Effects of air pollution exposure on adult bicycle commuters: an investigation of
respiratory health, motorised traffic proximity and the utility of commute re-routing
Doctor of Philosophy
Thesis by Publication
2012
Thomas A. Cole-Hunter
B. App. Sci. (Hons I)
International Laboratory for Air Quality and Health,
Institute of Health and Biomedical Innovation,
Faculty of Science and Engineering,
Queensland University of Technology,
Brisbane, Australia
2
TABLE of CONTENTS
LIST of PUBLICATIONS ......................................................................................................... 5
ABSTRACT ............................................................................................................................... 7
KEY WORDS ............................................................................................................................ 9
ABBREVIATIONS ................................................................................................................. 10
NOMENCLATURE ................................................................................................................ 10
STATEMENT of ORIGINAL AUTHORSHIP ....................................................................... 11
ACKNOWLEDGEMENT and DEDICATION ....................................................................... 12
1. INTRODUCTION ........................................................................................................ 13
1.1. Description of Research Problem Investigated ......................................................... 13
1.2. Overall Objectives of the Study ................................................................................ 13
1.3. Specific Aims of the Study ........................................................................................ 14
1.3.1. Project 1 ................................................................................................................. 14
1.3.2. Project 2 ................................................................................................................. 14
1.3.3. Project 3 ................................................................................................................. 15
1.4. Specific Hypotheses of the Study.............................................................................. 16
1.5. Account of Research Progress Linking the Research Papers .................................... 17
2. LITERATURE REVIEW ............................................................................................. 18
2.1. Introduction ................................................................................................................... 18
2.2. Ultrafine Particles (UFP) ............................................................................................... 19
2.3. In-transit UFP Exposure ................................................................................................ 20
2.4. UFP and Health Effects ................................................................................................. 26
2.5. Mechanisms of Cardiopulmonary Detriment ................................................................ 30
2.5.1. Measure of Dose: Surface Area vs. Number Concentration ...................................... 30
2.5.2. Effects of Chemical Composition .............................................................................. 31
2.5.3. Acute versus Chronic Exposure ................................................................................. 32
3
2.6. Knowledge Gaps ........................................................................................................... 34
3. GENERAL METHODS................................................................................................ 36
3.1. Project Design (Project 1, 2 & 3) .............................................................................. 36
3.2. Participant Recruitment (Project 1 & 3) .................................................................... 37
3.3. Questionnaire (Project 1 & 3) ................................................................................... 37
3.3.1. Perception Incidence and Magnitude .................................................................... 38
3.3.2. Symptom Incidence and Severity .......................................................................... 38
3.3.3. Risk Management Strategies ................................................................................. 39
3.4. Ultrafine Particle Monitoring (Project 2 & 3) ........................................................... 39
3.5. Ambient Pollution Data Monitoring (Project 1, 2 & 3) ............................................ 40
3.6. Heart Rate and Estimated Ventilation Rate (Project 2 & 3) ..................................... 40
3.7. Geographical Positioning (Project 2 & 3) ................................................................. 41
3.8. Meteorological Data Monitoring (Project 2 & 3) ...................................................... 41
4. PROJECT ONE: A questionnaire-based investigation of perceptions, symptoms and risk
management strategies for air pollution exposure and motorised traffic proximity of adult
bicycle commuters ................................................................................................................ 42
5. PROJECT TWO: Inhaled particle counts on bicycle commute routes of low and high
proximity to motorised traffic .............................................................................................. 73
6. PROJECT THREE: The reduction of ultrafine particle exposure by utilising bicycle
commute routes of low versus high proximity to major motorised traffic corridors ........... 76
7. GENERAL DISCUSSION ......................................................................................... 113
7.1. Introduction and Summary .......................................................................................... 113
7.2. Air Quality ............................................................................................................... 114
7.3. Air Pollution Exposure Symptoms and Susceptibility ............................................ 115
7.4. Air Pollution Exposure Perceptions and Risk Management ................................... 115
7.5. Novel Method Use .................................................................................................. 116
7.6. General Limitations ................................................................................................. 117
8. CONCLUSIONS......................................................................................................... 118
4
9. FUTURE DIRECTIONS ............................................................................................ 120
10. APPENDICES ............................................................................................................ 122
A. Questionnaire [Complete (Project 1)] ..................................................................... 122
B. Questionnaire [Amended (Project 3)] ..................................................................... 132
C. Bikeway Maps with Cyclist Counts (Project 2) ...................................................... 135
D. Media Releases (to assist participant recruitment for Projects 1 and 3) ................. 136
E. Publication of Project 2 ........................................................................................... 141
11. GENERAL REFERENCES ........................................................................................ 148
5
LIST of PUBLICATIONS
Journal Articles
Knibbs L.D., Cole-Hunter T., Morawska L. (2011). A review of commuter exposure to
ultrafine particles and its health effects. Atmospheric Environment, 45:2611-2622.
Cole-Hunter, T., Morawska, L., Stewart, I., Jayaratne, R., Solomon, C. (2012). Inhaled
particle counts on bicycle commute routes of low and high proximity to motorised traffic.
Atmospheric Environment, 61:197-203. doi: 10.1016/j.atmosenv.2012.06.041
Cole-Hunter, T., Morawska, L., Solomon, C. (In Review). A questionnaire-based
investigation of perceptions, symptoms, and risk management strategies for motorised traffic
exposure of adult bicycle commuters. PLoS ONE.
Cole-Hunter, T., Morawska, L., Stewart, I., Jayaratne, R., Solomon, C. (In Review). Utility of
bicycle commute route alteration to lower exposure to motorised traffic-emitted ultrafine
particles. Environmental Health.
International Conference Presentations
Cole-Hunter, T., Morawska, L., Stewart, I., Jayaratne, R., Solomon, C. Utility of bicycle
commute route alteration to lower exposure to motorised traffic-emitted ultrafine particles.
European Respiratory Society Annual Congress, Vienna, Austria, September 2012.
Cole-Hunter, T., Morawska, L., Solomon, C. Effects of Air Pollution on Lung Health and
Function: A Direct Investigation with Brisbane City Commuter Cyclists. European
Respiratory Society Annual Congress, Barcelona, Spain, September 2010.
6
National Conference Presentations
Cole-Hunter, T., Morawska, L., Solomon, C. Brisbane City Commuter Cyclists: Behaviour,
Reasoning and Risk Management. Public Health Association Australia State Conference,
Brisbane, QLD, March 2010.
Cole-Hunter, T., Stewart, I., Solomon, C. Effect of Exercise Mode on 40-50 Year Old Female
Triathletes. Sports Medicine Australia ‘Be Active’ National Conference, Brisbane, QLD,
August 2009.
7
ABSTRACT
Bicycle commuting has the potential to be an effective contributing solution to address some
of modern society’s biggest issues, including cardiovascular disease, anthropogenic climate
change and urban traffic congestion. However, individuals shifting from a passive to an
active commute mode may be increasing their potential for air pollution exposure and the
associated health risk. This project, consisting of three studies, was designed to investigate
the health effects of bicycle commuters in relation to air pollution exposure, in a major city in
Australia (Brisbane).
The aims of the three studies were to: 1) examine the relationship of in-commute air pollution
exposure perception, symptoms and risk management; 2) assess the efficacy of commute re-
routing as a risk management strategy by determining the exposure potential profile of
ultrafine particles along commute route alternatives of low and high proximity to motorised
traffic; and, 3) evaluate the feasibility of implementing commute re-routing as a risk
management strategy by monitoring ultrafine particle exposure and consequential
physiological response from using commute route alternatives based on real-world
circumstances; 3) investigate the potential of reducing exposure to ultrafine particles (UFP; <
0.1 µm) during bicycle commuting by lowering proximity to motorised traffic with real-time
air pollution and acute inflammatory measurements in healthy individuals using their typical,
and an alternative to their typical, bicycle commute route.
The methods of the three studies included: 1) a questionnaire-based investigation with regular
bicycle commuters in Brisbane, Australia. Participants (n = 153; age = 41 ± 11 yr; 28%
female) reported the characteristics of their typical bicycle commute, along with exposure
perception and acute respiratory symptoms, and amenability for using a respirator or re-
routing their commute as risk management strategies; 2) inhaled particle counts measured
along popular pre-identified bicycle commute route alterations of low (LOW) and high
(HIGH) motorised traffic to the same inner-city destination at peak commute traffic times.
During commute, real-time particle number concentration (PNC; mostly in the UFP range)
and particle diameter (PD), heart and respiratory rate, geographical location, and
meteorological variables were measured. To determine inhaled particle counts, ventilation
rate was calculated from heart-rate-ventilation associations, produced from periodic exercise
testing; 3) thirty-five healthy adults (mean ± SD: age = 39 ± 11 yr; 29% female) completed
8
two return trips of their typical route (HIGH) and a pre-determined altered route of lower
proximity to motorised traffic (LOW; determined by the proportion of on-road cycle paths).
Particle number concentration (PNC) and diameter (PD) were monitored in real-time in-
commute. Acute inflammatory indices of respiratory symptom incidence, lung function and
spontaneous sputum (for inflammatory cell analyses) were collected immediately pre-
commute, and one and three hours post-commute.
The main results of the three studies are that: 1) healthy individuals reported a higher
incidence of specific acute respiratory symptoms in- and post- (compared to pre-) commute
(p < 0.05). The incidence of specific acute respiratory symptoms was significantly higher for
participants with respiratory disorder history compared to healthy participants (p < 0.05). The
incidence of in-commute offensive odour detection, and the perception of in-commute air
pollution exposure, was significantly lower for participants with smoking history compared to
healthy participants (p < 0.05). Females reported significantly higher incidence of in-
commute air pollution exposure perception and other specific acute respiratory symptoms,
and were more amenable to commute re-routing, compared to males (p < 0.05). Healthy
individuals have indicated a higher incidence of acute respiratory symptoms in- and post-
(compared to pre-) bicycle commuting, with female gender and respiratory disorder history
indicating a comparably-higher susceptibility; 2) total mean PNC of LOW (compared to
HIGH) was reduced (1.56 x e4 ± 0.38 x e4 versus 3.06 x e4 ± 0.53 x e4 ppcc; p = 0.012).
Total estimated ventilation rate did not vary significantly between LOW and HIGH (43 ± 5
versus 46 ± 9 L•min; p = 0.136); however, due to total mean PNC, accumulated inhaled
particle counts were 48% lower in LOW, compared to HIGH (7.6 x e8 ± 1.5 x e8 versus 14.6
x e8 ± 1.8 x e8; p = 0.003); 3) LOW resulted in a significant reduction in mean PNC (1.91 x
e4 ± 0.93 x e4 ppcc vs. 2.95 x e4 ± 1.50 x e4 ppcc; p ≤ 0.001). Commute distance and
duration were not significantly different between LOW and HIGH (12.8 ± 7.1 vs. 12.0 ± 6.9
km and 44 ± 17 vs. 42 ± 17 mins, respectively). Besides incidence of in-commute offensive
odour detection (42 vs. 56 %; p = 0.019), incidence of dust and soot observation (33 vs. 47
%; p = 0.038) and nasopharyngeal irritation (31 vs. 41 %; p = 0.007), acute inflammatory
indices were not significantly associated to in-commute PNC, nor were these indices reduced
with LOW compared to HIGH.
The main conclusions of the three studies are that: 1) the perception of air pollution exposure
levels and the amenability to adopt exposure risk management strategies where applicable
will aid the general population in shifting from passive, motorised transport modes to bicycle
9
commuting; 2) for bicycle commuting at peak morning commute times, inhaled particle
counts and therefore cardiopulmonary health risk may be substantially reduced by decreasing
exposure to motorised traffic, which should be considered by both bicycle commuters and
urban planners; 3) exposure to PNC, and the incidence of offensive odour and
nasopharyngeal irritation, can be significantly reduced when utilising a strategy of lowering
proximity to motorised traffic whilst bicycle commuting, without significantly increasing
commute distance or duration, which may bring important benefits for both healthy and
susceptible individuals.
In summary, the findings from this project suggests that bicycle commuters can significantly
lower their exposure to ultrafine particle emissions by varying their commute route to reduce
proximity to motorised traffic and associated combustion emissions without necessarily
affecting their time of commute. While the health endpoints assessed with healthy individuals
were not indicative of acute health detriment, individuals with pre-disposing physiological-
susceptibility may benefit considerably from this risk management strategy – a necessary
research focus with the contemporary increased popularity of both promotion and
participation in bicycle commuting.
KEY WORDS
Air Pollution ; Ultrafine Particles ; Bicycle Commuting ; Inhaled Particle Count ;
Perceptions ; Symptoms ; Lung Function ; Inflammatory Mediators ;
Risk Assessment ; Risk Management
10
ABBREVIATIONS
BPM = Beats per minute
BUG = Bicycle user group
CBD = Central business district
EXP = Estimate exposure potential
NOX = Nitrogen oxides
NO2 = Nitrogen dioxide
PEF = Peak expiratory flow
PM = Particulate matter
PM2.5 = PM <2.5 micrometres diameter
PD = Particle diameter
PNC = Particle number concentration
PPCC = Particles per cubic centimetre
SE QLD = South-East Queensland
UFP = Ultrafine particle
NOMENCLATURE
Active transport = A mode of travel propelled by physical effort
Bicycle commuting = A mode of active transport, using a bicycle
11
STATEMENT of ORIGINAL AUTHORSHIP
The work contained in this thesis has not been previously submitted to meet requirements for
an award at this or any other higher education institution. To the best of my knowledge and
belief, the thesis contains no material previously published or written by another person
except where due reference is made.
Signature:
Date: / /
12
ACKNOWLEDGEMENT and DEDICATION
I would like to acknowledge my supervisors Dr Colin Solomon, Prof Lidia Morawska, Dr Ian
Stewart and Dr Rohan Jayaratne for their guidance and patience of my progress throughout a
challenging yet highly-rewarding candidature. You have all inspired me to better perform and
uphold the practice of scientific research for a more righteous path in the post-doctoral world.
Dear Colin, I will now endeavour to avoid taking the scenic route when writing academically.
A special acknowledgement should be made to Woody Pattinson from the University of
Canterbury’s Geography Department for assisting (without obligation) to tame the beast that
is geographic information systems software. Additionally, many thanks go to Prof Diana
Battistuta and Dr Dimitrios Vagenas for their assistance with my apparent statistical
disability. On this note, Camilla Tuttle and Matthew Hadaway, thank you very much for
shedding light on the more complicated laboratory procedures that were learnt with your
help. Further, I would like to thank Dr Luke Knibbs for providing me with the opportunity to
co-author a review article during my candidature.
As vital as a professional cooperative of superiors and peers is, the emotional support and
personal encouragement from family and close friends is just as valuable for any student. I
would like to thank and dedicate my journey so-far to my family and close friends. Mum and
my older brother Sean, you have been very supportive of my empty stomach, and tolerant of
my sometimes impatient urge to get things done (which is a habit I have acquired throughout
my candidature). Dad and my oldest brother Ross, you have been very supportive of my
empty wallet, and a personal inspiration to get things done and done well.
To my brothers and close mates (who know who they are), you have been a wonderful
distraction from the strains of post-graduate study, and a constant source of refreshment. The
last three years have gifted me a thirst for knowledge that will leave me drinking for the rest
of my life. Cheers!
13
1. INTRODUCTION
1.1. Description of Research Problem Investigated
Bicycle commuting is being encouraged in many cities around the world to improve public
health, air quality, and traffic congestion. Information concerning the effects of air pollution,
such as exposure perception, symptoms and risk management will contribute to responsible
advocacy and the implementation of appropriate infrastructure for bicycle commuters.
Frequent and chronic exposure to ultrafine particles (UFP; an indicator of air quality and a
major emission from motor vehicles) can generally result in cardiopulmonary detriment:
short term, through inflammation and reduced cardiopulmonary function; long term, through
accumulative toxic dose in the cardiopulmonary system. Dose of exposure and therefore
exposure risk is determined by commute duration, frequency, proximity to and extent of an
emission source. Bicycle commuters choosing to ride along major roads during peak traffic
times are sharing a microenvironment with high motorised traffic and associated exhaust
emissions [1]. A shorter route with higher particle number concentrations (PNC; mostly
including UFP), rather than a longer route with lower PNC, may facilitate a greater daily
PNC exposure [2]. Bicycle commuting in an urban environment of poor air quality has been
demonstrated as a potential health risk for susceptible participants; however, limited studies
have assessed the potential of exposure risk management strategies to minimise this potential
health risk for these individuals.
1.2. Overall Objectives of the Study
Project 1 aimed to investigate how personal and commute characteristics influenced air
pollution exposure perception, symptoms and adoption of risk management strategies.
Subsequently, Project 2 aimed to assess the improvement of air quality and consequential
reduction of inhaled ultrafine particle count by lowering proximity to motorised traffic, as a
risk management strategy, along popular bicycle commute routes.
14
Finally, Project 3 aimed to evaluate the utility of implementing commute re-routing as a risk
management strategy by monitoring air quality and physiological inflammatory
measurements with healthy frequent bicycle commuters and high or low proximity to
motorised traffic alterations of their typical bicycle commute activity.
1.3. Specific Aims of the Study
Project 1
The first project (Project 1) of this thesis was intended to assist in participant recruitment
and guide a large epidemiological field study (similar in design to Project 3).
Accordingly, the specific aims of Project 1 were to:
1. Characterise a large group (100+) of frequent bicycle commuter’s (i.e. participant’s)
travelling into and/or through a one kilometre radius of the Brisbane Central Business
District (CBD);
2. Determine if a participant’s personal and commute characteristics (including
estimated exposure) influence the incidence and level of perceived air pollution
exposure;
3. Determine if a participant’s personal and commute characteristics influence the
incidence and severity of acute respiratory symptoms attributable to air pollution
exposure;
4. Inquire if participant’s have previously used, or considered the use of, specific air
pollution exposure risk management strategies (i.e. commute re-routing or a
respirator), and the importance of specific factors related to their use;
5. Evaluate if a participant’s personal and commute characteristics, perceived exposure
level and acute symptom experience will influence their use of risk management
strategies.
Project 2
In conjunction with the first project’s proposition to guide a large epidemiological field
study, it was deemed appropriate to form an air pollution exposure profile of popular
bicycle commute routes from Brisbane suburbs to the CBD (driven by findings from the
15
first project). That is, the specific aims of the second project (Project 2) of this thesis were
to:
1. Define popular bicycle commute routes to Brisbane CBD from Northern, Eastern,
Southern and Western suburbs, identified using data from the first project; determine
commute route alterations of low and high proximity to motorised traffic;
2. Monitor and compare air quality and pulmonary ventilation rates along commute
route alterations;
3. Assess if a significant increase in air quality is experienced along routes of low,
compared to high, proximity to motorised traffic;
4. Determine if air quality is more influential than ventilation rate in deciding commute-
related inhaled particle count.
Project 3
Guided by findings of Project 2 and 3, the specific aims of the third project (Project 3)
were to:
1. Recruit healthy, frequent bicycle commuters (i.e. participant’s) of the Brisbane CBD;
2. Pre-determine routes of low and high proximity to motorised traffic, based on
recruited participant’s typical commute routes;
3. Monitor participant’s whilst bicycle commuting inbound and outbound along pre-
determined routes of low and high proximity to motorised traffic;
4. Assess if air pollution exposure is significantly reduced using a route of low,
compared to high, proximity to motorised traffic, and, if the presentation of
symptoms, lung function and markers of airway inflammation indicate a significantly
reduced biological inflammatory response;
5. Query which route alteration was favourable for participant’s and why.
16
1.4. Specific Hypotheses of the Study
The direct and testable hypotheses of this study, according to each project, are that:
1) For adult bicycle commuters: the incidence and severity of acute respiratory
symptoms will be greater in-commute compared to pre- and post-commute, and with
individuals of respiratory disorder history or female gender, and will be positively-
associated with estimated proximity to motorised traffic; the perceived exposure
levels of air pollution by adult bicycle commuters will be consistent with estimated
motorised traffic proximity levels; and, the amenability to adopt air pollution
exposure risk management strategies will be positively-associated with perceived
exposure levels and incidence of, or physiological pre-disposition to, acute respiratory
symptoms.
2) For the feasibility of commute re-routing: using a bicycle commute route of lower
exposure to motorised traffic (predominantly-determined by commute proximity to
motorised traffic) will facilitate a significant reduction in PNC, compared to a bicycle
commute route of higher motorised traffic exposure; mean heart rate and associated
pulmonary ventilation rate (as indices of physical effort) will not vary significantly
between bicycle commute routes of low and high motorised traffic exposure, and;
variation of inhaled particle counts will predominantly be determined by PNC levels
rather than physical effort (indicated by heart and ventilation rate).
3) For the practicality of commute re-routing: a route alteration designed to lower
proximity to motorised traffic during bicycle commuting will significantly reduce
exposure to combustion emissions [represented by the dominant ultrafine particle
(UFP; < 0.1 µm) number concentration (PNC)], compared to a high interaction route;
health outcomes (as incidence and severity of acute respiratory symptoms, peak flow
rate, and cell distribution in sputum) will be improved with the use of a route of lower
proximity to motorised traffic, compared to a route of high interaction, and; the
difference in the estimated inhaled UFP count between the two routes will be
attributable to the difference in PNC, rather than any differences in physical effort
(and therefore heart and ventilation rate).
17
1.5. Account of Research Progress Linking the Research Papers
Project 1 informed and guided Projects 2 and 3 by providing real-world characteristics of the
regions bicycle commuting population. The estimates, perceptions and attributable-symptoms
of air pollution exposure for participants’ typical bicycle commuting activity were inquired
of, along with importance of factors for risk management strategy adoption. The strategy of
lowering proximity to motorised traffic was profiled in Project 2 and evaluated in Project 3.
Project 2 utilised findings from Project 1 to accurately profile the potential of exposure to
ultrafine particles along popular bicycle commute routes from greater Brisbane to the inner-
city region, associated with the daily weekday work commute. Along with in-commute air
quality monitoring, heart rate and breathing frequency were monitored to calculate in-
commute ventilation rate as an indication of inhaled particle count. Exercise testing was
performed to produce heart rate – ventilation association equations. With the strategy of
lowering proximity to motorised traffic profiled and shown to be effective, the results of
Project 2 guided the evaluation of appropriateness for frequent bicycle commuters under real-
world circumstances as tested in Project 3.
Project 3 utilised findings from Projects 1 and 2 to define and evaluate circumstances
representative of the regions bicycle commuting participant’s. The risk management strategy
evaluated in Project 1, and trialled in Project 2, was applied to real-world circumstances.
Frequent bicycle commuters travelling to and from the inner-city region of Brisbane on
multiple days of the week performed alterations of their typical bicycle commute route, with
one facilitating high, and the other low, proximity to motorised traffic. The air quality and
subsequent physiological response of a participant was assessed when performing an inbound
and outbound trip of the high and low motorised traffic alterations of their typical bicycle
commute.
18
2. LITERATURE REVIEW
2.1. Introduction
Bicycle commuting, being a physically-active and fossil fuel-independent transport mode, is
increasingly promoted as an effective solution to address anthropogenic climate change,
motorised traffic congestion, and inactivity-related cardiovascular disease [3-5]. It is known
that elevated air pollution exposure is a substantial health risk [6-9], and that this risk is
increased with higher pulmonary ventilation [10-13] and greater proximity to motorised
traffic emissions [11, 14-19]; however, the overall associated benefits of reduced air pollution
emissions and increased physical activity can exceed such risks [6, 20-22]. Barriers may exist
for individuals choosing to shift from passive, motorised transport to bicycle commuting,
such as greater perceived or actual risk of exposure to air pollution [6, 19]. Fortunately,
insight is being provided for cause-effect mechanisms, and methods are being proposed to
reduce the degree of air pollution exposure whilst actively commuting [12, 23-26].
Air pollutants of potential concern for bicycle commuters in urban environments include
nitrogen dioxide (NO2) and particulate matter (PM) as they are emitted with motorised traffic
exhaust, can be consciously detected (i.e. perceived), can elicit acute respiratory symptoms
upon exposure [6], and are regionally-monitored. Typical examples of acute (i.e. quick-onset
and short-lived) symptoms which may arise in an individual exposed to such pollutants at
elevated concentrations, especially if an individual has a pre-disposing susceptibility, are
nasopharyngeal irritation, airway inflammation and broncho-constriction [27]. Short episodes
of PM exposure may only elicit acute health detriments, however frequent episodes (e.g.
daily commute-related exposure) could lead to heightened exposure sensitivity and increased
likelihood of chronic disease development in susceptible individuals [28].
Recent epidemiological evidence of mortality effects and long-term exposure to fine PM
suggest that adverse health effects are dependent on both exposure factors of concentration
and duration; long-term exposure can result in greater, more persistent cumulative effects
compared to short-term exposures [29]. Chronic exposure to elevated NO2 and PM can
suppress airway immune defences and consequently increase the incidence and severity of
(sometimes debilitating) upper respiratory tract infections [30, 31]. The methodological
19
reporting of acute respiratory symptoms has been used previously to assess air pollution
exposure in healthy and asthmatic children [32-34], as well as healthy and asthmatic adults
[35, 36], the elderly [37], and the general community [38, 39]. Further, questionnaires of
symptom inquiry have been used successfully with adults to investigate multi-modal air
pollution exposure perceptions for work-related commuting [40]. However, not all pollutants
elicit acute presentable symptoms, such as ultrafine particles (UFP; < 0.1 µm, the dominant
diameter range associated with PNC).
2.2. Ultrafine Particles (UFP)
An acknowledgement of ultrafine particles (UFP) and their exposure-related health
implications has garnered strength in the last two decades [41, 42], with general agreement
that UFP possess greater toxicity potential than coarser particles because of relatively higher
concentrations and surface area per mass [7, 43-45]. Further, UFP have a greater ability to
penetrate through the pulmonary system [46, 47] so far as to elicit detrimental effects on the
cardiovascular and nervous systems [48, 49]. UFP exposure, and thus associated health risk,
can vary with distance from a major emission source such as motorised traffic corridors [9,
12, 50]. Proximity to motorised traffic emissions is positively-correlated to pulmonary
dysfunction and biomarkers of systemic inflammation [51]. In addition to emission source
proximity, commute duration will influence UFP exposure potential of bicycle commuters
[2].
Risk management strategies of reducing air pollution exposure whilst actively commuting
can include lowering a commute’s proximity to motorised traffic by avoiding main roads at
peak traffic times [52]. Most popularly, this has been investigated by comparing micro-
environments of designated off-road and on-road bicycle paths to find that the former
generally facilitates a significantly lower air pollution exposure potential than the latter [2,
43, 50, 53, 54]. Health endpoints, including acute respiratory symptoms, impaired lung
function and cellular inflammatory mediators have been used to investigate a physiological
response to air pollution [55-57]. However, the practicality of an informed route alteration by
bicycle commuters to avoid major motorised traffic corridors and reduce their exposure risk
(thereby improving health endpoints) is yet to be investigated. Due to practical limitations
associated with sensitivity, size and weight of air pollution monitoring equipment, limited
20
knowledge is available on the UFP exposure potential of active transport participants.
