 5. The software can now graph the data directly, or it can be exported as a text file for more extensive statistics.  Exporting data - Excel (or other): 1. To export, select “File” then “Export”. It will save a .txt file under the same filename or another if you wish. You can also specify which type of delimiter to use. 2. Open the text file with Excel and specify which type of delimiter was used. A preview window shows what the data will look like to help with this stage. Design specifications:  222 In order to minimize the shock tilting the P-Trak suffers while being attached to a bicycle, a special cage was built. 1. A Swagman Phatt Folding Basket is used as the outer support for the device. The basket was bought from MEC (http://www.mec.ca). 2. A metal rod is attached to the basket and crosses the width (shorter) of the basket. Along the rod there are two (2) bearings. Electrical tape is wrapped around the rod at two points thick enough so that the bearings fit tightly onto the rod. 3. Hose clamps (from a hardware store) and used to secure thin bungee cord around the bearings, and the bungee cords support all the weight of the P-Trak. They feed around the outside of the clamps, and the loose ends come out in the middle. 4. The P-Trak should hang level with respect to gravity, and move freely forwards and backwards but not sideways (it should have pitch, but no roll, nor yaw). 5. The basket should be lined with foam to reduce the impact from large bumps and shakes resulting in contact to the sides of the basket; this should not be enough to reduce movement of the P-Trak. 6. The cage is secured to the bicycle using MEC Replacement Pannier Clips (http://www.mec.ca). The pannier clips were secured to the cage using various pieces of hardware such as metal strapping, nuts, and bolts.  223 Appendix I – SOP 3 GRIMM Dust Monitor Procedures In lab - Before sampling: 1. Charge the GRIMM Portable Dust Monitor battery by connecting the AC adapter to the both the unit and the wall. While charging the Battery Charge LED is solid red, and when the supply is fully charged the LED will switch off. Battery should last approximately 2 hours between charges.  2. Power on the GRIMM Portable Dust Monitor by holding the [ON/OFF] button. An audible beep will prompt you to press [+] if the filter has been changed or [-] if it has not. Press the [Standby] button until the LCD screen reads “Standby Mode Press 2nd Key”.  3. Set the clock manually by pressing the [Date/Time] button and then navigate using the [Date/Time], [+] and [-] buttons.  4. Connect GRIMM Portable Dust Monitor to PC using the provided RS 232 serial cable (must use the specific cable provided by GRIMM). Launch the DustMonitor software and select the “Dust Monitor Parameters” drop menu.  5. There are 3 tabs: select “Measurement Mode”. In this menu you may select between 4 types of data logging: Environmental (gives values of PM10  PM2.5 and PM1 in μg/m3), Occupational Health (gives values of Inhalable, Thoracic, and Alveolic dusts, in μg/m3), Mass Distribution (gives values for 15 particle size classes, in μg/m3), and Count Distribution (gives values for 15 particle size classes in units of particles per Litre). Multiplex allows you to select which size channels to select for, but only in “Count Distribution” mode.  Select Environmental.  6. Set the sampling period to every 6 seconds.  7. Select “Unit/C-Factor” Tab. The “Location” number in the software must match the location number displayed on the unit when you press the [Battery/Location] button. (When the “Read Memo Card Data from Location above” box is checked, the software will only search for data from  224 this location when downloading from the device. The device can store data in up to 99 Locations.)  8. Select the “Computer” Tab. Select COM1 and set the “Baud Rate” to 9600 baud. Click OK.  9. Select the “Automatic” drop menu. Select the “Presetting without attached PC” tab. IMPORTANT TIP: In this panel you will find the large button “Transfer Data to Dust Monitor”. Whenever you make changes in this or any other panel to preset the dust monitor, return to this screen and click this button. This makes sure the new settings are transferred to the unit.  10. Click “Transfer Data to Dust Monitor”. The software will prompt you to switch OFF the dust monitor; do so by pressing the “Stand by” button, followed by the “On/Off” button.  In Field - at sampling site (on the ride): 1. Attach the GRIMM portable Dust Monitor to the test bicycle pannier. Make sure the intake tube is sealed tight at the sample inlet port and that the intake tube is unobstructed (check the entire tube to make sure it is not crimped in any way).  2. Turn on the GRIMM Portable Dust Monitor and allow the unit to boot up. An audible beep will prompt you to press [+] if the filter has been changed and [-] if it has not. The unit will then power on and Self-Test.  3. The unit will begin recording data automatically once the Self-Test is completed, according the parameters defined in the software (see above section).  4. Sampling is typically 1 hour. When the bicycle ride is completed switch OFF the GRIMM Dust Monitor by pressing the “ Stand by” button, followed by the “On/Off” button.  Back at lab - downloading data after sampling: 1. Power ON the device, connect it to the PC, and set it to “Standby mode”. Select the “New Measurement” icon, which is in the upper left-hand corner of the program window. Select the Memory button to receive data from the unit.  225  2. The software will prompt you to give a file name. Provide the file name XXXR where XXX defines the subject ID, and R defines the route (ex 123D for subject 123, route D).  3. The PC and GRIMM Portable Dust Monitor will communicate (this sometimes takes a while) and sample data will appear. The information such as sampling start and end time will be displayed.  4. Select only ONE file at a time (i.e. the sample from the bike ride) and click save (otherwise multiple samples get saved with the same file name and different number identifiers).  5. Once saved, the screen will close automatically. To view the file, click “Open” and select the file you want.  6. Once it has been verified that the data is correctly downloaded, erase clear the data on the GRIMM. When the machine is On, press “Stand by”, and then simultaneously press the “Mean/Weight” button along with the “Factor C” button. The instrument will then ask you to confirm that the data should be cleared. If yes, select “+”. Doing so regularly will reduce the data download time after each trial.  Exporting data - Excel (or other): 1. Export the sample data by selecting the “Export” drop menu. The software will create a text delimited file (.txt) which can be opened in Excel. Select which delimiter you wish to use (tab or semicolon works best) and then click the export button. Select which file you wish to export (i.e. 123D). 2. Open Excel and then select “Open”. Select the (.txt) file in the main data directory of the GRIMM files on the PC. Excel will ask what kind of delimiter was used - match this to the delimiter of the file (a preview window shows what the data will look like to help with this stage).  226 Appendix J – SOP 2 – GPS Procedures  SOEH GlobalSat DG-100 GPS DataLogger procedures – rev-1.1, 18 August 2010 In lab - Before sampling: 1. Charge the battery of the GPS by either connecting the USB to a PC or to a AC converter. While charging the bottom LED will glow a solid amber. Fully charged batteries should last upwards of 24 hours. 2. Power ON the unit by pressing the large grey button until the top LED turns on solid green. Solid = searching for signal; Flashing = signal acquired and currently logging. 3. Connect the GPS unit to the USB on the PC. Open software “Data Logger PC Utility”. 4. Select the COM port corresponding to the USB port connected, in the top menu (most likely the largest number). 