Content uploaded by Ala Al Rajabi
Author content
All content in this area was uploaded by Ala Al Rajabi on Oct 16, 2019
Content may be subject to copyright.
Tob Regul Sci.™ 2019;5(5):463-479 463
Secondhand smoke exposure (SHSe) preva-
lence is a meaningful indicator of tobacco
control progress.1 In 2009, it was estimated
that on a global basis one-third of all adults were
regularly exposed to SHS from tobacco.2 One study
conducted between 2007-2009 found that 22% of
non-smoking Canadians were exposed to SHS on a
daily or near-daily basis.3 Similar to active tobacco
consumption, SHSe is associated with increased
risk of many cancers, cardiovascular events and dis-
ease, respiratory diseases, and other chronic condi-
tions.4,5 In light of scientic evidence, the World
Health Organization Framework Convention on
Tobacco Control (WHO FCTC) was the rst
evidence-based global public health treaty calling
for worldwide legislations and tobacco control ini-
tiatives to reduce and eliminate tobacco consump-
tion and SHSe.6 e treaty, which became active
in 2005, has 168 signatories, making it one of the
most widely accepted treaties in United Nations
history.7 Tobacco control initiatives such as cessa-
tion programs, tobacco products taxation, adver-
tising bans and promotion, and enforcement of
smoke-free environments are inexpensive, yet eec-
Tiany Haig, Health Promotion Facilitator II, Chronic Disease Prevention & Oral Health, Alberta Health Services, Calgary, AB (Canada). Alianu K.
Akawung, Biostatistician, Alberta’s Tomorrow Project, Alberta Health Services, Calgary, AB (Canada). Ala Al Rajabi, Senior Research Associate, Alber-
ta’s Tomorrow Project, Alberta Health Services, Calgary, AB (Canada). Jennifer E. Vena, Scientic Director, Alberta’s Tomorrow Project, Alberta Health
Services, Calgary, AB (Canada). Heather K. Whelan, Lecturer, Department of Health and Physical Education, Mount Royal University, Calgary, AB
(Canada). Karen A. Kopciuk, Research Scientist, Cancer Epidemiology and Prevention Research, Alberta Health Services, Calgary, AB (Canada).
Correspondence Dr Kopciuk; Karen.Kopciuk@ahs.ca
Reductions in Secondhand Smoke Exposure
among Non-smokers Post-legislation
Tiany R. Haig, BA
Alianu K. Akawung, MSc
Ala Al Rajabi, PhD
Jennifer E. Vena, PhD
Heather K. Whelan, MSc
Karen A. Kopciuk, PhD
Objectives: We evaluated associations between provincial smoke-free legislation implementa-
tion and changes in secondhand smoke exposure (SHSe) prevalence in a prospective cohort of
non-smokers in Alberta’s Tomorrow Project (ATP). Methods: We had 11,564 non-smoking ATP
participants with both enrollment (pre-legislation) and follow-up (post-legislation) self-report
questionnaire data. They comprised the main data set with a subset (N = 7133) employed at both
time-points who were utilized for the workplace evaluation. Participants reported frequency of
home and workplace SHSe at enrollment and follow-up. We used adjusted Generalized Estimat-
ing Equation modeling to evaluate the association between smoke-free legislation implemen-
tation and changes in home and workplace SHSe. Results: Pre-legislation workplace and home
SHSe prevalences were 19.8% and 9.7% respectively; these levels declined post-legislation by
33.3% and 47.4% respectively. Adjusted GEE models corroborated these reductions (adjusted
odds ratio (AOR) = 0.59, 95% CI: 0.55-0.64 and AOR= 0.58, 95% CI: 0.51-0.68, respectively). We
found a statistically signicant interaction between sex and policy period for home SHSe but not
workplace SHSe. Conclusion: Implementation of comprehensive smoke-free legislation target-
ing work and public places was associated with signicant reductions in self-reported home and
workplace SHSe among non-smokers with no evidence of exposure displacement in the home.
Key words: tobacco workplace legislation; secondhand smoke exposure; Alberta’s Tomorrow Project
Tob Regul Sci.™ 2019;5(5):463-479
DOI: https://doi.org/10.18001/TRS.5.5.7
Reductions in Secondhand Smoke Exposure among Non-smokers Post-legislation
464
tive public health methods for reducing the burden
of tobacco use and exposure.2 In particular, legis-
lations banning smoking remain one of the most
eective strategies for reducing SHSe.8–11
Implementation of smoke-free legislations is as-
sociated with substantial gains in overall population
health evident by reduced incidence of cardiovas-
cular-related events,12 respiratory conditions,12 hos-
pital admissions,13 and overall mortality.12 Indeed,
evidence supports the role of smoke-free legislations
as important public health interventions to improve
population health for both smokers and non-smok-
ers.14 For example, a smoke-free environment is a
supportive environment that encourages smokers
to quit,8,15 and stay smoke-free,16,17 while also reduc-
ing the chances that children and youth will take
up smoking.18 Further, smoke-free legislations that
promote smoking cessation and reduced cigarette
consumption in adult smokers also reduce SHSe in
non-smokers16,17 – smoke-free legislations have been
associated with reduced SHSe estimated by self-re-
port17,19-30 as well as measured circulating cotinine, a
biomarker and metabolite of nicotine, in children,
hospitality workers, and the public at large.20,24-26,29,31
Whereas smoke-free legislative actions have been
shown to be eective, they typically target public
and workplaces, whereas SHSe also can occur in
homes or private vehicles. Concerns have been
raised that implementation of smoke-free legisla-
tions targeting public and work spaces may displace
active smoking from these places into homes,32,33
and consequently aect SHSe of those most vul-
nerable – children and non-smoking adults.32,33
Studies evaluating the impact of various public and
workplace smoke-free legislations on home SHSe
have yielded inconsistent results ranging from re-
duced23,26,34,35 to unchanged25,31 home exposure.
Moreover, women are more likely to be exposed to
SHS compared to men,36,37and rural residents are
more likely to experience SHSe compared to ur-
ban residents.38 However, there has been little in-
quiry into the variability in SHSe before and after
smoke-free legislation implementation by sex and
geographic location. Furthering this understanding
may support future development of targetable and
eective tobacco control initiatives.
A comprehensive smoke-free places legislation
(Tobacco Reduction Act) was implemented in the
province of Alberta, Canada on January 1, 2008
(Bill 45/2007),39,40 with the aim of improving the
health of Albertans by reducing and eliminating to-
bacco use and SHSe. In addition to implementing
measures targeting active smoking (eg, restrictions
on tobacco products displays and sales), the legisla-
tion prohibited smoking in all public places, work-
places, public vehicles and within 5 meters from
any doorway, window, and air intake of a work-
place or public venue. Here, we evaluated overall
changes in SHSe prevalence at workplaces as well
as homes among middle-aged adult non-smokers
before (2001-2006) and after (2009-2015) pro-
vincial smoke-free legislation implementation in
2008. To our knowledge, this is the rst report on
the association of smoke-free legislation with SHSe
among non-smoking middle-aged adults from a
longitudinal Canadian cohort.
METHOD
Cohort Design and Data Collection
Alberta’s Tomorrow Project (ATP) is a prospec-
tive cohort of ~55,000 Albertans established in
2000 to study the etiology of cancer and chronic
diseases. Full description of study feasibility, design
and enrollment is detailed elsewhere.41,42 Briey, Al-
bertans aged 35-69 years, with no history of cancer
except non-melanoma skin cancer, were recruited
throughout the province by telephone-based ran-
dom digit dialing. At enrollment (2000-2008),
participants completed a Health and Lifestyle
Questionnaire (HLQ), Past-Year Total Physical Ac-
tivity Questionnaire (PYTPAQ),43,44 and Canadian
Diet History Questionnaire I(CDHQ-I).45,46 From
2009-2015, participants completed one of 2 fol-
low-up questionnaires containing similar content:
the Updated Health and Lifestyle Questionnaire
(UHLQ) or the CORE Questionnaire.47 All ques-
tionnaires were sent via postal mail to participants
who returned completed questionnaires to ATP in
pre-paid envelopes. HLQ and UHLQ/CORE col-
lected information on personal and family health
history including lifetime physician diagnosis of a
number of chronic conditions, reproductive his-
tory, anthropometric variables, sociodemographic
characteristics and lifestyle behaviors such as smok-
ing habits and SHSe. e present analysis includes
a subsample of ATP participants who self-reported
as non-smokers on the HLQ (pre-legislation) and
UHLQ/CORE (post-legislation) surveys.