However, the magnitude of PNC exposure during bicycle commuting proximal to motorised
traffic is believed to be associated with both motorised traffic intensity and proximity [8].
2.3. In-transit UFP Exposure
Several factors may determine the degree of exposure to UFP according to the mode of
transport used. For example, some transport modes facilitate shorter periods of exposure
because travel is faster, and some result in higher concentrations of exposure because travel is
physically-active (with higher pulmonary ventilation rates to satisfy oxygen demands). A
recent review of personal, micro-environmental and ambiental exposure studies using
different transport modes in urban street micro-environments highlighted that pedestrians and
cyclists experience lower fine PM concentrations compared to vehicle passengers despite the
physical barrier between cabin and outside air; PM concentrations were significantly
positively-correlated to pollutant source proximity and therefore it was recommended that
active commuters seek routes distal to high-traffic corridors [58]. Further, it was revealed that
fixed (that is, ambient or non-direct) monitoring stations were poor predictors of PM
concentrations at an individual level. Factors of significant influence included geography and
meteorology, yet a large proportion of variation attributable to other factors was considered
unaccountable. In conclusion, the authors called for further direct and personal exposure
assessment studies to be performed in aid of developing more appropriate exposure
minimisation strategies. More recently, a meta-analysis was performed using 47 exposure
studies across 6 transport modes, including bicycling and walking [59]. From this meta-
analysis, it was indicated that the mean UFP concentrations of bicycle, bus, automobile, rail,
walking and ferry modes are typically 3.4, 4.2, 4.5, 4.7, 4.9 and 5.7 × 104 particles per cubic
centimetre (ppcc), respectively, suggesting that bicycle commuting may facilitate the lowest
exposure to UFP compared to other common transport modes (see Figure 1, below).
21
Figure 1 - Mean PNC in-transit of different transport modes, including cycling and
walking [59].
The mean values of PNC per transit mode previously studied are trip-weighted, meaning that the
number of trips taken will influence the power of the findings - the number of trips taken in each mode
is shown in brackets. Error bars denote the trip-weighted standard deviation.
2.3.1. Active Transport: Bicycling
Cyclists are relatively unprotected from their external environment compared to individuals
travelling within a motor vehicle cabin ventilated with filtered-air. Accordingly, cyclists are
more prone to air pollution exposure due to this intimacy with the ambient atmosphere.
Furthermore, the increased aerobic demand and therefore higher pulmonary ventilation rates
associated with physical activity can increase the inhaled dose of an air pollutant [59].
Modest exposures to traffic-related emissions have elicited biological effects that suggest
adverse health detriment [59]. Personal UFP exposure in healthy non-smoking individuals
cycling in traffic has been associated with acute respiratory effects, including decreased lung
function (e.g. reduced peak expiratory flow rate) and acute respiratory symptoms (e.g. tussis
and chest wheezing) [9], although not with indicators of DNA damage [60].
22
2.3.1.1. Emission Source Proximity
Strong recommendations have been made for bicycle commuters to re-
consider their commute route, especially during certain weather conditions
(such as dry, warm weather), to reduce UFP exposure [2, 61]. To effectively
minimise PM exposure while bicycle commuting, it has been shown necessary
to design commute routes at a distance of 500 m from heavy traffic corridors
[18]. However, PM exposures along bicycle paths of differing urban zoning
and traffic volume has been mapped to find lower than expected exposures
(compared to previous findings), with day-to-day variability significantly
affected by meteorological conditions (e.g. PNC is negatively-correlated with
air temperature and wind speed) [61]. Urban zoning and traffic proximity
significantly affects exposure along bicycle routes, with highest exposures of
UFP associated with heavy traffic volumes [54]. Care should be given to local
conditions and methods when using central monitoring stations to characterise
wide-spread population exposures, in combination with an appropriate number
of samples, especially for long-term cohort epidemiological analyses assessing
population health [61].
2.3.1.2. Commute Duration and Time
In addition to commute location, commute departure time will influence the
exposure potential of cyclists. PM exposure from cycling the shortest and
highest exposure route compared to a longer and lower exposure route, inside
and outside peak traffic times, has been studied [2]. Passengers within a bus
following high-intensity motorised traffic corridors can experience greater PM
exposure than bicycle commuters riding along their equivalently-highest
exposure route, during peak traffic times [2]. A bicycle commuter’s exposure
to UFP and larger particles can be elevated during morning office hours
compared to moderate levels during the early afternoon, associated with
vehicle traffic and the proportion of heavy diesel vehicles [62]. A dust
mapping technique has been previously presented as a useful tool for town
planners and local policy makers, in addition to a photographic global
23
positioning (GPS) outlay to illustrate ‘hotspots’ such as traffic lights, as well
as an internet-based mapping program to help cyclist commuters choose a
route based on exposure with diurnal variations [62].
2.3.1.3. Bicycling versus other modes
Several studies have compared the exposure potential of cyclists against
motorists [13, 53, 63]. PNC has been measured simultaneously during cycling
and driving along pre-determined routes of 10-20 minutes duration over
eleven days, finding that motorists had an overall mean PNC exposure higher
than cyclists (by 5%) [63]. Variability of exposure was seen within and
between routes, with factors including passing motor vehicles and traffic
control infrastructure (such as intersections and lights) [63]. The ratio of mean
PNC between motorists and cyclists in this study (1.05) is in close agreement
with another multi-modal transport study (1.06; [53]). However, the estimated
ventilation rate of cyclists compared to motorists (i.e. car and bus passengers)
during transit, using real-time heart rates (with pre-determined individual
regression coefficients calculated between heart and ventilation rate during
sub-maximal cycle ergometry), showed cyclists to have a greater ventilation
rate (approximately twice-fold, with inter-individual and gender variability)
[13].
PNC and ventilatory parameters simultaneously measured for individuals
transiting via bicycle and motor vehicle along identical routes of three
different locations showed that mean bicycle-to-car ratios for PNC was
approximately equal and without significance, with any insignificant degree of
difference in PNC dependent on location [64]. While intra-modal
concentrations of PM can be consistent between locations, cyclists can
experience significantly greater inhaled count and deposition dose of PM, due
to greater minute ventilation (VE), by a factor of approximately four compared
to car passengers [64]. Accordingly, it is suggested that exposure estimates
(incorporating inhaled count and dose), rather than concentrations alone,
should be used when evaluating health risks of transport modes and policies.
24
Most studies comparing cycling and other transit modes have found PM exposure to be at
lower concentrations whilst cycling [14, 65, 66], however due to the consequential increased
ventilation associated with active transport, inhaled counts and doses can be similar, if not
greater, for cyclists [63, 64]. Internationally, it can be inferred that walking and cycling are
the most common forms of transport in an urban environment, yet relatively few personal
exposure studies have been performed for these active transport modes, probably due to
logistical constraints.
2.3.2. Active Transport: Walking
Few studies concerning pedestrian exposure to UFP have been published. Typically,
pedestrians travel at the lowest speed of any transport mode, however they experience higher
ventilation rates than motorists (although less than cyclists) due to physical demands. Also,
similar to cyclists, pedestrians are less protected from freshly-emitted motor vehicle
emissions of adjacent motorised traffic when walking along traffic corridors compared to
passengers within a motor vehicle cabin ventilated with filtered-air. Therefore, pedestrians
are potentially the most exposed of any transport participant, depending on duration of travel
and urban micro-environmental conditions.
2.3.2.1. Emission Source Proximity
Walking adjacent to heavy traffic corridors has been shown to facilitate a
median PNC of approximately 3.8 x 104 ppcc [67], with pedestrians found to
experience a lower fine PM exposure mean compared to cabin-bound
motorists (6.8 x 105 ppcc and 8.0 x 105 ppcc, respectively); however, no
significant differences were found between direction and peak-time of travel
for these pedestrians [68]. PNC recorded kerb-side (closer to the emission
source; 8.6 x 105 ppcc) was significantly greater than building-side recordings
(further from emission source; 7.3 x 105 ppcc) [68], highlighting the
importance of proximity to an emission source when considering exposure
concentrations. Similar circumstances have been described in previous studies
[49, 53].
25
2.3.2.2. Walking versus other modes
Pedestrians have been shown to experience greater exposure to UFP by a
factor of 1.4 compared to passengers within a motor vehicle cabin ventilated
with filtered-air, during trips of approximately thirty minutes duration [69].
When car trips are repeated to match walking trip durations for the same
distance, this factor increases to 4.4. However, car users can reduce their
exposure by using standard air-conditioning systems which will facilitate
individual micro-environments of reduced vehicle exhaust emissions [69]. If
in-vehicle air-conditioning systems are used, personal exposure of pedestrians
to UFP within urban areas can be approximately eight times higher than car
occupants of the same urban setting [70].
The magnitude and variance of individual exposure to fine PM and energy
expenditure associated with either motorised transport or walking may not
always be largely different between the two transport modes; however, higher
energy expenditures have been recorded for public light rail (2.03 Kcal/kg·hr,
including station transfers) and solely walking (2.6 Kcal/kg·hr) compared with
door-to-door driving of private automotives (1.51 Kcal/kg·hr) in an urban
environment [71]. As urban environments are increasing in congestion and
density, active transport such as cycling and walking are becoming more
popular to promote. Thus, it is apparent that a greater emphasis on in-transit
exposure assessment during walking and cycling, rather than driving, would
be warranted to properly assess public health implications of large-scale shifts
in inner-city transportation mode.
2.3.3. Measuring UFP In-transit
Due to limitations associated with size, sensitivity, and weight of monitoring equipment
needed for exposure assessments during active transport, some predictors of UFP exposure
and inhaled particle count or dose need to be employed to reduce interference with normal
commuting behaviour. Using exact time, GPS co-ordinates, visual media, traffic count and
26
meteorological data, estimations can be made of the presence of certain pollutants such as
fine PM and NO2 [63]. To account for the vast set of external influences unable to be
accounted or controlled for in a real-world study, investigators often utilise questionnaires to
accompany direct air pollution measurements, with the hope of explaining causes of inter-
individual variability associated with daily activities outside of exposure monitoring periods.
Previously, research participants have used symptom questionnaires to subjectively rate their
presentable respiratory symptoms (e.g. cough, wheeze) associated with commute exposure on
a scale of zero (no complaints) to three (severe complaints) [9]. In addition, each participant
may complete a baseline questionnaire, adapted from the European Community Respiratory
Health Survey, at the beginning of a study [9]. The use of a questionnaire can be a valid tool
when assessing the health effects of exposure to air pollution such as PM.
2.4. UFP and Health Effects
While the health effects of general air pollution have been noted for centuries, the biological
implications of exposure to UFP have only been recognised in the last two decades [41, 72].
Now, some of the adverse health effects of UFP exposure have been established in scientific
literature. In general, UFP have a greater toxicity than coarser particles [43] and a more
efficient ability to penetrate through the pulmonary system [46] so much as to elicit
detrimental effects on the cardiovascular system [49].
In 2003, a review was compiled for the Australian Government which summarised the health
impacts of UFP indicated by established research [73]. These findings included: 1) increases
in mortality associated with UFP; 2) delayed morbidity and mortality of UFP due to a role in
cardiovascular disease development; 3) decreased respiratory function and increased
respiratory symptoms, and medication use, in asthmatics exposed to elevated PNC; 4)
stronger acute effects on respiratory health from UFP rather than coarser particles; 5) more
severe UFP-related acute respiratory health effects in asthmatic adults rather than asthmatic
children; 6) cumulative effects shown to be stronger than same-day effects of lung
inflammatory events from UFP exposure; and, 7) the association of chronic heart disease
morbidity and UFP exposure.
Recent years have seen much progress in the fields of epidemiological, toxicological and
clinical exposure science, deepening the knowledge base of this expansive, multi-disciplinary
27
subject [59]. An expert elicitation on UFP potential mechanisms of health effects has been
produced to assess the evidence for causal relationships between UFP exposure, health
endpoints and cardiac events: fourteen European experts, including epidemiologists,
toxicologists and clinicians, rated the likelihood of independent causal relationships between
increased short-term UFP exposure and all-cause cardiopulmonary morbidity and mortality
[74]. Respiratory inflammation and subsequent thrombotic effects were rated the most likely
causal pathway for the aggravation of asthma symptoms (e.g. lung function decrements), with
the likelihood rated as moderate to high. The expert panel stressed the importance of
considering UFP in future health-impact assessments of transport exposure studies and health
effects due to exposure [74]. More recently, eleven European experts from the fields of
epidemiology, toxicology, and clinical medicine were brought together to address the lack of
a quantitative summary for the specific exposure-response function and health impact
assessment of UFP exposure related to mortality and hospital admissions [75]. It was found
that substantial differences and uncertainty in the estimated health effects of UFP exposure
existed, with the most important source of uncertainty due to a lack of studies observing long-
term exposure to UFP and effects on hospital admissions [75]. While all experimental
methods are important for increasing knowledge of effects, the scope of the current project is
to perform real-world studies in the aim of sustaining relevance to human health and
environmental interactions.
2.4.1. Clinical Studies
Clinical studies provide insight on cause-effect mechanisms by extensively characterising
individuals in controlled exposure conditions [76]. Through clinical studies, it has been
shown that UFP have a higher predicted deposition rate in the alveolar space (compared to
larger PM) due to their relatively smaller size [76-78]. Further, deposited UFP can evade
alveolar macrophage clearance from the lung and enter lung cells, interstitium and the
pulmonary vascular bed [79], then can travel from the lung to other organs via the
cardiovascular system [46]. Moreover, UFP may have a higher oxidant activity due to a
relatively greater surface area (and thus bioavailability) compared to larger particles of
similar collective mass [80]. However, limitations exist with clinical-based studies, such as
the typically-small statistical sample size due to design-associated constraints, and the
28
inability to account for changes in ‘real-world’ particle characteristics and size through
agglomeration [76].
2.4.2. Epidemiological Studies
Through epidemiological investigation, consistent associations have been shown between
cardiovascular morbidity or mortality and outdoor PM concentrations, particularly relating to
finer particles [81]. Relations between particle mass concentrations from fossil fuel
combustion and adverse cardiopulmonary outcomes have been demonstrated, with new
evidence showing that smaller particle diameter (with increased surface area) and particle
composition (e.g. adsorbed oxidant gases, organic compounds and transition metals) are
considered more important than particle mass [72, 82]. While the biological mechanisms are
still to be fully characterised, studies with repeated measures are supporting time-series
results. Associations between ischaemia and arrhythmias, hypertension, decreased heart-rate
variability, and increased biomarkers of inflammation and thrombosis have been shown with
exposure to UFP from fossil fuel combustion [59].
PM is being shown to cause both acute (i.e. days) and chronic (i.e. months to years)
cardiovascular health effects [83]. Daily concentration elevations of fine PM have been
strongly-associated with hospital admission rates for cardiopulmonary disease, especially
heart failure [28]. Children living close to (within 200 metres of) major traffic corridors (such
as highways) have shown an elevated risk of developing asthma with reduced lung function;
however, the exact mechanism and magnitude of risk is not yet known [84]. The link between
PM and health detriment is becoming clearer because of analytical characterisation in
epidemiological studies [42]. However, questions of exposure-response relations still remain
unclear, which will hinder effective policy implementation.
2.4.3. Field Studies
For effectively assessing individual exposure-responses to real-world circumstances,
individuals need to be monitored in their natural environment with real-world conditons, such
as is facilitated with field studies. Through field studies, young and elderly lung-function-
29
impaired patients have shown associations with decreased heart-rate variability indices,
indicating disturbances on autonomic function by sub-micrometre particles [48]. The acute or
direct health effects of diesel exhaust exposure in non-smoking adult asthmatics can include
lung function decrements, with physically-active exposure eliciting the most sustained
decrease in lung function and increase in respiratory symptoms of asthmatics, related to the
rise in PNC [49].
2.4.4. Toxicological Studies
The principle of toxicology is that a substance’s toxicity is related to its’ circumstantial dose.
Various toxicological studies have shown that UFP are more toxic than coarser particles [7,
43-45] and that individual particles are capable of inducing inflammation and oxidative stress
at adequate doses [45]. Therefore, it is suggested that PNC, which is dominated by UFP, may
be more indicative of potential health impacts (i.e. toxicity) than particle mass concentrations.
At a specified mass, UFP have profoundly greater surface areas than coarser particles, which
is believed to affect the relative toxicity towards the respiratory system, in combination with
a higher deposition efficiency in the alveolar region and potential to trans-locate into the
cardiovascular system [85, 86]. Quantification of nitro-polyaromatic-hydrocarbon’s adsorbed
to UFP in a roadside atmosphere has shown their important contribution to high direct-acting
mutagenicity of UFP (however only with bacterial, rather than human, cell line assays) [87].
Indications of physiological aggravation from UFP exposure have been given through
monitoring of biological markers in humans. Pro-thrombotic and pro-inflammatory
responses, as well as alterations of cardiac re-polarisation, have resulted from UFP exposure
[85]. Significant increases of fibrinogen and regulatory T-cells expressing inflammatory cell
mediators have been elicited with subway micro-environment exposure, corresponding to a
biological response with symptoms of irritation in the lower airways [88].
2.4.5. Further Afield
Research focus is shifting from respiratory to cardiac effects as incidents of elevated UFP
exposure are now known to increase cardiac morbidity and/or mortality [81]. The
30
associations between urban PM distribution and heart-rate variability have been assessed
during repeated (bi-weekly) sub-maximal exercise testing over 6 months with stable coronary
heart disease patients [89]. It was found that elevated UFP levels two days previous to a
clinic visit had significant association to increased risk of heart-rate variability depression
during sub-maximal exercise testing; the association was stronger in non-β-blocker-user
participants, suggesting the cardiovascular morbidity associated with PM is mediated by an
increased susceptibility to myocardial ischaemia [89]. It has been postulated that different
sized PM may act independently due to different sources or different mechanisms of effect,
and that there’s a plausible link between elevated ambient PM and increased risk of mortality
due to ischaemic heart disease [89].
2.5. Mechanisms of Cardiopulmonary Detriment
Knowledge on the biological mechanisms of UFP exposure-related cardiopulmonary insult
and response is still relatively limited. The causation and exacerbation of asthma due to UFP
exposure has gained the most attention, with the highest number of published studies to date
[59]. PM is known to induce biological inflammatory processes by recruitment of
inflammatory mediators such as cytokines, which are important components in the
homeostasis of respiratory structure [59]. Pre-disposing circumstances, such as infantile or
elderly status, or an underlying health condition, such as asthma or chronic obstructive
pulmonary disease (COPD), will heighten the influence PM has on an individual from
exposure during transit [59].
2.5.1. Measure of Dose: Surface Area vs. Number Concentration
As UFP constitutes the major particle count of freshly-emitted combustion exhaust, it is
considered more important than larger-sized particles when assessing public health
implications of in-transit air pollution exposure [59]. A consequential characteristic for higher
particle counts of smaller particles is a relatively-greater surface area, which provides a
higher potential for toxic agents to adsorb to deeper-depositing particles [90]. Notably, it is
believed by some that the toxic agents attached to UFP, rather than the particles themselves,
are the cause of cardiopulmonary insult [90].
31
UFP in concentrations as high as 100,000 ppcc may induce alveolar inflammation and
inflammatory mediator release, consequently exacerbating lung disease and increasing blood
coagulability in susceptible individuals [72]. Cardiovascular mortality of individuals with
indoor-based occupations has been observed to be lower than out-door workers (because of
relatively-higher PNC exposure) during elevated air pollution episodes [72]. Smaller particles
in ambient aerosols, for the same particle mass concentration, are capable of causing more
damage to pulmonary cells due to higher interaction of reactive oxygen species content [91].
The individual response of an elderly population exposed to unfiltered freeway PM has been
shown as significantly-related to particle count or PNC rather than particle mass [92].
However, due to insufficient studies focusing on PNC, further investigation is warranted to
elucidate the mechanisms of effect of PNC exposure, as well as the chemical composition of
motorised transport emissions, for in-transit micro-environments.
2.5.2. Effects of Chemical Composition
It has been stated that an explanation of UFP toxicity should not only rely on particle
count but also the surface properties of the particle [44]. Investigation of the interaction
between particle surface molecular characteristics and epithelial, mucosal and nervous tissue
in exposed individuals can inform air quality and pollution emission guidelines.
Epidemiological associations between health detriment and low exposure to NO2 have a
confounding effect linked to UFP count and surface area [72]. It has been proposed that the
biological response to high UFP count exposure is from the lung’s evolutionary defence
system against microorganisms (which were inhaled in large numbers, not high mass);
macrophages and endothelium release mediators (including fibrinogen) which systemically
reduce blood coagulability and destabilise atheromatous plaques causing acute cardiac effects
and increasing cardiac risk in susceptible individuals [90]. Further, lipid peroxidation due to
the oxidative properties of iron-rich particles has been proposed as a mechanism for increased
particle uptake into the cardiovascular system [93, 94]. More recently, it has been suggested
that the elemental oxidation state determines toxicological importance of PM iron
components [95]. Higher toxicity may present from higher content of iron and manganese in
PM, typical of emissions from metro train wheel and brake system abrasion [96]. A murine
model has shown that non-cytotoxic concentrations of PM can induce a transient time- and
32
dose-dependent increase in TNF-a and MIP-2 (associated with oxidative stress) thought
plausible due to the high iron content of the particles [97].
PM sampling of varying distances (i.e. 20 and 275 m) from a high-intensity traffic corridor
has shown that higher proximity samples comprise of greater amounts of metals,
significantly-associated with increases in inflammatory proteins of BALF and reduced
pulmonary function [98]. Interestingly, extra-pulmonary responses but not cardiac effects or
detectable lung inflammation and injury have been observed after UFP exposure, in
agreement with earlier human exposure studies [85, 99]. Metallic nano-particles attached to
larger host particles (such as organic matter or sulphate) can behave differently because of
deposition in shallower respiratory zones, remaining once host particles have dissolved [100].
The components of modern urban air mixtures have been shown as complex through the use
of transmission electron microscopy (TEM) with non-automotive-emitted nanoparticles of
metal composition (including iron, lead and/or zinc) common in the Mexico City area and
other large cities [100].
2.5.3. Acute versus Chronic Exposure
While short episodes of exposure may only produce acute health detriments, regular
occurrence of these episodes (for example, twice a day for five days a week as work-related
commuting) could lead to a heightened exposure sensitivity and an increased risk of chronic
disease development. Recent epidemiological evidence of long-term exposures to fine PM
and mortality effects suggest that adverse health effects are dependent on both exposure
factors of concentration and duration, with long-term exposures causing greater, more
persistent cumulative effects compared to short-term exposures [29]. Evidence suggests that
air quality standards controlling for 1-hour averages, in addition to 24-hour averages, of
ambient PM would be more effective at reducing associated health risks; symptom severity in
asthmatics during elevated PM are indicated to be more positively-associated with 1-hour
averages compared to 8- or 24-hour averages [101].
33
2.5.4. Pre-disposing Physiological Susceptibilities
2.5.4.1. Older Age
The elderly are more susceptible to the effects of in-transit UFP exposure due to
natural decrements in cardiopulmonary function [59]. An elderly yet healthy
population exposed to both particle-filtered and non-filtered air (particle-rich, of
approximately three times greater PNC) over a 48-hour period exemplified a
biological inflammatory response (via various secondary health endpoints, such as
blood-based proteins that are indicators of endothelial function and inflammation)
associated with PM exposure and an increased risk of cardiovascular disease [102].
Decreases (of ~20%) in atrial ectopic beat (arrhythmia) incidence both during and
after exposure to Los Angeles freeway unfiltered air, in addition to decreases (of
~30%) in N-terminal pro B-type natriuretic peptide (indicative of cardiopulmonary
stress) have been seen [92]. Consequently, the implementation of PM filters in vehicle
cabins have been called upon to prevent sustained arrhythmias triggered by premature
atrial beats from automotive exhaust exposure [92].
2.5.4.2. Asthma
Exacerbations of asthma symptoms can be caused by oxidative stress and
inflammation (possibly from the allergic immune response characteristic of asthma) in
the lungs of susceptible individuals [19]. Asthmatic-reaction promotion (i.e. allergen
sensitivity several hours later, measured as the increase in specific airway resistance)
has not been seen in mild allergic asthmatics through subjective symptoms (i.e.
reduced lung function) during moderate tunnel exposure (PM2.5 > 100 µg·m-3) [103].
However, short-term effects of exposure to diesel exhaust with significantly-higher
PNC in mild or moderate asthmatics can lead to asymptomatic yet consistent
reductions in lung function, and increases in inflammatory neutrophilic biomarkers as
well as airway acidification [49].
34
2.5.4.3. Cardiovascular Disease
Acute cardiovascular effects in Type-2 Diabetics exposed to in-vehicle traffic-related
pollution during highway trips include increases in low frequency heart-rate
variability post-commute, and decreases in high frequency heart-rate variability the
next day, associated with inter-quartile range increases in PNC [104]. Associations
between daily PNC fluctuations, heart rate variability (representing cardiac autonomic
control), and sympathovagal balance (representing the low-to-high frequency ratio) in
elderly stable coronary artery disease patients have been shown during paced
breathing [105]. UFP and PM2.5 cardiovascular effects are independent of each other
and may be modified by individual and source exposure characteristics. Measuring
HRV using ambulatory ECG recordings during exposure events is difficult but valid
for large epidemiological studies [104].
2.6. Knowledge Gaps
Along with epidemiological and toxicological investigation, it is apparent that further work is
required to characterise personal exposure of bicycle commuters to real-world motorised
traffic emissions of UFP, best done through field studies. The potential inhaled count or dose
of UFP resulting from changes in physical effort and therefore pulmonary ventilation while
bicycle commuting has not been investigated extensively; however, some studies have
compared multiple modes and predicted ventilation rates. This thesis intended to address such
a knowledge gap by accurately measuring in-commute, real-time ventilation rate while using
popular bicycle commute routes of altered motorised traffic exposure and different directions
in an urban setting.
The exposure and consequential response of individuals to automotive emissions, particularly
UFP, is a novel field of research (due to technological advancements of measurement
devices) with many frontiers left to explore. Therefore, the strategies which could manage
such an exposure response are yet to be addressed. The use of respirators for filtering inhaled
air pollutants is not yet known to be effective for particulate matter in the ultrafine range, and
so their efficacy in this regard warrants attention. Several studies have compared bicycle
35
commuters using routes of high and low exposure to motorised traffic. Such routes were
defined by the study investigators to facilitate a coherent research question and protocol;
however, as this definition was not reflecting the variance of real-world commute routes, the
benefits of utilising these (or any) alternative routes are not applicable to the wider
population. To address such a knowledge gap, this thesis intended to apply similar protocols
to personally-identified and altered routes of high and low motorised traffic exposure, in real-
world circumstances (for example, of typical commute departure time and speed or physical
effort and therefore pulmonary ventilation rate).
Ultimately, if certain circumstances are deemed to require strategies to be implemented to
mitigate the exposure effects of air pollution on health, then such strategies should be guided
by the needs and desires of the strategy beneficiary. For example, if a certain cohort of
patients are particularly susceptible to certain pollutants, however would benefit from the
frequent, moderate physical activity associated with bicycle commuting that would be
otherwise difficult to obtain, then the use of a respirator or altered commute route may
provide better odds for the cost-benefit ratio of such activity. To an extent, this thesis
intended to provide insight on the desires of bicycle commutes when considering such risk
management strategies as respirator use and more particularly commute re-routing.