5. Select the “Configuration” menu under the “Settings” drop menu. Set the sampling to include position, time, date, and speed. Set the sampling period to per 1 second. Check the box beside “Enable WAAS/EGNOS/MSAS”. Make sure that the sampling mode you intend to use (A, B or C) matches the physical switch found on the side of the GPS unit. 6. Make sure the “Metric System” is checked in the “Settings” drop menu. 7. Clear the memory of the GPS unit by selecting “TrackRecord” -> “Delete All”.  In the field - on the bike ride: 1. Power ON the unit in an exposed outdoor area. The top LED should light up and stay a solid green while the GPS acquires a signal. Once the signal is acquired, the LED will begin to flash green. Data will be automatically logged once a signal is acquired.  227 2. Place the GPS in the outer pocket of the pannier attached to the subject’s bicycle. 3. Sampling lasts approximately 1 hour. Power OFF the unit at the end of sampling upon return to the lab by holding the large grey button until the lights power off.  Back at the lab - uploading the data to the PC: 1. Power ON the GPS unit and connect it to the USB port on the PC. Open software “Data Logger PC Utility”. 2. Select “TrackRecord” -> “Load Track points”. The GPS stores data points in groups of 95 points, so the entire track must be merged. If the memory was cleared before beginning, select all  tracks and load. Otherwise, select the tracks that apply and load. 3. Save. Give the (.gsd) file a name in the format XXXR, where XXX represents the subjects ID and R represents the route ID (ex 123D for subject 123 and D for route D). 4. Select “File” -> “Select all files”. Select “Map” -> “View Point”. This brings up all the points from different track files into one continuous track. 5. Export the data as a (.klm) file for viewing in Google Earth, and as a (.csv) file for viewing in Excel (and merging with other air quality data).    228 Appendix K – SOP 1 – PowerTap Procedures  SOEH CycleOps PowerTap Comp procedures – rev-1.0, 26 August 2010 In lab - Before sampling: 1. Request the subject place the heart rate chest-strap across their chest, centered just below the breasts/pectoral muscles. 2. Set up the wiring on the bicycle according to specifications set out by CycleOps. Install the wheel with the built-in hub on the test bicycle. 3. The PowerTap computer should have a clear memory, but if it does not download the contained data onto the PC using the USB connector and the “Power Agent 7” software to clear the memory. 4. Rest the yellow PowerTap computer into the reciever “shoe” on the handlebar and make sure the computer is connected. Spin the back wheel until the symbol in the left-upper corner of the computer appears. This means the computer is communicating with the wheel. The heart rate should also appear if you bring up the correct menu option.  In the Field - At the sampling site (on the ride): 1. Just before cycling, make sure the yellow computer is turned on (it may have switched off while not riding). 2. Proceed to ride the bike, and sample for roughly one (1) hour. 3. When returned from the ride, simply remove the computer from the receiver “shoe” and it will switch itself off.  Back at lab - downloading data after sampling: 1. Launch the “Power Agent 7” software on the PC and connect the PowerTap computer to the PC via the USB connector. The screen of the PowerTap computer should display the word “host” if properly connected. 2. Create a new user by selecting “File” then “New” then “Add new user”. Enter the subject ID number (ie 123) in the ‘name’ and ‘display’ fields, and enter the individuals information such as height and weight.  229 3. The subject should appear in the left-hand pannel, select it. 4. Download the activity data by selecting the correct icon on the top menu (“Download Activity Data”). The data should appear on the PC and may be fragmented. Select all the files from the ride and select “Merge”. Then select the one file and select “Save”. The PC will ask if you want to clear the data on the device. Click “Yes”. 5. Bring up the Activity Data and enter the Route ID into the “Comments” field (either D or C).   230 Appendix L – Form 6 Velotron test data form             231 Appendix M – Nimacizer file preparation and analysis  A separate excel spreadsheet was created for each trial (ex. one for subject 100 on the residential route, and a separate one for subject 100 on the downtown route). The first section of columns holding the GRIMM data (GRIMM time, PM10, PM2.5, PM1), followed by a section of columns containing the GPS data (GPS time, latitude in degrees, longitude in degrees, universal transverse mercator X and Y coordinators, speed (in kilometers per hour), followed by the section of P-trak data (P-trak time, UFP concentration in particles per cubic centimetre), and lastly the PowerTap section which contained columns for time (minutes elapsed), torque in Newtonmetres, speed in kilometers per hour, power output in watts, distance travelled in kilometers, cadence in revolutions per minute, and HR in bpm.   Due to variable time intervals between data points (dependent on each instrument), a simple computer program was written by a UBC computer science student (Nima Hazar) in 2011 to assist with data processing. This software aligned data from our four on-bike instruments, which did not all record data using 1 second measurement intervals due to instrument design.  The Nimacizer program produced a new excel spreadsheet, with the addition of a new column at the beginning called “True Time”. The first row of this column was the same time recorded by the first point for the GPS; each row that followed showed time at one-second intervals. The columns for the other instruments were adjusted around the GPS data, to put all of the instruments on the same time scale, so for instruments that recorded at time intervals larger than one second apart, the previous measurement (as long as it was within the previous six seconds) was  232 carried into the five rows (five seconds) that followed. The “True Time” is indicated in bold, with the times for each other instrument underlined, for comparison.  233 True Time GRIMM time PM10 PM2.5 PM1.0 GPS Time      Latitude Longitude UTMx UTMy Speed (km/ hour) Ptrak hh:mm:ss pt/cc PowrtapMinutes  Torq (N-m) Km/h Watts Km Cadence HR 10:05:20 10:04:06 3.4 2.6 1.3 10:05:20 4915.75 -12307.39 491037.50 5456641.93 0.4 10:21:16 11833 0:09:01 0.00 4.25 0 0.17 0 92 10:05:21 10:04:06 3.4 2.6 1.3 10:05:20 4915.75 -12307.39 491037.50 5456641.93 0.4 10:21:16 11833 0:09:01 0.00 4.25 0 0.17 0 92 10:05:22 10:04:06 3.4 2.6 1.3 10:05:20 4915.75 -12307.39 491037.50 5456641.93 0.4 10:21:16 11833 0:09:01 0.00 4.25 0 0.17 0 92 10:05:23 10:04:06 3.4 2.6 1.3 10:05:20 4915.75 -12307.39 491037.50 5456641.93 0.4 10:21:16 11833 0:09:04 0.11 9.06 1 0.18 0 92 10:05:24 10:04:12 3.4 2.6 1.3 10:05:20 4915.75 -12307.39 491037.50 5456641.93 0.4 10:21:22 11550 0:09:04 0.11 9.06 1 0.18 0 92 10:05:25 10:04:12 3.4 2.6 1.3 10:05:20 4915.75 -12307.39 491037.50 5456641.93 0.4 10:21:22 11550 0:09:06 9.04 10.67 80 0.19 68 93 10:05:26 10:04:12 3.4 2.6 1.3 10:05:26 4915.75 -12307.39 491037.50 5456640.82 3.8 10:21:22 11550 0:09:06 9.04 10.67 80 0.19 68 93 10:05:27 10:04:12 3.4 2.6 1.3 10:05:26 4915.75 -12307.39 491037.50 5456640.82 3.8 10:21:22 11550 0:09:06 9.04 10.67 80 0.19 68 93 10:05:28 10:04:12 3.4 2.6 1.3 10:05:26 4915.75 -12307.39 491037.50 5456640.82 3.8 10:21:22 11550 0:09:09 15.59 15.88 206 0.20 70 93 10:05:29 10:04:12 3.4 2.6 1.3 10:05:26 4915.75 -12307.39 491037.50 5456640.82 3.8 10:21:22 11550 0:09:09 15.59 15.88 206 0.20 70 93 10:05:30 10:04:18 3.3 2.5 1.3 10:05:26 4915.75 -12307.39 491037.50 5456640.82 3.8 10:21:28 10783 0:09:11 9.60 19.34 155 0.21 64 94 10:05:31 10:04:18 3.3 2.5 1.3 10:05:26 4915.75 -12307.39 491037.50 5456640.82 3.8 10:21:28 10783 0:09:11 9.60 19.34 155 0.21 64 94 10:05:32 10:04:18 3.3 2.5 1.3 10:05:32 4915.74 -12307.39 491042.34 5456635.06 9.6 10:21:28 10783 0:09:11 9.60 19.34 155 0.21 64 94 10:05:33 10:04:18 3.3 2.5 1.3 10:05:32 4915.74 -12307.39 491042.34 5456635.06 9.6 10:21:28 10783 0:09:14 5.88 21.05 103 0.23 75 97 10:05:34 10:04:18 3.3 2.5 1.3 10:05:32 4915.74 -12307.39 491042.34 5456635.06 9.6 10:21:28 10783 0:09:14 5.88 21.05 103 0.23 75 97 10:05:35 10:04:18 3.3 2.5 1.3 10:05:32 4915.74 -12307.39 491042.34 5456635.06 9.6 10:21:28 10783 0:09:16 5.76 22.30 107 0.24 77 97 10:05:36 10:04:24 3.5 2.7 1.4 10:05:32 4915.74 -12307.39 491042.34 5456635.06 9.6 10:21:34 11416 0:09:16 5.76 22.30 107 0.24 77 97 10:05:37 10:04:24 3.5 2.7 1.4 10:05:32 4915.74 -12307.39 491042.34 5456635.06 9.6 10:21:34 11416 0:09:16 5.76 22.30 107 0.24 77 97 10:05:38 10:04:24 3.5 2.7 1.4 10:05:38 4915.74 -12307.36 491071.44 5456630.76 18.2 10:21:34 11416 0:09:19 2.83 22.92 54 0.26 116 96 10:05:39 10:04:24 3.5 2.7 1.4 10:05:38 4915.74 -12307.36 491071.44 5456630.76 18.2 10:21:34 11416 0:09:19 2.83 22.92 54 0.26 116 96 10:05:40 10:04:24 3.5 2.7 1.4 10:05:38 4915.74 -12307.36 491071.44 5456630.76 18.2 10:21:34 11416 0:09:21 5.08 22.67 96 0.27 72 93   234 Appendix N – PM2.5 and PNC time series plots from all trials  Figure 23. Time series plot for subject 100 along the Residential route.   