Haig et al
Tob Regul Sci.™ 2019;5(5):463-479 465 DOI: https://doi.org/10.18001/TRS.5.5.7
Study Population
Inclusion in the current study was restricted to
participants who completed the enrollment HLQ
before January 1, 2007 and the follow-up UHLQ/
CORE after December 31, 2008 with a minimum
of 2 years between these questionnaire comple-
tions. e date of January 1, 2007 was selected to
account for municipal bylaw implemented prior
to provincial implementation, and December 31,
2008 was selected to allow for sucient time for
province-wide legislation to be in full eect. e
minimum of 2 years between completions provides
a clear distinction between pre- and post-legislation
to allow for an accurate assessment of the associa-
tion between comprehensive province-wide legisla-
tion and SHSe. Participants at enrollment and/or
follow-up who were residing outside Alberta (N =
526), pregnant (N = 49 women), current tobacco
users (N = 2575; tobacco includes cigarettes, ci-
gars, cigarillos, pipes, sheesha, betel nut, paan and
chewing tobacco gum and snu), did not answer
all SHSe questions (N = 273), or who were diag-
Table 1
Comparable Exposure Domains Pre- and Post- 2008 Alberta Tobacco Reduction
Act Legislation Implementation
Pre-legislation Post-legislationa
Original Categories Original Categories Revised Categories
In the past year were you exposed to
SHSb at home?
1=Yes
2=No
At home how often you are usually exposed?
5=Every day
4=Almost everyday
3=At least once a week
2=At least once a month
1=less than once a month
0=Never
Yes (2, 3, 4, 5)c
No (0, 1)d
In the past year were you exposed to
SHS at work?
1=Yes
2=No
At work how often are you usually exposed?
5=Every day
4=Almost everyday
3=At least once a week
2=At least once a month
1=less than once a month
0=Never
Yes (2, 3, 4, 5)
No (0, 1)
In the past year were you exposed to
SHS in public?
1=Yes
2=No
During leisure time undertaken outside of your
home, how often are you usually exposed to SHS?
5=Every day
4=Almost everyday
3=At least once a week
2=At least once a month
1=less than once a month
0=Never
Yes (2, 3, 4, 5)
No (0, 1)
Note.
a: There were 6 response categories on the post-legislation survey (UHLQ and CORE) compared to only 2 response
categories on the pre-legislation survey (HLQ) and therefore follow-up survey categories were collapsed into
comparable categories.
b: Secondhand smoke.
c: Refers to at least once a month, at least once a week, almost every day and every day.
d: Refers to never and less than once a month
Reductions in Secondhand Smoke Exposure among Non-smokers Post-legislation
466
nosed with cancer prior to enrollment (N = 82)
were excluded from this analysis. CORE data were
used for participants who completed both UHLQ
and CORE (N = 17).
Prior to the implementation of the 2008 pro-
vincial Tobacco Reduction Act, 2 large munici-
palities in Alberta, Edmonton and Calgary, had
comprehensive smoking bylaws in place that
banned smoking in all public places and pro-
hibited designated smoking rooms as of July 1,
2005 and January 1, 2007 respectively.48 To ac-
count for municipal bylaws implemented prior to
provincial implementation (January 1, 2008) and
the potential impact of within household correla-
tion, we ran 2 sensitivity analyses excluding Ed-
monton participants who completed HLQ after
July 1, 2005 (N = 528) or participants recruited
as second in a household (N = 133). e results
were not signicantly dierent (data not shown);
therefore, these participants were included in the
Table 2
Percentages of Participants’ Sociodemographic Characteristics at Enrollment Stratied
by Sex or Geographic Location (N = 11,564)a
Sex Geographic Locationc
Men Women Urban Rural Total
N = 4100
%b
N = 7464
%b
N = 8952
%b
N = 2612
%b
N = 11,564
%b
Sex
Men n/a n/a 36.0 33.5 35.5
Women n/a n/a 49.5 66.5 64.6
Geographic Locationc
Urban 78.7 76.7 n/a n/a 77.4
Rural 21.3 23.3 n/a n/a 22.6
Age (years)
35-44 27.1 28.6 29.0 24.9 28.1
45-54 35.5 36.0 36.1 35.0 35.8
55-64 27.9 26.4 26.2 29.4 26.9
65-69 9.5 9.0 6.7 10.8 9.2
Marital Status
Married/Living with a partner 86.4 79.4 80.1 88.0 81.9
Single/never married 5.3 5.2 5.8 3.2 5.2
Divorced/Separated/Widowed 8.3 15.4 14.1 8.7 12.9
Education Leveld
≤ High school 19.4 24.9 20.1 32.8 22.9
Some postsecondary 46.1 46.9 45.8 49.5 46.6
Post-secondary 34.5 28.2 34.2 17.8 30.5
Employment Status
Employede82.8 69.4 75.2 70.6 74.1
Not Employedf17.2 30.6 24.8 29.4 25.9
Incomeg
<50,000 19.5 30.3 24.0 35.0 26.5
50,000 - <100,000 45.3 41.2 42.4 43.3 42.6
≥100,000 33.7 25.7 31.3 19.0 28.6
(continued on next page)
Haig et al
Tob Regul Sci.™ 2019;5(5):463-479 467 DOI: https://doi.org/10.18001/TRS.5.5.7
study. In Calgary, the bylaw was implemented one
year earlier than the provincial bylaw, on Janu-
ary 1, 2007; as part of the inclusion criteria for
this analysis, participants living in Calgary were
included only if they completed the enrollment
questionnaire prior to this date and therefore
implementation of this bylaw does not aect the
present study population and analyses.
e nal sample size was N = 11,564 adults and
used for home exposure analyses. Participants who
reported being unemployed (not employed but
looking for work, homemaker, or retired) at either
time point (N = 4431) were excluded from work-
place SHSe analyses; therefore, the nal workplace
sample size was N = 7133. e median (IQR) time
between completions of questionnaires was 5.9
(4.1) years.
Study Outcome: SHSe
e HLQ (pre-legislation) asked about past-year
SHSe on a binary scale (yes/no) across home, work-
place and public domains (bars, restaurants, shop-
ping malls, arenas, bingo halls, bowling alleys). e
UHLQ/CORE (post-legislation) did not specify a
recall timeframe and asked about SHSe on a 6-point
scale (never, less than once a month, at least once
a month, at least once a week, almost every day
and every day) across 3 domains: home, workplace
and during leisure time undertaken outside of the
home. e dierent recall timeframes (past-year vs
not-specied) were deemed comparable because it
was assumed participants are more likely to report
recent exposures even when asked to report on a
long-term basis. e 6 response categories on the
UHLQ/CORE were collapsed into a yes/no binary
Respiratory Diseaseh
Yes 2.7 3.0 2.8 3.4 2.9
No 97.2 96.9 97.2 96.6 97.0
Cardiovascular Diseasei
Yes 27.5 22.7 23.9 26.2 24.4
No 72.4 77.3 76.1 73.8 75.6
Note.
a: A total of 283 participants (62 men, 221 women) had missing sociodemographic data.
b: Values are presented as column percentages.
c: Geographic location was determined using postal codes, where “0” as the second character of the rst 3 digits of the
postal code indicates rural residence.
d: ≤ High school: Did not complete grade 8, or completed grade 8 but not high school, or completed high school; Some
post-secondary: Completed some technical school/college training, or completed technical school/college training, or
completed some part of university degree; Post-secondary completed: Completed university degree, or completed
some part of post-graduate university degree, or completed university post-graduate degree.