Accordingly, the proposed hypotheses of this thesis attempted to address the knowledge gaps
identified. The first and third project intended to provide insight on the perception of in-
commute air pollution exposure and any related incidence of respiratory symptoms, along
with preferences towards air pollution exposure risk management strategies. The second and
third project then added to this knowledge base by investigating how effective and practical
the risk management strategy of commute re-routing was for reducing in-commute exposure
to newly-emitted motor vehicle exhaust (directly represented by PNC) in real-world
conditions of commute characteristics and emission circumstances.
36
3. GENERAL METHODS
3.1. Project Design (Project 1, 2 & 3)
Project 1 was intended to help guide the design of Project 2 and 3. Accordingly, Project 1
was itself designed to characterise a large group (100+) of frequent bicycle commuters
travelling into and/or through a one kilometre radius of the Brisbane CBD, and determine
typical commute characteristics and attitudes to the risk management strategy of commute re-
routing to be evaluated in Project 2 and 3. The questionnaire used in Project 1 was purpose-
designed with review and input from fellow researchers and a sub-set of intended
participants.
With typical commute characteristics detailed, and a receptive attitude to commute re-routing
recognised, Project 2 was designed to produce an air pollution exposure profile of popular
bicycle commute routes from Brisbane suburbs to the CBD based on a single participant
model. Pre-determined popular bicycle commute routes traversing Brisbane from the North,
East, South and Western suburbs to the CBD were to be repeatedly monitored, with
alterations of both high and low proximity to motorised traffic at morning peak traffic time,
to quantify real-time and total mean PNC, heart and ventilation rates and to determine and
compare PNC exposure and inhaled particle count.
Project 3 was similar to Project 2, although designed to include an appropriately-sized
participant group model (to satisfy statistical power needs). To replicate the aims of Project 2
under real-world circumstances while monitoring for a physiological inflammatory response
related to in-commute PNC exposure levels, adults were to be recruited to perform their
regular commute route (determined as high proximity to motorised traffic) and an alternative
route of low proximity to motorised traffic. One return trip of low, and one of high, proximity
to motorised traffic was to be performed while carrying geo-location, heart rate and ultrafine
particle monitoring instruments. Further, symptom-experience reporting, peak flow metering
and spontaneous sputum sampling was to be performed before and after commutes to relate
in-commute exposure with post-commute physiological inflammatory response. In addition,
route alteration preference (and features) was to be queried (between regular / ‘high’ and
37
‘low’ route) to evaluate whether a conscious decision to lower proximity to motorised traffic
(and therefore emissions) is feasible as an exposure risk management strategy.
3.2. Participant Recruitment (Project 1 & 3)
Initially, contacts were established with major Brisbane bicycle user groups (BUGs) and
dedicated bicycle commuter end-of-trip facilities in Brisbane city. Questionnaires were
distributed both physically (by providing printed copies to BUG members via BUG
administration, and attending active transport promotional events) and electronically (via
BUG email lists and electronic newsletters). Further, a media campaign was engaged to
reach a broader, non-BUG audience (incorporating a radio interview and several newspaper
articles) reaching the greater south-east Queensland (SE QLD) region. Participants expressed
informed-consent by returning the completed questionnaire. Further, participants were able to
express interest and provide consent to be later contacted regarding subsequent research, such
as the third study in this thesis. While the majority of previous participants were available for
the third study, only 33 were deemed appropriate (according to their propensity for bicycle
commuting proximal to motorised traffic). In addition, 120 new potential participants were
gained through a second recruitment drive. Project 2 was a single participant model - that is,
the principal investigator performed all tasks required and therefore participant recruitment
was not required.
3.3. Questionnaire (Project 1 & 3)
The questionnaire used in this investigation was novel and constructed with the help of
review by fellow researchers and a sub-set of intended participants. The complete
questionnaire used for Project 1 is included in the appendix (A) of this thesis. A total of 77
questions were delivered and based on a five-grade Likert scale (sub-totalling 43 questions
with the scale of: 1 = “very low”; 2 = “low”; 3 = “moderate”; 4 = “high”; 5 = “very high”), or
were of categorical (sub-totalling 5 questions), continuous (sub-totalling 12 questions) or
nominal (sub-totalling 17 questions) format. A shortened version of the questionnaire used
for Project 1 was supplied to participants of Project 3. This amended questionnaire included
the same inquiry of pre-, in- and post-commute perceptions and symptoms, as well as the risk
38
management strategy features of importance for commute re-routing only. The amended
questionnaire used for Project 3 is also included in the appendix (B) of this thesis.
The qualitative Likert scale response data, concerning the incidence of in-commute air
pollution exposure perception and symptoms, as well as the importance of factors concerning
risk management strategy use, were converted to a group fractional mean between 1 and 5
(as: 1 = “very low”; 2 = “low”; 3 = “moderate”; 4 = “high”; 5 = “very high”). Determined
attempts were not made to control for pre-disposing susceptibilities (e.g. health history, age
and gender), however these factors were analysed and compared for increased incidence and
severity of symptoms experienced both in- or post-commute and pre-commute (using the
Likert scale ranking).
3.3.1. Perception Incidence and Magnitude
Perception of air pollution exposure, defined as the conscious detection of poor air
quality, was reported by the participant for incidence and magnitude (as “very low” to
“very high”) for their typical bicycle commute. A participant was deemed to properly
perceive air pollution when they reported the incidence and subsequent magnitude of
perception as “moderate” or higher. Although the detection of offensive odour could
be better defined as a sensory experience and the perception of poor air quality, it is
considered a mild acute respiratory symptom by the American Thoracic Society
[106].
3.3.2. Symptom Incidence and Severity
The format of inquiry for acute respiratory symptoms attributable to air pollution
exposure (including the detection of offensive odour) was based on recommendations
by the American Thoracic Society [106] and previous research [107]. The incidence
of acute respiratory symptoms of varying severity (including offensive odour
detection, nasopharyngeal or ‘eye, nose and throat’ irritation, tussis or ‘coughing’,
chest tightness and wheezing) were inquired of specifically for bicycle commuting
(one hour pre-commute, in-commute, and one post-commute): frequency of incidence
was ranked by participants from “very low” to “very high”. A participant was deemed
39
to experience an acute respiratory symptom when they ranked its incidence as of
“moderate” or greater frequency.
3.3.3. Risk Management Strategies
Air pollution risk management strategies, including commute re-routing and respirator
use, had specific factors of importance for use ranked (from “very low” to “very
high”) by participants to guide strategy implementation, if found appropriate and
effective for susceptible individuals by future research. Brisbane bikeway maps [108]
were appended to the end of the complete questionnaire (used for Project 1) for
participant reference of bicycle commute routing (see Appendix A).
3.4. Ultrafine Particle Monitoring (Project 2 & 3)
To monitor exposure concentrations and inhaled particle count, a portable UFP recorder
(Aerasense Nanotracer, Phillips, The Netherlands) was used for the second and third project.
Logging the mean PNC and PD every 16 seconds (as the maximum frequency), an air quality
index (of PNC and PD) was provided approximately every 5 metres (based on an average
cycling speed of 20 km/hr, as indicated from Project 1).
In-commute PNC and PD means, as well as minimum and maximum values, were calculated
for comparison between identified commute routes of high and low proximity to motorised
traffic. When PNC readings were below 100 ppcc, and when subsequent readings changed by
more than a factor of 10, they were carefully considered and removed if deemed unrealistic
[12, 109] – this happened only on rare occasion. Participants of the third project were
instructed to maintain air quality (along with heart rate) monitoring for three hours post-
commute, if practical, to observe any possible events affecting the three-hour-post-commute
time-point testing – again, this only happened on rare occasion.
40
3.5. Ambient Pollution Data Monitoring (Project 1, 2 & 3)
Ambient hourly mean PNC of Brisbane CBD were recorded by a WCPC 3781 (TSI Inc.,
USA) and logged in a meta-database at the air monitoring research station of the Queensland
University of Technology (QUT). The station is located on the sixth floor of a building in
QUT’s south-eastern CBD campus and is of equivalent height to the Southeast Freeway
approximately 100 m south-west of the station. The freeway, which experiences mild
congestion during peak times, consists of four lanes each inbound and outbound. In-commute
trip means (as an indication of localised concentrations) were referenced against ambient
hourly means (as an indication of background concentrations) to evaluate commute-
attributable UFP exposure.
3.6. Heart Rate and Estimated Ventilation Rate (Project 2 & 3)
To help estimate in-commute ventilation rate (VE) and inhaled particle count, heart rate (FH)
was recorded in real-time during bicycle commuting. Two different instruments were used
for Project 2 (Equivital) and Project 3 (Polar) (with details to follow in relevant sections).
Along with HR, in-commute breathing rate (FB) was recorded in the second project (although
disregarded with the estimation of VE). Participants of the third project were instructed to
maintain FH (along with air quality) monitoring for three hours post-commute, if practical, to
observe any possible events affecting the three-hour-post-commute condition inflammatory
response testing.
FH data collected and logged during participant commute is applied to a standardised heart
rate-ventilation curve, adjusted for age and gender. Data from previous laboratory studies
indicate VE increases faster than FH when a participant performs upper body exercise
compared with lower body exercise [110]. Using this method, as most cycling activities do
not involve upper body exertion, it has been concluded that FH interpretation is a feasible
approach to estimate VE in field studies [85].
41
3.7. Geographical Positioning (Project 2 & 3)
To localise air quality readings, a global positioning system device (BT-Q1000X, Qstarz,
Taiwan) was used for the second and third project. Logging the longitude, latitude and
altitude of a participants every 4 seconds, 20 metre lengths of a commute route could be
allocated to a single mean PNC and PD. ArcGIS Desktop (Esri, USA) was used to
graphically represent mean PNC and PD recorded along designated commute routes.
3.8. Meteorological Data Monitoring (Project 2 & 3)
The Australian Bureau of Meteorology Climate Database [111] was accessed for hourly
regional measures of temperature, humidity, wind direction and speed, air pressure, and
precipitation. Wind direction was considered as either ‘downwind’ or ‘upwind’ of adjacent
motorised traffic, along with low or high wind speeds. Meteorological variable data was
collated and analysed to help explain any particle measurement anomalies.
42
4. PROJECT ONE: A questionnaire-based investigation of perceptions,
symptoms and risk management strategies for air pollution exposure
and motorised traffic proximity of adult bicycle commuters
A questionnaire-based investigation of perceptions, symptoms and risk management
strategies for air pollution exposure and motorised traffic proximity of adult bicycle
commuters
Tom Cole-Huntera,b, Lidia Morawskab, Colin Solomonc,d*
PLoS ONE, In Review.
a Institute of Health and Biomedical Innovation, Queensland University of Technology, 60
Musk Avenue, QLD 4059, Australia.
b International Laboratory for Air Quality and Health, Queensland University of Technology,
2 George Street, QLD 4001, Australia.
c School of Life Sciences, Queensland University of Technology, 2 George Street, QLD 4001,
Australia.
d School of Health and Sport Sciences, University of the Sunshine Coast, Sippy Downs Drive,
QLD 4556, Australia.
*Corresponding Author: Dr Colin Solomon, School of Health and Sport Sciences, University
of the Sunshine Coast, Sippy Downs Drive, Sippy Downs, QLD 4556, Australia.
Telephone: +61754301128. E-mail: [email protected]
43
44
ABSTRACT
Bicycle commuting is encouraged in many cities around the world to improve public
health, air quality, and traffic congestion. Information concerning the effects of air
pollution, such as exposure perception, symptoms and risk management is necessary to
responsibly advocate and sustain bicycle commuting participation. To determine reported
air pollution exposure perceptions, symptoms and amenability for specific risk
management strategies, and relate these to estimated levels of motorised traffic proximity,
a questionnaire-based investigation was conducted with regular bicycle commuters in
Brisbane, Australia. Participants (n = 153; age = 41 ± 11 yr; 28% female) reported the
characteristics of their typical bicycle commute, along with exposure perception and acute
respiratory symptoms, and amenability for using a respirator or re-routing their commute
as risk management strategies. Healthy individuals reported a higher incidence of specific
acute respiratory symptoms in- and post- (compared to pre-) commute (p < 0.05). The
incidence of specific acute respiratory symptoms was significantly higher for participants
with respiratory disorder history compared to healthy participants (p < 0.05). The
incidence of in-commute offensive odour detection, and the perception of in-commute air
pollution exposure, was significantly lower for participants with smoking history
compared to healthy participants (p < 0.05). Females reported significantly higher
incidence of in-commute air pollution exposure perception and other specific acute
respiratory symptoms, and were more amenable to commute re-routing, compared to
males (p < 0.05). Healthy individuals have indicated a higher incidence of acute
respiratory symptoms in- and post- (compared to pre-) bicycle commuting, with female
gender and respiratory disorder history indicating a comparably-higher susceptibility. The
perception of air pollution exposure levels and the amenability to adopt exposure risk
45
management strategies where applicable will aid the general population in shifting from
passive, motorised transport modes to bicycle commuting.
Key words: air pollution; bicycle commuting; perception; symptom; risk management.
1. Introduction
Bicycle commuting requires physical effort and is a fossil fuel-independent transport mode,
therefore it is being increasingly promoted as a solution to help alleviate physical-inactivity-
related cardiovascular disease and anthropogenic climate change, as well as intra-urban
motorised traffic congestion [3-5, 25]. However, barriers (either physical or psychological)
may exist for individuals choosing to shift from passive, motorised transport to bicycle
commuting such as the greater perceived or actual risk of exposure to air pollution [6, 19]. It
is known that elevated air pollution exposure is a health risk which can be increased with
heightened pulmonary ventilation [10-13] and proximity to motorised traffic emissions [11,
14-19]; however, the overall benefits associated with improved air quality and public health
from a major uptake in bicycle commuting have been shown to negate such risks [6, 25, 26].
Regionally-monitored air pollutants of potential interest for bicycle commuting in urban
environments include particulate matter (PM) and nitrogen dioxide (NO2), which are emitted
with motorised traffic exhaust, can be consciously detected (i.e. perceived) and elicit acute
respiratory symptoms upon exposure [6]. Generally, nasopharyngeal irritation, airway
inflammation and bronchoconstriction (manifesting as cough and phlegm production, and
chest tightness or wheezing) are acute (i.e. rapid-onset and short-lived) symptoms which may
arise in an individual exposed to such pollutants at elevated concentrations, especially if a
pre-disposing respiratory disorder such as asthma exists [27]. Further, chronic exposure to
elevated PM and NO2 can suppress airway immune defences and consequently increase the
46
incidence and severity of (sometimes debilitating) upper respiratory tract infections [30, 31].
The reporting of such symptoms has been used previously to assess air pollution exposure in
healthy and asthmatic children [32-34], as well as healthy and asthmatic adults [35, 36], the
elderly [37], and the general community [38, 39]. Further, symptom questionnaires have been
used successfully with adults to investigate multi-modal air pollution exposure perceptions
for work-related commuting [40].
Investigation of perceived and actual exposure risk is warranted for an informed transition
from passive, motorised transport modes to bicycle commuting, especially for individuals
with a physiological pre-disposition to exposure effects. Further, if risk management
strategies (such as commute re-routing or respirator use) are deemed appropriate and
effective, the successful adoption of these strategies will rely on the features meeting the
needs and desires of the potential user. The aims of this project were to investigate: air
pollution exposure perceptions, symptoms and amenability for specific risk management
strategies of frequent adult bicycle commuters, and to relate these with participant history of
respiratory disorder or smoking and also estimated proximity to motorised traffic; if this
perception is influenced by factors of typical bicycle commute or personal characteristics,
and; if perception of exposure can facilitate self-managed exposure risk strategies. This study
did not intend to compare estimated and perceived exposures to actual exposures, but to
inform subsequent studies for which personal exposure measurements are taken.
Accordingly, it was hypothesised that in adult bicycle commuters: 1) the incidence and
severity of acute respiratory symptoms will be greater in-commute compared to pre- and
post-commute, and with individuals of respiratory disorder history or female gender, and will
be positively-associated with estimated proximity to motorised traffic; 2) the perceived
exposure levels of air pollution by adult bicycle commuters will be consistent with estimated
motorised traffic proximity levels; and, 3) the amenability to adopt air pollution exposure risk
47
management strategies will be positively-associated with perceived exposure levels and
incidence of, or physiological pre-disposition to, acute respiratory symptoms.
2. Methods
2.1. Project Design
A questionnaire-based investigation was performed with the aim to: 1) evaluate the incidence
and severity of acute respiratory symptoms associated with estimated proximity to motorised
traffic, and determine if pre-disposing factors of age, gender and respiratory disorder history
are associated with the incidence and severity of these acute respiratory symptoms; 2) assess
if participants’ reported in-commute perception of air pollution exposure is consistent with
estimated proximity to motorised traffic (derived from commute duration, frequency and
proximity to motorised traffic); and, 3) evaluate if the amenability of participants to adopt
commute re-routing or respirator use as air pollution exposure risk management strategies
(along with the importance of specific strategy features) depends on individual
characteristics, exposure perception levels, incidence of (and physiological-susceptibility to)
acute respiratory symptoms.
2.2. Questionnaire Design
The questionnaire used in this investigation was purpose-designed with review and input
from researchers and a sub-set of intended participants. Further, the format of assessment for
acute respiratory signs and symptoms attributable to air pollution exposure was based on
recommendations by the American Thoracic Society [106] and previous research [107]. The
complete questionnaire is available online from the journal as an additional electronic file.
The questions (total of 77) used a qualitative five-grade scale of equal variance (43 questions:
48
1 = “very low”; 2 = “low”; 3 = “moderate”; 4 = “high”; 5 = “very high”), or were of
categorical (5 questions), continuous (12 questions) or nominal (17 questions) format.
The incidence of acute respiratory signs and symptoms (including offensive odour detection,
eye, nose and throat or ‘nasopharyngeal’ irritation, tussis or coughing, chest tightness and
wheezing) of varying severity were assessed specifically for bicycle commuting at conditions
of one hour pre-commute, in-commute, and one hour post-commute: the frequency of
incidence was ranked by participants from “very low” / ‘1’ to “very high” / ‘5’. Perception of
in-commute air pollution exposure (defined as the reporting and therefore conscious detection
of moderately-poor air quality), was reported by participants for incidence (“yes”/ “no”) and
level (“very low” / ‘1’ to “very high” / ‘5’) of their typical bicycle commute. To address
individual response subjectivity, a participant was considered to perceive air pollution
exposure only when they reported the level of perception as “moderate” / ‘3’ or higher. As
well as an indicator of perception, the detection of offensive odour is considered an acute
respiratory sign [106] and could aid the self-management of air pollution exposure incidence
and level. An inquiry of upper respiratory tract infection (URTI; either mild or debilitating
and interfering with normal daily duties) along with commute history was made to elucidate a
possible effect of chronic exposure to motorised traffic emissions [30, 31]. To address
individual response subjectivity, a participant was considered to experience an acute
respiratory symptom only when they reported its’ incidence as of “moderate” / ‘3’ or greater
frequency.
Air pollution risk management strategies, including commute re-routing and respirator use,
had the importance of specific strategy features rated (from “very low” / ‘1’ to “very high” /
‘5’) by participants to guide future implementation of such strategies, if found to be
appropriate and effective by future research. The intention of this inquiry is to indicate which
strategy might warrant the most attention for future research, and which properties of each
49
strategy might require the most attention for developers. Brisbane bikeway maps [108] were
appended to the end of the questionnaire for participant reference of bicycle commute
routing.
2.3. Participant Recruitment and Sample
Potential participants were initially contacted through the university and major Brisbane
bicycle user groups (BUGs), or through newspaper and radio segments (noting the research
and recruitment contact details) to reach a larger group of potential participants.
Questionnaires were distributed as a paper copy (with a reply paid envelope; via BUG
facilities administration, mail-out, and attending an active transport promotional event) or an
electronic copy (via return E-mail). Eligible participants were adults and regular bicycle
commuters of the Brisbane inner-city region (defined as completing two or more return trips
in a five week-day period to a destination within a one kilometre radius of Brisbane’s Central
Business District). Potential participants with a current smoking status, or smoking cessation
of less than twenty-four months prior to the study, were not eligible for participation. Eligible
participants were supplied with the questionnaire and instructed to provide responses which
most accurately and completely recollected the cumulative experience of their own typical
bicycle commuting. The project was conducted with a protocol approved by the Queensland
University of Technology. Participants indicated informed consent by returning the
completed questionnaire. See Table 1 for recruited participant characteristics. The mean
frequency of return bicycle commutes per week and the mean historical duration of indicated
bicycle commute route suggests the participants of this study were regular bicycle commuters
with a variety of commuting experience (Table 1). Participants were divided into groups
according to history of respiratory disorder or smoking, as well as gender, for comparative
analysis of responses (Table 2, 3, 4 and 5).
50
INSERT TABLE 1 HERE
2.4. Data Calculation and Statistical Analyses
The qualitative five-grade scale data, concerning frequency of in-commute air pollution
exposure perception and symptoms, as well as importance of the features evaluated for the
use of risk minimisation strategies, were converted from an interval scale of measurement (of
‘very low’ to ‘very high’) to an ordinal scale (of 1.0 to 5.0) as: 1.0 = “very low”; 2.0 = “low”;
3.0 = “moderate”; 4.0 = “high”; 5.0 = “very high” to allow group mean values for collation
and statistical analysis. Attempts were not made to control for pre-dispositions of respiratory
disorder history or smoking history, or gender, but participants with these characteristics were
grouped, analysed and compared against each other (eg. female versus male gender) for
incidence (using ordinal scale data) of symptoms assessed between either in-commute or one
hour post-commute and one hour pre-commute conditions.
To provide an objective estimation of in-commute proximity to motorised traffic (PROX), a
ranking was calculated for individual participants as the product of their bicycle commute
duration, frequency (per week) and use of motorised traffic corridors (being the proportion of
route ‘on-road’, or route shared with motorised traffic) for their typical bicycle commute. The
proportion of on-road paths taken for a typical bicycle commute was converted to a fraction
from 0.1 to 0.9, with 0-10% on-road use ranked as the minimum (‘0.1’), ~50% on-road use
ranked as the median (‘0.5’), and 90-100% on-road use ranked as the maximum (‘0.9’),
etcetera. The on-road proportional fraction was then multiplied by the duration (as minutes)
and the frequency (as number of return trips per week) for a total time of motorised traffic
51
proximity whilst bicycle commuting per week. The product of this process allocated
participants a rank of either 1 (“very low”; n = 30), 2 (“low”; n = 32), 3 (“moderate”; n = 30),
4 (“high”; n = 32) or 5 (“very high”; n = 32) to represent the level of estimated proximity to
motorised traffic (as a proportion of total commute adjacent to motorised traffic corridors)
and associated exposure to air pollution emissions. The interval value (of approximately 25)
for each PROX rank allowed near-equal numbers of participants per rank for better control of
covariates when performing statistical analyses.
The questionnaire responses were analysed using predictive analytics software (PASW
Statistics Data Editor, V18.0; IBM Corporation, USA). One-way analysis of variance
(ANOVA) were performed to identify differences between group mean responses (of
perception and symptom incidence) at pre-, in- and post-commute conditions, along with
group mean differences between participant characteristic groups (of respiratory disorder or
smoking history and gender), and commute behaviour (including PROX). Subsequently,
Tukey HSD Post Hoc comparisons were performed with these ANOVA to identify specific
pair-wise differences. Further, Fisher’s Exact Test and Pearson’s Chi Square Test were
performed to signify the effect of participant and commute characteristics (of PROX,
respiratory disorder or smoking history, gender and historical duration of commute) on
participant responses, and the associated incidence of one participant’s response to another
response (within an individual). Statistical significance was indicated at the 95% confidence
interval (i.e. p < 0.05), which was not adjusted for repeated measures.
52
3. Results
3.1. Participant recruitment and response rate
The estimated target population size, according to BUG membership of targeted
organisations during recruitment, was 500. Approximately 200 potentially eligible
individuals expressed interest to participate in this study. Of these, 160 were confirmed as
eligible and therefore supplied with their preferred choice of a physical or electronic copy of
the questionnaire: 60 of 61 (98%) electronic copies and 93 of 99 (94%) printed copies
supplied to eligible participants were returned within a three month period from March to
June of 2010.
3.2. Symptom incidence and exposure perception reporting
Healthy participants reported significantly higher incidence of specific acute respiratory
symptoms in- and post-commute compared to pre-commute (p < 0.05; Table 2). The
incidence of in-commute offensive odour detection was positively-associated with incidence
of in-commute nasopharyngeal irritation [F(29,99) = 11.22, p < 0.001], tussis [F(18,151) =
4.50, p = 0.002], chest tightness [F(10,87) = 4.39, p = 0.002] and wheezing [F(6,79) = 2.82, p
= 0.027]; and, post-commute nasopharyngeal irritation [F(13,178) = 2.84, p = 0.026], tussis
[F(8,107) = 2.97, p = 0.022] and chest tightness [F(3,51) = 2.50, p = 0.045] of healthy
individuals.
The majority of participants (80%) reported in-commute perception of exposure to moderate
or higher levels of air pollution (Table 3), which was positively-associated with the incidence
of in-commute offensive odour detection [F(43,136) = 48.25, p < 0.001], nasopharyngeal
irritation [F(8,120) = 10.63, p = 0.001] and chest wheeze [F(2,82) = 4.59, p = 0.034].
Additionally, in-commute air pollution exposure perception was positively-associated with
53
the number of weekly return trips performed [F(9,256) = 5.50, p = 0.020], and general
concern for Brisbane’s ambient air pollution levels [F(3,28) = 18.59, p = 0.001].
3.3. Factors of physiological susceptibility
3.3.1. History of respiratory disorder or smoking
The incidence of acute respiratory symptoms (nasopharyngeal irritation, tussis, chest
tightness and wheezing) was significantly higher for participants with respiratory disorder
history compared to healthy participants (p < 0.05; Table 4). The incidence of upper
respiratory tract infection (URTI) was significantly higher for participants with respiratory
disorder history compared to healthy participants, both mild URTI (p < 0.001) and
debilitating URTI (p = 0.002). The incidence of in-commute offensive odour detection, and
the perception of in-commute air pollution exposure level (of moderate or above), was
significantly lower for participants with smoking history compared to healthy participants (p
< 0.05; Table 4 and 5).
3.3.2. Gender
Females reported significantly higher incidence of in-commute nasopharyngeal irritation (p =
0.009) and chest wheeze (p = 0.046), and post-commute nasopharyngeal irritation (p =
0.006), as well as in-commute air pollution exposure perception (p = 0.039), compared to
males (Table 2 and 3).