Figure 24. Time series plot for subject 100 along the Downtown route.  0100002000030000400005000060000700000510152025303510:54 11:01 11:08 11:16 11:23 11:30 11:37 11:44 11:52 11:59PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 100 ResidentialPM2.5 UFPPNC0100002000030000400005000060000700000510152025303512:04 12:11 12:18 12:25 12:33 12:40 12:47 12:54 13:01PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 100 DowntownPM2.5 UFP 235   Figure 25. Time series plot for subject 113 along the Residential route.   Figure 26. Time series plot for subject 113 along the Downtown route.0100002000030000400005000060000700000510152025303514:29 14:36 14:43 14:50 14:57 15:05 15:12 15:19 15:26 15:33PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 113 ResidentialPM2.5 UFPPNC0100002000030000400005000060000700000510152025303514:08 14:15 14:22 14:29 14:36 14:44 14:51 14:58 15:05PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 113 DowntownPM2.5 UFPPNC 236  Figure 27. Time series plot for subject 116 along the Residential route.   Figure 28. Time series plot for subject 116 along the Downtown route.  010000200003000040000500006000070000051015202530358:33 8:40 8:47 8:54 9:02 9:09 9:16 9:23 9:30PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 116 ResidentialPM2.5 UFPPNC010000200003000040000500006000070000051015202530359:53 10:00 10:07 10:14 10:22 10:29 10:36 10:43 10:50 10:58PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 116 DowntownPM2.5 UFPPNC 237  Figure 29. Time series plot for subject 119 along the Residential route.   Figure 30. Time series plot for subject 119 along the Downtown route.  0100002000030000400005000060000700000510152025303514:19 14:26 14:34 14:41 14:48 14:55 15:02 15:10 15:17 15:24PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 119 ResidentialPM2.5 UFPPNC010000200003000040000500006000070000051015202530353:33 3:41 3:48 3:55 4:02 4:09 4:17 4:24PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 119 DowntownPM2.5 UFPPNC 238  Figure 31. Time series plot for subject 122 along the Residential route. *PM2.5 values are above the values listed on the y-axis for 122 Residential  Figure 32. Time series plot for subject 122 along the Downtown route. 0100002000030000400005000060000700000510152025303514:34 14:49 15:03 15:18 15:32 15:46PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 122 ResidentialPM2.5 UFPPNC0100002000030000400005000060000700000510152025303514:02 14:09 14:16 14:24 14:31 14:38 14:45 14:52 15:00 15:07PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 122 DowntownPM2.5 UFPPNC 239  Figure 33. Time series plot for subject 129 along the Residential route.   Figure 34. Time series plot for subject 129 along the Downtown route.  010000200003000040000500006000070000051015202530359:22 9:29 9:37 9:44 9:51 9:58 10:05 10:13 10:20PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 129 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:54 11:01 11:08 11:16 11:23 11:30 11:37 11:44 11:52 11:59PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 129 DowntownPM2.5 UFP 240  Figure 35. Time series plot for subject 130 along the Residential route.   Figure 36. Time series plot for subject 130 along the Downtown route.  010000200003000040000500006000070000051015202530359:48 9:55 10:02 10:09 10:17 10:24 10:31 10:38 10:45PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 130 ResidentialPM2.5 UFP010000200003000040000500006000070000051015202530359:48 9:55 10:02 10:09 10:17 10:24 10:31 10:38 10:45 10:53PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 130 DowntownPM2.5 UFP 241  Figure 37. Time series plot for subject 133 along the Residential route.   Figure 38. Time series plot for subject 133 along the Downtown route.  0100002000030000400005000060000700000510152025303514:06 14:13 14:20 14:27 14:34 14:42 14:49 14:56 15:03PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 133 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303515:07 15:14 15:21 15:28 15:36 15:43 15:50 15:57 16:04PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 133 DowntownPM2.5 UFP 242  Figure 39. Time series plot for subject 134 along the Residential route.   Figure 40. Time series plot for subject 134 along the Downtown route.  0100002000030000400005000060000700000510152025303510:00 10:07 10:14 10:21 10:28 10:36 10:43 10:50 10:57PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 134 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:13 10:20 10:27 10:34 10:41 10:49 10:56 11:03 11:10 11:17PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 134 DowntownPM2.5 UFP 243  Figure 41. Time series plot for subject 141 along the Residential route.   Figure 42. Time series plot for subject 141 along the Downtown route.  0100002000030000400005000060000700000510152025303514:45 14:52 15:00 15:07 15:14 15:21 15:28 15:36 15:43PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 141 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303515:02 15:09 15:16 15:24 15:31 15:38 15:45 15:52 16:00 16:07PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 141 DowntownPM2.5 UFP 244  Figure 43. Time series plot for subject 142 along the Residential route.   Figure 44. Time series plot for subject 142 along the Downtown route.  010000200003000040000500006000070000051015202530359:58 10:05 10:12 10:19 10:27 10:34 10:41 10:48 10:55PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 142 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:09 10:17 10:24 10:31 10:38 10:45 10:53 11:00 11:07 11:14PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 142 DowntownPM2.5 UFP 245  Figure 45. Time series plot for subject 145 along the Residential route.   Figure 46. Time series plot for subject 145 along the Downtown route.  0100002000030000400005000060000700000510152025303514:13 14:20 14:27 14:34 14:41 14:49 14:56 15:03 15:10 15:17PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 145 ResidentialPM2.5 UFP010000200003000040000500006000070000051015202530359:59 10:06 10:14 10:21 10:28 10:35 10:42 10:50 10:57PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 145 DowntownPM2.5 UFP 246  Figure 47. Time series plot for subject 147 along the Residential route.   Figure 48. Time series plot for subject 147 along the Downtown route. *Time period was incorrectly set on GRIMM to record one measurement per minute for PM2.5.0100002000030000400005000060000700000510152025303514:03 14:10 14:17 14:24 14:31 14:39 14:46 14:53 15:00PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 147 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303514:34 14:42 14:49 14:56 15:03 15:10 15:18 15:25 15:32PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 147 DowntownPM2.5 UFP 247  Figure 49. Time series plot for subject 150 along the Residential route. * Above trial was missing PNC data.   Figure 50. Time series plot for subject 150 along the Downtown route. 