e: Working full-time (30 hours or more per week), or working part-time (Less than 30 hours per week), or student, or
self-employed.
f: Not employed but looking for work, or homemaker, or retired.
g: Annual household income ($CAD).
h: Self-reported personal history of respiratory disease; “Yes” indicates participant had ever been told by a doctor that
they had one of the following conditions: emphysema, chronic bronchitis and/or chronic obstructive pulmonary
disease (COPD).
i: Self-reported personal history of cardiovascular disease; “Yes” indicates participant had ever been told by a doctor
that they had one of the following conditions: high blood pressure, heart attack, stroke or diabetes.
n/a = Not applicable
Table 2 (continued)
Percentages of Participants’ Sociodemographic Characteristics at Enrollment Stratied
by Sex or Geographic Location (N = 11,564)a
Sex Geographic Locationc
Men Women Urban Rural Total
N = 4100
%b
N = 7464
%b
N = 8952
%b
N = 2612
%b
N = 11,564
%b
Reductions in Secondhand Smoke Exposure among Non-smokers Post-legislation
468
scale to allow for comparability to the response cat-
egories on the HLQ (Table 1). Home and work-
place exposure domains were directly comparable
between HLQ and UHLQ/CORE due to similar-
ity in question content. However, the questions on
public and leisure domains were not phrased in a
similar fashion and therefore were deemed not di-
rectly comparable between enrollment and follow-
up surveys. As such, our primary analyses focused
on exposure at home and in the workplace.
Data Analysis
Pre- and post-legislation SHSe rates are pre-
sented using descriptive statistics and geographic
information system (GIS) maps. e impact of
time-varying covariates was determined using the
Kappa coecient to evaluate agreement between
pre- and post-legislation timepoints on sociode-
mographic characteristics, and by the adjusted
Generalized Methods of Moments (GMM) mod-
els to account for pre- and post-legislation covari-
Table 3
Percentage of Participants’ Characteristics from the Workplace Subset (N = 7133) who
Reported Secondhand Smoke Exposure in the Workplace Pre- (N = 1410) versus
Post-legislation (N = 944), Stratied by Sex or Geographic Locationa
Pre-legislation Post-legislation
Men
N = 864
%b
Women
N = 546
%b
Urban
N = 1059
%b
Rural
N = 351
%b
Total
N = 1410
%b
Men
N = 604
%b
Women
N = 340
%b
Urban
N = 708
%b
Rural
N = 236
%b
Total
N = 944
%b
29.9 12.9 18.9 23.1 19.8 20.9** 8.0** 12.6** 15.4** 13.2**
Sex
Men n/a n/a 28.2 36.2 29.9 n/a n/a 19.8** 25.0** 20.9**
Women n/a n/a 12.6 13.8 12.9 n/a n/a 7.8** 8.8* 8.0**
Geographic Locationc
Urban 28.2 12.6 n/a n/a 18.9 19.8** 7.8** n/a n/a 12.6**
Rural 36.2 13.8 n/a n/a 23.1 25.0** 8.8* n/a n/a 15.4**
Age (years)
35-44 32.6 12.8 19.4 25.4 20.6 22.5** 8.5* 13.2** 17.4* 14.0**
45-54 29.4 13.1 18.3 23.4 19.4 20.6** 7.8** 11.9** 16.0* 12.7**
55-64 26.7 12.9 19.4 18.1 19.1 18.9* 7.9* 13.1* 11.8* 12.8**
65-69 24.1 7.7 14.3 26.8 18.0 19.5 3.9 14.3 12.2 13.7
Marital Statusd
Married 29.9 11.8 18.9 22.3 19.7 20.4** 6.6** 11.7** 15.4** 12.6**
Single 26.2 17.4 19.0 28.1 20.3 20.0 17.4 18.7 15.6 18.2
Divorced 32.1 15.8 18.7 28.5 20.0 27.0 10.8* 14.7* 16.3* 14.9*
Education Levele
≤ High school 43.4 18.6 27.7 27.5 27.6 32.7* 9.8** 17.8** 19.1* 18.2**
Some post-secondary 38.2 13.7 23.1 25.4 23.6 26.5** 8.6** 15.3** 17.4* 15.8**
Post-secondary 12.1 8.0 9.6 11.3 9.8 7.7* 6.1* 6.8* 6.1* 6.8*
Incomef
<50,000 35.8 20.3 24.2 25.4 24.5 27.5* 12.6** 17.2* 15.7* 16.7**
50, 000 – <100, 000 33.8 12.1 20.5 23.7 21.2 24.2** 7.6** 14.3** 15.5* 14.6**
≥100, 000 23.2 7.8 14.2 19.7 15.1 14.6** 4.7* 8.2** 15.8 9.4**
Respiratory Diseaseg
Yes 39.3 17.3 25.6 25.0 25.5 29.5 11.5 16.0 25.0 18.2
No 29.7 12.8 18.7 23.1 19.6 20.7** 7.9** 12.5** 15.3** 13.1**
(continued on next page)
Haig et al
Tob Regul Sci.™ 2019;5(5):463-479 469 DOI: https://doi.org/10.18001/TRS.5.5.7
ates. Kappa values ≥0.7 were considered in good
agreement indicating that participant’s sociodemo-
graphic characteristics were similar between pre-
and post-legislation periods; therefore, analysis of
time-varying covariates would not be necessary.
Unadjusted pairwise comparisons using McNe-
mar’s test and adjusted Generalized Estimating
Equations (GEE) models were used to evaluate the
SHSe rates pre- versus post-legislation. e GEE
models, assuming an independent working covari-
ance structure with pairwise log odds ratios for the
association between responses, were adjusted for
the following categorical covariates measured at en-
rollment: age (modelled on continuous scale), sex,
cardiovascular and related disease (CVD), respira-
tory disease, marital status, educational level, em-
ployment status (home SHSe model only), annual
household income and geographic location (urban/
rural). Statistical interactions between time-period
(pre- vs post-legislation) and each control variable
were added into the model to test eect modi-
cation. Interactions were dropped if found to be
statistically insignicant, using backward elimina-
tion (starting with the highest p ≥ .05). Covariate
categorization schemes are described in more detail
in the footnotes in Tables 2-5.
To compare trends in SHSe among ATP partici-
pants to other population-based surveys in Alberta,
we examined agreement with the Canadian Com-
munity Health Survey (CCHS) cross-sectional data-
sets collected from 2001-2014.49 e CCHS cycles
matching the timeframe of current study did not
capture workplace SHSe; therefore, comparisons of
workplace SHSe were not feasible. CCHS data was
restricted to Albertans aged 35-69 years to coincide
with ATP inclusion criteria. Data from 2001-2006
and from 2009-2014 was combined as pre- and
post-legislation data, respectively and exposure prev-
alence estimates from both cohorts were adjusted for
age using Alberta 2001 census data as the standard
Cardiovascular Diseaseh
Yes 31.3 14.2 21.2 24.8 22.0 25.3* 8.0* 15.5* 17.1* 15.9*
No 29.5 12.6 18.3 22.7 19.3 19.8** 8.0** 12.0** 15.2* 12.6**
Note.
a: A total of 145 participants (40 men, 105 women) had missing sociodemographic data.
b: Percentage of men (N = 2889), women (N = 4244), urban (N = 5614), rural (N = 1519) and total (N = 7133) participants exposed to
SHS at workplace within each sociodemographic domain.
c: Geographic location was determined using postal codes, where “0” as the second character of the rst 3 digits of the postal code
indicates rural residence.
d: Married: Married/Living with a partner; Single: Single/never married; Separated: Divorced/Separated/Widowed.
e: ≤ High school: Did not complete grade 8, or completed grade 8 but not high school, or completed high school; Some post-secondary:
Completed some technical school/college training, or completed technical school/college training, or completed some part of
university degree; Post-secondary: Completed university degree, or completed some part of post-graduate university degree, or
completed university post-graduate degree.
f: Annual household income ($CAD).