INSERT TABLE 2 HERE
54
INSERT TABLE 3 HERE
INSERT TABLE 4 HERE
INSERT TABLE 5 HERE
3.4. Historical duration of bicycle commute
There was a positive-association between historical duration of bicycle commute and PROX
[F(17976,144390) = 2.26, p = 0.027]. The incidence of debilitating URTI was near-
significantly associated with longer historical duration of bicycle commute [F(22,53) = 1.51,
p = 0.058], and debilitating URTI incidence was positively-associated with in-commute
nasopharyngeal irritation [F(7,120) = 2.98, p = 0.033] and tussis [F(9,158) = 2.84, p = 0.040],
and post-commute tussis [F(7,108) = 3.03, p = 0.031]. Amenability to utilise a higher
proportion of off-road paths as an air pollution exposure risk-management strategy, if found
to be appropriate and effective, was negatively-associated with historical duration of bicycle
commute [F(11,23) = 1.69, p = 0.021].
3.5. Estimated in-commute proximity to motorised traffic
The total group mean proportion of estimated in-commute proximity to motorised traffic
(PROX; the product of commute duration, frequency and proportional fraction of commute
55
using motorised traffic corridors) was 51.8 ± 2.8 % (Table 3). There was no significant
difference between groups according to gender, smoking history and respiratory disorder
history (Table 3 and 5). PROX was positively-associated with in-commute air pollution
exposure perception [F(3,22) = 2.31, p = 0.023] and incidence of in-commute offensive odour
detection [F(18,160) = 2.08, p = 0.041]. However, PROX was not associated with the
incidence of any acute respiratory symptoms, either in- (p ≥ 0.113) or post-commute (p ≥
0.095). A higher general concern for Brisbane’s ambient air pollution level was not
associated with a lower PROX (p = 0.42).
3.6. Air pollution exposure risk management strategies
3.6.1. Commute re-route
Most participants (68%) reported that they were amenable to re-route their commute to
reduce proximity to motorised traffic as an exposure risk management strategy, if proven to
be appropriate and effective, which varied with gender (Table 3) and health status (Table 5).
Females, compared to males, were significantly more amenable to commute re-routing (p <
0.05; Table 3). The group mean importance of strategy features (out of 5.0) for the adoption
of commute re-routing were, from highest to lowest, “safety” (4.0), “time” (3.8), “fitness”
(2.9), “health” (2.9) and “social” (1.5). Further, a participant’s PROX level was negatively-
associated with the importance for commute re-routing strategy features of “health”
[F(34,196) = 3.25, p = 0.002] and “safety” [F(31,182) = 3.10, p = 0.003].
3.6.2. Respirator use
Zero participants currently used a respirator during their bicycle commute; however,
approximately one fifth of participants (21%) had previously considered such use, and this
56
consideration was positively-associated with in-commute air pollution exposure perception
[F(1,23) = 6.33, p = 0.013] and offensive odour detection [F(3,22) = 4.52, p = 0.002]. The
majority of participants (75%) indicated that they would use a respirator as an exposure risk
management strategy if proven to be appropriate and effective, which (similar to commute re-
routing) varied with gender (Table 3) and health status (Table 5). The group mean importance
of strategy features (out of 5.0) for using a respirator while bicycle commuting were, from
highest to lowest, “breathing impedance” (3.9), “wear comfort” (3.7), “appearance” (2.9) and
“expense” (2.5).
4. Discussion
The major results of this study suggest that in healthy individuals, the incidence of specific
acute respiratory symptoms is higher in- and post-commute compared to pre-commute
conditions. Further, the incidence of acute respiratory symptoms in association with bicycle
commuting is higher with respiratory disorder history (compared to healthy) and female
(compared to male) gender participant cohorts. A significant positive-association exists
between the perceived level of in-commute air pollution exposure and the estimated level of
in-commute proximity to motorised traffic (PROX; a product of commute duration,
frequency and proportion of commute on-road) by healthy participants. However, PROX was
not associated with the reported incidence of acute respiratory symptoms in or post typical
bicycle commuting in healthy participants. The majority of participants indicated that they
were amenable to the risk management strategies of commute re-routing and respirator use if
these strategies were shown to be necessary and effective by future research and they were
developed with specified practical features.
57
The detection of offensive odours, associated with vapour gases such as nitrogen dioxide
(NO2), has previously been positively-associated with perceived health risk and prevalence of
acute respiratory symptom reporting [112]. Populations living near environmental odour
sources have reported consistent patterns of subjective symptoms, including exacerbation of
underlying medical conditions and stress-induced illness from offensive odour exposure [38,
113, 114]. However, air pollution exposure limits may not coincide with odour detection or
irritation, as some pollution constituents can cause irritation or harm below perceivable limits
[115]. Previously, exposed individuals have been shown to be capable of both under- and
over-estimation of exposure according to self-reported perception and symptoms compared to
direct air quality measurements [116, 117]. Therefore, communication of accurate air
pollution levels and consequential exposure risk could help to effectively facilitate self-
managed exposure risk strategies for elevated air pollution events, unfavourable
meteorological conditions and physiologically-susceptible individuals, if found to be
appropriate and effective by future research. The increased incidence of acute respiratory
symptoms in participants with physiological susceptibility has also been observed in past
research [32, 35] and other questionnaire-based studies [32, 34-39, 118, 119]. Participants in
the current study with a history of respiratory disorder were more susceptible to acute
respiratory symptoms (and chronic URTI): a history of respiratory disorder increased , and a
history of smoking decreased, the incidence of perception to moderate or higher levels of in-
commute air pollution, compared to healthy participants, which has been observed elsewhere
[120].
Particulate matter (PM) can trigger inflammation in the airways, exacerbate respiratory
disorders, and suppress airway antimicrobial defences [31, 121]. During the study period,
south-east Queensland (SE QLD; surrounding Brisbane, Australia) ambient PM10 and PM2.5
(particulate matter with diameters of 10 and 2.5 micrometres, respectively) maximum daily
58
mean particle mass concentrations were 37 and 19 μg/m3, respectively [122]. Previous
recordings of roadside PM10 showed one-hour mean concentrations of 25 ± 13 μg/m3 and a
maximum of 90 μg/m3, associated with high traffic counts and large proportions of heavy
duty vehicles [123]. PM2.5 one-hour mean concentrations of 21 ± 11 μg/m3 and a maximum of
195 μg/m3 were also shown, with the highest values on week-days (Monday to Friday)
believed to be due to greater traffic counts and proportion of heavy duty vehicles [123].
Short-term exposure to PM10 and PM2.5 at regionalised outdoor mass concentrations of 14 ± 7
and 11 ± 5 μg/m3, respectively, has not been associated with detrimental health effects or
detectable systemic inflammation in young, healthy participants performing exercise [124].
As roadside PM concentrations in this project were higher than that previously shown to be
non-detrimental, in-commute PM exposure could be the cause of acute respiratory symptoms
in physiologically-susceptible individuals in this study; however, the greatest concern
regarding PM is considered to be the particle number concentration (that is, particle count)
rather than particle mass concentration [90]. For a specific mass concentration, ultrafine
particles (UFPs; < 0.1 μm diameter) are the main diameter range of motorised traffic
particulate emissions [59]. As UFPs are not routinely monitored in SE QLD, it is difficult to
consider the effects on bicycle commuters indirectly.
Similar to PM, NO2 can trigger inflammation in the lower airways, exacerbate asthma and
chronic bronchitis, and suppress upper respiratory tract antimicrobial defences such as
macrophage function [31, 121]. During the study period, Brisbane’s ambient NO2 annual
mean was 7 parts per billion (ppb), with a daily peak 1-hour mean of 37 ppb (DERM, 2011).
At such levels, acute respiratory symptoms in healthy adults have not been shown. However,
as NO2 is a major emission component of motorised traffic, exposure concentrations are
expected to be much higher when adjacent to major traffic corridors. Brisbane’s roadside
mean NO2 concentrations have been recorded between 18 and 34 ppb with peaks of
59
approximately 60 ppb positively-correlated with morning (7.00-8.00 AM) and afternoon
(4.00-6.00 PM) commute traffic flow rates, indicating traffic emissions as a dominant
emission source [123]. Asthmatic adults are twice as sensitive as non-asthmatics to short-
term exposures of NO2, however significantly increased airway resistance (due to
inflammation) has not been observed below 500 ppb [30]. Further, acute exposure of very
high concentrations (~5,000 - 10,000 ppb) are necessary to elicit symptoms due to
inflammation such as nasopharyngeal irritation, dyspnoea and tussis in healthy adults [30].
Historical duration of commute and the incidence and severity of URTI were not significantly
linked in this study, nor were acute symptom incidence (such as nasopharyngeal irritation,
dyspnoea and tussis) and estimated exposure level, most likely due to the relatively low
regional traffic counts and associated emissions compared to previously studied regions [30,
125, 126]. Future research involving direct investigation of bike-path and roadside air quality
is warranted to advise the appropriateness and efficacy of implementing risk management
strategies such as commute re-routing.
Preferences of participants for air pollution exposure risk management strategy features were
evaluated to highlight which features of a commute route or respirator are necessary or
desirable to help strategy adoption, if shown to be appropriate and effective. Currently, less
than half of the participant cohort used the highest proportion of off-road paths and nil used a
respirator. It may be the case that some bicycle commute routes do not have a high proportion
of off-road paths available for use, or that there is limited knowledge of respirator availability
for bicycle commuters. The feature of “time (defined to the participant as “more convenient /
quickest route”) was ranked as more important than the feature of “health” (defined to the
participant as “to avoid air pollution”) by participants when choosing a commute route, which
should be considered when new bike paths are being developed. The most important factors
to be addressed for respirator use by participants in this study were “breathing impedance”
60
and “wear comfort”. There are commercially-available respirators that are recommended for
use by urban bicycle commuters due to their design accommodating increased heat
production, perspiration and ventilation rate associated with moderate physical activity (e.g.
‘City Mask’ by Respro Ltd, UK). To support the desired increased participation of bicycle
commuting as a response to public health and infrastructure concerns, desirable properties of
self-managed risk strategies such as direct and safe bicycle paths, or comfortable and
functional respirators, may need to be implemented in some circumstances.
Commuters using different travel modes in SE QLD have previously indicated that they
thought of air pollution exposure as a substantial health concern [40]. Regardless, the
participants of this previous study did not consider air pollution exposure to be a major
barrier to participating in active transport (such as bicycle commuting) [40]. However, a
limitation highlighted by the authors of this previous study was the relatively small sub-set of
active commuters (n = 64 of 745 / 9%) surveyed. The current study (with a cohort of twice
that number) contributes with the suggestion that healthy bicycle commuters can perceive in-
commute air pollution exposure levels consistent with estimated proximity to motorised
traffic, and are generally amenable to reduce their exposure to air pollution by adopting risk
management strategies, if known to be appropriate and effective.
Limitations of the current study include the design of a unique questionnaire - to the authors’
knowledge, a precedent model was not available for reference - however, this design process
was rigorously performed with the review and input of respiratory scientists, epidemiological
statisticians and a sample of the intended participant cohort. Bias for participation of potential
respondents is possible due to the nature of the questionnaire (i.e. bringing focus to a subject
which may discourage the act of bicycle commuting by highlighting associated dangers),
however potential participants spoken-to expressed a positive attitude towards the issue and
hence recruitment did not prove difficult. As the symptoms were self-reported,
61
misunderstandings could have arisen by question misinterpretation; however, again, due to
the review process of questionnaire design, this was believed to be minimised. Finally, the
specifics of human exposure and the associated biological responses (reported as symptom
experience) to ambient air pollution are difficult to assess due to atmospheric mixing effects
and meteorological influence [127], which were beyond the scope of this study.
The merit of responsibly encouraging increased participation of bicycle commuting, and a
strength of this study, is indicated by the fact that the mean one-way commute duration of
participants was approximately 30 minutes, which coincides with daily physical-activity
recommendations to reduce the risk of cardiopulmonary disease development [20-22].
Further, participants typically performed this activity twice a day on four days per week.
Eligibility for participation only required two or more return trips per week (to satisfy the
definition of a regular bicycle commuter) but the group mean was four trips per week,
suggesting that this study represented a dedicated and experienced population of bicycle
commuters. However, seasonal variation of bicycle commuting participation was not
investigated.
5. Conclusions
Healthy individuals have indicated a higher incidence of acute respiratory symptoms in- and
post- (compared to pre-) bicycle commute, with respiratory disorder history and female
gender indicating a comparably-higher susceptibility to air pollution exposure, in this study.
The perception of air pollution exposure levels, combined with an amenability of susceptible
populations to adopt exposure risk management strategies, will aid the informed transition
62
from passive, motorised transport modes to bicycle commuting required for contemporary
urban growth and the increasing popularity of bicycle commuting.
Abbreviations: PROX (estimation ranking of in-commute proximity to motorised traffic);
ppb (parts per billion); SE QLD (South East Queensland); URTI (upper respiratory tract
infection).
ACKNOWLEDGEMENTS
The authors declare they have no actual or potential competing personal or financial
interests. The project was conducted with a protocol approved (#1000001175) by the
University Human Research Ethics Committee (UHREC: +61 7 3138 5123 /
[email protected]), Queensland University of Technology. TCH performed all data
acquisition, analyses and manuscript preparation under the supervision of CS and LM. TCH
is supported by an Australian Postgraduate Award (Department of Innovation, Industry,
Science and Research, Australian Government) scholarship. CS and LM hold academic
positions at the University of the Sunshine Coast and the Queensland University of
Technology, respectively. The authors acknowledge and sincerely thank Dr Ian Stewart
(Institute of Health and Biomedical Innovation) and Dr Rohan Jayaratne (International
Laboratory for Air Quality and Health) of Queensland University of Technology for research
consultation, along with Andrew Onley (Cycle2City Centre, Brisbane, Australia), Sabrina
von Bayer (Active Transport Unit, Brisbane City Council, Australia), and Jan Bell
(Department of Environment and Resource Management, Queensland Government,
Australia) for research participant recruitment. Finally, we sincerely thank all research
participants for their vital role in this study.
63
PROJECT ONE REFERENCES
1. de Nazelle A, Nieuwenhuijsen M (2010) Integrated health impact assessment of cycling.
Occupational and Environmental Medicine 67: 76.
2. de Nazelle A, Nieuwenhuijsen M, Antó J, Brauer M, Briggs D, et al. (2011) Improving
health through policies that promote active travel: a review of evidence to support integrated
health impact assessment. Environment International 37: 766.
3. Woodcock J, Edwards P, Tonne C, Armstrong BG, Ashiru O, et al. (2009) Public health
benefits of strategies to reduce greenhouse-gas emissions: urban land transport. The Lancet
374: 1930-1943.
4. Rabl A, de Nazelle A (2012) Benefits of shift from car to active transport. Transport Policy
19: 121-131.
5. de Hartog JJ, Boogaard H, Nijland H, Hoek G (2010) Do the health benefits of cycling
outweigh the risks? Environmental Health Perspectives 118: 1109.
6. Weichenthal S, Kulka R, Dubeau A, Martin C, Wang D, et al. (2011) Traffic-Related Air
Pollution and Acute Changes in Heart Rate Variability and Respiratory Function in Urban
Cyclists. Environmental health perspectives 119: 1373–1378.
7. Int Panis L, de Geus B, Vandenbulcke G, Willems H, Degraeuwe B, et al. (2010) Exposure
to particulate matter in traffic: A comparison of cyclists and car passengers. Atmospheric
Environment 44: 2263-2270.
8. McNabola A, Broderick BM, Gill LW (2008) Relative exposure to fine particulate matter
and VOCs between transport microenvironments in Dublin: Personal exposure and uptake.
Atmospheric Environment 42: 6496-6512.
9. Zuurbier M, Hoek G, Oldenwening M, Lenters V, Meliefste K, et al. (2010) Commuters’
exposure to particulate matter air pollution is affected by mode of transport, fuel type, and
route. Environmental Health Perspectives 118: 783.
10. Zuurbier M, Hoek G, Oldenwening M, Meliefste K, Van den Hazel P, et al. (2009) Acute
Health Effects of Commuters' Exposure to Air Pollution. Epidemiology 20: S197.
64
11. Adams HS, Nieuwenhuijsen MJ, Colvile RN (2001) Determinants of fine particle (PM2.
5) personal exposure levels in transport microenvironments, London, UK. Atmospheric
Environment 35: 4557-4566.
12. Briggs DJ, de Hoogh K, Morris C, Gulliver J (2008) Effects of travel mode on exposures
to particulate air pollution. Environment International 34: 12-22.
13. Kaur S, Nieuwenhuijsen MJ, Colvile RN (2005) Pedestrian exposure to air pollution
along a major road in Central London, UK. Atmospheric Environment 39: 7307-7320.
14. Kaur S, Nieuwenhuijsen MJ, Colvile RN (2007) Fine particulate matter and carbon
monoxide exposure concentrations in urban street transport microenvironments. Atmospheric
Environment 41: 4781-4810.
15. Zhou Y, Levy J (2007) Factors influencing the spatial extent of mobile source air
pollution impacts: a meta-analysis. BMC Public Health 7: 89.
16. Rojas-Rueda D, de Nazelle A, Tainio M, Nieuwenhuijsen MJ (2011) The health risks and
benefits of cycling in urban environments compared with car use: health impact assessment
study. BMJ: British Medical Journal 343.
17. Künzli N, Jerrett M, Garcia-Esteban R, Basagaña X, Beckermann B, et al. (2010)
Ambient air pollution and the progression of atherosclerosis in adults. PLoS One 5: e9096.
18. Lee K, Yanagisawa Y, Spengler JD, Nakai S (1994) Carbon monoxide and nitrogen
dioxide exposures in indoor ice skating rinks. Journal of Sports Sciences 12: 279-279.
19. Ghio AJ, Devlin RB (2001) Inflammatory lung injury after bronchial instillation of air
pollution particles. American Journal of Respiratory and Critical Care Medicine 164: 704.
20. Belanger K, Gent JF, Triche EW, Bracken MB, Leaderer BP (2006) Association of
indoor nitrogen dioxide exposure with respiratory symptoms in children with asthma.
American journal of respiratory and critical care medicine 173: 297.
21. Johns DO, Svendsgaard D, Linn WS (2010) Analysis of the concentration-respiratory
response among asthmatics following controlled short-term exposures to sulfur dioxide.
Inhalation Toxicology 22: 1184-1193.
65
22. Wilhelm M, Qian L, Ritz B (2009) Outdoor air pollution, family and neighborhood
environment, and asthma in LA FANS children. Health & Place 15: 25-36.
23. Jarvis D, Chinn S, Luczynska C, Burney P (1996) Association of respiratory symptoms
and lung function in young adults with use of domestic gas appliances. The Lancet 347: 426-
431.
24. Kjærgaard T, Cvancarova M, Steinsvåg SK (2011) Cigarette smoking and self-assessed
upper airway health. European Archives of Oto-Rhino-Laryngology 268: 219-226.
25. van der Zee SC, Hoek G, Boezen MH, Schouten JP, van Wijnen JH, et al. (2000) Acute
effects of air pollution on respiratory health of 50-70 yr old adults. European Respiratory
Journal 15: 700.
26. Aatamila M, Verkasalo PK, Korhonen MJ, Suominen AL, Hirvonen MR, et al. (2010)
Odour annoyance and physical symptoms among residents living near waste treatment
centres. Environmental Research 111: 164-170.
27. Stenlund T, Liden E, Andersson K, Garvill J, Nordin S (2009) Annoyance and health
symptoms and their influencing factors: a population-based air pollution intervention study.
Public Health 123: 339-345.
28. Badland HM, Duncan MJ (2009) Perceptions of air pollution during the work-related
commute by adults in Queensland, Australia. Atmospheric Environment 43: 5791-5795.
29. ATS (2000) What constitutes an adverse health effect of air pollution? American Journal
of Respiratory and Critical Care Medicine 161: 665-673.
30. Rudell B, Ledin M, Hammarström U, Stjernberg N, Lundbäck B, et al. (1996) Effects on
symptoms and lung function in humans experimentally exposed to diesel exhaust.
Occupational and Environmental Medicine 53: 658.
31. Council) BBC (2010) Bikeway and shared pathway maps.
32. Cone JE, Shusterman D (1991) Health effects of indoor odorants. Environ Health Persp
95: 53.
33. Lercher P, Schmitzberger R, Kofler W (1995) Perceived traffic air pollution, associated
behavior and health in an alpine area. Science of The Total Environment 169: 71-74.
66
34. Shusterman D (1992) Community health and odor pollution regulation. American Journal
of Public Health 82: 1566.
35. Paustenbach DJ, Gaffney SH (2006) The role of odor and irritation, as well as risk
perception, in the setting of occupational exposure limits. International Archives of
Occupational and Environmental Health 79: 339-342.
36. Dalton P (1999) Cognitive influences on health symptoms from acute chemical exposure.
Health Psychology 18: 579-590.
37. Righi E, Aggazzotti G, Fantuzzi G, Ciccarese V, Predieri G (2002) Air quality and well-
being perception in subjects attending university libraries in Modena (Italy). Science of the
Total Environment 286: 41-50.
38. Hirsch T, Weiland SK, Von Mutius E, Safeca AF, Gräfe H, et al. (1999) Inner city air
pollution and respiratory health and atopy in children. European Respiratory Journal 14: 669-
677.
39. Simoni M, Annesi-Maesano I, Sigsgaard T, Norback D, Wieslander G, et al. (2010)
School air quality related to dry cough, rhinitis and nasal patency in children. European
Respiratory Journal 35: 742.
40. Nikolopoulou M, Kleissl J, Linden P, Lykoudis S (2011) Pedestrians' perception of
environmental stimuli through field surveys: Focus on particulate pollution. The Science of
the total environment 409: 2493-2502.
41. Nel AE, Diaz-Sanchez D, Ng D, Hiura T, Saxon A (1998) Enhancement of allergic
inflammation by the interaction between diesel exhaust particles and the immune system.
Journal of Allergy and Clinical Immunology 102: 539-554.
42. Management) DDoEaR (2011) Queensland 2010 Air Monitoring Report.
43. Ayoko G, Jamriska M, Jayaratne R, Morawska L. Air pollution levels measured at traffic
hot spots: Brisbane urban corridor study; 2005. CASANZ (Clean Air Society of Australia &
New Zealand). pp. 1-7.
44. Bräuner EV, Møller P, Barregard L, Dragsted LO, Glasius M, et al. (2008) Exposure to
ambient concentrations of particulate air pollution does not influence vascular function or
inflammatory pathways in young healthy individuals. Particle and Fibre Toxicology 5: 13.
67
45. Oberdörster G, Oberdörster E, Oberdörster J (2005) Nanotoxicology: an emerging
discipline evolving from studies of ultrafine particles. Environmental Health Perspectives
113: 823.
46. Knibbs LD, Cole-Hunter T, Morawska L (2011) A review of commuter exposure to
ultrafine particles and its health effects. Atmospheric Environment 45: 2611-2622.
47. Belanger E, Kielb C, Lin S (2006) Asthma Hospitalization Rates Among Children, and
School Building Conditions, by New York State School Districts, 1991 2001. Journal of
School Health 76: 408-413.
48. Pribyl CR, Racca J (1996) Toxic gas exposures in ice arenas. Clinical Journal of Sport
Medicine 6: 232.
49. Blair SN, Morris JN (2009) Healthy hearts--and the universal benefits of being physically
active: Physical activity and health. Annals of Epidemiology 19: 253-256.
50. Shiroma EJ, Lee I (2010) Physical activity and cardiovascular health: lessons learned
from epidemiological studies across age, gender, and race/ethnicity. Circulation 122: 743.
51. Warburton DER, Nicol CW, Bredin SSD (2006) Health benefits of physical activity: the
evidence. Canadian Medical Association Journal 174: 801.
52. Brunekreef B, Holgate ST (2002) Air pollution and health. The Lancet 360: 1233-1242.
68
TABLES
Table 1 - Characteristics of Bicycle Commuter Participants
CHARACTERISTIC
MEAN SD MIN / MAX
Gender (% female)
Age (years)
Single Trip Distance (km)
Single Trip Duration (min)
Return Trips (per week)
Commute History (month)
Inbound Start Time (24 hr)
Outbound Start Time (24 hr)
28.2
41
11
31
4
27
07:15
17:10
--
11
7
15
1
32
0:56
1:00
--
19 / 64
4 / 32
10 / 65
2 / 5
6 / 180
05:30 / 10:00
15:30 / 19:00
69
Table 3 - Perceptions and preferences of regular bicycle commuters for the total group and
gender groups
RESPONSE TOTAL GENDER
[n = 153 (100%)]
Female
[n = 43 (28%)]
Male
[n = 110 (72%)]
Ambient Air Concern (%Yes) 71 80 68
Perceived Exposure (%Yes) 80 91* 76
Re-route Amenable (%Yes) 68 80* 63
Respirator Amenable (%Yes) 75 77 74
Estimated Exposure (%Commute) 52 ± 2.8 46 ± 3.4 54 ± 2.6
‘%Yes’ = proportion of positive response from Total / Gender group. ‘%Commute’ = group mean proportional
time in proximity to motorised traffic of Total / Gender group; values are group mean ± standard deviation (SD).
* p < 0.05, higher positive response than Male. Participants [n = participant number (percentage of total cohort)]
reported their general concern for ambient air pollution, their perception of in-commute air pollution exposure,
their amenability to use risk management strategies of commute re-routing or respirator use. Estimated exposure
is given as a percentage of time bicycle commuting adjacent to motorised traffic corridors. All other percentages
are group total of positive responses. One-way analysis of variance (ANOVA) were performed between groups
of female and male participants to highlight any significant differences of Gender group mean responses.
70
Table 2 – Acute respiratory symptoms of regular bicycle commuters for the total group and
gender groups
RESPONSE TOTAL GENDER
[n = 153 (100%)]
Female
[n = 43 (28%)]
Male
[n = 110 (72%)]
Pre In Post Pre In Post Pre In Post
Offensive Odour
1.38
±
0.08
2.70
±
0.12
1.44
±
0.08
1.45
±
0.12
2.77
±
0.14
1.50
±
0.11
1.35
±
0.06
2.67
±
0.11
1.41
±
0.07
Nasopharyngeal
Irritation
1.39
±
0.07
2.05
±
0.10
1.72
±
0.12
1.50
±
0.11
2.27**
±
0.13
2.11**
±
0.25
1.35
±
0.06
1.96
±
0.09
1.57
±
0.07
Cough and/or
Phlegm
1.35
±
0.07
1.96
±
0.11
1.66
±
0.09
1.39
±
0.11
2.00
±
0.15
1.75
±
0.13
1.33
±
0.06
1.94
±
0.10
1.63
±
0.08
Chest Tightness
1.21
±
0.05
1.48
±
0.09
1.30
±
0.07
1.27
±
0.09
1.52
±
0.12
1.34
±
0.11
1.18
±
0.04
1.47
±
0.08
1.28
±
0.05
Chest Wheeze
1.18
±
0.05
1.45
±
0.08
1.30
±
0.05
1.30
±
0.09
1.64*
±
0.12
1.39
±
0.01
1.14
±
0.04
1.38
±
0.07
1.27
±
0.06
Group mean ± standard deviation (SD) * p < 0.05, ** p < 0.01, higher incidence than Male. Participants [n =
participant number (percentage of total cohort)] reported the incidence of air pollution perception and acute
respiratory symptoms of increasing severity one hour before (Pre), during (In) and one hour after (Post) their
standard bicycle commute, using an ordinal scale of 1.00 to 5.00 converted from the group mean reported
interval scale (as: 1 = Very Low; 2 = Low; 3 = Moderate; 4 = High; 5 = Very High). One-way analysis of
variance (ANOVA) were performed between groups of female and male participants to highlight any significant
differences of Gender group mean responses.