010000200003000040000500006000070000051015202530359:44 9:51 9:59 10:06 10:13 10:20 10:27 10:35 10:42PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 150 ResidentialPM 2.5 UFP010000200003000040000500006000070000051015202530359:43 9:51 9:58 10:05 10:12 10:19 10:27 10:34 10:41PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 150 DowntownPM2.5 UFP 248  Figure 51. Time series plot for subject 152 along the Residential route.    Figure 52. Time series plot for subject 152 along the Downtown route.0100002000030000400005000060000700000510152025303511:17 11:24 11:31 11:39 11:46 11:53 12:00 12:07 12:15PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 152 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303511:10 11:17 11:25 11:32 11:39 11:46 11:53 12:01 12:08PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 152 DowntownPM2.5 UFP 249  Figure 53. Time series plot for subject 154 along the Residential route.   Figure 54. Time series plot for subject 154 along the Downtown route.  0100002000030000400005000060000700000510152025303510:08 10:16 10:23 10:30 10:37 10:44 10:52 10:59 11:06PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 154 ResidentialPM2.5 UFP010000200003000040000500006000070000051015202530359:53 10:00 10:07 10:15 10:22 10:29 10:36 10:43 10:51 10:58PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 154 DowntownPM2.5 UFP 250 Figure 55. Time series plot for subject 157 along the Residential route.   Figure 56. Time series plot for subject 157 along the Downtown route.0100002000030000400005000060000700000510152025303514:03 14:10 14:18 14:25 14:32 14:39 14:46 14:54 15:01 15:08PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 157 DowntownPM2.5 UFP0100002000030000400005000060000700000510152025303514:42 14:49 14:56 15:03 15:10 15:18 15:25 15:32 15:39 15:46PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 157 ResidentialPM2.5 UFP 251   Figure 57. Time series plot for subject 158 along the Residential route.   Figure 58. Time series plot for subject 158 along the Downtown route. 0100002000030000400005000060000700000510152025303510:36 10:43 10:50 10:58 11:05 11:12 11:19 11:26 11:34 11:41 11:48PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 158 ResidentialPM2.5 UFP010000200003000040000500006000070000051015202530359:52 9:59 10:06 10:14 10:21 10:28 10:35 10:42 10:50PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 158 DowntownPM2.5 UFP 252   Figure 59. Time series plot for subject 159 along the Residential route.   Figure 60. Time series plot for subject 159 along the Downtown route. 0100002000030000400005000060000700000510152025303514:09 14:16 14:24 14:31 14:38 14:45 14:52 15:00 15:07PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 159 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303514:30 14:37 14:44 14:51 14:58 15:06 15:13 15:20 15:27 15:34PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 159 DowntownPM2.5 UFP 253  Figure 61. Time series plot for subject 160 along the Residential route.  Figure 62. Time series plot for subject 160 along the Downtown route. 010000200003000040000500006000070000051015202530359:41 9:48 9:55 10:02 10:09 10:17 10:24 10:31 10:38PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 160 ResidentialPM2.5 UFP010000200003000040000500006000070000051015202530359:23 9:30 9:38 9:45 9:52 9:59 10:06 10:14 10:21PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 160 DowntownPM2.5 UFP 254   Figure 63. Time series plot for subject 161 along the Residential route.   Figure 64. Time series plot for subject 161 along the Downtown route. 0100002000030000400005000060000700000510152025303510:26 10:33 10:40 10:48 10:55 11:02 11:09 11:16 11:24PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 161 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:22 10:29 10:36 10:43 10:50 10:58 11:05 11:12 11:19 11:26PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 161 DowntownPM2.5 UFP 255   Figure 65. Time series plot for subject 162 along the Residential route.   Figure 66. Time series plot for subject 162 along the Downtown route. 010000200003000040000500006000070000051015202530359:44 9:51 9:59 10:06 10:13 10:20 10:27 10:35 10:42PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 162 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:02 10:09 10:17 10:24 10:31 10:38 10:45 10:53 11:00PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 162 DowntownPM2.5 UFP 256   Figure 67. Time series plot for subject 163 along the Residential route.   Figure 68. Time series plot for subject 163 along the Downtown route. 0100002000030000400005000060000700000510152025303514:26 14:34 14:41 14:48 14:55 15:02 15:10 15:17 15:24 15:31PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 163 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303514:04 14:11 14:18 14:25 14:33 14:40 14:47 14:54 15:01 15:09PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 163 DowntownPM2.5 UFP 257   Figure 69. Time series plot for subject 165 along the Residential route.   Figure 70. Time series plot for subject 165 along the Downtown route.010000200003000040000500006000070000051015202530359:48 9:56 10:03 10:10 10:17 10:24 10:32 10:39 10:46PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 165 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:19 10:26 10:33 10:40 10:48 10:55 11:02 11:09 11:16PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 165 Downtown PM2.5 UFP 258   Figure 71. Time series plot for subject 166 along the Residential route.   Figure 72. Time series plot for subject 166 along the Downtown route. 0100002000030000400005000060000700000510152025303510:18 10:25 10:32 10:40 10:47 10:54 11:01 11:08 11:16PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 166 ResidentialPM2.5 UFP010000200003000040000500006000070000051015202530359:44 9:51 9:58 10:05 10:13 10:20 10:27 10:34 10:41PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 166 DowntownPM2.5 UFP 259   Figure 73. Time series plot for subject 169 along the Residential route.   Figure 74. Time series plot for subject 169 along the Downtown route.010000200003000040000500006000070000051015202530359:33 9:41 9:48 9:55 10:02 10:09 10:17 10:24 10:31PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 169 ResidentialPM2.5 UFP010000200003000040000500006000070000051015202530359:37 9:44 9:51 9:59 10:06 10:13 10:20 10:27 10:35 10:42PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 169 DowntownPM2.5 UFP 260  Figure 75. Time series plot for subject 170 along the Residential route.   Figure 76. Time series plot for subject 170 along the Downtown route.   0100002000030000400005000060000700000510152025303514:25 14:32 14:39 14:46 14:53 15:01 15:08 15:15 15:22 15:29PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 170 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303515:38 15:45 15:53 16:00 16:07 16:14 16:21 16:29 16:36 16:43PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 170 DowntownPM2.5 UFP 261  Figure 77. Time series plot for subject 172 along the Residential route.   Figure 78. Time series plot for subject 172 along the Downtown route.   0100002000030000400005000060000700000510152025303514:40 14:47 14:54 15:02 15:09 15:16 15:23 15:30 15:38PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 172 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303514:08 14:15 14:23 14:30 14:37 14:44 14:51 14:59 15:06PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 172 DowntownPM2.5 UFP 262  Figure 79. Time series plot for subject 178 along the Residential route.   Figure 80. Time series plot for subject 178 along the Downtown route. 0100002000030000400005000060000700000510152025303510:22 10:30 10:37 10:44 10:51 10:58 11:06 11:13 11:20 11:27PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 178 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:04 10:12 10:19 10:26 10:33 10:40 10:48 10:55 11:02 11:09PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 178 DowntownPM2.5 UFP 263  Figure 81. Time series plot for subject 184 along the Residential route.   Figure 82. Time series plot for subject 184 along the Downtown route. 