g: Self-reported personal history of respiratory disease; “Yes” indicates participant had ever been told by a doctor that they had one of
the following conditions: emphysema, chronic bronchitis and/or chronic obstructive pulmonary disease (COPD).
h: Self-reported personal history of cardiovascular disease; “Yes” indicates participant had ever been told by a doctor that they had
one of the following conditions: high blood pressure, heart attack, stroke or diabetes.
n/a = Not applicable
*- p-values were obtained by the McNemar’s test comparing pre- versus post-legislation exposure rates: * indicates p < .05 and ≥ .001,
** indicates p < .001, no symbol indicates p ≥ .05
Table 3 (continued)
Percentage of Participants’ Characteristics from the Workplace Subset (N = 7133) who
Reported Secondhand Smoke Exposure in the Workplace Pre- (N = 1410) versus
Post-legislation (N = 944), Stratied by Sex or Geographic Locationa
Pre-legislation Post-legislation
Men
N = 864
%b
Women
N = 546
%b
Urban
N = 1059
%b
Rural
N = 351
%b
Total
N = 1410
%b
Men
N = 604
%b
Women
N = 340
%b
Urban
N = 708
%b
Rural
N = 236
%b
Total
N = 944
%b
29.9 12.9 18.9 23.1 19.8 20.9** 8.0** 12.6** 15.4** 13.2**
Reductions in Secondhand Smoke Exposure among Non-smokers Post-legislation
470
population framework. e Geographical Informa-
tion System (GIS) maps were created using ArcMap
version 10.6 on SQL 2014-R2 server database. All
statistical analyses were conducted using SAS version
9.3 on a Linux interface and the criterion for statisti-
cal signicance was set at alpha < .05 (2-tailed).
RESULTS
Sociodemographic Characteristics
Pre-legislation
Table 2 summarizes participants’ sociodemo-
graphic characteristics at enrollment, stratied by
sex or residential geographic location. e mean
Table 4
Percentage of Participants’ Characteristics in the Study Population (N = 11,564) who
Reported Secondhand Smoke Exposure in the Home Pre- (N = 1122) versus
Post-legislation (N = 593), Stratied by Sex or Geographic Locationa
Pre-legislation Post-legislation
Men
N = 338
%b
Women
N = 784
%b
Urban
N = 772
%b
Rural
N = 350
%b
Total
N = 1122
%b
Men
N = 210
%b
Wowmen
N = 383
%b
Urban
N = 420
%b
Rural
N = 173
%b
Total
N = 593
%b
8.2 10.5 8.6 13.4 9.7 5.1** 5.1** 4.7** 6.6** 5.1**
Sex
Men n/a n/a 7.5 11.0 8.2 n/a n/a 4.5** 7.5* 5.1**
Women n/a n/a 9.3 14.6 10.5 n/a n/a 4.8** 6.2** 5.1**
Geographic Locationc
Urban 7.5 9.3 n/a n/a 8.6 4.5** 4.8** n/a n/a 4.7**
Rural 11.0 14.6 n/a n/a 13.4 7.5* 6.2** n/a n/a 6.6**
Age (years)
35-44 6.2 9.1 7.4 10.9 8.1 3.4* 4.3** 3.9** 4.5** 4.0**
45-54 8.9 11.2 9.1 14.6 10.4 4.8** 5.3** 4.6** 6.8** 5.1**
55-64 9.4 11.7 9.7 14.6 10.9 6.5* 6.0** 5.4** 8.6* 6.2**
65-69 8.2 8.7 7.2 12.1 8.5 7.2 4.8* 5.7 5.7* 5.7*
Marital Statusd
Married 8.2 10.7 8.6 13.5 9.8 4.9** 5.0** 4.3** 6.7** 4.9**
Single 7.4 8.5 7.9 9.5 8.1 5.6 4.9* 5.2 4.8 5.1*
Divorced 8.8 10.1 9.0 14.0 9.8 7.6 6.1* 6.4* 6.6* 6.4*
Education Levele
≤ High school 13.0 14.4 12.5 17.2 14.0 7.6* 6.5** 6.5** 7.6** 6.8**
Some post-secondary 9.6 11.3 10.0 13.2 10.7 6.0** 5.4** 5.2** 6.9** 5.6**
Post-secondary 3.8 5.7 4.6 7.1 4.9 2.6 3.5* 3.0* 4.1* 3.2*
Employment Status
Employedf8.0 11.1 8.9 13.5 9.9 4.8** 5.2** 4.6** 6.6** 5.1**
Not employedg9.2 9.1 7.8 13.1 9.1 6.5* 5.0** 4.9* 6.8** 5.4**
Incomeh
<50, 000 10.0 13.8 12.2 14.2 12.8 7.1* 7.1** 7.3** 6.7** 7.1**
50, 000 – <100, 000 8.8 10.5 8.9 13.1 9.9 5.2** 5.1** 4.7** 6.4** 5.1**
≥100, 000 6.2 6.0 5.1 11.7 6.1 3.6* 2.5** 2.3** 6.9* 3.0**
Respiratory Diseasei
Yes 8.9 13.8 10.5 16.9 12.2 8.0 7.1* 6.5 10.1 7.4*
No 8.2 10.4 8.6 13.3 9.6 5.0** 5.1** 4.6** 6.5** 5.1**
(continued on next page)
Haig et al
Tob Regul Sci.™ 2019;5(5):463-479 471 DOI: https://doi.org/10.18001/TRS.5.5.7
(±SD) age of participants was 51.0 (±9.1) years and
BMI was 27.4 (±5.2) kg/m2. At enrollment, most
participants lived in an urban setting (77.4%), with
a partner (81.9%), were well-educated (77.1%),
and employed (74.1%). Greater proportions of
men and urban participants reported higher levels
of education, being employed, and greater house-
hold income, compared to women and rural par-
ticipants, respectively.
Prevalence of SHSe Pre-legislation versus
Post-legislation
e prevalence of workplace SHSe was signi-
cantly reduced from 19.8% to 13.2% post-leg-
islation (p < .001). We also observed signicant
reductions in post-legislation workplace SHSe
prevalence among men, women, urban and rural
participants, compared to pre-legislation (Table 3).
At both pre- and post-legislation, greater propor-
tions of men (29.9% and 20.9%, p < .001) and
rural (23.1% and 15.4%, p < .001) participants
reported workplace SHSe compared to women
(12.9% and 8.0%, p < .001) and urban (18.9%
and 12.6%, p < .001) participants, respectively.
e relative reduction in SHSe in the workplace
was higher for women compared to men (38.0% vs
30.1%, p < .001, results not shown), but was iden-
tical between urban and rural participants (33.3%,
non-signicant).
Similar to workplace exposure, we observed sig-
nicant reduction in home SHSe from 9.7% to
5.1% post-legislation (p < .001). Signicant re-
ductions in post-legislation home SHSe prevalence
were also observed among men, women, urban
Cardiovascular Diseasej
Yes 11.3 12.7 10.8 16.1 12.1 7.7* 6.0** 5.9** 9.2* 6.7**
No 7.1 9.9 7.9 12.5 8.9 4.1** 4.9** 4.3** 5.7** 4.6**
Note.
a: A total of 283 participants (62 men (M), 221 women (W)) had missing sociodemographic data.
b: Percentage of men (N = 4100), women (N = 7464), urban (N = 8952), rural (2612) and total (N = 11,564) participants exposed to SHS
at home within each sociodemographic domain. P-values were obtained by the McNemar’s test comparing pre- vs. post-legislation
exposure rates: * indicates p < .05 and ≥ .001, ** indicates p < .001, no symbol indicates p ≥ .05.
c: Geographic location was determined using postal codes, where “0” as the second character of the rst 3 digits of the postal code
indicates rural residence.
d: Married: Married/Living with a partner; Single: Single/never married; Separated: Divorced/Separated/Widowed.
e: ≤ High school: Did not complete grade 8, or completed grade 8 but not high school, or completed high school; Some post-secondary:
Completed some technical school/college training, or completed technical school/college training, or completed some part of
university degree; Post-secondary: Completed university degree, or completed some part of post-graduate university degree, or
completed university post-graduate degree.