71
Table 5 – Perceptions and preferences for regular bicycle commuters of health status groups
RESPONSE HEALTH STATUS
Healthy Smoking History Disorder History
[n = 93 (62%)] [n = 24 (15%)] [n = 36 (23%)]
Ambient Air Concern (% Yes) 74 64 65
Perceived Exposure (% Yes) 82 68* 72
Re-route Amenable (% Yes) 68 59 65
Respirator Amenable (% Yes) 75 71 73
Estimated Exposure (% Commute) 55 ± 2.6 44 ± 5.2 45
‘%Yes’ = proportion of positive response from Health Status groups. ‘%Commute’ = group mean proportional
time in proximity to motorised traffic of Health Status groups; values are group mean ± standard deviation (SD).
* p < 0.05, lower positive response than Healthy. Participants [n = participant number (percentage of total
cohort)] reported their general concern for ambient air pollution, their perception of in-commute air pollution
exposure, their amenability to use risk management strategies of commute re-routing or respirator use.
Estimated exposure is given as a percentage of time bicycle commuting adjacent to motorised traffic corridors.
All other percentages are group total of positive responses. Smoking History is defined as a participant who
ceased habitual smoking greater than 24 months previously but is otherwise healthy. Disorder History is defined
as a participant who reported any history of a respiratory disorder. One-way analysis of variance (ANOVA)
were performed between groups of Healthy, Smoking History and Disorder History participants to highlight any
significant differences of Health Status group mean responses.
72
Table 4 – Acute respiratory symptoms of regular bicycle commuters (according to health
status)
RESPONSE HEALTH STATUS
Healthy
[n = 93 (62%)]
Smoking History
[n = 24 (15%)]
Disorder History
[n = 36 (23%)]
Pre In Post Pre In Post Pre In Post
Offensive Odour 1.31
±
0.06
2.67
± 0.11
1.38
±
0.07
1.45
±
0.14
2.40+
±
0.22
1.45
±
0.14
1.53
±
0.15
2.76
±
0.19
1.51
±
0.13
Nasopharyngeal
Irritation
1.34
±
0.06
1.96*
± 0.09
1.63*
±
0.12
1.60
±
0.21
1.95*
±
0.21
1.70
±
0.16
1.44
±
0.10
2.24##
±
0.16
1.97#
±
0.17
Cough and/or Phlegm 1.28
±
0.06
1.80*
± 0.09
1.47*
±
0.07
1.45
±
0.15
1.95*
±
0.25
1.90*
±
0.25
1.47
±
0.14
2.41##
±
0.21
2.05##
±
0.16
Chest Tightness 1.15
±
0.04
1.35
± 0.06
1.18
±
0.05
1.25
±
0.10
1.55*
±
0.15
1.40
±
0.13
1.28
±
0.09
1.84##
±
0.18
1.51#
±
0.11
Chest Wheeze 1.14
±
0.05
1.36
± 0.07
1.18
±
0.05
1.20
±
0.09
1.45
±
0.15
1.40
±
0.15
1.28
±
0.09
1.76##
±
0.16
1.57#
±
0.13
Group mean ± standard deviation (SD) *p < 0.05, higher symptom incidence than pre-commute (‘Pre’). #p <
0.05, ##p < 0.01, higher incidence than Healthy. + p < 0.05, lower incidence than Healthy. Participants [n =
participant number (percentage of total cohort)] reported the incidence of air pollution perception and acute
respiratory symptoms of increasing severity one hour before (Pre), during (In) and one hour after (Post) their
standard bicycle commute, using an ordinal scale of 1.00 to 5.00 converted from the group mean reported
interval scale (as: 1 = Very Low; 2 = Low; 3 = Moderate; 4 = High; 5 = Very High). Smoking History is defined
as a participant who ceased habitual smoking greater than 24 months previously but is otherwise healthy.
Disorder History is defined as a participant who reported any history of a respiratory disorder. One-way analysis
of variance (ANOVA) were performed between groups of healthy, smoking history and disorder history
participants to highlight any significant differences of Health Status group mean responses.
73
5. PROJECT TWO: Inhaled particle counts on bicycle commute routes of
low and high proximity to motorised traffic
Inhaled particle counts on bicycle commute routes of low and high proximity to motorised
traffic
Tom Cole-Huntera,b, Lidia Morawskab, Ian Stewarta, Rohan Jayaratneb, Colin Solomonc, d*
Atmospheric Environment, (2012) 61:197-203. doi: 10.1016/j.atmosenv.2012.06.041
a Institute of Health and Biomedical Innovation, Queensland University of Technology, 60
Musk Avenue, QLD 4059, Australia.
b International Laboratory for Air Quality and Health, Queensland University of Technology,
2 George Street, QLD 4001, Australia.
c School of Life Sciences, Queensland University of Technology, 2 George Street, QLD 4001,
Australia.
d School of Health and Sport Sciences, University of the Sunshine Coast, Sippy Downs Drive,
QLD 4556, Australia.
*Corresponding Author: Dr Colin Solomon, School of Health and Sport Sciences, University
of the Sunshine Coast, Sippy Downs Drive, Sippy Downs, QLD 4556, Australia.
Telephone: +61 7 54301128. E-mail: [email protected]
74
75
SEE APPENDIX E FOR PUBLICATION
76
6. PROJECT THREE: The reduction of ultrafine particle exposure by
utilising bicycle commute routes of low versus high proximity to major
motorised traffic corridors
The reduction of ultrafine particle exposure by utilising bicycle commute routes of low
versus high proximity to major motorised traffic corridors
Tom Cole-Hunter1,2, Lidia Morawska2, Ian Stewart1, Rohan Jayaratne2, Matthew Hadaway1,
Colin Solomon3,4,*
Environmental Health, In Review.
1 Institute of Health and Biomedical Innovation, Queensland University of Technology, 60
Musk Avenue, QLD 4059, Australia.
2 International Laboratory for Air Quality and Health, Queensland University of
Technology, 2 George Street, QLD 4001, Australia.
3 School of Life Sciences, Queensland University of Technology, 2 George Street, QLD 4001,
Australia.
4 School of Health and Sport Sciences, University of the Sunshine Coast, Sippy Downs Drive,
QLD 4556, Australia.
*Corresponding Author: Dr Colin Solomon, School of Health and Sport Sciences, University
of the Sunshine Coast, Sippy Downs Drive, Sippy Downs, QLD 4556, Australia.
Telephone: +61754301128. E-mail: [email protected]
77
78
Abstract
Background
Bicycle commuting in an urban environment of high levels of air pollution is known as a
potential health risk, especially for susceptible individuals. While risk management strategies
aimed to reduce motorised traffic emissions exposure have been suggested, limited studies
have assessed the utility of such strategies in real-world circumstances.
Objectives
The potential of lowering exposure to ultrafine particles (UFP; < 0.1 µm) during bicycle
commuting by reducing proximity to major motorised traffic corridors (without significantly
affecting commute distance or duration) was investigated using continuous UFP and periodic
respiratory symptom and inflammation measurements in healthy individuals using their
typical, and an alternative purposely-designed, bicycle commute route.
Methods
Thirty-five healthy adults (mean ± SD: age = 39 ± 11 yr; 29 % female) each completed two
return trips of their typical route (HIGH) and a purposely-designed route alteration of lower
proximity to major motorised traffic corridors (LOW). Particle number concentration (PNC)
and diameter (PD) were monitored continuously in-commute. Acute inflammatory indices of
respiratory symptom occurrence, pulmonary function and spontaneous sputum (for
inflammatory cell analyses) were collected immediately pre-commute, and one and three
hours post-commute.
Results
In-commute mean PNC was significantly reduced in LOW compared to HIGH [mean ± SD:
1.91 x e4 ± 0.93 x e4 parts per cubic centimetre (ppcc) vs. 2.95 x e4 ± 1.50 x e4 ppcc; p ≤
0.001]. Commute distance and duration were not significantly different between LOW and
HIGH (12.8 ± 7.1 vs. 12.0 ± 6.9 km, p = 0.325; 44 ± 17 vs. 42 ± 17 mins, p = 0.196;
79
respectively). Self-reported occurrences of acute health-associated signs and symptoms were
reduced in LOW compared to HIGH, including in-commute offensive odour detection (42 vs.
56 %; p = 0.019), dust and soot observation (33 vs. 47 %; p = 0.038) and nasopharyngeal
irritation (31 vs. 41 %; p = 0.007) .
Conclusions
Exposure level of UFP PNC (and the occurrence of offensive odour and nasopharyngeal
irritation) can be significantly reduced (without significantly affecting commute distance or
duration) if an individual uses a purposely-designed route of reduced proximity to major
motorised traffic corridors whilst bicycle commuting, which will provide health benefits for
both healthy and susceptible individuals.
Key words
Air pollution, bicycle commuting, route alteration, ultrafine particle, respiratory symptom,
peak expiratory flow, inflammatory cell
Introduction
The health benefits of physical activity associated with active transport are well-established
[1-3], as are the negative health effects associated with elevated levels of air pollution
exposure [4-6]. Subsequently, there has been investigation of approaches to reduce the degree
of air pollution exposure, along with mechanisms of health effects, whilst actively
commuting [7-11]. Risk management strategies for reducing air pollution exposure whilst
actively commuting can include reducing proximity to motorised traffic by avoiding major
motorised traffic corridors at peak traffic times [12]. The majority of projects on this topic
have utilised micro-environments of designated off-road and on-road bicycle paths, and have
determined that the former generally facilitates a significantly lower potential for exposure to
80
air pollution, mainly from motorised traffic emissions such as ultrafine particles (UFP) [13-
17]. Health indices, including acute respiratory symptoms, impaired pulmonary function and
inflammation-associated cell distribution have been used to investigate the physiological
response to components of air pollution including particle number concentration (PNC;
which is dominated by UFP) [18-20].
Questionnaires that assess the influence of exposure on the respiratory system have been used
previously [29, 30], including specific symptoms attributable to acute air pollution exposure
recommended by the American Thoracic Society [31]. For instance, airway narrowing due to
inflammation and excessive mucous secretion (as an immune response to airway irritation by
pollutants) can induce coughing and chest tightness or wheezing, as well as reduce
pulmonary function indicated by lowered peak expiratory flow rates [32]. Further, an increase
in the number of leukocytes, and specifically neutrophils, found in the airways and systemic
circulation can indicate an inflammatory response to exposure from pollutants such as
ultrafine particles [33, 34]. Yet to be investigated is the feasibility to reduce exposure to
elevated PNC levels (and thereby decrease any associated negative health effects) with
bicycle commuters using a purposely-designed alteration of their typical route to avoid major
motorised traffic corridors.
Bicycle commuters may not have the amenity of a route which allows complete use of off-
road bicycle paths. Therefore, it is not expected to be practical for a bicycle commuter to
completely alter their commute route and altogether avoid exposure to motorised traffic
emissions, particularly due to factors such as road crossings which dissect off-road paths;
however, it is feasible to decrease exposure to UFP by selecting a route which has reduced
proximity to major motorised traffic corridors. A previous study by the current investigators,
using a single participant model to determine inhaled particle counts along popular bicycle
commute routes of high and low proximity to major motorised traffic corridors, has shown
81
that air quality (indicated by PNC) can be significantly improved by taking such a purposely-
designed route alteration [49]. For the current project, it was hypothesised that: 1) a bicycle
commute route alteration purposely-designed to be of reduced proximity to major motorised
traffic corridors will significantly reduce exposure to combustion emissions [represented by
the dominant ultrafine particle (UFP; < 0.1 µm) number concentration (PNC)], compared to a
higher proximity route; 2) health outcomes (as occurrence and severity of acute respiratory
symptoms, peak flow rate, and cell distribution in sputum) will be improved with the use of a
route of reduced proximity to major motorised traffic corridors, compared to a route of high
proximity; 3) the difference in the estimated inhaled UFP count between the two routes will
be attributable to the difference in PNC, rather than any differences in physical effort
(represented by heart rate and reflecting ventilation rate).
Methods
Project Design
This project was intended to determine whether the use of a bicycle commute route
purposely-designed to reduce proximity to major motorised traffic corridors (and therefore
exposure to associated emissions) is practical as an air pollution exposure reduction and risk
management strategy. Thirty-five healthy adults were recruited to perform their typical
workday commute along both their typical route (selected as being of high proximity to major
motorised traffic corridors; labelled ‘HIGH’) and an altered route (designed to be of reduced
proximity to major motorised traffic corridors; labelled ‘LOW’). The participants and their
bicycles were instrumented to measure real-time exposure variables of geolocation, heart
rate, and particle number concentration and size while in-commute. Participants performed
symptom-occurrence reporting, peak expiratory flow (PEF) metering and spontaneous
82
sputum sampling immediately pre-commute, and immediately and three hours post-commute.
Data collection occurred in Brisbane, Australia between Autumn and Spring (April to
September) of 2011, on consecutive days if practical for participants and not including the
weekend.
Participants
The participants of this project were healthy adults (N: 35; 29% female. Mean ± SD:
age = 39 ± 11 yr; PFR, female = 447± 66, male = 584 ± 89, |total = 558 ± 105| L·min) with
no history of cardiopulmonary disease and no recent history of smoking (cessation > twenty-
four months prior) or respiratory infection (symptoms > two weeks prior). Participants were
required to be frequent bicycle commuters of the Brisbane inner-city region [defined as
completing two or more return trips in a five day period to a destination within a 1 km radius
of the Brisbane Central Business District (CBD)] and have a typical commute route of high
proximity to major motorised traffic corridors. Recruitment was conducted from participants
who provided consent as part of a previous unpublished study, and eligible respondents of a
regional media release. Participants were requested to avoid any air pollution sources where
possible, such as second-hand smoke and traffic congestion during the one hour pre-commute
and three hour post-commute monitoring period. This request may have affected a
participant’s typical daily exposure, however was included to minimise any confounding of
an acute inflammatory response from non-commute exposure and in-commute exposure.
Project Locality
The current project was performed in Brisbane, which is the state capital of
Queensland and the third largest city in Australia. The Brisbane CBD is located at 27º3´
South, 153º9´ East, approximately 20 km inland from the Pacific Ocean. A large river runs
through the city of Brisbane (located within a low-lying floodplain) with several large hills of
83
up to 300 metres in height within the area, bordered to the west by a coastal mountain range.
The regional climate is sub-tropical, being cool and dry in winter (June to August), and
humid and wet in summer (December to February) [23]. The city of Brisbane has a
population of approximately two million, which has been increasing for the last two decades
by approximately two percent annually [24]. Motorised traffic volume (along with population
growth) is rapidly increasing, particularly due to outer-city residential development [24]. The
number of motor vehicles registered within Brisbane in 2011 was approximately 1 million,
however the greater region of South-East Queensland includes a total of 2.8 million motor
vehicles [25]. Industrial air pollution sources include a major airport, seaport, and oil
refineries (approximately 15 km north-east of the CBD), a coal power station (approximately
30 km south-west of the CBD), and various manufacturing companies in the outer suburbs.
Brisbane is in compliance with the standards and goal of the Ambient Air Quality National
Environment Protection Measure (AAQ NEPM). The Queensland air monitoring report 2011
[70] confirms compliance of the South-East Queensland Air-shed (containing Brisbane) for
CO (daily AAQ NEPM standard upper limit ≥ 9.0 ppm; number of days exceeding = 0 days),
NO2 (≥ 0.12 ppm; 0 days), PM10 (≥ 50 µg/m3; 1 day), PM2.5 (≥ 25 µg/m3; 1 day), SO2 (≥ 0.20
ppm; 0 days) in 2011.
Routes of High and Low Proximity to Major Motorised Traffic Corridors
Participants, in consultation with the primary investigator, designed an altered route
of reduced proximity to major motorised traffic corridors (LOW) based on their typical
bicycle commute route (HIGH). Each participant rode a return trip [inbound (morning) and
outbound (evening)] of HIGH and LOW on consecutive days. An equal number of
participants performed either HIGH or LOW first, to counter-balance and negate any
influence of the order of the route alteration. Therefore, a total of 140 trips were performed as
84
a result of 35 participants each completing an inbound-HIGH, outbound-HIGH, inbound-
LOW and outbound-LOW trip.
In-commute Particle Concentration and Diameter
To measure and record real-time particle number concentration (PNC) and particle
diameter (PD) in-commute, an Aerasense NanoTracer (Philips, The Netherlands) with a 16-
second logging frequency was carried by each participant; three units were used to process
three participants simultaneously, with one unit assigned to each participant. The sampling
tube of the NanoTracer was attached to the participant in their immediate breathing zone, for
example on their shirt collar or upper backpack strap. The NanoTracer is a compact and
portable device capable of measuring PNC [0 – e6 particles per cubic centimetre (ppcc)] and
PD (0.01 – 0.3 µm) in real-time via diffusion charging [26, 65]; thus, tilt errors that may be
experienced with fluid-reliant instruments (such as a condensation particle counter) during
vigorous use (such as active transport monitoring) are avoided with the NanoTracer.
Correction factors for the NanoTracer are regularly determined by laboratory calibration
testing at the International Laboratory for Air Quality and Health (ILAQH) against a water-
based condensation particle counter (WCPC 3781; TSI Inc., USA) and a scanning mobility
particle sizer (SMPS 3934, TSI Inc., USA) in atmospheric air and at outdoor locations for 4
hours to derive a ratio of the two average values. The WCPC and SMPS are themselves
regularly calibrated in the laboratory using standard aerosols of known size and
concentration. For the current project, correction factors (of PNC ± 500 ppcc and PD ± 0.01
µm) were applied to raw particle measurement data prior to statistical analyses; the three
units generally have a close correlation (r2 = 0.94).
In-commute PNC and PD means, medians and range were calculated with NanoReporter
software (Philips, The Netherlands) for comparison between HIGH and LOW. PNC or PD
85
16-second-means below 100 ppcc or 0.01 µm, respectively, have been previously considered
as unrealistic and thus were removed prior to analyses [10, 27].
In-commute Heart Rate
Heart rate (FH) was monitored in-commute using a telemetry unit (Polar Electro,
Finland) logging at five second intervals. In-commute FH was compared between HIGH and
LOW to determine if there is a difference in physical effort when performing the two routes
and therefore to indicate if an inhaled particle count for the two routes is due primarily to
variation in air quality or ventilation rate. As an individual’s FH and ventilation rate are
associated [28], a higher mean trip FH would produce a higher mean trip ventilation rate and
therefore a higher total number of inhaled particles at any PNC.
Meteorology
The Australian Bureau of Meteorology Climate Database [23] was accessed for
hourly regional measures of temperature, humidity, wind direction and speed, air pressure,
and precipitation. Meteorological data was collated and analysed to determine any particle
measurement differences between commute monitoring days due to changing atmospheric
conditions.
Physiological Inflammatory Responses
The participants performed three self-administered tests to assess an acute biological
inflammatory response attributable to air pollution exposure. A verbal demonstration and
written explanation for the performance of each test was provided to participants at an
induction meeting, either one day before or on the day of monitoring commencement.
Symptom experience and peak flow rates were self-administered immediately pre- and post-
commute, and three hours post-commute either at the participant’s home or work location.
86
Sputum samples were collected immediately pre-commute and three hours post-commute
only.
Symptom Experience Questionnaire
Participants were supplied with a questionnaire to report the occurrence of
specific signs and symptoms of differing severity, including offensive odour, dust
or soot detection (such as the smell or sight of motor vehicle exhaust or roadway
dust), nasopharyngeal or “eye, nose and throat” irritation, tussis or “cough”, chest
tightness and/or wheezing, on a five-grade scale (1 = ‘Very Low’, 2 = ‘Low’, 3 =
‘Moderate’, 4 = ‘High’, and 5 = ‘Very High’). The questionnaire used in this
investigation was purpose-designed with review and input from researchers and a
sub-set of intended participants. Further, the format of assessment for acute
respiratory signs and symptoms attributable to air pollution exposure was based
on recommendations by the American Thoracic Society [31] and previous
research [64]. The same questions were used for each time period of one hour pre-
commute, in-commute, and three hours post-commute to attribute symptom
occurrence to air pollution data of each monitored trip. See appendix for symptom
questionnaire and written instructions provided to participants.
Peak Expiratory Flow Rates
To obtain an indication of airway diameter and therefore pulmonary function,
participants were supplied with peak expiratory flow meters (MicroPeak,
CareFusion, UK). Participants were instructed to perform and record three peak
expiratory flow tests (to obtain a best value) immediately pre-commute,
immediately post-commute, and three hours post-commute to relate airway
87
diameter changes to UFP data of each monitored trip. See appendix for written
instructions provided to participants.
Sputum Cell Counts
To collect sputum samples, participants were supplied with plastic (Falcon) tubes
(15 mL) containing 2 mL of RNAlater (Ambion, USA), as well as the relevant
RNAlater material safety data sheet and instructions for spontaneous sputum
production. Approximately 2 mL of sputum was collected in the RNAlater and
immediately refrigerated (at approximately 4 ºC) by participants, and then frozen
at -80 ºC within 24 hours by investigators for later analysis. Total and differential
cell counts were performed (double-blinded) via haemocytometry (Olympus light
microscope), using 20 µL of cell suspension (cell pellet plus 2 mL PBS). Cell
count reference values in induced sputum of healthy adults were consulted from a
previous study [35]. See appendix for written instructions provided to participants
for sputum production.
Total Cell Counts
Samples were removed from the -80°C freezer and thawed, then centrifuged
for 15 mins at 500 x G at 25°C. The supernatant was removed and the cell
pellet maintained, adding 2 mL of PBS to suspend the cells and then briefly
vortexed. The cell suspension was aliquoted (2 x 10 µL) to a haemocytometer
and then viewed under a 40X lens and the cells counted by a blinded
investigator. The proportion of squamous epithelial cells (SEC) was
determined to indicate validity or saliva contamination of each sputum sample
(as ≤ 400 SEC per 100 leukocytes).
88
Differential Cell Counts
200 µL of cell suspension was cytospun (for 5 mins at 100 x G at 25°C), fixed
with methanol, stained (Diff-Quick) and mounted (Permount). Satisfactory
differential cell counts (from 2 slides with > 50% viability and < 25%
squamous cells, 100 non-squamous cells counted per slide) were conducted by
a blinded investigator.
Statistical Data Analysis
Due to the expected variation of typical commute characteristics (including commute
time, distance and duration) within the participant group, the in-commute trip variable means
and medians of the four different individual data sets (i.e. both inbound and outbound trips of
HIGH and LOW) of a single participant were initially compared. Subsequently, group means
or medians of both inbound and outbound trips of HIGH and LOW were compared within an
individual to determine if lower exposure to motorised traffic-emitted UFP occurred because
of an individual’s purposely-designed route alteration; however, no attempts were made to
compare in-commute exposure between individual participants. All analyses were performed
with predictive analytics software (PASW v18.0; IBM, New York).
Estimated marginal means of personal and commute exposure factors, along with descriptive
values, were produced. Pearson bivariate correlations were performed for particle
measurements (PNC, PD) with independent variables of meteorology, and symptom
occurrence with participant characteristics such as age. Pearson bivariate correlations were
also performed between CBD proximity (indicative of motorised traffic level) and HIGH or
LOW values of PNC, PD and FH. Multivariate repeated measure ANOVA was performed
with the mean and median of the dependent variables of PNC and PD for both inbound and
outbound HIGH and LOW to signify intra-individual variability. One-way ANOVA (plus
89
Tukey Post-Hoc, where applicable) were performed with PNC, PD (now as independent
variables), gender and the dependent variable of symptom reporting at the three different
time-points. Mixed Effects Models analysis was performed with PNC, PD and participant
symptom reporting, peak flow and cell counts to determine the effect of air quality on a
physiological inflammatory response between inbound and outbound HIGH and LOW.
Further, this analysis was performed to signify the difference across the three commute-
related time-points (i.e. one-hour pre-commute, one-hour post-commute and three-hours post-
commute) in relation to in-commute PNC and PD. Statistical significance was accepted at a
confidence interval of 95% (i.e. p < 0.05).
Results
Bicycle Commute Characteristics
Due to the local and regional location of the bicycle paths, it was not practically
possible to produce exactly the same proportion of off-road paths; therefore, as expected,
there was a range in the distribution of path type within HIGH and LOW. For example,
popular South and West LOW routes ran adjacent (but physically-separated from) two
different major motorised traffic corridors and therefore had lower proportions of off-road
paths. Conversely, popular North and East LOW routes ran adjacent to parklands and a major
river, respectively, allowing higher proportions of off-road paths.
In-commute Particle Measurements
The group mean commute particle number concentration (PNC) of HIGH was
significantly higher than in LOW [F-statistic (degrees of freedom) and p-value: F(1,35) =
21.079 and p ≤ 0.001], and the group mean commute PNC in HIGH was significantly higher
90
with inbound compared to outbound trips [F(1,35) = 8.441; p = 0.007]. See Table 1.
Additionally, the group median commute PNC in HIGH was significantly higher than in
LOW [F(1,35) = 14.025; p = 0.001], however there was no significant difference between
inbound and outbound trips in HIGH [F(1,35) = 23.154; p = 0.085] or LOW [F(1,35) =
19.237; p = 0.122]. See Table 1.
The group median commute particle diameter (PD) was not significantly different between
HIGH and LOW, nor between inbound and outbound trips of either HIGH or LOW [F(1,35)
= 4.843; p = 0.083]. However, group mean PNC and PD were significantly negatively-
correlated (r = -0.645; p = 0.048). See Table 1.
Commute Speed and Heart Rate
The group mean commute distance and duration, and therefore commute speed, for
HIGH and LOW, were not significantly different (12.0 ± 6.9 vs. 12.8 ± 7.1 km and 42 ± 17
vs. 44 ± 17 mins, respectively). The group mean commute heart rate between HIGH and
LOW (136 ± 11 vs. 133 ± 9 bpm), or between inbound and outbound trips in either HIGH or
LOW, was not significantly different (See Table 1).
Meteorology
All meteorological variables were not significantly different between HIGH and
LOW. Due to natural diurnal variation, the group mean inbound (morning) commute
temperature was significantly lower [F(1,35) = 47.085; p ≤ 0.001] and the humidity
significantly higher [F(1,35) = 54.114; p ≤ 0.001], compared to the outbound (afternoon)
trip. See Table 1.
Group mean regional commute temperature was negatively-correlated with PNC (r = -0.83; p
= 0.005) and positively-correlated with PD (r = 0.79; p = 0.014). However, group mean
91
regional commute humidity was not significantly correlated with mean PNC or PD. While
regional wind direction was not correlated to particle measurements, general wind speed was
negatively-correlated to PNC (r = -0.77; p = 0.018) and PD (r = -0.74; p = 0.021).
Inflammatory Response
Air Quality Detection and Symptom Experience
Offensive odour detection occurrence in HIGH was significantly greater than in
LOW [F(1,406) = 5.515; p = 0.019], as was dust and soot detection [F(1,140) = 4.340;
p = 0.038], nasal irritation [F(1,140) = 7.266; p = 0.007] and throat irritation [F(1,140)
= 8.876; p = 0.003]. All other specific acute respiratory symptoms reported were not
significantly different between HIGH and LOW. See Table 2.