0100002000030000400005000060000700000510152025303514:44 14:52 14:59 15:06 15:13 15:20 15:28 15:35 15:42PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 184 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303513:48 13:55 14:03 14:10 14:17 14:24 14:31 14:39 14:46 14:53PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 184 DowntownPM2.5 UFP 264  Figure 83. Time series plot for subject 186 along the Residential route.   Figure 84. Time series plot for subject 186 along the Downtown route. 010000200003000040000500006000070000051015202530359:37 9:44 9:51 9:58 10:05 10:13 10:20 10:27 10:34PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 186 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:02 10:09 10:17 10:24 10:31 10:38 10:45 10:53 11:00 11:07PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 186 DowntownPM2.5 UFP 265  Figure 85. Time series plot for subject 187 along the Residential route.   Figure 86. Time series plot for subject 187 along the Downtown route.   0100002000030000400005000060000700000510152025303513:55 14:02 14:10 14:17 14:24 14:31 14:38 14:46 14:53PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 187 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303513:48 13:55 14:02 14:10 14:17 14:24 14:31 14:38 14:46 14:53PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 187 DowntownPM2.5 (ug/m^3) UFP (pt/cc) 266  Figure 87. Time series plot for subject 190 along the Residential route. Note: above trial was missing PNC data.   Figure 88. Time series plot for subject 190 along the Downtown route. 0100002000030000400005000060000700000510152025303510:14 10:21 10:29 10:36 10:43 10:50 10:57 11:05 11:12 11:19PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 190 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:05 10:12 10:20 10:27 10:34 10:41 10:48 10:56 11:03 11:10PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 190 DowntownPM2.5 UFP 267  Figure 89. Time series plot for subject 193 along the Residential route.   Figure 90. Time series plot for subject 193 along the Downtown route. 010000200003000040000500006000070000051015202530359:36 9:43 9:50 9:57 10:04 10:12 10:19 10:26 10:33PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 193 ResidentialPM2.5 UFP0100002000030000400005000060000700000510152025303510:07 10:15 10:22 10:29 10:36 10:43 10:51 10:58 11:05 11:12PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 193 DowntownPM2.5 UFP 268  Figure 91. Time series plot for subject 197 along the Residential route.   Figure 92. Time series plot for subject 197 along the Downtown route.  010000200003000040000500006000070000051015202530359:55 10:02 10:10 10:17 10:24 10:31 10:38 10:46 10:53PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 197 ResidentialPM2.5 UFP010000200003000040000500006000070000051015202530359:48 9:56 10:03 10:10 10:17 10:24 10:32 10:39 10:46 10:53PNC (particles/cc)PM 2.5 (ug/m^3)Time Series Plot: Subject 197 DowntownPM2.5 UFP 269 Appendix O – NASA Satellite August 4, 2010 image of British Columbia forest fires  http://earthobservatory.nasa.gov/NaturalHazards/view.php?id=45056  Acquired August 4, 2010 download large image (2 MB, JPEG, 2200x2800)   Several large forest fires burned in British Columbia, Canada on August 4, 2010. The fires are outlined in red in this true-color image taken by the Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA’s Aqua satellite. The fires shown in this image are in the Cariboo region of the province, where 120 fires were burning on August 3. Many of the large fires ignited in a lightning storm on July 28, and additional lightning-caused fires started on August 3, said the wildfire management branch of the forest service in British Columbia. More than 400 fires burned throughout British Columbia on August 3, reported the Canadian Broadcasting Corporation. The large image is the highest resolution (most detailed) version of the image. The image is available in additional resolutions from the MODIS Rapid Response System.  Canadian Broadcasting Corporation News. (2010, August 3). Over 400 wildfires scorch B.C. Accessed August 5, 2010. Wildfire Management Branch. (2010, August 3). Thunderstorms continue to hit the Cariboo. Province of British Columbia. Accessed August 5, 2010.  270 NASA image courtesy Jeff Schmaltz, MODIS Rapid Response Team at NASA GSFC. Caption by Holli Riebeek. Instrument: Aqua - MODIS  Fires in British Columbia, Canada More in this Event (view all)  The Earth Observatory is part of the EOS Project Science Office located at NASA Goddard Space Flight Center webmaster: Paul Przyborski | NASA official: Warren Wiscombe       271 Appendix P – Press Release from British Columbia Wildfire Management Branch “Weather Produces Smoky Skies on the Coast” From the British Columbia Wildfire Management Branch [retrieved February 3, 2013] http://bcwildfire.ca/hprScripts/WildfireNews/DisplayArticle.asp?SearchTerm=fire&ID=1747  Ministry of Forests and Range, Wildfire Management Branch WEATHER PRODUCES SMOKY SKIES ON THE COAST 8/4/2010 5:13 PM Weather conditions are forcing smoke from wildfires in the interior of British Columbia to the coast, producing smokey skies throughout the Coastal Fire Centre.  Outflow winds and an easterly flow aloft is moving the smoke down the Fraser Canyon, Pemberton Valley, and coastal inlets and rivers. Inversion conditions are holding the smoke down to ground level. Heavy smoke has been reported on the lower mainland from the border to Bella Coola, and in all areas of Vancouver Island.  These conditions will continue today, and while the smoke may dissipate overnight it is expected the smoke will reform tomorrow. These conditions may persist for several days. People who are concerned with their health due to air quality should contact their local health authority. The Coastal Fire Centre continues in high or extreme fire danger rating, and the public is cautioned to exercise care with any activity that may produce a spark. All open fire is prohibited within the Coastal Fire Centre jurisdiction.  To report a wildfire or unattended campfire please call *xxxx on your cell or toll-free 1-800-xxx-xxxx  For the latest information on fire activity, conditions and prohibitions, visit the Wildfire Management Branch website at www.bcwildfire.ca. For Facebook and Twitter updates, as well as details about evacuation orders and alerts, road conditions and air quality advisories, go to www.firesafebc.ca.  -30-   Contact:  Mike McCulley Fire Information Offic  Donna MacPherson Fire Information Offic News Release 1747  272 Appendix Q – Air Quality Health Index Categories  Air Quality Health Index Categories and Health Messages The AQHI uses a scale to show the health risk associated with the air pollution we breathe. The following table provides the health messages for ‘at risk’ individuals and the general public for each of the AQHI Health Risk Categories. Health Risk Air Quality Health Index Health Messages     At Risk Population* General Population Low 1 - 3 Enjoy your usual outdoor activities. Ideal air quality for outdoor activities. Moderate 4 - 6 Consider reducing or rescheduling strenuous activities outdoors if you are experiencing symptoms. No need to modify your usual outdoor activities unless you experience symptoms such as coughing and throat irritation. High 7 - 10 Reduce or reschedule strenuous activities outdoors. Children and the elderly should also take it easy. Consider reducing or rescheduling strenuous activities outdoors if you experience symptoms such as coughing and throat irritation. Very High Above 10 Avoid strenuous activities outdoors. Children and the elderly should also avoid outdoor physical exertion. Reduce or reschedule strenuous activities outdoors, especially if you experience symptoms such as coughing and throat irritation. * People with heart or breathing problems are at greater risk. Follow your doctor's usual advice about exercising and managing your condition.  