f: Working full-time (30 hours or more per week), or working part-time (Less than 30 hours per week), or student, or self-employed.
g: Not employed but looking for work, or homemaker, or retired.
h: Annual household income ($CAD).
i: Self-reported personal history of respiratory disease; “Yes” indicates participant had ever been told by a doctor that they had one of
the following conditions: emphysema, chronic bronchitis and/or chronic obstructive pulmonary disease (COPD).
j: Self-reported personal history of cardiovascular disease; “Yes” indicates participant had ever been told by a doctor that they had
one of the following conditions: high blood pressure, heart attack, stroke or diabetes.
n/a = Not Applicable
*- p-values were obtained by the McNemar’s test comparing pre- versus post-legislation exposure rates: * indicates p < .05 and ≥ .001,
** indicates p < .001, no symbol indicates p ≥ .05
Table 4 (continued)
Percentage of Participants’ Characteristics in the Study Population (N = 11,564) who
Reported Secondhand Smoke Exposure in the Home Pre- (N = 1122) versus
Post-legislation (N = 593), Stratied by Sex or Geographic Locationa
Pre-legislation Post-legislation
Men
N = 338
%b
Women
N = 784
%b
Urban
N = 772
%b
Rural
N = 350
%b
Total
N = 1122
%b
Men
N = 210
%b
Wowmen
N = 383
%b
Urban
N = 420
%b
Rural
N = 173
%b
Total
N = 593
%b
8.2 10.5 8.6 13.4 9.7 5.1** 5.1** 4.7** 6.6** 5.1**
Reductions in Secondhand Smoke Exposure among Non-smokers Post-legislation
472
Table 5
Adjusted Odds Ratios (AOR) and 95% Condence Intervals (95% CI) from Generalized
Estimating Equation (GEE) for Home (N = 11,281) and Workplace (N = 6988) SHSea
Workplace exposure Home exposure
AOR [95% CI] AOR [95% CI]
Policy Periodb
Post vs PrecMen Women
0.59 [0.55 – 0.64] 0.59 [0.51-0.68] 0.45 [0.41-0.50]
Sex
Women vs Menc0.28 [0.25 – 0.32] n/a
Geographic Locationd
Rural vs Urbanc 1.04 [0.92 – 1.18] 1.33 [1.16 – 1.52]
Age
(continuous) 0.98 [0.97 – 0.99] 1.01 [1.00 – 1.01]
Marital Status
Divorced/Separated/Widowed vs
Married/Living with a partnerc
1.17 [0.99 – 1.39] 0.91 [0.76 – 1.09]
Single (never married) vs
Married/Living with a partnerc
1.26 [1.00 – 1.58] 0.83 [0.63 – 1.10]
Employment Status
Not employedf vs Employedcg n/a 0.72 [0.61 – 0.84]
Education Levele
≤High school vs Some post-secondaryc1.28 [1.12 – 1.46] 1.17 [1.02 – 1.34]
Post-secondary completed vs Some post-secondaryc0.37 [0.32 – 0.43] 0.56 [0.47 – 0.66]
Incomeh
<50, 000 vs 50, 000 - <100, 000c 1.28 [1.11 – 1.48] 1.27 [1.10 – 1.47]
≥100, 000 vs 50, 000 - <100, 000c0.70 [0.62 – 0.80] 0.66 [0.55 – 0.78]
Respiratory Diseasei
Yes vs Noc 1.27 [0.92 – 1.75]1.15 [0.84 – 1.57]
Cardiovascular Diseasej
Yes vs Noc1.16 [1.02 – 1.32] 1.32 [1.16 – 1.51]
Note.
a: A total of 283 (62 men, 221 women) and 145 (40 men, 105 women) had missing covariate data for home and
workplace SHSe, respectively, and were excluded from the corresponding analyses.
b: Binary time variable (1 = pre-legislation period, 2 = post-legislation period).
c: Reference category.
d: Geographic location was determined using postal codes, where “0” as the second character of the rst 3 digits of
the postal code indicates rural residence.
e: ≤ High school: Did not complete grade 8, or completed grade 8 but not high school, or completed high school;
Some post-secondary: Completed some technical school/college training, or completed technical school/college
training, or completed some part of university degree; Post-secondary completed: Completed university degree,
or completed some part of post-graduate university degree, or completed university post-graduate degree.
f: Not employed, but looking for work, or homemaker, or retired.
g: Working full-time (30 hours or more per week), or working part-time (Less than 30 hours per week), or student,
or self-employed.
h: Annual household income ($CAD).
i: Participant had ever been told by a doctor that they had one of the following conditions: emphysema, chronic
bronchitis and/or chronic obstructive pulmonary disease (COPD).
j: Participant had ever been told by a doctor that they had one of the following conditions: high blood pressure,
heart attack, stroke or diabetes.
n/a = Not applicable to the exposure domain.
Haig et al
Tob Regul Sci.™ 2019;5(5):463-479 473 DOI: https://doi.org/10.18001/TRS.5.5.7
and rural participants, compared to pre-legislation
(Table 4). At both pre-and post-legislation, greater
proportions of rural (13.4% and 6.6%) partici-
pants reported home SHSe compared to urban
participants (8.6% and 4.7%, both p < .001).
Greater proportions of women reported home
SHSe compared to men pre-legislation (10.5% vs
8.2%, p < .0324); the proportions were similar be-
tween men and women post-legislation (5.1% vs
5.1%, non-signicant). e relative decrease (re-
sults not shown) in SHSe in the home was higher
for women compared to men (51.4% vs 37.8 %, p
< .001) and was also higher for rural compared to
urban participants (50.7% vs 45.3 %, p < .001),
respectively.
Association between the 2008 Smoke-free
Legislation and Reduction in Workplace and
Home SHSe
With the exception of household income and
employment status (Kappa values = 0.5), all Kap-
pa values were ≥ 0.7, indicating that participants’
sociodemographic characteristics did not substan-
Figure 1
GIS Maps Showing Workplace (A) and Home (B) SHS Exposure Rates before and after the
2008 Alberta Tobacco Reduction Act Implementation
Note.
Municipalities with < 20 participants were combined with adjacent municipalities.
Minimum number of participants per municipality, N = 20; maximum number, N = 424; Mean (SD) = 94 (72).
“Percent exposed” in each municipality was obtained by dividing the number of participants exposed to SHS by the
total number of participants within that municipality.
Reductions in Secondhand Smoke Exposure among Non-smokers Post-legislation
474
tially change between pre- and post-legislation pe-
riods. e GMM models that adjusted for pre- and
post-legislation time-dependent covariates yielded
similar results to GEE models adjusted for pre-leg-
islation covariates only (results not shown). ere-
fore, the results from the GEE models that only
adjusted for covariates at enrollment are presented.
Compared to pre-legislation, the odds of work-
place SHSe during post-legislation signicantly de-
clined (AOR = 0.59, 95% CI: 0.55-0.64) (Table 5).
ere was a statistically signicant interaction be-
tween sex and policy period for home exposure. For
post-legislation SHSe at home, compared to pre-
legislation, women had an AOR of 0.45 (Table 5)
whereas men had an AOR of 0.59. Hence, women
had a decreased odds of exposure for home SHSe
compared with men post-legislation. e spatial
distributions of these ndings are depicted in the
GIS maps, where the post-legislation maps indicate
generally lower exposure rates (lighter shades) for
home and workplace settings (Figure 1).