The occurrence of offensive odour, and dust or soot, detection was significantly
higher in-commute, compared to pre-commute and post-commute, for both HIGH and
LOW [F(1,406) = 4.165 ; p = 0.031]; however, the occurrence of nasal and throat
irritation was significantly higher in-commute, compared to pre-commute and post-
commute, only for HIGH [F(1,140) = 7.545; p = 0.006].
The group mean total occurrence of symptoms in-commute, compared to pre-
commute and post-commute, was significantly higher for offensive odour detection (p
≤ 0.001), dust or soot detection (p ≤ 0.001), eye irritation (p ≤ 0.001), nasal irritation
(p ≤ 0.001); throat irritation (p ≤ 0.001); phlegm production (p ≤ 0.001); and, chest
tightness (p = 0.003). Tussis and chest wheeze were significantly higher in occurrence
for the in-commute time period (p = 0.012 and p = 0.017), but not the post-commute
time period (p = 0.070 and p = 0.176), compared to the pre-commute time period.
92
Age was positively-correlated with the in-commute occurrence of throat irritation (r =
0.78, p = 0.049) and phlegm production (r = 0.83, p = 0.024). Further, female
participants, compared to males, reported significantly higher in-commute occurrence
of throat irritation (1.57 ± 0.88 versus 1.33 ± 0.68; p ≤ 0.001) and headache (1.14 ±
0.49 versus 1.06 ± 0.35; p = 0.005).
Peak Flow Rate
As an indication of airway diameter, peak flow rate was not significantly different
from pre-commute to one or three hours post-commute in HIGH or LOW. Further,
there was no significant difference between post-commute HIGH and LOW
measurements. Female, compared to male, group mean baseline peak flow rate was
significantly lower (447 ± 66 versus 584 ± 89 L·min; p ≤ 0.001); however, the
percentage change from pre-commute and post-commute measures were not
significantly different when females were compared to males (1.105 ± 0.189 versus
1.103 ± 0.197 %Δ; p = 0.394). Intra-individual PFR reproduction variability as a
group mean was 20.3 ± 11.3 L·min. See Table 2.
Sputum Cell Counts
Total and differential cell counts of valid participant sample sets (when compared
to previously established reference values of healthy adults [35] as stated in the
Methods, 22 out of 35, or 63%) were not significantly different between pre-commute
in either HIGH or LOW and post-commute in either HIGH or LOW (p > 0.07).
Further, cell counts were not significantly different between post-commute sampling
in HIGH and post-commute sampling in LOW (p = 0.09). There was no correlation
between the percent change of pre-commute to post-commute cell count group means
and in-commute PNC group means (r = 0.54, p = 0.08). See Table 3.
93
Discussion
Within the scope of this project, the results suggest that a purposely-designed bicycle
commute route alteration designed to reduce proximity to major motorised traffic corridors
can significantly reduce exposure to combustion UFP PNC without necessarily affecting
factors of route utility such as commute distance or duration. Subjective symptoms of nasal
and throat irritation occurrence were reduced in LOW compared to HIGH; however, an
objectively-measured physiological inflammatory response (represented by the reduction in
peak expiratory flow and an increase of inflammatory cells within sputum) was not seen in
these healthy individuals. Due to the group mean heart rate not being significantly different
between HIGH and LOW, it is suggested that an inhaled UFP count would be typically
determined by a variation in air quality (that is, PNC) rather than a difference in physical
effort (and thus ventilation rate) of alternative bicycle commute routes.
While off-road routes allow for reduced proximity to major motorised traffic corridors and
thus reduced air pollution exposure, these routes can be less direct and increase commute
duration; however, this was not the case in the current project, as commute distance and
duration were not significantly affected by a commute route alteration to reduce proximity to
major motorised traffic corridors. Published work by another research group of the same
region found that while local commuters recognised air pollution exposure as a health risk, it
did not deter them from bicycle commuting [29]. Regardless, the development of appropriate
infrastructure (separating traffic types) and educational schemes (indicating best air quality
routes) would be desirable to sustain the increased popularity of bicycle commuting, and to
assist an individual in managing their own air pollution exposure risk.
94
In-commute Particle Measurements
The strategy of lowering proximity to major motorised traffic corridors while bicycle
commuting to improve air quality has been supported as feasible, with mean and median
PNC significantly reduced in LOW compared to HIGH, in agreement with similar previous
research [6, 10, 15]. Further, this mean reduction was most marked for the inbound,
compared to outbound, commute in HIGH, which is also in agreement with previous urban
measurement studies and reflects the expectation that morning peak hour traffic is more time-
concentrated than the afternoon peak [36-38]. The median reduction of PNC being smaller in
magnitude than the mean reduction (however still significant) indicates the influence peak
PNC events (such as road crossing with traffic control lights) have on total commute PNC
exposure. There was no significant difference between LOW inbound and outbound trips,
indicating the influence of proximity to motorised traffic on PNC.
Previously, a mean PNC of 7.4 x e3 ppcc and a median PD of 40 nm (a diameter strongly-
associated with motor vehicle emissions) have been shown in Brisbane [39]. More recently,
PNC in Brisbane has been shown to have marginally increased to a mean of 10.0 x e3 ppcc,
and PD slightly decreased to a median of 38 nm. However, these PNC levels are relatively-
low compared with other studied cities worldwide [40, 66] and generally would not reflect in-
commute exposure [41, 42]. A meta-analysis performed with 71 UFP studies of different
environments showed typical mean PNC of 7.3 x e3 ppcc for urban background and 42.1 x e3
ppcc for roadside measurements, and indicated that greater proximity to major motorised
traffic corridors is positively-associated with PNC [43]. In the current project, the group
median particle diameter (PD) was not significantly different between HIGH and LOW or
between inbound and outbound trips of either HIGH or LOW, suggesting that the detection of
95
fresh petrol emissions from either more (HIGH) or less (LOW) motorised traffic was not
necessarily occurring.
A recent meta-analytical review of UFP exposure in-transit across different modes of
transport indicated that cyclists are generally exposed to the lowest PNC mean values of any
mode [44]. Studies specifically comparing bicycle commute routes with high and low
proximity to major motorised traffic corridors have indicated a PNC mean of 3.5 x e4 and 2.6
x e4 ppcc, respectively [44]. In comparison to other commute modes, motor vehicle
passengers can be exposed to PNC means of 1.3 times greater than that of cyclists [45, 46].
While interest in particulate matter (such as PNC) for in-commute air pollution exposure
studies is increasing, further investigation of the dynamic, heterogenous mixture (which was
beyond the scope of this study) is warranted; for example, PNC may be negatively-correlated
with, or be adsorbed with, other toxic components or reaction products of motorised traffic
emissions [67]. It has been shown that ozone and proximity to major motorised traffic
corridors are negatively-correlated and so conditions of low proximity to major motorised
traffic corridors may facilitate low UFP but high ozone exposure [68].
Heart Rate and Physical Effort
As bicycle commuting requires physical exercise, pulmonary ventilation rates of
participants can be an important factor when determining the inhaled dose of UFP and
therefore a toxic biological interaction. Ventilation rates of cyclists observed during
epidemiological studies have been approximately 2 to 4 times greater than motor vehicle
passengers, though this rate is believed to be conservative [46, 47, 69]. An experimental
study showed that particle deposition can be 4.5 times higher during moderate bicycling
exercise compared to rest in healthy individuals [48]. Further, it has been shown that inhaled
mean PNC dose can be halved by using a pre-determined route alteration of low, compared to
96
high, proximity to major motorised traffic corridors [49], and also suggested by the current
project according to mean PNC exposure concentrations. Importantly, the group mean
distance and duration were not significantly increased from the alteration of LOW from
HIGH, therefore not increasing overall exposure to motorised traffic emissions due to an
increased exposure time. Further, as commute distance or duration is not increased, the utility
of an altered bicycle commute route to reduce proximity to motorised traffic emissions has
been demonstrated as practical for individuals rating time as an important feature of a
commute route.
The correlation between heart rate and pulmonary ventilation rate during exercise is high and,
while it varies between individuals, a predictable association can be made for an individual
once a heart-rate ventilation association equation has been produced [58, 59]. The current
project did not include exercise testing to provide values for input to such an equation;
however, the intention was to make intra-individual comparison of heart rates between route
alterations and suggest if inhaled particle count was determined by PNC rather than
pulmonary ventilation level. As heart rates did not significantly differ between HIGH and
LOW, it could be inferred that any potential difference in inhaled particle count would be
attributable to differences in PNC rather than physical exercise, and therefore the ventilation
rate, required to use HIGH or LOW. A previous study in the same geographical region by the
current research group showed that estimated ventilation rates (via heart rate-ventilation
associated curves produced with exercise testing) did not significantly differ between popular
bicycle commute routes of low and high proximity to major motorised traffic corridors [49].
Meteorological Variation
In the current project, the measured meteorological variables were not significantly
different between HIGH and LOW; however, the diurnal variation of climate facilitated a
97
lower mean temperature and higher mean humidity for the inbound (morning), compared to
the outbound (afternoon), commute. The influence that these differences in temperature and
humidity have when comparing air quality measures of HIGH and LOW is negligible as
inbound and outbound trips were performed in equal measure for HIGH and LOW; however,
the significantly-higher PNC in HIGH inbound compared to HIGH outbound could be
attributable to nucleation inhibition which can occur in the circumstance of higher
temperature and lower humidity [60]. Previous study conditions using lower temperature, and
higher relative humidity, have been shown to facilitate secondary particle production and an
increase in PNC [61-63], such as the circumstances in the HIGH inbound (morning) commute
in the current project.
Physiological Inflammatory Response
Despite significantly higher PNC in HIGH compared to LOW, the occurrence of
acute respiratory symptoms (beyond offensive odour detection and nasal irritation) was not
increased in-commute or post-commute. While personal NOX exposure was not monitored, it
is likely that concentrations were substantially higher in HIGH compared to LOW due to the
strong association of NOX and PNC to motorised traffic emissions [51]. An increased
occurrence of offensive odour detection in HIGH compared to LOW, acute respiratory
symptoms associated with elevated NOX exposure concentrations (including nasopharyngeal
irritation, dyspnoea and tussis) seen in previous research [52] were not observed post-
commute or reduced in LOW compared to HIGH. Further, there was no significant change in
peak expiratory flow rate or neutrophil counts, either pre-commute to post-commute, or in
LOW compared to HIGH. Previously, healthy and asthmatic adults exposed to a mean PNC
of 1.45 x e5 ppcc during 2 hours of intermittent exercise did not exhibit significant differences
in sputum neutrophil counts immediately and four hours post-exposure [33]. Similarly,
98
healthy and asthmatic adults exposed to a mean PNC of 4.77 x e6 ppcc during rest and
exercise did not exhibit significant differences in respiratory symptoms or sputum neutrophil
counts, however there was a decrease in maximal mid-expiratory flow rate (not measured in
the current project) twenty-one hours post-exposure [4].
Sputum neutrophils, obtained from the lower airways, have been used previously as a
biomarker of airway inflammation but can have a low association with pulmonary function
and respiratory symptom reporting [20]. However, the utility of repeated sputum induction on
cell counts over a 24-hour period has been questioned [55]. Some research has shown no
significant changes in sputum cell differential counts of healthy individuals in response to
PNC exposure during rest and exercise (≤ 6.9 x e6 ppcc, 120 mins). However, in asthmatics
following a similar protocol, PNC was associated with a significant increase in alveolar
macrophage percentage of 11% compared to filtered air [4]. Similar to acute respiratory
symptoms and pulmonary function, a physiological inflammatory response was not indicated
by a significant change in neutrophil count in this study. Significant effects of UFP exposure
on symptoms, pulmonary function, and markers of airway or systemic inflammation are not
yet confirmed [56]. While the mechanisms of these effects for inhaled UFP are not yet
known, these particles have been shown to have significantly greater pulmonary
inflammatory effects compared to coarser particles at equal mass dose [7, 44, 57].
While this study did not indicate any acute health implications from the variables measured,
the PNC exposure levels surpass previous levels observed to increase systemic markers of
inflammation in healthy individuals exercising intermittently for a longer duration (1.1 x e4
ppcc, 120 mins) [53]. Further, exposure at higher levels and longer durations than the current
project has produced increases in lower airway inflammatory mediators [54] and systemic
markers of inflammation [34], oxidative DNA damage [42], and decreased airway diameter
99
[6]. In summary, it is suggested that mean PNC exposure levels or exposure durations in
HIGH and LOW were too low or too short, respectively, to significantly affect the acute
health-associated variables measured in the current project.
Limitations
The acute inflammatory tests used in this study were purposefully simple and
performed soon (one to three hours) after exposure, with this limited protocol being necessary
to not influence participant commute behaviour such as by the logistics of laboratory testing
or the over-lapping of inbound (morning) and outbound (afternoon) commute monitoring.
Also, while PNC exposure levels and commute duration were realistic, they may not have
been of a sufficient level or duration, respectively, to allow observable effects of the health-
associated variables measured. Further, this study was based on the available portable Philips
NanoTracer which measures PNC, a key component of motor vehicle exhaust, although not
the perfect marker.
The design of a unique questionnaire (such as that used for sign and symptom reporting in the
current project) without a precedent model available for reference will have a factor of
unknown validity and reliability. Additionally, as the questionnaire was self-administered,
respondent misunderstandings could have arisen due to question misinterpretation. Further,
questionnaire response bias may have resulted as participants could not be made blind to the
routes of HIGH or LOW; however, the limitation of respondent misunderstanding is believed
to have been minimised due to the review process of questionnaire design (see Methods).
Similar to symptom reporting, the performance of peak flow measurement and spontaneous
sputum sampling were reliant on participant competence. While verbal and written
instructions were provided to participants at induction, field performance was not supervised
and therefore cannot be validated.
100
The use of a new personal UFP monitor for field research was novel and therefore precedent
reference was not available. However, the majority of data collected was deemed valid and
the measurement accuracy was calibrated in controlled conditions against an accepted device
(see Methods). It is possible that the instrument may have not detected the smallest diameter
range (approaching 10 nm) of particles typically associated with traffic exhaust emissions and
thus explaining the low PNC means reported in the current project, relative to other studies of
on-road environments. The lower range cut-off of the device is 10 nm, while some particles
between 10 and 20 nm are also not detected.
Conclusions
Exposure to ultrafine particles, typically associated with combustion emissions from
motorised traffic, can be significantly reduced by lowering proximity to major motorised
traffic corridors without necessarily increasing commute distance or duration whilst bicycle
commuting. Governing authorities encouraging bicycle commuting participation should
educate participants in air quality and risk management, but also give proper consideration to
creating bicycle commuting routes, to reduce proximity to major motorised traffic corridors
and therefore minimise any health risk as a consequence of frequent in-commute exposure to
motorised traffic emissions.
List of abbreviations
bpm; beats per minute. CBD; central business district. PEF; peak expiratory flow. PM;
particulate matter. ppcc; particles per cubic centimetre. PNC; particle number concentration.
UFP; ultrafine particle.
101
Competing interests
The authors declare that they have no real or perceived competing interests.
Authors’ contributions
TCH managed participants, data collection, data analyses and manuscript preparation; LM
assisted with project supervision and manuscript revision; IS assisted with project supervision
and manuscript revision; RJ assisted with project supervision and manuscript revision; MH
assisted with laboratory procedures; CS assisted with project supervision and manuscript
revision .
Acknowledgements
We thank the participants for their enthusiasm an involvement in this project. Further, we
thank Camilla Tuttle for detailed guidance with laboratory procedures, and Dr Dimitrios
Vagenas for his patient help with statistical analyses.
Project Three References
1. Blair SN, Morris JN: Healthy hearts‐‐and the universal benefits of being physically
active: Physical activity and health. Ann Epidemiol 2009, 19:253‐256.
2. Shiroma EJ, Lee I: Physical activity and cardiovascular health: lessons learned from
epidemiological studies across age, gender, and race/ethnicity. Circulation 2010,
122:743.
102
3. Warburton DER, Nicol CW, Bredin SSD: Health benefits of physical activity: the
evidence. Can Med Assoc J 2006, 174:801.
4. Pietropaoli AP, Frampton MW, Hyde RW, Morrow PE, Oberdörster G, Cox C, Speers
DM, Frasier LM, Chalupa DC, Huang LS: Pulmonary function, diffusing capacity, and
inflammation in healthy and asthmatic subjects exposed to ultrafine particles.
Inhal Toxicol 2004, 16:59‐72.
5. Knibbs LD, Cole‐Hunter T, Morawska L: A review of commuter exposure to ultrafine
particles and its health effects. Atmospheric Environment 2011, 45:2611‐2622.
6. Strak M, Boogaard H, Meliefste K, Oldenwening M, Zuurbier M, Brunekreef B, Hoek
G: Respiratory health effects of ultrafine and fine particle exposure in cyclists.
Occup Environ Med 2010, 67:118.
7. Singh S, Shi T, Duffin R, Albrecht C, van Berlo D, Höhr D, Fubini B, Martra G, Fenoglio
I, Borm PJA, Schins, RPF: Endocytosis, oxidative stress and IL‐8 expression in human
lung epithelial cells upon treatment with fine and ultrafine TiO2: Role of the
specific surface area and of surface methylation of the particles. Toxicol Appl
Pharmacol 2007, 222:141‐151.
8. de Hartog J, Boogaard H, Nijland H, Hoek G: Do The Health Benefits Of Cycling
Outweigh The Risks? Environ Health Perspect 2010, 118:1109.
9. Rabl A, de Nazelle A: Benefits of shift from car to active transport. Transport Policy
2012, 19:121‐131.
10. Zuurbier M, Hoek G, Oldenwening M, Lenters V, Meliefste K, Van Den Hazel P,
Brunekreef B: Commuters’ Exposure to Particulate Matter Air Pollution Is Affected
by Mode of Transport, Fuel Type, and Route. Environ Health Perspect 2010,
118:783.
103
11. Rojas‐Rueda D, de Nazelle A, Tainio M, Nieuwenhuijsen MJ: The health risks and
benefits of cycling in urban environments compared with car use: health impact
assessment study. BMJ: British Medical Journal 2011, 343.
12. Giles LV, Barn P, Künzli N, Romieu I, Mittleman MA, van Eeden S, Allen R, Carlsten C,
Stieb D, Noonan C: From good intentions to proven interventions: effectiveness of
actions to reduce the health impacts of air pollution. Environ Health Perspect 2011,
119:29.
13. Hertel O, Hvidberg M, Ketzel M, Storm L, Stausgaard L: A proper choice of route
significantly reduces air pollution exposure‐‐A study on bicycle and bus trips in
urban streets. Sci Total Environ 2008, 389:58‐70.
14. 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 2005, 39:3629‐3641.
15. Pattinson W: Cyclist exposure to traffic pollution: microscale variance, the impact
of route choice and comparisons to other modal choices in two New Zealand cities.
2009.
16. Thai A, McKendry I, Brauer M: Particulate matter exposure along designated bicycle
routes in Vancouver, British Columbia. Sci Total Environ 2008, 405:26‐35.
17. Wåhlin P, Palmgren F, Van Dingenen R: Experimental studies of ultrafine particles in
streets and the relationship to traffic. Atmos Environ 2001, 35:63‐69.
18. Nightingale JA, Maggs R, Cullinan P, Donnelly LE, Rogers DF, Kinnersley R, Fan Chung
K, Barnes PJ, Ashmore M, Newman‐Taylor A: Airway Inflammation after Controlled
Exposure to Diesel Exhaust Particulates. Am J Respir Crit Care Med 2000, 162:161‐
166.
104
19. Nordenhäll C, Pourazar J, Blomberg A, Levin JO, Sandström T, Ädelroth E: Airway
inflammation following exposure to diesel exhaust: a study of time kinetics using
induced sputum. Eur Respir J 2000, 15:1046‐1051.
20. Singh D, Edwards L, Tal‐Singer R, Rennard S: Research Sputum neutrophils as a
biomarker in COPD: findings from the ECLIPSE study. 2010.
21. Organisation WH: Global Database on Body Mass Index. 2012.
23. Climate Data Online. In.: Australian Bureau of Meteorology; 2011.
24. Regional Population Growth, Australia, 2009‐10
25. Roads DoTaM: Vehicles on Register in Queensland's South East Region. In. Edited by
Roads DoTaM: Queensland Government; 2011.
26. Marra J, Den Brink W, Goossens H, Kessels S: Nanoparticle monitoring for exposure
assessment. Nanotechnology Magazine, IEEE 2009, 3:6‐37.
27. Puustinen A, Hämeri K, Pekkanen J, Kulmala M, de Hartog J, Meliefste K, ten Brink H,
Kos G, Katsouyanni K, Karakatsani A: Spatial variation of particle number and mass
over four European cities. Atmos Environ 2007, 41:6622‐6636.
28. Mermier CM, Samet JM, Lambert WE, Chick TW: Evaluation of the Relationship
between Heart Rate and Ventilation for Epidemiologic Studies. Archives of
Environmental Health: An International Journal 1993, 48:263‐269.
29. Badland HM, Duncan MJ: Perceptions of air pollution during the work‐related
commute by adults in Queensland, Australia. Atmos Environ 2009, 43:5791‐5795.
30. Kjærgaard T, Cvancarova M, Steinsvåg SK: Cigarette smoking and self‐assessed
upper airway health. Eur Arch Otorhinolaryngol 2011, 268:219‐226.
31. ATS: What constitutes an adverse health effect of air pollution? Am J Respir Crit
Care Med 2000, 161:665‐673.
105
32. Stanek LW, Brown JS, Stanek J, Gift J, Costa DL: Air Pollution Toxicology—A Brief
Review of the Role of the Science in Shaping the Current Understanding of Air
Pollution Health Risks. Toxicol Sci 2011, 120:S8.
33. Gong Jr H, Linn WS, Clark KW, Anderson KR, Sioutas C, Alexis NE, Cascio WE, Devlin
RB: Exposures of healthy and asthmatic volunteers to concentrated ambient
ultrafine particles in Los Angeles. Inhal Toxicol 2008, 20:533‐545.
34. Jacobs L, Nawrot T, de Geus B, Meeusen R, Degraeuwe B, Bernard A, Sughis M,
Nemery B, Panis L: Subclinical responses in healthy cyclists briefly exposed to
traffic‐related air pollution: an intervention study. Environmental Health 2010,
9:64.
35. Belda J, Leigh R, Parameswaran K, O'BYRNE PM, Sears MR, Hargreave FE: Induced
sputum cell counts in healthy adults. Am J Respir Crit Care Med 2000, 161:475.
36. Molnár P, Janhäll S, Hallquist M: Roadside measurements of fine and ultrafine
particles at a major road north of Gothenburg. Atmos Environ 2002, 36:4115‐4123.
37. Weichenthal S, Dufresne A, Infante‐Rivard C, Joseph L: Determinants of ultrafine
particle exposures in transportation environments: findings of an 8‐month survey
conducted in Montreal, Canada. Journal of Exposure Science and Environmental
Epidemiology 2008, 18:551‐563.
38. Karellas N, Whitelaw C, Debosz J, Stacey N: Ultrafine Particle Measurements in Real
Time at an Urban Site in Ontario. 2011.
39. Morawska L, Thomas S, Bofinger N, Wainwright D, Neale D: Comprehensive
characterization of aerosols in a subtropical urban atmosphere: particle size
distribution and correlation with gaseous pollutants. Atmos Environ 1998, 32:2467‐
2478.
106
40. Mejıa J, Morawska L, Mengersen K: Spatial variation in particle number size
distributions in a large metropolitan area. Atmos Chem Phys 2008, 8:1127‐1138.
41. Kaur S, Nieuwenhuijsen M, Colvile R: Fine particulate matter and carbon monoxide
exposure concentrations in urban street transport microenvironments. Atmos
Environ 2007, 41:4781‐4810.
42. Vinzents P, Møller P, Sørensen M, Knudsen L, Hertel O, Jensen F, Schibye B, Loft S:
Personal exposure to ultrafine particles and oxidative DNA damage. Environ Health
Perspect 2005, 113:1485.
43. Morawska L, Ristovski Z, Jayaratne E, Keogh DU, Ling X: Ambient nano and ultrafine
particles from motor vehicle emissions: Characteristics, ambient processing and
implications on human exposure. Atmos Environ 2008, 42:8113‐8138.
44. Knibbs LD, Cole‐Hunter T, Morawska L: A review of commuter exposure to ultrafine
particles and its health effects. Atmos Environ 2011, 45:2611‐2622.
45. 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
2009, 43:4234‐4242.
46. Int Panis L, de Geus B, Vandenbulcke G, Willems H, Degraeuwe B, Bleux N, Mishra V,
Thomas I, Meeusen R: Exposure to particulate matter in traffic: A comparison of
cyclists and car passengers. Atmos Environ 2010, 44:2263‐2270.
47. van Wijnen J, Verhoeff A, Jans H, van Bruggen M: Exposure of Cyclists, Car Drivers
and Pedestrians to Motor Vehicle Exhaust Emissions. Epidemiology 1995, 6:S25.
48. Daigle CC, Chalupa DC, Gibb FR, Morrow PE, Oberdörster Gn, Utell MJ, Frampton
MW: Ultrafine Particle Deposition in Humans During Rest and Exercise. Inhal
Toxicol 2003, 15:539‐552.
107
49. Cole‐Hunter T, Morawska, L., Stewart, I., Jayaratne, J., Solomon, C.: Inhaled particle
counts on bicycle commute routes of low and high proximity to
motorised traffic. Atmos Environ 2012, 61:197‐203
50. Lercher P, Schmitzberger R, Kofler W: Perceived traffic air pollution, associated
behavior and health in an alpine area. Sci Total Environ 1995, 169:71‐74.
51. Ayoko G, Jamriska M, Jayaratne R, Morawska L: Air pollution levels measured at
traffic hot spots: Brisbane urban corridor study. In: 2005: CASANZ (Clean Air Society
of Australia & New Zealand); 2005: 1‐7.
52. Lee K, Yanagisawa Y, Spengler JD, Nakai S: Carbon monoxide and nitrogen dioxide
exposures in indoor ice skating rinks. J Sports Sci 1994, 12:279‐279.
53. Nyström AK, Svartengren M, Grunewald J, Pousette C, Rödin I, Lundin A, Sköld CM,
Eklund A, Larsson BM: Health effects of a subway environment in healthy
volunteers. Eur Respir J 2010, 36:240‐248.
54. Larsson B, Sehlstedt M, Grunewald J, Sköld C, Lundin A, Blomberg A, Sandström T,
Eklund A, Svartengren M: Road tunnel air pollution induces bronchoalveolar
inflammation in healthy subjects. Eur Respir J 2007, 29:699.
55. Nightingale JA, Rogers DF, Barnes PJ: Effect of repeated sputum induction on cell
counts in normal volunteers. Thorax 1998, 53:87.
56. Frampton MW, Utell MJ, Zareba W, Oberdörster G, Cox C, Huang LS, Morrow PE, Lee
FE, Chalupa D, Frasier LM: Effects of exposure to ultrafine carbon particles in
healthy subjects and subjects with asthma. Research report (Health Effects Institute)
2004:1.