Retrieved February 3, 2013 from Environment Canada http://www.ec.gc.ca/cas-aqhi/default.asp?lang=En&n=79A8041B-1   273 Appendix R – The HR - minute ventilation relationship curve for each 2011 participant. Table 28- Subject 101 Velotron test data points and predicted minute ventilation values   Figure 93. Subject 101 curve and predictive equation.  y = 0.0007x2.2253R² = 0.978401020304050600 20 40 60 80 100 120 140 160 180Minute Ventilation (litres per minute)Heart Rate (beats per minute)Subject 101 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation lineProgrammed Power Output (watts) Measured Heart Rate (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 0.0007 (HR)^2.2253 0 73 8.72 9.8 20 98 18.80 18.9 40 101 21.40 20.2 60 106 24.44 22.5 80 112 24.36 25.4 100 120 32.16 29.6 120 130 34.80 35.4 140 140 37.16 41.8 160 150 49.64 48.7 180 156 48.88 53.1  274 Table 29- Subject 154 Velotron test data points and predicted minute ventilation values Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 0.0169 (HR)^1.6522 0 57 13.76 1.3 30 77 21.84 2.8 60 88 27.92 4.1 90 98 32.56 5.3 120 105 36.72 6.5 150 117 39.76 7.4 180 124 47.82 10.1 210 132 60.24 14.7   Figure 94. Subject 154 curve and predictive equation.  y = 0.0169x1.6522R² = 0.98360102030405060700 20 40 60 80 100 120 140Minute Ventilation (Litres per minute)Heart Rate (beats per minute)Subject 154 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation trendline 275 Table 30- Subject 157 Velotron test data points and predicted minute ventilation values Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 0.0113(HR)^1.7904 0 57 14.14 15.7 30 n/a n/a n/a 60 86 35.32 32.9 90 98 44.72 41.5 120 104 52.92 46.2 150 113 52.68 53.6 180 126 62.80 65.1 210 145 73.80 83.7 240 154 94.28 93.2   Figure 95. Subject 157 curve and predictive equation.  y = 0.0113x1.7904R² = 0.974901020304050607080901000 50 100 150 200Minute Ventilation (Litres per minute)Heart Rate (beats per minute)157 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation line 276 Table 31- Subject 158 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred. = -0.0071(HR)2 +2.1814(HR) – 90.485 0 57 14.14 15.7 30 n/a n/a n/a 60 86 35.32 32.9 90 98 44.72 41.5 120 104 52.92 46.2 150 113 52.68 53.6 180 126 62.80 65.1 210 145 73.80 83.7 240 154 94.28 93.2    Figure 96. Subject 158 curve and predictive equation.  y = -0.0071x2 + 2.1814x - 90.485R² = 0.9695010203040506070800 20 40 60 80 100 120 140Minute Ventilation (Litres per minute)Heart rate (beats per minute)158 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation line 277 Table 32- Subject 159 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred.  = -0.0048(HR)2 +1.6751(HR) – 99.3 0 94 8.64 15.7 20 n/a n/a n/a 40 98 25.16 18.8 60 109 28.12 26.3 80 120 32.08 32.6 100 130 35.44 37.3 120 143 38.44 42.1 140 156 46.52 45.2    Figure 97. Subject 159 curve and predictive equation.  y = -0.0048x2 + 1.6751x - 99.3R² = 0.8698051015202530354045500 20 40 60 80 100 120 140 160 180Minute Ventilation (Litres per minute)Heart Rate (beats per minute)Subject 159 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation line 278 Table 33- Subject 160 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred. = 0.0148(HR)2 +1.4824 (HR) + 53.664 0 54 17.44 16.8 30 73 22.16 24.3 60 83 35.56 32.6 90 95 43.88 46.4 120 99 55.2 52.0 150 111 71.4 71.5 180 124 n/a n/a    Figure 98. Subject 160 curve and predictive equation.  y = 0.0148x2 - 1.4824x + 53.664R² = 0.9855010203040506070800 20 40 60 80 100 120Minute Ventilation (Litres per minute)Heart Rate (beats per minute)Subject 160 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation trendline 279 Table 34- Subject 161 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred. = 0.0021(HR)2 +0.038(HR) + 0.6164 0 63 9.20 11.4 20 74 16.40 14.9 40 82 19.20 17.9 60 97 23.40 24.1 80 108 30.24 29.2 100 122 33.76 36.5 120 133 40.68 42.8 140 143 50.00 49.0 160 154 56.04 56.3    Figure 99. Subject 161 curve and predictive equation.  y = 0.0021x2 + 0.038x + 0.6164R² = 0.988901020304050600 20 40 60 80 100 120 140 160 180Minute Ventilation (Litres per minute)Heart Rate (beats per minute)Subject 161 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation trendline 280 Table 35- Subject 162 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred. = -0.0151(HR)2 + 3.9554 (HR) – 219.81 0 93 7.92 17.4 30 93 25.84 17.4 60 103 30.96 27.4 90 104 25.72 28.2 120 114 33.72 34.9 150 121 36.52 37.7 180 126 39.32 38.8 210 136 n/a n/a    Figure 100. Subject 162 curve and predictive equation.  y = -0.0151x2 + 3.9554x - 219.81R² = 0.72040510152025303540450 20 40 60 80 100 120 140Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 162 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation trendline 281 Table 36- Subject 163 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred. = 0.0051(HR)2 – 0.4077 (HR) + 16.297 0 73 14.56 13.7 30 96 22.24 24.2 60 107 32.44 31.1 90 114 37.72 36.1 120 124 44.29 44.2 150 135 57.16 54.2 180 143 60.60 62.3 210 152 73.76 72.2    Figure 101. Subject 163 curve and predictive equation.  y = 0.0051x2 - 0.4077x + 16.297R² = 0.9931010203040506070800 20 40 60 80 100 120 140 160Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 163 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation trendline 282 Table 37- Subject 164 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 0.0005(HR)^2.3064 0 80 11.48 12.3 20 101 21.28 21.0 40 104 23.16 22.4 60 111 27.92 26.1 80 120 32.32 31.2 100 128 40.56 36.2 120 136 42.2 41.7 140 145 47.64 48.3 160 156 57.04 57.2 180 167 60.24 66.9 200 174 78.08 73.5    Figure 102. Subject 164 curve and predictive equation.  y = 0.0005x2.3064R² = 0.987801020304050607080900 50 100 150 200Minute ventilation (Litres per minute)Heart rate (beats per minute)Subject 164 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation trendline 283 Table 38- Subject 165 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 2.4183e^0.0249(HR) 0 65 12.52 12.2 30 79 17.44 17.3 60 86 19.92 20.6 90 98 27.92 27.8 120 106 31.92 33.9 150 115 42.52 42.4 180 125 56.48 54.4 210 133 66.16 66.3    Figure 103. Subject 165 curve and predictive equation.  y = 2.4183e0.0249xR² = 0.99710102030405060700 20 40 60 80 100 120 140Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 165 Minute Ventilation Relationship Measured minute ventilation Predictive equation trendline 284 Table 39- Subject 166 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred. = 0.0036(HR)^2 -0.3676(HR) + 31.608 0 65 20.56 22.9 30 93 34.92 28.6 60 107 32.36 33.5 90 114 36.08 36.5 120 125 40.24 41.9 150 135 45.44 47.6 180 143 51.32 52.7 210 160 67.26 65.0    Figure 104. Subject 166 curve and predictive equation.  y = 0.0036x2 - 0.3676x + 31.608R² = 0.9547010203040506070800 20 40 60 80 100 120 140 160 180Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 166 Minute Ventilation Relationship Measured minute ventilation Predictive equation trendline 285 Table 40- Subject 169 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred. = 6.9369e^0.017(HR) 0 71 22.96 23.2 30 85 30.00 29.4 60 91 35.24 32.6 90 97 35.56 36.1 120 106 40.16 42.0 150 113 45.40 47.4 180 119 53.52 52.4 210 128 58.12 61.1 240 136 75.68 70.0    Figure 105. Subject 169 curve and predictive equation.  y = 6.9369e0.017xR² = 0.9814010203040506070800 20 40 60 80 100 120 140 160Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 169 Minute Ventilation RelationshipMeasured minute ventilation Predictive equation trendline 286 Table 41- Subject 170 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred. = 0.0006(HR)^2 + 0.2484(HR) – 9.1433 0 82 12.96 15.3 20 90 18.60 18.1 40 92 20.80 18.8 60 106 24.36 23.9 80 113 29.60 26.6 100 129 28.32 32.9 120 140 36.00 37.4 140 148 43.68 40.8    Figure 106. Subject 170 curve and predictive equation.  