Age-adjusted Prevalence of SHSe in ATP
Compared to Alberta CCHS Data
Similar to ATP, Alberta-specic CCHS results
showed signicant reductions in post-legislation
Table 6
Age-adjusted Pre- and Post-legislation Secondhand Smoke Exposure at Home and Public
Places Reported by Alberta’s Tomorrow Project (ATP) Participants and Albertan Respondents
to the Canadian Community Health Survey (CCHS), Stratied by Sex
Exposure at home
ATP
(N = 11,564)
Alberta CCHSd
(N = 18,394 (pre), N = 24,764 (post))
Men Women Total Men Women Total
%a%a%a%b%b%b
Pre-legislationf8.0 10.4 9.6 17.3 16.8 17.0
Post-legislationg4.8 5.1 5.0 9.3 8.9 9.1
Exposure in public placese
ATP
(N = 11,564)
Alberta CCHS
( N = 14,889 (pre))
Men Women Total Men Women Total
%a%a%a%c%c%c
Pre-legislationf77.6 73.7 75.2 27.2 18.9 22.2
Post-legislation - - -
Note.
a: Age-adjusted percentage of ATP participants exposed to secondhand smoke within men (N = 4100), women (N = 7464)
and total (N = 11,564).
b: Age-adjusted percentage of Albertan CCHS participants exposed to secondhand smoke within men (pre-legislation
N = 8719, post-legislation N = 11,367), women (pre-legislation N = 9675, post-legislation N = 13,397), total (pre-
legislation N = 18,394, post-legislation N = 24,764).
c: Age-adjusted percentage of Albertan CCHS participants exposed to secondhand smoke within men (pre-legislation
N = 6971), women (pre-legislation N = 7918) and total (pre-legislation N = 14,889).
d: The CCHS home SHSe was captured using the following question: “Including both household members and regular
visitors, does anyone smoke inside your home, every day or almost every day.” This question was used as an
approximation for exposure to secondhand smoke in the home.
e: At post-legislation, questions about public place SHSe was not comparable between CCHS and ATP because they
were asked differently.
f: Alberta CCHS pre-legislation data includes cycles 2001, 2003, 2005 and 2007.
g: Alberta CCHS post-legislation data includes cycles 2009, 2010, 2011, 2012, 2013 and 2014.
Haig et al
Tob Regul Sci.™ 2019;5(5):463-479 475 DOI: https://doi.org/10.18001/TRS.5.5.7
home SHSe among men and women, compared to
pre-legislation (Table 6). Despite overall similari-
ties, smaller proportions of men (8.0% and 4.8%)
and women (10.4% and 5.1%) in the ATP cohort
reported home SHSe pre- and post-legislation
compared to CCHS men (17.3% and 9.3%) and
women (16.8% and 8.9%), respectively.
DISCUSSION
In the current study, we evaluated the association
between implementation of the 2008 Alberta To-
bacco Reduction Act and changes in self-reported
SHSe among non-smokers using longitudinal
cohort data. We observed signicant reductions
in SHSe following legislation implementation at
both workplaces and homes, even though smoking
at home was not directly targeted by the legisla-
tion. Moreover, results from both ATP and Alberta
CCHS cohorts showed comparable declines in
home SHSe (48% vs 46%; unadjusted values).
Our ndings are comparable to previously pub-
lished studies. Indeed, several studies have demon-
strated that tobacco control legislation is associated
with a reduction in SHSe;17,19-27,29,30,50-53 however,
only a few followed the same participants over
time.17,21-23,29,52 Additionally, a number of these stud-
ies were limited by small sample size,21,22,29,50 short
follow-up time (eg, 6 weeks - 12 months),17,21,22,29,30
or a single exposure domain.21-23,29,50,51,53,54 Finally, a
limited number of studies evaluated SHSe among
non-smokers19,26,27,30,55 compared to active smokers
only17,23,56 or a combination of both smokers and
non-smokers.24,25,28 On a population level, a repeat-
ed cross-sectional study of non-smokers in Spain
found larger declines of self-reported workplace
and home SHSe post-legislation (60% and 64%
compared to 41% and 42% in the current study,
respectively).19 However, the Spanish study only
adjusted for age in their analysis, whereas we have
adjusted for multiple covariates. Another repeated
cross-sectional study in Scotland reported 68%
decline of self-reported workplace SHSe among
non-smokers. However, unlike the ndings in the
present study, the Scottish study did not observe a
statistically signicant change in home SHSe. e
authors concluded that there was no evidence of
displacement of smoking into the home, whereas
we have shown signicant reduction in home SHSe
in addition to that observed in the workplace. e
dierence observed in home SHSe could be attrib-
uted to adjustments for fewer known confounders
(eg, sex, years in education, and deprivation catego-
ry of residence) and shorter follow-up time (eg, one
year) in the Scotland study, compared to the ATP
cohort.57 e ATP cohort is expected to better cap-
ture the association between smoke-free legislation
and SHSe rate given its longitudinal nature, which
compares the same participants over time. More-
over, the Scotland study queried participants about
past week exposure which may not be an accurate
indicator of typical exposure compared to the ATP
cohort which asked about past year exposure (pre-
legislation). In the present analysis, we observed
that both home and workplace SHSe were higher
in rural compared to urban areas, which is consis-
tent with previously published literature in Alber-
ta58 and elsewhere.59,60 ese ndings suggest that
rural communities may face specic barriers that
needs to be addressed by policymakers to increase
eectiveness of smoke-free legislation. Our ndings
agree with previous reports that associated rurality
of residence with reduced eectiveness of smoking
bans.38,58 We also found that greater proportions of
women reported pre-legislation home SHSe com-
pared to men, whereas similar proportions of men
and women reported home SHSe post-legislation.
On the other hand, greater proportions of men ex-
perienced workplace SHSe compared to women,
both pre- and post-legislation. ese ndings sup-
port other published literature and also suggest
that smoke exposure is sex divisive and more work
needs to be done to understand the complexity of
sex in tobacco use and exposure.36,37
Our ndings are consistent with cross-sectional
studies but provide further evidence from a longitu-
dinal cohort supporting the benecial inuence of
tobacco control legislations on SHSe. Furthermore,
the data suggest that a reduction in exposure in
workplaces does not correspond to increased home
exposure and actually home exposure is reduced as
well, similar to previous studies.20,23,25,31,35,52,53,61 In
fact, smoke-free legislation targeting work and pub-
lic places may encourage smokers to refrain from
smoking at home as well.23,34,53 Overall, our results
support a positive association between smoke-free
legislation and reductions in SHSe in workplaces,
and even in homes, a domain which is not directly
regulated by the legislation. ese ndings support
the ‘social diusion hypothesis’ which suggests that
Reductions in Secondhand Smoke Exposure among Non-smokers Post-legislation
476
changes in social norms due to public and work-
place smoking bans may result in an increased up-
take in voluntary home smoking restrictions.62,63
e timing of this research is particularly relevant
to ongoing discussions around potential conse-
quences of the Canada-wide legalization of recre-
ational cannabis that took eect on October 17,
2018 (e Cannabis Act, Bill C-45).65 e Can-
nabis Act allows local jurisdictions to designate
cannabis consumption areas in public places. is
likely will increase secondhand cannabis smoke
exposure among non-users at work and public
settings, but especially at home since people will
be allowed to consume cannabis on their private
properties. Indeed, it is expected that the Cannabis
Act implementation will result in increased second-
hand cannabis smoke exposure among children, es-
pecially under 6 years of age.64
Strengths and Limitations
e current study followed the same large sam-
ple of non-smokers who completed both pre- and
post-legislation questionnaires. is longitudinal
nature represents a signicant strength compared
to studies that were restricted to assessing cross-
sectional data,24-27,30,54,56 sometimes with dierences
in demography30 and low response/compliance
rates.22,23,30,52,53 e median (IQR) follow-up time
of 5.9 (4.1) years with at least 2 years between col-
lections is long compared to other studies,21,22,30
and provides a clear distinction between pre- and
post-legislation SHSe.
One other strength is that we captured informa-
tion on domains of SHSe that are often absent in
other studies evaluating SHSe. Unfortunately, we
were unable to accurately quantify the associated
change in exposure rates in public places pre- and
post-legislation due to the dierent ways questions
were asked at the 2 time points in the ATP cohort.
Descriptive results from CCHS Alberta showed
lower prevalence of pre-legislation public SHSe
compared to ATP data. However, it is possible that
ATP may have overestimated pre-legislation SHSe
in the public places.