108
57. Oberdörster G, Oberdörster E, Oberdörster J: Nanotoxicology: an emerging
discipline evolving from studies of ultrafine particles. Environ Health Perspect 2005,
113:823.
58. Vai F, Bonnet J, Ritter P, Pioger G: Relationship between heart rate and minute
ventilation, tidal volume and respiratory rate during brief and low level exercise.
Pacing Clin Electrophysiol 1988, 11:1860‐1865.
59. Treese N, MacCarter D, Akbulut O, Coutinho M, Baez M, Liebrich A, Meyer J:
Ventilation and heart rate response during exercise in normals: Relevance for rate
variable pacing. Pacing Clin Electrophysiol 1993, 16:1693‐1700.
60. Kulmala M, Vehkamäki H, Petäjä T, Dal Maso M, Lauri A, Kerminen VM, Birmili W,
McMurry P: Formation and growth rates of ultrafine atmospheric particles: a
review of observations. Journal of Aerosol Science 2004, 35:143‐176.
61. Holmes NS: A review of particle formation events and growth in the atmosphere in
the various environments and discussion of mechanistic implications. Atmos
Environ 2007, 41:2183‐2201.
62. Jeong CH, Hopke PK, Chalupa D, Utell M: Characteristics of nucleation and growth
events of ultrafine particles measured in Rochester, NY. Environmental Science &
Technology 2004, 38:1933‐1940.
63. Kaur S, Nieuwenhuijsen M: Determinants of personal exposure to PM2. 5, ultrafine
particle counts, and CO in a transport microenvironment. Environmental Science &
Technology 2009, 43:4737‐4743.
Jamriska, M, Morawska, L, Mergersen, K: The effect of temperature and humidity
on size segregated traffic exhaust particle emissions. Atmos Environ 2008, 42: 2369‐
2382.
109
64. Rudell, B., Ledin, M., Hammarström, U., Stjernberg, N., Lundbäck, B., Sandström, T:
Effects on symptoms and lung function in humans experimentally exposed to
diesel exhaust. Occup Environ Med 1996, 53: 658.
65. Marra J, Voetz M, Kiesling HJ: Monitor for detecting and assessing exposure to
airborne nanoparticles. Journal of Nanoparticle Research 2010, 12:21‐37.
66. Zuurbier M, Hoek G, Oldenwening M, Lenters V, Meliefste K, Van Den Hazel P,
Brunekreef B: Commuters’ exposure to particulate matter air pollution is
affected by mode of transport, fuel type, and route. Environ Health Perspect 2010,
118:783.
67. Kelly FJ, Fussell JC: Size, source and chemical composition as determinants of
toxicity attributable to ambient particulate matter. Atmos Environ 2012, 60:504-
526.
68. McConnell R, Berhane K, Yao L, Lurmann FW, Avol E, Peters JM: Predicting
residential ozone deficits from nearby traffic. Sci Total Environ 2006, 363:166-
174.
69. Zuurbier M, Hoek G, Hazel P, Brunekreef B: Minute ventilation of cyclists, car and
bus passengers: an experimental study. Environmental Health 2009, 8:48.
70. Department of Science, Information Technology, Innovation and the Arts.
Queensland air monitoring report 2011. Available:
http://www.ehp.qld.gov.au/air/pdf/reports/2011-air-monitoring-report.pdf
110
Tables
Table 1. Commute variables for routes of high (HIGH) and low (LOW) proximity to major
motorised traffic corridors both Inbound and Outbound
Condition HIGH LOW
Inbound Outbound Inbound Outbound
Time of Day (24:00) 8:20 ± 0:22 16:39 ± 0:23 8:04 ± 0:22 16:33 ± 0:23
Distance (km) 12.3 ± 6.9 11.7 ± 6.9 12.9 ± 7.2 12.6 ± 7.0
Duration (min) 42 ± 18 41 ± 15 45 ± 17 43 ± 16
Speed (km·hr-1) 17.3 ± 4.3 16.7 ± 4.8 17.1 ± 4.6 17.1 ± 4.7
Heart Rate (bpm) 137 ± 11 135 ± 11 134 ± 9 131 ± 9
Temperature (ºC) 17.9 ± 3.5 21.1 ± 3.0## 17.7 ± 3.5 21.5 ± 3.2##
Humidity (%) 61 ± 14 48 ± 19## 62 ± 13 49 ± 19##
Wind Speed (km·hr-1) 5.8 ± 3.2 9.5 ± 4.8 7.1 ± 3.2 8.5 ± 4.4
Air Pressure (hPa) 1019 ± 6 1016 ± 5 1019 ± 6 1016 ± 5
Precipitation (mL·day) 0.33 ± 0.01 0.25 ± 0.01 0.31 ± 0.01 0.26 ± 0.02
PNC Mean ( x e4; ppcc) 3.30 ± 1.57 2.60 ± 1.35## 1.99 ± 1.02** 1.84 ± 0.84**
PNC Median ( x e4; ppcc) 2.20 ± 1.02 1.77 ± 1.08 1.34 ± 0.79** 1.38 ± 0.67**
PNC Max ( x e4; ppcc) 10.82 ± 2.14 10.57 ± 2.06 5.70 ± 1.44 4.20 ± 1.26
PNC Min ( x e4; ppcc) 1.11 ± 0.98 0.80 ± 0.77 0.71 ± 0.84 0.73 ± 0.95
PD Median (nm) 47 ± 8 49 ± 10 50 ± 11 52 ± 11
PD Min (nm) 30 ± 6 35 ± 7 31 ± 4 34 ± 5
PD Max (nm) 67 ± 5 69 ± 5 79 ± 4 83 ± 7
Values are Means (or Median, and Ranges, where indicated) ± Standard Deviation. Significance [from
multivariate repeated measure ANOVA]: *p < 0.05, ** p < 0.01 compared to HIGH; #p < 0.05, ##p < 0.01
compared to Inbound. PNC (ppcc) = particle number concentration (particles per cubic centimetre).
111
Table 2. Time point comparison of health-associated variables for routes of high (HIGH) and
low (LOW) proximity to major motorised traffic corridors
Condition HIGH LOW
Inbound Outbound Inbound Outbound
Timepoint Pre In Post Pre In Post Pre In Post Pre In Post
Offensive Odour
1.09 ±
0.12
2.71*,## ± 0.73
1.21 ±
0.15
1.12 ±
0.13
2.88*,## ± 0.83
1.06 ±
0.11
1.18 ±
0.13
2.06## ± 0.42
1.09 ±
0.15
1.18 ±
0.14
2.18## ± 0.48
1.09 ±
0.11
Dust, Soot 1.06
± 0.11
2.35*,## ± 0.55
1.06 ±
0.11
1.06 ±
0.11
2.21*,## ± 0.49
1.03 ±
0.12
1.18 ±
0.14
1.65*,## ± 0.27
1.06 ±
0.11
1.15 ±
0.13
1.65*,## ± 0.27
1.09 ±
0.21
Eye Irritation
1.06 ±
0.11
1.56 ± 0.24
1.06 ±
0.11
1.18 ±
0.14
1.65 ± 0.27
1.06 ±
0.10
1.18 ±
0.10
1.26 ± 0.16
1.09 ±
0.12
1.18 ±
0.14
1.35 ± 0.18
1.06 ±
0.12
Nose Irritation
1.38 ±
0.19
1.82** ± 0.33
1.24 ±
0.15
1.24 ±
0.16
1.74 ± 0.30
1.12 ±
0.13
1.24 ±
0.15
1.53 ± 0.23
1.12 ±
0.13
1.09 ±
0.12
1.38 ± 0.19
1.12 ±
0.13
Throat Irritation
1.56 ±
0.24
2.00** ± 0.40
1.41 ±
0.19
1.35 ±
0.18
2.09 ± 0.44
1.26 ±
0.16
1.38 ±
0.19
1.56 ± 0.24
1.26 ±
0.16
1.24 ±
0.15
1.56 ± 0.25
1.35 ±
0.18
PFR (%∆) 0.00
± 0.00
1.28 ± 0.16
1.18 ±
0.14
0.00 ±
0.00
1.65 ± 0.27
1.63 ±
0.27
0.00 ±
0.00
1.76 ± 0.31
1.41 ±
0.20
0.00 ±
0.00
2.38 ± 0.57
2.14 ±
0.46
Values are Group Mean ± Standard Deviation. Significance [from Linear Mixed Models]: * p < 0.05, ** p <
0.01 compared to LOW; # p < 0.05, ## p < 0.01 compared to pre-commute / ‘Pre’. Values presented are the self-
reported group means, on a frequency scale from 1 (very low) to 5 (very high). Timepoint is the period relative
to bicycle commute trip performance: immediately pre-commute = Pre; in-commute = In; three hours post-
commute = Post. Peak flow rate (PFR) is expressed as the percentage change from pre-commute / ‘Pre’ values.
112
Table 3. Group Means of Total and Differential Cell Counts: Time Point Comparison for
Routes of High (HIGH) and Low (LOW) Proximity to major motorised traffic corridors
Condition
HIGH LOW
Inbound Outbound Inbound Outbound
Timepoint Pre Post Pre Post Pre Post Pre Post
Leukocyte (x e6 cells·g-1) 1.36 ± 0.42
1.38 ± 0.43
1.23 ± 0.38
1.28 ± 0.39
1.40 ± 0.43
1.37 ± 0.42
1.44 ± 0.45
1.44 ± 0.45
Epithelial (x e6 cells·g-1) 1.16 ± 0.30
1.19 ± 0.31
1.05 ± 0.27
1.10 ± 0.28
1.20 ± 0.31
1.17 ± 0.30
1.23 ± 0.32
1.23 ± 0.30
Columnar (x e6 cells·g-1) 0.58 ± 0.18
0.59 ± 0.18
0.53 ± 0.16
0.55 ± 0.17
0.60 ± 0.19
0.59 ± 0.15
0.62 ± 0.22
0.62 ± 0.21
Squamous (x e6 cells·g-1) 0.78 ± 0.30
0.79 ± 0.31
0.70 ± 0.27
0.73 ± 0.28
0.80 ± 0.31
0.78 ± 0.20
0.82 ± 0.29
0.82 ± 0.30
Macrophage (%) 59 ± 18
58 ± 18
59 ± 18
59 ± 18
59 ± 18
59 ± 20
58 ± 17
59 ± 18
Lymphocyte (%) 1.4 ± 0.4
1.5 ± 0.5
0.8 ± 0.3
0.9 ± 0.3
0.8 ± 0.2
1.3 ± 0.5
1.0 ± 0.3
1.0 ± 0.4
Neutrophil (%) 39 ± 12
40 ± 12
40 ± 12
40 ± 12
40 ± 12
39 ± 10
41 ± 11
40 ± 11
Eosinophil (%) 0.6 ± 0.2
0.5 ± 0.1
0.2 ± 0.1
0.1 ± 0.1
0.2 ± 0.1
0.7 ± 0.2
0.1 ± 0.1
0.1 ± 0.1
Values are Group Mean ± Standard Deviation. Significance [from Linear Mixed Models]: * p < 0.05, ** p <
0.01 compared to LOW, # p < 0.05, ## p < 0.01 of three hours post-commute (Post) compared to immediately
pre-commute (Pre). Total cell count values (of leukocyte – squamous) are presented as number of cells per gram
of spontaneous sputum sample. Differential cell count values (of macrophage – eosinophil) are given as
percentage of the total leukocyte cell count. Timepoint is relative to bicycle commute trip performance: Pre =
sampled immediately pre-commute; Post = sampled three hours post-commute.
113
7. GENERAL DISCUSSION
7.1. Introduction and Summary
The proposed risk management strategy of bicycle commute re-routing to lower proximity to
motorised traffic and thus reduce exposure to elevated roadside air pollution emissions for
participants has been shown effective and feasible by this research project, although most
appropriately for susceptible individuals. The results of Project 1 and 3 suggest that air
pollution exposure levels may be adequately perceived by healthy individuals, although not
reflected by the incidence of the assessed acute physiological inflammatory response during
or after typical bicycle commuting in healthy participants. However, as identified with
Project 1 (concurrent with previous literature), the incidence of such symptoms were
increased with personal characteristics of female gender and respiratory disorder history
[128]. The findings of Project 2 and 3 suggest that using an informed bicycle commute route
alteration to lower proximity to motorised traffic (determined by commute duration,
frequency and proximity to motorised traffic) will facilitate a significant reduction in particle
number concentration (PNC; dominated by particles in the ultrafine diameter range, UFP),
which is concurrent with previous literature [128]. Further, the potential UFP inhaled particle
count (and thus dose) from proximity to motorised traffic could be attributable to variation in
air quality rather than variation in physical effort (and thus ventilation rate) of an alternative
route.
However, as identified by Project 3, a physiological inflammatory response (represented by
the incidence and severity of acute respiratory symptoms, the change in lung function, and
the presence of inflammatory mediators in sputum) is not associated with current recorded
PNC levels in healthy individuals – while this finding is concurrent with some previous
literature [128], it is not in agreement with other studies using higher exposure levels [128],
longer exposure periods [60, 88], or more sensitive detection methods [51, 60]. Regardless,
Project 1 and 3 have highlighted that the majority of participants are amenable to adopting
risk management strategies, particularly commute re-routing, as deemed potentially effective
by Project 2 and 3. Commute re-routing as a risk management strategy could be
complemented with improved bicycle commuting infrastructure, including more
114
considerately planned off-road paths, and community education schemes, in addition to real-
time roadside air quality or motorised traffic congestion broadcasts.
7.2. Air Quality
The quality of air adjacent to major traffic corridors, attributable to automotive emissions, has
previously been indicated as poorer than regional background levels [129]. However, road-
side air quality in the current study is favourable when compared with that of other cities
studied previously by different research groups [128], possibly attributable to the current
regions relatively small (although quickly growing) motor vehicle fleet. As identified by
Project 1, previously measured ambient and roadside levels of nitrogen dioxide (NO2; which
is a pollutant strongly-associated with motor vehicle emissions) was not high enough to elicit
the specific symptoms inquired by the supplied questionnaire; however, particulate matter
was. As discussed by Project 1, the greatest concern when considering particulate matter
(PM) is PNC, which is strongly-correlated to proximity to motorised traffic [59].
As PNC, dominated by particles in the ultrafine range (UFP; < 0.1 μm diameter), is not
regionally-monitored, Project 2 and 3 aimed to both profile the exposure and inhaled particle
count of UFP and any consequential physiological inflammatory response between popular
bicycle commute routes of high and low motorised traffic proximity. In general, an
association of measured air quality and the incidence and severity of specific acute
respiratory symptoms was not shown, despite a significant reduction of PNC exposure levels
from lowering proximity to motorised traffic practically (that is, conserving the values of
commute distance and duration for viability as a risk management strategy, being applicable
to commuters desiring direct or short-duration routes). An improvement of health endpoints
was not exhibited in Project 3, although importantly, only healthy individuals were
monitored; susceptible individuals such as those with a history of respiratory disorder may
present differently under the same circumstances.
115
7.3. Air Pollution Exposure Symptoms and Susceptibility
In general, healthy individuals participating in frequent bicycle commuting activity are not
indicated to be acutely affected from in-commute exposure to particulate pollution. However,
results from susceptible individuals participating in Project 1 suggest they possess higher
sensitivity to exposure effects. Participants with a history of cardiopulmonary disorder or of
female gender indicated a higher incidence of mild and more severe respiratory symptoms
from in-commute exposure, which is in agreement with past research [32, 35] and other
questionnaire-based studies [128].
In the current study, females (compared to males) were more likely to report the incidence of
moderate air pollution exposure during their typical bicycle commute, as well as chest
wheeze in-commute with nasopharyngeal irritation in- and post-commute. Individuals with a
history of respiratory disorder (compared to healthy individuals) reported a higher incidence
of acute respiratory symptoms during and a few hours after their typical bicycle commute
(along with chronic upper respiratory tract infection), but also had a compromised perception
of exposure compared to estimated levels. A history of respiratory disease increased (and a
history of smoking decreased) the incidence of exposure perception to moderate or higher
levels of in-commute air pollution, compared to healthy respondents, which has been
observed elsewhere [130].
7.4. Air Pollution Exposure Perceptions and Risk Management
The perception of in-commute air quality is suggested to be reasonably-accurate compared to
estimated and measured air pollution exposure levels. In effect, susceptible individuals
identified from current and previous research, and other individuals stressed by their frequent
exposure to elevated air pollution levels, could sensibly manage their exposure risk. Two risk
management strategies were addressed in Project 1, including commute re-routing and
respirator use. The majority of participants perceived moderate in-commute air pollution
exposure, however less than half of the participants used commute routes of low proximity to
motorised traffic and nil used a respirator. Due to resource constraints, Project 2 and 3
focused on the strategy of commute re-routing to minimise proximity to motorised traffic.
116
In Project 1, the willingness of a participant to use a respirator or re-route was addressed to
determine what factors would have to be present in a risk management strategy for it to be
adopted, if further research indicates frequent in-commute air pollution exposure to be a
significant health risk. The most important factors to be addressed for respirator use by
participants, being “breathing impedance” and “wear comfort”, were both functional and not
financial or psychological. The respirators recommended for adoption by urban bicycle
commuters, if deemed appropriate and effective, are designed to accommodate moderate
physical activity with neoprene material and two ventilation valves.
When rating the importance of factors for choosing a commute route in Project 1, “time (e.g.
more convenient, quickest route)” was ranked as more important than the factor of “health
(e.g. to avoid air pollution)” by participants. For Project 3, when participants were asked if
they preferred their original or LOW route alteration and the reasons for this, “time (e.g. more
convenient, quickest route)” and “safety (e.g. greater riding space, visibility)” were of the
highest importance. Infrastructure which allows direct (that is, time efficient) yet safe (that is,
adequate riding space and visibility of other traffic) commuting. The factors of time and
safety can be antagonistic, however they are both necessary for bicycle commuters and public
health advocates to accommodate greater participation rates and lower proximity to
motorised traffic and associated air pollution emissions.
7.5. Novel Method Use
This project was possible because of the use of two novel instruments. The first, which was
used in both Project 1 and 3, is a questionnaire inquiring of perceptions, symptoms and
preference of risk management strategies associated with in-commute air pollution exposure.
While novel in application, the design of the questionnaire was guided by recommendations
[106] and previous research [107]. Further, the question format used was rigorously reviewed
by respiratory scientists, epidemiological statisticians and a sample of the intended
respondent cohort before provision to research participants.
Secondly, to produce an exposure profile of popular bicycle commute routes (Project 2) and
to associate questionnaire responses with actual in-commute air quality (Project 3), a
compact, light-weight and portable device (Philips Nanotracer) capable of measuring
particles of very high concentrations and at very low diameters [131] was used. To the
117
authors’ knowledge, the use of this instrument in field research such as the current study is
novel; however, the authors’ hope that the current study will serve as an example of the
practicality for such an instrument in future field research.
7.6. General Limitations
The use of novel methods will present certain limitations. For example, the design of a
unique questionnaire (used in Project 1 and 3) without a complete precedent model available
for reference has a factor of unknown reliability. Additionally, as the questionnaires were
self-administered, respondent misunderstandings could have arisen due to question
misinterpretation. However, due to the review process of questionnaire design (noted in
General Methods), these limitations are believed to be minimised.
In Project 2, a single participant model was used to estimate inhaled particle count.
Therefore, broader application of findings to the general public is limited; however, commute
behaviour such as route, distance, speed, and departure times were informed by the real-
world results of Project 1. Additionally, Project 2 monitored only morning peak commute
times; however, this represented what is expected to be more consistent commute
performance times and capitalised on more reliable meteorological conditions.
The performance of spontaneous sputum sampling and peak flow metering in Project 3 were
reliant on participant competence. While training was given to participants, performance was
not supervised and therefore is hard to validate. Total cell counts of sputum were performed
to indicate validity of spontaneous sampling. Similarly, the standard deviations of peak flow
rate performance were used to indicate reproducibility of tests.
A lesser limitation was with the use of a novel particle measurement device, and its’
application to field research. Precedent reference of reliability in field research had not yet
been established; however, the majority of data collected in this study was deemed valid, with
measurement accuracy calibrated in controlled conditions against a benchmark device. On the
contrary, tilt errors experienced with fluid-reliant instruments (being the previous technology
of condensation particle counters) during vigorous use were not an issue with this novel
device.
118
8. CONCLUSIONS
This project has shown that bicycle commute route alterations of low proximity to motorised
traffic facilitate improves air quality (represented by particle number concentrations, typically
emitted from petrol-powered motorised traffic) compared to routes high proximity to
motorised traffic. The consistent perception of estimated in-commute air pollution exposure
levels and the willingness to reduce proximity to motorised traffic during frequent, inner-city
commuting has been indicated in this project; therefore, a self-managed commute re-routing
strategy could be effective at reducing acute respiratory symptoms in more susceptible groups
such as females and respiratory health-compromised individuals. It is recommended that
frequent bicycle commuters with physiological-susceptibility to air pollution exposure, more
prone to health detriment, consider reducing their proximity to motorised traffic for their
frequent bicycle commute. While healthy individuals were observed to be frequently exposed
to elevated in-commute air pollution concentrations, significant incidence or severity of acute
respiratory symptoms has not been indicated; however, recurrent assessment of this may be
necessary during contemporary regional urban growth and increasing popularity of active
transport such as bicycle commuting. Appropriate governmental bodies wanting to increase
bicycle commuting participation rates should both educate susceptible participants about air
quality and risk management and apply similar knowledge when creating bicycle commuting
infrastructure.
The body of work contained within this thesis contributes multiple findings to current
knowledge. It has been made apparent that individuals who bicycle commute frequently to
and from the city centre can reasonably assess their own exposure to air pollution and are
willing to alter their commute route if deemed appropriate. Altering a commute route to
lower proximity to motorised traffic has been shown to be effective at significantly reducing
PNC exposure, as a function of significantly improving air quality rather than an increase in
pulmonary ventilation rate (and therefore potential inhaled particle count or dose), without
significantly increasing commute distance or duration. If possible, authorities involved with
planning new bicycle infrastructure (as being done in many places across the world) are
urged to consider the proximity of bicycle paths to major motorised traffic corridors. Further,
appropriate education to assist self-managed risk strategies against elevated air pollution
exposure, such as what constitutes a sign or symptom of elevated air pollution exposure,
119
would be appropriate as indicated by this thesis. Particularly, individuals with susceptibilities
to air pollution exposure (such as asthmatics) and/or individuals who wish to continue (or do
not have an alternative to) using paths proximal to major motorised traffic corridors could be
assisted by such education in the alleviation of any side-effects of exposure associated with
frequent bicycle commuting.
120
9. FUTURE DIRECTIONS
In this project, healthy individuals have not exhibited acute health detriment from frequent,
elevated in-commute PNC (such as with bicycle commuting alongside motorised traffic), and
therefore an alleviation of such detriment by lowering proximity to a primary PNC emission
source (such as motorised traffic) has not been found as appropriate for a healthy population.
Therefore, future research should include individuals representing pre-disposed and
physiologically-susceptible populations, such as asthmatics [35, 36] who are especially
prevalent in the region of SE QLD [132]. Such research direction is further supported by the
fact that Project 1 included a sub-population of 36 asthmatic individuals (as 23% of all
participants), suggesting that asthmatics are a significant component of regional frequent
bicycle commuters and deserve the appropriate attention. Further still, Project 1 showed that
asthmatics may experience a higher incidence of acute respiratory symptoms compared to
healthy participants.
The efficacy of respirators (for which amenability of use, found to be high, with certain
features was evaluated by participants of Project 1) has not been investigated with PNC. A
commercially-available unit is currently used by fire services personnel; while the unit is
accommodating of increased perspiration and ventilation rates, it has only been shown to be
effective with larger particles and certain vapour gases [133]. In-commute use of a respirator
was slightly more popular than commute re-routing as an air pollution exposure risk
management strategy, with 75% compared to 68% of participants willing to adopt either
strategy, respectively. Further, 21% of participants had already considered using a respirator
for bicycle commuting. The use of a respirator might be a more viable option to individuals
who identify time as important when choosing a commute route, or who are commuting from
directions that have a low proportion of designated off-road paths (i.e. low proximity to
motorised traffic) available for use.
A line of investigation not taken in this thesis, although worthy of future research, is
extrapolation of the real-time PNC, heart rate and geographical location for the third project
(as considered with the second project) plotted using geospatial visualisation (in software
such as ArcGIS). The benefit of such analysis would be the provision of additional
information concerning the levels and timing of high and low in-commute motorised traffic
emission exposures, and therefore the highlighting of any exposure ‘hot-spots’ that may
121
warrant attention by the exposed individual (e.g. to avoid by re-routing) or by the individuals
responsible for infrastructure development (e.g. to mitigate by constructing isolation
barriers). However, such plotting of individual data sets would prove challenging – instead, a
network of locations commonly encountered by multiple participants may give a broader
overview of regional hot-spots. The second project of this thesis could succeed in such
plotting as the routes were defined with repeated measurements by the one participant,
without significant deviation/alteration of route.
In summary, the exploration of other air pollution risk management strategies to support the
desire of individuals, and especially the needs of susceptible populations, participating in
bicycle commuting (as a method to improve their own health and that of the public and the
environment) should be supported with future research into alternative, appropriate and
effective risk management strategies - such an example would be the encouragement of
respirator use by asthmatics. Additionally, more in-depth analysis and visualisation of real-
time data of real-world bicycle commute routes (collected in the third project) would help to
identify hot-spots of poor air quality which could be useful information for both route users
and developers.
122
10. APPENDICES
A. Questionnaire [Complete (Project 1)]
123
124
125
126
127
128
129
130
131
(Example of Bikeway Maps Appended to Questionnaire. Bikeway Map #5 of a total 12)
132
B. Questionnaire [Amended (Project 3)]
133
134
135
C. Bikeway Maps with Cyclist Counts (Project 2)
Popular Bicycle Commute Routes of Brisbane, Australia [134]
136
D. Media Releases (to assist participant recruitment for Projects 1 and 3)
137
138
139
140
141
E. Publication of Project 2
148
11. GENERAL REFERENCES
1. Knibbs LD, Cole-Hunter T, Morawska L: A review of commuter exposure to ultrafine particles and its health effects. Atmospheric Environment 2011, 45(16):2611-2622.
2. Hertel O, Hvidberg M, Ketzel M, Storm L, Stausgaard L: A proper choice of route significantly reduces air pollution exposure--A study on bicycle and bus trips in urban streets. Sci Total Environ 2008, 389(1):58-70.
3. de Nazelle A, Nieuwenhuijsen M: Integrated health impact assessment of cycling. Occup Environ Med 2010, 67(2):76.
4. de Nazelle A, Nieuwenhuijsen M, Antó J, Brauer M, Briggs D, Braun-Fahrlander C, Cavill N, Cooper A, Desqueyroux H, Fruin S: Improving health through policies that promote active travel: a review of evidence to support integrated health impact assessment. Environment International 2011, 37(4):766.
5. Woodcock J, Edwards P, Tonne C, Armstrong BG, Ashiru O, Banister D, Beevers S, Chalabi Z, Chowdhury Z, Cohen A: Public health benefits of strategies to reduce greenhouse-gas emissions: urban land transport. The Lancet 2009, 374(9705):1930-1943.