y = 0.0006x2 + 0.2484x - 9.1433R² = 0.9264051015202530354045500 20 40 60 80 100 120 140 160Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 170 Measured Minute VentilationMeasured minute ventilation Predictive equation trendline 287 Table 42- Subject 172 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 0.018 (HR)^1.6229 0 76 18.24 20.3 30 102 36.32 32.7 60 112 42.40 38.1 90 130 50.12 48.5 120 148 58.24 59.9 150 160 57.48 68.0 180 171 81.80 76.0    Figure 107. Subject 172 curve and predictive equation. y = 0.018x1.6229R² = 0.948801020304050607080900 20 40 60 80 100 120 140 160 180Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 172 Measured Minute VentilationMeasured minute ventilation Predictive equation trendline 288 Table 43- Subject 178 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 0.0071(HR)^2 – 1.1033(HR) + 64.241 0 82 17.92 21.5 30 98 33.88 24.3 60 102 23.20 25.6 90 114 29.16 30.7 120 122 36.70 35.3 150 135 44.96 44.7 180 148 53.52 56.5 210 160 70.88 69.5 240 169 83.08 80.6    Figure 108. Subject 178 curve and predictive equation.  y = 0.0071x2 - 1.1033x + 64.241R² = 0.966401020304050607080900 20 40 60 80 100 120 140 160 180Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 178 Measured Minute VentilationMeasured minute ventilation Predictive equation trendline 289 Table 44- Subject 181 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "polynomial" formula for trendline); pred. = -0.0001(HR)^2 + 0.3594(HR) – 10.881 0 62 10.32 11.0 20 76 17.96 15.9 40 85 17.44 18.9 60 92 20.16 21.3 80 102 25.52 24.7 100 112 28.12 28.1 120 121 30.36 31.1 140 134 35.36 35.5    Figure 109. Subject 181 curve and predictive equation.  y = -0.0001x2 + 0.3594x - 10.881R² = 0.979505101520253035400 20 40 60 80 100 120 140 160Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 181 Measured Minute VentilationMeasured minute ventilation Predictive equation trendline 290 Table 45- Subject 184 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 0.007(HR)^2 – 0.754 (HR) + 30.883 0 89 18.88 19.2 30 99 27.82 24.8 60 102 27.68 26.8 90 112 30.06 34.2 120 120 41.14 41.2 150 132 57.76 53.3 180 148 72.04 72.6    Figure 110. Subject 184 curve and predictive equation.     y = 0.007x2 - 0.754x + 30.883R² = 0.9791010203040506070800 20 40 60 80 100 120 140 160Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 184 Measured Minute VentilationMeasured minute ventilation Predictive equation trendline 291 Table 46- Subject 186 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 57.13ln(HR) – 229.87 0 88 23.60 25.9 30 101 36.40 33.8 60 110 39.78 38.7 90 114 42.70 40.7 120 128 44.32 47.3 150 140 49.46 52.4 180 150 58.98 56.4 210 162 60.80 60.8    Figure 111. Subject 186 curve and predictive equation.  y = 57.13ln(x) - 229.87R² = 0.95910102030405060700 20 40 60 80 100 120 140 160 180Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 186 Measured Minute VentilationMeasured minute ventilation Predictive equation trendline 292 Table 47- Subject 187 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 1.7296 e^ 0.0256(HR) 0 49 5.46 6.1 30 75 14.68 11.8 60 83 15.42 14.5 90 88 14.90 16.5 120 101 19.98 23.0 150 109 22.10 28.2 180 117 43.86 34.6 210 125 52.76 42.4 240 136 55.00 56.2 270 148 66.88 76.5    Figure 112. Subject 187 curve and predictive equation.  y = 1.7296e0.0256xR² = 0.951401020304050607080900 20 40 60 80 100 120 140 160Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 187 Measured Minute VentilationMeasured minute ventilation Predictive equation trendline 293 Table 48- Subject 190 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 2.2025 e^ 0.0192(HR) 0 86 10.92 11.5 30 125 29.36 24.3 60 135 30.84 29.4 90 143 29.80 34.3 120 152 39.96 40.8 150 162 44.20 49.4 180 167 51.00 54.4 210 174 67.88 62.2 240 179 72.40 68.5    Figure 113. Subject 190 curve and predictive equation.  y = 2.2025e0.0192xR² = 0.9658010203040506070800 50 100 150 200Minute Ventilation(Litres per minute)Heart rate (beats per minute)Subject 190 Measured Minute Ventilation Measured minute ventilation Predictive equation trendline 294 Table 49- Subject 193 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 0.0427 e^ 1.4758(HR) 0 69 21.76 22.1 30 86 32.90 30.6 60 98 34.96 37.1 90 105 40.76 41.0 120 115 46.16 46.9 150 128 52.32 55.0 180 141 70.52 63.4 210 158 72.16 75.0    Figure 114. Subject 193 curve and predictive equation.  y = 0.0427x1.4758R² = 0.9783010203040506070800 20 40 60 80 100 120 140 160 180Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 193 Measured Minute VentilationMeasured minute ventilation Predictive equation trendline 295 Table 50- Subject 197 Velotron test data points and predicted minute ventilation values. Programmed Power Output (watts) Measured HR (bpm) Measured  (L/min) Predicted  (using "power" formula for trendline); pred. = 0.0427 e^ 1.4758(HR) 0 67 14.74 18.5 30 79 28.50 25.5 60 90 33.46 31.8 90 94 39.94 34.1 120 102 36.16 38.5 150 111 40.40 43.5 180 122 45.78 49.4 210 132 53.56 54.6 240 141 63.04 59.2 270 156 66.46 66.6    Figure 115. Subject 197 curve and predictive equation.  y = -0.0006x2 + 0.6749x - 24.05R² = 0.9535010203040506070800 20 40 60 80 100 120 140 160 180Minute Ventilation (Litres per minute)Heart rate (beats per minute)Subject 197 Measured Minute VentilationMeasured minute ventilation Predictive equation trendline 296 Appendix S – Additional comparisons of Intakes 1, 2, and 3 Table 51- "Intake 2" of PMs by trial Subject and trial PM1  Intake  PM2.5 Intake  PM10 Intake  100C 9.0 14.8 45.5 100D    101C    101D 15.1 23.7 34.3 113C 5.0 8.3 24.5 113D 6.5 10.4 32.4 116C 13.9 32.4 61.6 116D 19.1 25.1 44.7 119C 3.9 8.1 15.3 119D*    122C 282.9 310.4 793.8 122D 38.6 48.1 71.3 129C 9.0 14.0 36.4 129D 21.5 34.0 63.6 130C 25.9 32.7 60.9 130D 7.6 12.4 33.6 133C 3.8 5.6 10.7 133D 8.9 13.1 38.8 134C 7.8 12.3 24.4 134D 3.1 10.3 26.3 141C 3.2 5.1 13.9 141D 8.3 16.1 27.8 142C 3.0 8.4 27.0 142D 8.0 11.7 22.1 145C 3.4 5.2 10.7 145D 5.4 9.5 41.8 147C 10.3 19.6 37.5 147D    150C 5.6 11.7 21.4 150D 13.7 19.9 50.8 152C 46.4 57.3 73.6 152D 7.0 15.3 39.9 154C 3.3 5.2 7.8 154D 6.2 8.4 11.7 157C 16.3 22.2 28.7 157D 13.6 18.1 25.1 158C 2.3 5.5 10.8 158D 25.2 37.0 55.0 159C 4.6 9.1 18.2 159D 4.0 5.4 8.9  297 Subject and trial PM1  Intake  PM2.5 Intake  PM10 Intake  160C 12.5 16.0 22.3 160D 13.1 23.9 34.5 161C 15.7 21.8 33.0 161D 29.5 34.3 41.5 162C 2.9 5.3 8.9 162D 8.4 14.3 21.9 163C 1.8 3.4 6.5 163D 3.9 6.1 14.1 164C 13.3 21.2 33.9 164D 21.8 29.3 42.9 165C 4.8 8.8 22.8 165D 4.8 9.1 18.9 166C 15.6 29.1 40.4 166D 4.6 7.7 12.6 169C 8.6 17.1 27.3 169D 101.4 106.0 119.0 170C 12.8 17.3 25.2 170D 4.1 5.9 10.4 172C 13.7 23.4 33.1 172D 5.1 9.5 17.6 178C 10.0 15.0 19.1 178D 6.4 10.9 17.7 181C 6.8 9.8 15.5 181D 6.0 8.4 13.4 184C 12.0 16.2 22.0 184D 5.0 7.7 13.4 186C 29.5 35.5 42.9 186D 10.5 15.7 26.5 187C 15.9 23.4 35.6 187D 11.4 16.2 30.5 190C 1.9 2.9 4.9 190D 3.9 5.4 8.8 193C 26.2 39.8 62.2 193D 26.7 35.4 51.2 197C 9.3 12.4 21.6 197D 233.9 245.5 252.2 Note 1: * data from this trial was only partial, this data was not included because the instrument ceases to function after approximately 30 minutes of the trial had passed.   298 Figure 115 shows the PM2.5 intake estimates (with jittered points) for the Downtown (DT) and Residential (Res) routes (Intake 1 = DT1 and Res1, Intake 2 = DT2 and Res2, Intake 3= DT3 and Res3). Statistical bars show the central point (the median), and the 25th and 75th percentile range. Intake 3 was not available for a total of 32 trials for PM2.5, while five intake estimates were not available for PM2.5 for Intake 2, and four estimates were not available for PM2.5 for Intake 1. The Downtown route showed a higher SD for the Downtown Intake 2 estimates, and “Max” values were also higher for all three Downtown Intake estimates. Two trials resulted in Intake 2 estimates beyond the visible y-axis (106 µg from 169-Downtown, and 245 ug from 197- Downtown), however they were included in analyses because they were the product of  and PM2.5 measurements within the upper range of normal. Table 18 shows the median intake values between the two routes were relatively comparable (ex. Intake 1 Downtown was 19.0 µg compared with Intake 1 Residential, which was 17.0 µg), though the SD of the Downtown intakes were higher, especially for Intake 2. 