ATP has well documented quality assurance pro-
cedures and validation of the data with few miss-
ing values.42 SHSe was captured using self-report
without biomarker validation. Despite this, self-re-
ported SHSe has been previously validated against
cotinine in epidemiological studies.65-68 Although
there are strengths to using cotinine measurement
for nicotine exposure, it is logistically burdensome,
time-delayed and economically infeasible for large
cohort studies.68 Future studies would benet from
pairing biochemical validation with self-reported
data. Moreover, self-report is susceptible to varying
forms of bias including recall and underreporting
bias. However, collection of pre-legislation SHSe
data occurred prospectively prior to legislation im-
plementation; therefore, the problem of recall bias
and reverse causality, one of the major limitations
in cross-sectional studies, is unlikely to be a concern
in our study. It is possible that there is underreport-
ing in our estimates but the degree of underreport-
ing is believed to be relatively constant; therefore,
underreporting is believed to have limited eect on
estimates of exposure reduction overtime.69 Unfor-
tunately, we were unable to evaluate other known
inuencers of SHSe including temporal changes
in smoking habits and prevalence and changes in
social norms surrounding tobacco use.70 Addition-
ally, we were also unable to evaluate whether re-
duction in both active and passive smoke exposure
following legislation is associated with lower inci-
dence of chronic disease, due to insucient follow-
up time. Lastly, our study inclusion criteria meant
some ATP participants were not included in our
analyses. However, the longitudinal nature of this
cohort supports future analyses evaluating trends
in health outcomes over time for both smokers and
non-smokers.
We were able to adjust for a wide range of po-
tential covariates using dierent statistical models.
Similar results were obtained from the GMM and
GEE statistical models, and after excluding partici-
pants who were recruited as second in household
or those who could have been aected by munici-
pality bylaws implemented before the provincial
bylaw, highlighting the robustness of the results.
Moreover, the decision to limit the analysis to non-
smokers allowed for accurate quantication of true
SHSe without concern for reverse causality or con-
founding that may arise by the inclusion of current
smokers. However, caution should be taken when
interpreting the ndings of this study; these results
are associations only and as such cannot confer
causation, like most epidemiologic studies. Our
study design precludes a direct measurement of the
eect of the law.
Haig et al
Tob Regul Sci.™ 2019;5(5):463-479 477 DOI: https://doi.org/10.18001/TRS.5.5.7
IMPLICATIONS FOR TOBACCO
REGULATION
Our ndings provide strong evidence that the
2008 Alberta Tobacco Reduction Act is associ-
ated with declines in SHSe among non-smoking
individuals at home and at work. As there are few
longitudinal research studies in North American
populations, we were able to evaluate the impact
on non-smokers in the general population across
time and a variety of exposure settings following
an implementation of smoke-free legislation. Post-
legislation SHSe rates were signicantly lower than
pre-legislation rates, including domains targeted
(workplace) and not targeted (home) by smoke-
free legislation. Our novel ndings include sup-
port against exposure displacement into homes.
ese ndings support further eorts to develop
stringent smoke-free legislation, as called for in the
WHO FCTC, as a population-level intervention
with strict enforcement practices to protect vulner-
able populations from the eects of SHSe and re-
duce future burden of cancer and chronic diseases.
Human Subjects Statement
is study received ethics approval from the
Health Research Ethics Board of Alberta (HREBA)
– Cancer Committee, Certication le number
HREBA.CC-16-0609. All participants included in
the ATP cohort returned a completed and signed
consent form and consented to future active and
passive follow-up for up to 50 years or until the age
of 85. Participants also consented to future linkage
with administrative healthcare and cancer registry
data and for their data to be used in future research.
ATP participants were free to withdraw at any time
during the duration of their study participation.
Conict of Interest Disclosure Statement
All authors of this article declare they have no
conicts of interest.
Acknowledgements
Alberta’s Tomorrow Project is only possible due
to the commitment of its research participants, its
sta and its funders: Alberta Health and the Alber-
ta Cancer Prevention Legacy Fund, Alberta Cancer
Foundation, Canadian Partnership Against Can-
cer and substantial in kind funding from Alberta
Health Services. e views expressed herein repre-
sent the views of the authors and not of Alberta’s
Tomorrow Project or any of its funders.
References
1. World Health Organization. Report on the Global Tobacco
Epidemic: the MPOWER package. Available at: https://
www.who.int/tobacco/mpower/gtcr_download/en/. Ac-
cessed May 10, 2018.
2. World Bank Group. Global Monitoring Report 2009. Avail-
able at: http://siteresources.worldbank.org/INTGLO-
MONREP2009/Resources/5924349-1239742507025/
GMR09_book.pdf. Accessed July 26, 2019.
3. Wong SL, Malaison E, Hammond D, Leatherdale ST.
Secondhand smoke exposure among Canadians: cotinine
and self-report measures from the Canadian Health Mea-
sures Survey 2007-2009. Nicotine Tob Res. 2013;15:693-
700.
4. World Health Organization. Tobacco Fact Sheet. Available
at: https://www.who.int/news-room/fact-sheets/detail/
tobacco. Accessed January 22 2019.
5. World Health Organization. WHO Report on the Global
Tobacco Epidemic 2017: Monitoring Tobacco Use and Pre-
vention Policies. Available at: https://www.who.int/tobac-
co/global_report/2017/en/. Accessed July 26, 2019.
6. World Health Organization. WHO Framework Conven-
tion on Tobacco Control (FCTC). Geneva, Switzerland:
WHO Press; 2003:1-44.
7. Valdés Salgado R. e framework convention on tobacco
control. Salud Publica Mex. 2003;45:153-154.
8. Frazer K, Callinan JE, McHugh J, et al. Legislative smok-
ing bans for reducing harms from secondhand smoke
exposure, smoking prevalence and tobacco consumption.
Cochrane Database Syst Rev. 2016;2:CD005992.
9. Gleich F, Mons U, Pötschke-Langer M. Air contamina-
tion due to smoking in German restaurants, bars, and
other venues--before and after the implementation of a
partial smoking ban. Nicotine Tob Res. 2011;13:1155-
1160.
10. Naiman AB, Glazier RH, Moineddin R. Is there an im-
pact of public smoking bans on self-reported smoking
status and exposure to secondhand smoke? BMC Public
Health. 2011;11:146.
11. López MJ, Nebot M, Schiano A, et al. Two-year impact
of the Spanish smoking law on exposure to secondhand
smoke: evidence of the failure of the ‘Spanish model’. Tob
Control. 2012;21:407-411.
12. Vicedo-Cabrera AM, Röösli M, Radovanovic D, et al.
Cardiorespiratory hospitalisation and mortality reduc-
tions after smoking bans in Switzerland. Swiss Med Wkly.
2017;146:w14381.
13. Gaudreau K, Sanford CJ, Cheverie C, McClure C. e
eect of a smoking ban on hospitalization rates for car-
diovascular and respiratory conditions in Prince Edward
Island, Canada. PLoS One. 2013;8:e56102.
14. IARC Working Group on the Evaluation of Carcinogenic
Risks to Humans. Household use of solid fuels and high-
temperature frying. IARC Monogr Eval Carcinog Risks
Hum. 2010;95:1-430.
15. Azagba S, Asbridge M. e association between smoking
Reductions in Secondhand Smoke Exposure among Non-smokers Post-legislation
478
bans and nicotine dependence: a longitudinal analysis of
current smokers in Canada. Addict Behav. 2013;38:2817-
2820.
16. Mills AL, Messer K, Gilpin EA, Pierce JP. e eect of
smoke-free homes on adult smoking behavior: a review.
Nicotine Tob Res. 2009;11:1131-1141.
17. Fong GT, Hyland A, Borland R, et al. Reductions in
tobacco smoke pollution and increases in support for
smoke-free public places following the implementation
of comprehensive smoke-free workplace legislation in the
Republic of Ireland: ndings from the ITC Ireland/UK
Survey. Tob Control. 2006;15(Suppl 3):iii51-iii-58.