6. de Hartog JJ, Boogaard H, Nijland H, Hoek G: Do the health benefits of cycling outweigh the risks? Environ Health Perspect 2010, 118(8):1109.
7. Pietropaoli AP, Frampton MW, Hyde RW, Morrow PE, Oberdörster G, Cox C, Speers DM, Frasier LM, Chalupa DC, Huang LS: Pulmonary function, diffusing capacity, and inflammation in healthy and asthmatic subjects exposed to ultrafine particles. Inhal Toxicol 2004, 16(s1):59-72.
8. Knibbs LD, Cole-Hunter T, Morawska L: A review of commuter exposure to ultrafine particles and its health effects. Atmospheric Environment 2011, 45(16):2611-2622.
9. Strak M, Boogaard H, Meliefste K, Oldenwening M, Zuurbier M, Brunekreef B, Hoek G: Respiratory health effects of ultrafine and fine particle exposure in cyclists. Occup Environ Med 2010, 67(2):118.
10. Int Panis L, de Geus B, Vandenbulcke G, Willems H, Degraeuwe B, Bleux N, Mishra V, Thomas I, Meeusen R: Exposure to particulate matter in traffic: A comparison of cyclists and car passengers. Atmos Environ 2010, 44(19):2263-2270.
11. 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 2008, 42(26):6496-6512.
12. Zuurbier M, Hoek G, Oldenwening M, Lenters V, Meliefste K, Van Den Hazel P, Brunekreef B: Commuters’ Exposure to Particulate Matter Air Pollution Is Affected by Mode of Transport, Fuel Type, and Route. Environ Health Perspect 2010, 118(6):783.
13. Zuurbier M, Hoek G, Oldenwening M, Meliefste K, Van den Hazel P, Brunekreef B: Acute Health Effects of Commuters' Exposure to Air Pollution. Epidemiology 2009, 20(6):S197.
14. Adams HS, Nieuwenhuijsen MJ, Colvile RN: Determinants of fine particle (PM2. 5) personal exposure levels in transport microenvironments, London, UK. Atmos Environ 2001, 35(27):4557-4566.
15. Briggs DJ, de Hoogh K, Morris C, Gulliver J: Effects of travel mode on exposures to particulate air pollution. Environment International 2008, 34(1):12-22.
149
16. Kaur S, Nieuwenhuijsen MJ, Colvile RN: Pedestrian exposure to air pollution along a major road in Central London, UK. Atmos Environ 2005, 39(38):7307-7320.
17. Kaur S, Nieuwenhuijsen MJ, Colvile RN: Fine particulate matter and carbon monoxide exposure concentrations in urban street transport microenvironments. Atmos Environ 2007, 41(23):4781-4810.
18. Zhou Y, Levy J: Factors influencing the spatial extent of mobile source air pollution impacts: a meta-analysis. BMC Public Health 2007, 7(1):89.
19. 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 Cyclists. Environ Health Perspect 2011, 119(10):1373–1378.
20. Blair SN, Morris JN: Healthy hearts--and the universal benefits of being physically active: Physical activity and health. Ann Epidemiol 2009, 19(4):253-256.
21. Shiroma EJ, Lee I: Physical activity and cardiovascular health: lessons learned from epidemiological studies across age, gender, and race/ethnicity. Circulation 2010, 122(7):743.
22. Warburton DER, Nicol CW, Bredin SSD: Health benefits of physical activity: the evidence. Can Med Assoc J 2006, 174(6):801.
23. Singh S, Shi T, Duffin R, Albrecht C, van Berlo D, Höhr D, Fubini B, Martra G, Fenoglio I, Borm PJA et al: Endocytosis, oxidative stress and IL-8 expression in human lung epithelial cells upon treatment with fine and ultrafine TiO2: Role of the specific surface area and of surface methylation of the particles. Toxicol Appl Pharmacol 2007, 222(2):141-151.
24. de Hartog J, Boogaard H, Nijland H, Hoek G: Do The Health Benefits Of Cycling Outweigh The Risks? Environ Health Perspect 2010, 118(8):1109.
25. Rabl A, de Nazelle A: Benefits of shift from car to active transport. Transport Policy 2012, 19(1):121-131.
26. Rojas-Rueda D, de Nazelle A, Tainio M, Nieuwenhuijsen MJ: The health risks and benefits of cycling in urban environments compared with car use: health impact assessment study. BMJ: British Medical Journal 2011, 343.
27. Künzli N, Jerrett M, Garcia-Esteban R, Basagaña X, Beckermann B, Gilliland F, Medina M, Peters J, Hodis HN, Mack WJ: Ambient air pollution and the progression of atherosclerosis in adults. PLoS One 2010, 5(2):e9096.
28. Dominici F, Peng RD, Bell ML, Pham L, McDermott A, Zeger SL, Samet JM: Fine particulate air pollution and hospital admission for cardiovascular and respiratory diseases. Journal of the American Medical Association 2006, 295(10):1127.
29. Pope III CA: Mortality effects of longer term exposures to fine particulate air pollution: review of recent epidemiological evidence. Inhal Toxicol 2007, 19(S1):33-38.
30. Lee K, Yanagisawa Y, Spengler JD, Nakai S: Carbon monoxide and nitrogen dioxide exposures in indoor ice skating rinks. J Sports Sci 1994, 12:279-279.
31. Ghio AJ, Devlin RB: Inflammatory lung injury after bronchial instillation of air pollution particles. Am J Respir Crit Care Med 2001, 164(4):704.
32. Belanger K, Gent JF, Triche EW, Bracken MB, Leaderer BP: Association of indoor nitrogen dioxide exposure with respiratory symptoms in children with asthma. American journal of respiratory and critical care medicine 2006, 173(3):297.
150
33. Johns DO, Svendsgaard D, Linn WS: Analysis of the concentration-respiratory response among asthmatics following controlled short-term exposures to sulfur dioxide. Inhal Toxicol 2010, 22(14):1184-1193.
34. Wilhelm M, Qian L, Ritz B: Outdoor air pollution, family and neighborhood environment, and asthma in LA FANS children. Health & Place 2009, 15(1):25-36.
35. Jarvis D, Chinn S, Luczynska C, Burney P: Association of respiratory symptoms and lung function in young adults with use of domestic gas appliances. The Lancet 1996, 347(8999):426-431.
36. Kjærgaard T, Cvancarova M, Steinsvåg SK: Cigarette smoking and self-assessed upper airway health. Eur Arch Otorhinolaryngol 2011, 268(2):219-226.
37. van der Zee SC, Hoek G, Boezen MH, Schouten JP, van Wijnen JH, Brunekreef B: Acute effects of air pollution on respiratory health of 50-70 yr old adults. Eur Respir J 2000, 15(4):700.
38. Aatamila M, Verkasalo PK, Korhonen MJ, Suominen AL, Hirvonen MR, Viluksela MK, Nevalainen A: Odour annoyance and physical symptoms among residents living near waste treatment centres. Environ Res 2010, 111(1):164-170.
39. Stenlund T, Liden E, Andersson K, Garvill J, Nordin S: Annoyance and health symptoms and their influencing factors: a population-based air pollution intervention study. Public Health 2009, 123(4):339-345.
40. Badland HM, Duncan MJ: Perceptions of air pollution during the work-related commute by adults in Queensland, Australia. Atmos Environ 2009, 43(36):5791-5795.
41. Wichmann H, Peters A: Epidemiological evidence of the effects of ultrafine particle exposure. Philosophical Transactions: Mathematical, Physical and Engineering Sciences 2000, 358(1775):2751-2769.
42. Russell AG, Brunekreef B: A focus on particulate matter and health. Environmental Science & Technology 2009, 43(13):4620-4625.
43. Wåhlin P, Palmgren F, Van Dingenen R: Experimental studies of ultrafine particles in streets and the relationship to traffic. Atmos Environ 2001, 35:63-69.
44. Donaldson K, Brown D, Clouter A, Duffin R, MacNee W, Renwick L, Tran L, Stone V: The pulmonary toxicology of ultrafine particles. J Aerosol Med 2002, 15(2):213-220.
45. Li N, Sioutas C, Cho A, Schmitz D, Misra C, Sempf J, Wang M, Oberley T, Froines J, Nel A: Ultrafine particulate pollutants induce oxidative stress and mitochondrial damage. Environ Health Perspect 2003, 111(4):455.
46. Elder A, Gelein R, Silva V, Feikert T, Opanashuk L, Carter J, Potter R, Maynard A, Ito Y, Finkelstein J: Translocation of inhaled ultrafine manganese oxide particles to the central nervous system. Environ Health Perspect 2006, 114(8):1172.
47. Nyström AK, Svartengren M, Grunewald J, Pousette C, Rödin I, Lundin A, Sköld CM, Eklund A, Larsson BM: Health effects of a subway environment in healthy volunteers. Experimental data. Eur Respir J 2009:09031936.00099909 v09031931.
48. Chan CC, Chuang KJ, Shiao GM, Lin LY: Personal exposure to submicrometer particles and heart rate variability in human subjects. Environ Health Perspect 2004, 112(10):1063.
49. McCreanor J, Cullinan P, Nieuwenhuijsen M, Stewart-Evans J, Malliarou E, Jarup L, Harrington R, Svartengren M, Han I, Ohman-Strickland P: Respiratory effects of exposure to diesel traffic in persons with asthma. N Engl J Med 2007, 357(23):2348.
151
50. Pattinson W: Cyclist exposure to traffic pollution: microscale variance, the impact of route choice and comparisons to other modal choices in two New Zealand cities. 2009.
51. Jacobs L, Nawrot T, de Geus B, Meeusen R, Degraeuwe B, Bernard A, Sughis M, Nemery B, Panis L: Subclinical responses in healthy cyclists briefly exposed to traffic-related air pollution: an intervention study. Environmental Health 2010, 9(1):64.
52. Giles LV, Barn P, Künzli N, Romieu I, Mittleman MA, van Eeden S, Allen R, Carlsten C, Stieb D, Noonan C: From good intentions to proven interventions: effectiveness of actions to reduce the health impacts of air pollution. Environ Health Perspect 2011, 119(1):29.
53. 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 2005, 39(20):3629-3641.
54. Thai A, McKendry I, Brauer M: Particulate matter exposure along designated bicycle routes in Vancouver, British Columbia. Sci Total Environ 2008, 405(1-3):26-35.
55. Nightingale JA, Maggs R, Cullinan P, Donnelly LE, Rogers DF, Kinnersley R, Fan Chung K, Barnes PJ, Ashmore M, Newman-Taylor A: Airway Inflammation after Controlled Exposure to Diesel Exhaust Particulates. Am J Respir Crit Care Med 2000, 162(1):161-166.
56. Nordenhäll C, Pourazar J, Blomberg A, Levin JO, Sandström T, Ädelroth E: Airway inflammation following exposure to diesel exhaust: a study of time kinetics using induced sputum. Eur Respir J 2000, 15(6):1046-1051.
57. Singh D, Edwards L, Tal-Singer R, Rennard S: Research Sputum neutrophils as a biomarker in COPD: findings from the ECLIPSE study. 2010.
58. Kaur S, Nieuwenhuijsen M, Colvile R: Fine particulate matter and carbon monoxide exposure concentrations in urban street transport microenvironments. Atmos Environ 2007, 41(23):4781-4810.
59. Knibbs LD, Cole-Hunter T, Morawska L: A review of commuter exposure to ultrafine particles and its health effects. Atmos Environ 2011, 45:2611-2622.
60. Vinzents P, Møller P, Sørensen M, Knudsen L, Hertel O, Jensen F, Schibye B, Loft S: Personal exposure to ultrafine particles and oxidative DNA damage. Environ Health Perspect 2005, 113(11):1485.
61. Wilson JG, Kingham S, Pearce J, Sturman AP: A review of intraurban variations in particulate air pollution: Implications for epidemiological research. Atmos Environ 2005, 39(34):6444-6462.
62. Berghmans P, Bleux N, Panis L, Mishra V, Torfs R, Van Poppel M: Exposure assessment of a cyclist to PM10 and ultrafine particles. Sci Total Environ 2009, 407(4):1286-1298.
63. 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 2009, 43(27):4234-4242.
64. Panis L, Bleux N, Torfs R, Mishra V, Geus B, Meeusen R, Vandenbulcke G, Thomas I: Exposure of Cyclists to Ultra Fine Particles. Highway and Urban Environment 2010:115-122.
65. 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 2005, 39(20):3629-3641.
152
66. van Wijnen J, Verhoetf A, Jans H, van Bruggen M: Exposure of Cyclists, Car Drivers and Pedestrians to Motor Vehicle Exhaust Emissions. Epidemiology 1995, 6(2):S25.
67. Seaton A, Dennekamp M: Hypothesis: Ill health associated with low concentrations of nitrogen dioxide—an effect of ultrafine particles? Thorax 2003, 58(12):1012.
68. Kaur S, Nieuwenhuijsen M, Colvile R: Pedestrian exposure to air pollution along a major road in Central London, UK. Atmos Environ 2005, 39(38):7307-7320.
69. Briggs D, de Hoogh K, Morris C, Gulliver J: Effects of travel mode on exposures to particulate air pollution. Environment International 2008, 34(1):12-22.
70. Cattaneo A, Garramone G, Taronna M, Peruzzo C, Cavallo D: Personal exposure to airborne ultrafine particles in the urban area of Milan. In: 2009: IOP Publishing; 2009: 012039.
71. Morabia A, Amstislavski PN, Mirer FE, Amstislavski TM, Eisl H, Wolff MS, Markowitz SB: Air pollution and activity during transportation by car, subway, and walking. Am J Prev Med 2009, 37(1):72-77.
72. Seaton A, Godden D, MacNee W, Donaldson K: Particulate air pollution and acute health effects. The Lancet 1995, 345(8943):176-178.
73. Morawska L, Moore, M. and Ristovski, Z.: Desktop Literature Review and Analysis of Health Impacts of Ultrafine Particles. In.; 2003.
74. Knol AB, De Hartog JJ, Boogaard H, Slottje P, Van Der Sluijs JP, Lebret E, Cassee FR, Wardekker JA, Ayres JG, Borm PJ: Expert elicitation on ultrafine particles: likelihood of health effects and causal pathways. Particle and Fibre Toxicology 2009, 6(1):19.
75. Hoek G, Boogaard H, Knol A, de Hartog J, Slottje P, Ayres JG, Borm P, Brunekreef B, Donaldson K, Forastiere F: Concentration response functions for ultrafine particles and all-cause mortality and hospital admissions: results of a European expert panel elicitation. Environmental Science & Technology 2009, 44(1):476-482.
76. Frampton MW: Does Inhalation of Ultrafine Particles Cause Pulmonary Vasular Effects in Humans? Inhal Toxicol 2007, 19(S1):75-79.
77. Daigle CC, Chalupa DC, Gibb FR, Morrow PE, Oberdörster Gn, Utell MJ, Frampton MW: Ultrafine Particle Deposition in Humans During Rest and Exercise. Inhal Toxicol 2003, 15(6):539-552.
78. Phalen RF, Mendez LB: Dosimetry considerations for animal aerosol inhalation studies. Biomarkers 2009, 14(S1):63-66.
79. Geiser M, Rothen-Rutishauser B, Kapp N, Schürch S, Kreyling W, Schulz H, Semmler M, Im Hof V, Heyder J, Gehr P: Ultrafine particles cross cellular membranes by nonphagocytic mechanisms in lungs and in cultured cells. Environ Health Perspect 2005, 113(11):1555.
80. Donaldson K, Warheit DB: From Ambient Ultrafine Particles to Nanotechnology and Nanotoxicology. Cardiovascular Effects of Inhaled Ultrafine and Nanosized Particles 2011:525-543.
81. Delfino R, Sioutas C, Malik S: Potential role of ultrafine particles in associations between airborne particle mass and cardiovascular health. Environ Health Perspect 2005, 113(8):934.
82. Delfino RJ, Staimer N, Tjoa T, Gillen DL, Polidori A, Arhami M, Kleinman MT, Vaziri ND, Longhurst J, Sioutas C: Air pollution exposures and circulating biomarkers of effect in a susceptible population: clues to potential causal component mixtures and mechanisms. Environ Health Perspect 2009, 117(8):1232.
153
83. Tonne C, Yanosky J, Gryparis A, Melly S, Mittleman M, Goldberg R, Von Klot S, Schwartz J: Traffic particles and occurrence of acute myocardial infarction: a case–control analysis. Occup Environ Med 2009, 66(12):797.
84. Brugge D, Durant JL, Rioux C: Near-highway pollutants in motor vehicle exhaust: a review of epidemiologic evidence of cardiac and pulmonary health risks. Environmental Health 2007, 6(1):23.
85. Samet J, Rappold A, Graff D, Cascio W, Berntsen J, Huang Y, Herbst M, Bassett M, Montilla T, Hazucha M: Concentrated ambient ultrafine particle exposure induces cardiac changes in young healthy volunteers. Am J Respir Crit Care Med 2009, 179(11):1034.
86. Charron A, Harrison RM: Atmospheric Nanoparticles. Environmental and Human Health Impacts of Nanotechnology 2010:163-209.
87. Kawanaka Y, Matsumoto E, Wang N, Yun SJ, Sakamoto K: Contribution of nitrated polycyclic aromatic hydrocarbons to the mutagenicity of ultrafine particles in the roadside atmosphere. Atmos Environ 2008, 42(32):7423-7428.
88. Nyström AK, Svartengren M, Grunewald J, Pousette C, Rödin I, Lundin A, Sköld CM, Eklund A, Larsson BM: Health effects of a subway environment in healthy volunteers. Eur Respir J 2010, 36(2):240-248.
89. Pekkanen J, Peters A, Hoek G, Tiittanen P, Brunekreef B, de Hartog J, Heinrich J, Ibald-Mulli A, Kreyling WG, Lanki T: Particulate air pollution and risk of ST-segment depression during repeated submaximal exercise tests among subjects with coronary heart disease. Circulation 2002, 106(8):933-938.
90. Oberdörster G, Oberdörster E, Oberdörster J: Nanotoxicology: an emerging discipline evolving from studies of ultrafine particles. Environ Health Perspect 2005, 113(7):823.
91. Hung H, Wang C: Experimental determination of reactive oxygen species in Taipei aerosols. Journal of Aerosol Science 2001, 32(10):1201-1211.
92. Cascio W, Katwa L, Linn W, Stram D, Zhu Y, Cascio J, Hinds W: Effects of Vehicle Exhaust in Aged Adults Riding on Los Angeles Freeways. Am J Respir Crit Care Med 2009, 179(1 MeetingAbstracts):A1175.
93. Keeling B, Li KY, Churg A: Iron enhances uptake of mineral particles and increases lipid peroxidation in tracheal epithelial cells. Am J Respir Cell Mol Biol 1994, 10(6):683.
94. Fubini B, Mollo L: Role of iron in the reactivity of mineral fibers. Toxicol Lett 1995, 82:951-960.
95. Karlsson HL, Nilsson L, Möller L: Subway particles are more genotoxic than street particles and induce oxidative stress in cultured human lung cells. Chem Res Toxicol 2005, 18(1):19-23.
96. Nieuwenhuijsen M, Gomez-Perales J, Colvile R: Levels of particulate air pollution, its elemental composition, determinants and health effects in metro systems. Atmos Environ 2007, 41(37):7995-8006.
97. Bachoual R, Boczkowski J, Goven D, Amara N, Tabet L, On D, Leçon-Malas V, Aubier M, Lanone S: Biological effects of particles from the paris subway system. Chem Res Toxicol 2007, 20(10):1426-1433.
98. Cho SH, Tong H, McGee JK, Baldauf RW, Krantz QT, Gilmour MI: Comparative toxicity of size-fractionated airborne particulate matter collected at different distances from an urban highway. Environ Health Perspect 2009, 117(11):1682.
99. Gong Jr H, Sioutas C, Linn W: Controlled exposures of healthy and asthmatic volunteers to concentrated ambient particles in metropolitan Los Angeles. Research report (Health Effects Institute) 2003(118):1.
154
100. Adachi K, Buseck P: TEM study of soot, organic aerosol, and sea-salt particles collected during CalNex. In: 2010; 2010: 0111.
101. Michaels RA, Kleinman MT: Incidence and apparent health significance of brief airborne particle excursions. Aerosol Science & Technology 2000, 32(2):93-105.
102. Bräuner E, Møller P, Barregard L, Dragsted L, Glasius M, Wåhlin P, Vinzents P, Raaschou-Nielsen O, Loft S: Exposure to ambient concentrations of particulate air pollution does not influence vascular function or inflammatory pathways in young healthy individuals. Particle and Fibre Toxicology 2008, 5(1):13.
103. Svartengren M, Strand V, Bylin G, Jarup L, Pershagen G: Short-term exposure to air pollution in a road tunnel enhances the asthmatic response to allergen. Eur Respir J 2000, 15(4):716-724.
104. Laumbach R, Rich D, Gandhi S, Amorosa L, Schneider S, Zhang J, Ohman-Strickland P, Gong J, Lelyanov O, Kipen H: Acute changes in heart rate variability in subjects with diabetes following a highway traffic exposure. J Occup Environ Med 2010, 52(3):324.
105. Timonen K, Vanninen E, de Hartog J, Ibald-Mulli A, Brunekreef B, Gold D, Heinrich J, Hoek G, Lanki T, Peters A: Pekkanen. Effects of ultrafine and fine particulate J control in subjects with coronary artery disease: the ULTRA study. J Expo Sci Environ Epidemiol 2006, 16:332-341.
106. ATS: What constitutes an adverse health effect of air pollution? Am J Respir Crit Care Med 2000, 161:665-673.
107. Rudell B, Ledin M, Hammarström U, Stjernberg N, Lundbäck B, Sandström T: Effects on symptoms and lung function in humans experimentally exposed to diesel exhaust. Occup Environ Med 1996, 53(10):658.
108. Bikeway and shared pathway maps [http://www.brisbane.qld.gov.au/traffic-transport/cycling/bikeway-and-shared-pathway-maps/index.htm]
109. Puustinen A, Hämeri K, Pekkanen J, Kulmala M, de Hartog J, Meliefste K, ten Brink H, Kos G, Katsouyanni K, Karakatsani A: Spatial variation of particle number and mass over four European cities. Atmos Environ 2007, 41(31):6622-6636.
110. Treese N, MacCARTER D, AKBULUT O, COUTINHO M, BAEZ M, LIEBRICH A, MEYER J: Ventilation and heart rate response during exercise in normals: Relevance for rate variable pacing. Pacing Clin Electrophysiol 1993, 16(8):1693-1700.
111. Meteorology ABo: Climate Data Online. In.: Australian Bureau of Meteorology; 2011.
112. Cone JE, Shusterman D: Health effects of indoor odorants. Environ Health Persp 1991, 95:53.
113. Lercher P, Schmitzberger R, Kofler W: Perceived traffic air pollution, associated behavior and health in an alpine area. Sci Total Environ 1995, 169(1-3):71-74.
114. Shusterman D: Community health and odor pollution regulation. Am J Public Health 1992, 82(11):1566.
115. Paustenbach DJ, Gaffney SH: The role of odor and irritation, as well as risk perception, in the setting of occupational exposure limits. Int Arch Occup Environ Health 2006, 79(4):339-342.
116. Dalton P: Cognitive influences on health symptoms from acute chemical exposure. Health Psychol 1999, 18:579-590.
117. Righi E, Aggazzotti G, Fantuzzi G, Ciccarese V, Predieri G: Air quality and well-being perception in subjects attending university libraries in Modena (Italy). Sci Total Environ 2002, 286(1-3):41-50.
155
118. Hirsch T, Weiland SK, Von Mutius E, Safeca AF, Gräfe H, Csaplovics E, Duhme H, Keil U, Leupold W: Inner city air pollution and respiratory health and atopy in children. Eur Respir J 1999, 14(3):669-677.
119. Simoni M, Annesi-Maesano I, Sigsgaard T, Norback D, Wieslander G, Nystad W, Canciani M, Sestini P, Viegi G: School air quality related to dry cough, rhinitis and nasal patency in children. Eur Respir J 2010, 35(4):742.
120. Nikolopoulou M, Kleissl J, Linden P, Lykoudis S: Pedestrians' perception of environmental stimuli through field surveys: Focus on particulate pollution. The Science of the total environment 2011, 409(13):2493-2502.
121. Nel AE, Diaz-Sanchez D, Ng D, Hiura T, Saxon A: Enhancement of allergic inflammation by the interaction between diesel exhaust particles and the immune system. J Allergy Clin Immunol 1998, 102(4):539-554.
122. Queensland 2010 Air Monitoring Report [http://www.derm.qld.gov.au/air/pdf/reports/2010-air-monitoring-report.pdf]
123. Ayoko G, Jamriska M, Jayaratne R, Morawska L: Air pollution levels measured at traffic hot spots: Brisbane urban corridor study. In: 2005: CASANZ (Clean Air Society of Australia & New Zealand); 2005: 1-7.
124. Bräuner EV, Møller P, Barregard L, Dragsted LO, Glasius M, Wåhlin P, Vinzents P, Raaschou-Nielsen O, Loft S: Exposure to ambient concentrations of particulate air pollution does not influence vascular function or inflammatory pathways in young healthy individuals. Particle and Fibre Toxicology 2008, 5(1):13.
125. Belanger E, Kielb C, Lin S: Asthma Hospitalization Rates Among Children, and School Building Conditions, by New York State School Districts, 1991 2001. J Sch Health 2006, 76(8):408-413.
126. Pribyl CR, Racca J: Toxic gas exposures in ice arenas. Clin J Sport Med 1996, 6(4):232.
127. Brunekreef B, Holgate ST: Air pollution and health. The Lancet 2002, 360(9341):1233-1242.
128. Knibbs LD, Cole-Hunter T, Morawska L: A review of commuter exposure to ultrafine particles and its health effects. Atmospheric Environment 2011, 45(16):2611-2622.
129. Morawska L, Ristovski Z, Jayaratne E, Keogh DU, Ling X: Ambient nano and ultrafine particles from motor vehicle emissions: Characteristics, ambient processing and implications on human exposure. Atmos Environ 2008, 42(35):8113-8138.
130. Nikolopoulou M, Kleissl J, Linden P, Lykoudis S: Pedestrians' perception of environmental stimuli through field surveys: Focus on particulate pollution. Sci Total Environ 2011.
131. Marra J, Den Brink W, Goossens H, Kessels S: Nanoparticle monitoring for exposure assessment. Nanotechnology Magazine, IEEE 2009, 3(2):6-37.
132. Derrick E: Asthma and the Brisbane climate. Aust N Z J Med 1972, 2(3):235-246. 133. Urban Commuter Cycling Respirator [http://www.respro.com/products/urban-
commuting/cycling/] 134. Burke M, Sipe N, Hatfield E: Evaluation of King George Square Cycle Centre. In.:
Research Paper. Brisbane, Australia: Urban Research Program, Griffith University. Available from: http://www.griffith. edu. au/__data/assets/pdf_file/0009/208368/urp-rp30-burke-et-a l-2010. pdf; 2010.