299  Figure 116. Scatter plot of PM2.5 Intake estimates one (1), two (2) and three (3) along each route.  Table 52- Summary statistics for estimated mass of PM2.5 intakes for Downtown and Residential routes.  PM2.5 mass µg Downtown Residential Intake 1 Intake 2 Intake 3 Intake 1 Intake 2 Intake 3 Mean 19.0 26.0 18.4 17.0 16.6 16.4 Median 13.4 14.3 14.7 14.0 14.4 15.7 SD 16.2 42.4 14.6 12.1 11.9 10.2 Min 4.8 5.4 5.3 3.3 2.9 2.8 Max 68.9 245 66.0 56.7 57.3 40.1  The correlation scatter plots and coefficients for PM2.5 are visible in Figure 117 and 118. Only the correlation coefficient between Intake 1 and Intake 3 is above 0.9. Residential trial 122 was already shown to be an unusual exposure concentration due to very high PM2.5 concentrations, so it has been removed from the analysis in Figure 117. By removing this trial, the correlation coefficient improved from 0.742 to 0.982 for Intake 2 and Intake 3.    300    Figure 117. Scatter plots, and correlation coefficients of each intake estimate of PM2.5 shown in Table 10. Trial 122C was removed from this analysis completely, while 119D and 193D (missing HR data) could not be correlated as matched pairs due to incomplete data.    Figure 118. Scatter plots, and correlation coefficients of each intake estimate of PM2.5 shown in Table 10. Trial 122C was removed from this analysis completely, while 119D and 193D (missing HR data) could not be correlated as matched pairs due to incomplete data.  Trials 169D and 197D also removed as data points were highly influential.  Intake 1 0.677 0.924050100150200250Intake 2 0.74210 20 30 40 50 60 701020304050600 50 100 150 200 250Intake 3Correlation of PM2.5 for Intake Estimates  for 1, 2, and 3- Trial 122C removedIntake 1 0.941 0.8921020304050Intake 2 0.98210 20 30 4 5102030401 20 30 40 50Intake 3C rrelation f PM2.5 for Intake Estimates  for 1, 2, and 3- Trials 122C, 169D, 197D removed 301 Appendix T – Paired t-test tables for other route comparisons Table 53- Paired t-test of Power Output, comparing the Residential and Downtown routes. Paired t-test comparing the mean power outputs of the two routes t = -2.5 df = 35 p-value = 0.018 alternative hypothesis: true difference in means is not equal to 0 95% CI: -18 -1.8 mean of the differences -9.9    Table 54- Paired t-test of HR (in bpm), comparing the Residential and Downtown routes. Paired t-test comparing the mean HRs of the two routes t = -2.2 df = 35 p-value = 0.035 alternative hypothesis: true difference in means is not equal to 0 95% CI: -8.5 -0.33 mean of the differences -4.4    Table 55- Paired t-test of cadences along the Residential and Downtown Routes Paired t-test comparing the mean cadence of the two routes t = -6.8 df = 34 p-value = 8.5 x 10-8 alternative hypothesis: true difference in means is not equal to 0 95% CI:  -7.5 -4.0  mean of the differences -5.8   302 Appendix U – Spirometry results by trial. Table 56- Spirometry results by trial, showing Post minus Pre differences for each parameter. Subject Route Post – Pre measurement FVC (L) FEV1 (L) FEF25-75% FEV1/FVC ratio 100 C -0.08 -0.10 -0.12 0.00 100 D -0.02 -0.06 0.44 -0.01 101 C -0.24 -0.41 -0.76 -0.05 101 D -0.15 -0.27 -0.34 -0.03 113 C 0.07 0.08 0.14 0.00 113 D -0.12 0.00 0.19 0.01 116 C -0.05 -0.18 -0.32 -0.03 116 D -0.14 -0.04 0.18 0.01 119 C 0.08 -0.07 -0.30 -0.03 119 D 0.03 0.01 0.06 -0.01 122 C 0.10 0.27 0.41 0.03 122 D n/a n/a n/a n/a 129 C 0.06 0.13 0.14 0.02 129 D 0.15 0.14 0.17 0.00 130 C -0.96 -0.58 -0.06 0.05 130 D -0.24 0.06 0.62 0.05 133 C 0.33 0.13 -0.11 -0.03 133 D -0.28 -0.19 -0.16 0.00 134 C 0.26 0.33 0.39 0.02 134 D -0.09 -0.11 -0.08 -0.01 141 C -0.13 0.27 0.64 0.05 141 D -0.02 -0.05 -0.03 -0.01 142 C 0.05 0.12 0.28 0.01 142 D 0.23 0.12 -0.15 -0.01 145 C 0.03 -0.10 -0.37 -0.03 145 D 0.19 0.13 0.18 -0.01 147 C -0.05 0.25 0.50 0.04 147 D 0.20 0.22 0.25 0.02 150 C 0.03 0.11 0.15 0.03 150 D 0.00 0.08 0.33 0.02 152 C -0.16 0.24 0.90 0.06 152 D 0.15 0.13 0.22 0.00 154 C -0.26 0.09 0.74 0.04 154 D -0.04 0.03 0.00 0.01 157 C 0.08 0.00 -0.15 -0.01 157 D -0.02 0.05 0.13 0.01 158 C 0.03 0.01 0.01 0.00  303 Subject Route Post – Pre measurement FVC (L) FEV1 (L) FEF25-75% FEV1/FVC ratio 158 D 0.24 0.26 0.33 0.01 159 C 0.25 -0.05 -0.43 -0.06 159 D 0.01 0.04 0.35 0.01 160 C -0.18 0.10 0.49 0.04 160 D 0.13 0.21 0.41 0.02 161 C -0.03 0.08 0.38 0.03 161 D -0.24 -0.09 0.20 0.03 162 C -0.03 0.13 0.45 0.03 162 D 0.19 0.07 -0.09 -0.01 163 C -0.04 0.03 0.14 0.01 163 D 0.26 0.19 0.09 -0.01 164 C -0.18 0.03 0.19 0.04 164 D -0.01 0.09 0.23 0.02 165 C 0.20 -0.02 -0.39 -0.03 165 D -0.18 0.10 0.57 0.04 166 C -0.14 0.02 0.13 0.02 166 D -0.01 -0.05 -0.05 -0.01 169 C 1.03 0.24 -0.69 -0.07 169 D 0.27 0.19 0.22 0.00 170 C 0.41 0.22 0.00 -0.02 170 D 0.00 0.00 0.14 0.00 172 C 0.03 0.20 0.46 0.03 172 D 0.16 0.16 0.30 0.01 178 C 0.29 0.09 -0.36 -0.03 178 D 0.09 0.01 -0.09 -0.01 181 C -0.20 0.17 0.73 0.06 181 D 0.08 -0.02 0.03 -0.01 184 C 0.71 0.06 -0.84 -0.11 184 D 0.13 -0.04 -0.39 -0.03 186 C 0.01 0.00 0.02 0.00 186 D 0.14 -0.12 -0.43 -0.03 187 C -0.11 0.03 0.33 0.02 187 D 0.12 0.11 0.10 0.00 190 C 0.01 0.15 0.39 0.02 190 D 0.21 0.20 0.34 0.01 193 C -0.42 -0.11 0.27 0.04 193 D 0.16 0.12 -0.20 -0.01 197 C -0.02 -0.01 0.07 0.00 197 D 0.14 0.03 -0.15 -0.02  304 Appendix V – Testing RHI data for normality Other studies have log-transformed their RHI data. This was not done for these RHI measures, due to the evidence seen in the QQ plots of residuals and histograms below. There is a more symmetrical distribution in the histogram of the loge-transformed data, but Shapiro-Wilkes test results are both p> 0.05.    Figure 119. RHI measurements plotted in histograms using the raw data values (left pane), and log-transformed values (right pane)  Shapiro-Wilk normality test, where p-value <0.05 means one should reject the NULL hypothesis (that the samples came from a normal distribution)  Original RHI Data (DTpost - DTpre) - (Respost - Respre)    305 Shapiro-Wilk test of normality W = 0.9738, p-value = 0.5905 Loge-transformed RHI Data (DTpost - DTpre) - (Respost - Respre)  Shapiro-Wilk test of normality W = 0.9738, p-value = 0.5905  Figure 120. Raw (left pane) and log-transformed (right pane) RHI data, in quantile-quantile plots  Both Shapiro-Wilk tests have a p-value larger than 0.05, therefore the null hypothesis should be accepted for both versions of the data, and it is likely both sets of data came from a normally distributed data set.    The histogram of the log-transformed data is slightly more symmetrical, however the Quantile-Quantile plots show very little difference between the original and the log-transformed data. For these reason, the original form of the RHI data has been analyzed in the form it was collected.  306 Appendix W – Correlation of PMs, with removal of trial 122C   Figure 121. Correlation of GMs of exposure data, including (left) and then excluding (right) the Residential trial by subject 122. PM1, PM2.5 and PM10 are measured in the unit of µg/m3, while PNC is measured in the unit of pt/cc.   When the data were analyzed by correlating all of the trials (including 122C) compared to all of data (except 122C), very different correlation coefficients emerged, as shown in Figure 14. The Pearson product-moment correlation coefficient between PM1, PM2.5, and PM10 continued to be above r = 0.8, and the correlation coefficients between PNC and the larger PM sizes increased to being weakly correlated (r = 0.156 to r = 0.292) after the exclusion of trial 122C. The data from participant 122 along the Residential route appears to exert a large influence on this analysis compared to those of the other trials, which is unlike the other trials that were completed on the surrounding dates (August 9th, 11th, and 17th, 2011). Knowing the data for trial 122C was collected under uncharacteristic air quality conditions for the city of Vancouver, this was justification to exclude the larger PM data (PM1, PM2.5, PM10) in particular from most analyses. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0107223/manifest

Comment

Related Items