18. Farkas AJ, Gilpin EA, White MM, Pierce JP. Association
between household and workplace smoking restrictions
and adolescent smoking. JAMA. 2000;284:717-722.
19. Fernández E, Fu M, Pérez-Ríos M, et al. Changes in
secondhand smoke exposure after smoke-free legislation
(Spain, 2006-2011). Nicotine Tob Res. 2017;19:1390-
1394.
20. Tsai Y-W, Chang L-C, Sung H-Y, et al. e impact of
smoke-free legislation on reducing exposure to second-
hand smoke: dierences across gender and socioeconomic
groups. Tob Control. 2015;24:62-69.
21. Mulcahy M, Evans DS, Hammond SK, et al. Secondhand
smoke exposure and risk following the Irish smoking ban:
an assessment of salivary cotinine concentrations in ho-
tel workers and air nicotine levels in bars. Tob Control.
2005;14:384-388.
22. Farrelly MC, Nonnemaker JM, Chou R, et al. Changes in
hospitality workers’ exposure to secondhand smoke fol-
lowing the implementation of New York’s smoke-free law.
Tob Control. 2005;14:236-341.
23. Mons U, Nagelhout GE, Allwright S, et al. Impact
of national smoke-free legislation on home smoking
bans: ndings from the International Tobacco Control
Policy Evaluation Project Europe Surveys. Tob Control.
2013;22:e2-e9.
24. Azagba S. Eect of smoke-free patio policy of restaurants
and bars on exposure to second-hand smoke. Prev Med
(Baltim). 2015;76:74-78.
25. Ye X, Yao Z, Gao Y, et al. Second-hand smoke exposure in
dierent types of venues: before and after the implemen-
tation of smoke-free legislation in Guangzhou, China.
BMJ Open. 2014;4:e004273.
26. Sureda X, Martínez-Sánchez JM, Fu M, et al. Impact of
the Spanish smoke-free legislation on adult, non-smoker
exposure to secondhand smoke: cross-sectional surveys
before (2004) and after (2012) legislation. PLoS One.
2014;9:e89430.
27. Martínez-Sánchez JM, Gallus S, Zuccaro P, et al. Ex-
posure to secondhand smoke in Italian non-smokers 5
years after the Italian smoking ban. Eur J Public Health.
2012;22:707-712.
28. Hyland A, Hassan LM, Higbee C, et al. e impact of
smokefree legislation in Scotland: results from the Scot-
tish ITC: Scotland/UK longitudinal surveys. Eur J Public
Health. 2009;19:198-205.
29. Bondy SJ, Zhang B, Kreiger N, et al. Impact of an indoor
smoking ban on bar workers’ exposure to secondhand
smoke. J Occup Environ Med. 2009;51:612-619.
30. Haw SJ, Gruer L. Changes in exposure of adult non-
smokers to secondhand smoke after implementation of
smoke-free legislation in Scotland: national cross section-
al survey. BMJ. 2007;335:549.
31. Jarvis MJ, Sims M, Gilmore A, Mindell J. Impact of
smoke-free legislation on children’s exposure to second-
hand smoke: cotinine data from the Health Survey for
England. Tob Control. 2012;21:18-23.
32. Adda J, Comaglia F. e eect of bans and taxes on pas-
sive smoking. Am Econ J Appl Econ. 2010;2:1-32.
33. Ho SY, Wang MP, Lo WS, et al. Comprehensive smoke-
free legislation and displacement of smoking into the
homes of young children in Hong Kong. Tob Control.
2010;19:129-133.
34. Cheng K-W, Glantz SA, Lightwood JM. Association be-
tween smokefree laws and voluntary smokefree-home
rules. Am J Prev Med. 2011;41:566-572.
35. Bhaumik S. Ban on smoking in workplaces in India has
led to more smoke free homes. BMJ. 2013;346:f2186.
36. Greaves LJ, Hemsing NJ. Sex, gender, and secondhand
smoke policies: implications for disadvantaged women.
Am J Prev Med. 2009;37:S131-S137.
37. World Health Organization. Global Estimate of
the Burden of Disease from Second-hand Smoke.
Available at: http://apps.who.int/iris/bitstream/
handle/10665/44426/9789241564076_eng.
pdf?sequence=1. Accessed May 15, 2018.
38. American Lung Association. Cutting Tobacco’s Rural Roots:
Tobacco Use in Rural Communities. Available at: https://
www.lung.org/assets/documents/research/cutting-tobac-
cos-rural-roots.pdf. Accessed July 26, 2019.
39. Government of Alberta. Tobacco and Smoking Reduction
Act, Statutes of Alberta. 2005. Available at: http://www.
qp.alberta.ca/documents/Acts/T03P8.pdf. Accessed Jan-
uary 22, 2018.
40. Government of Alberta. Bill 45: Smoke-free Places (To-
bacco Reduction). 2007. Available at: http://www.as-
sembly.ab.ca/ISYS/LADDAR_les/docs/bills/bill/legis-
lature_26/session_3/20070307_bill-045.pdf. Accessed
January 22, 2018.
41. Bryant H, Robson PJ, Ullman R, et al. Population-based
cohort development in Alberta, Canada: a feasibility
study. Chronic Dis Can. 2006;27:51-59.
42. Robson PJ, Solbak NM, Haig TR, et al. Design, methods
and demographics from phase I of Alberta’s Tomorrow
Project cohort: a prospective cohort prole. CMAJ Open.
2016;4:E515-E527.
43. Friedenreich CM, Courneya KS, Bryant HE. e lifetime
total physical activity questionnaire: development and re-
liability. Med Sci Sports Exerc. 1998;30:266-274.
44. Friedenreich CM, Courneya KS, Neilson HK, et al. Reli-
ability and validity of the Past Year Total Physical Activity
Questionnaire. Am J Epidemiol. 2006;163:959-970.
45. National Institutes of Health. Diet History Questionnaire.
Available at: https://epi.grants.cancer.gov/DHQ/. Ac-
cessed July 26, 2019.
46. Csizmadi I, Kahle L, Ullman R, et al. Adaptation and
evaluation of the National Cancer Institute’s Diet History
Questionnaire and nutrient database for Canadian popu-
lations. Public Health Nutr. 2007;10:88-96.
47. Ye M, Robson PJ, Eurich DT, et al. Cohort Prole: Al-
berta’s Tomorrow Project. Int J Epidemiol. 2017;46:1097-
1098l.
48. Statistics Canada. Appendix Table A: Smoking Ban Legisla-
Haig et al
Tob Regul Sci.™ 2019;5(5):463-479 479 DOI: https://doi.org/10.18001/TRS.5.5.7
tion in Canadian Provinces and Municipal Bylaws in Select-
ed Cities. Available at: http://www.statcan.gc.ca/pub/82-
003-x/2006008/article/smoking-tabac/t/4060721-eng.
htm. Accessed January 22, 2018.
49. Statistics Canada. Canadian Community Health Survey:
Annual Component. 2013. Available at: https://www150.
statcan.gc.ca/n1/daily-quotidien/140612/dq140612b-
eng.htm. Accessed January 22, 2018.
50. Valente P, Forastiere F, Bacosi A, et al. Exposure to ne
and ultrane particles from secondhand smoke in public
places before and after the smoking ban, Italy 2005. Tob
Control. 2007;16:312-317.
51. Fernando D, Fowles J, Woodward A, et al. Legislation re-
duces exposure to second-hand tobacco smoke in New
Zealand bars by about 90%. Tob Control. 2007;16:235-
238.
52. Hyland A, Higbee C, Hassan L, et al. Does smoke-free
Ireland have more smoking inside the home and less in
pubs than the United Kingdom? Findings from the inter-
national tobacco control policy evaluation project. Eur J
Public Health. 2008;18:63-65.
53. Edwards R, omson G, Wilson N, et al. After the smoke
has cleared: evaluation of the impact of a new national
smoke-free law in New Zealand. Tob Control. 2008;17:e2.
54. Akhtar PC, Haw SJ, Currie DB, et al. Smoking restric-
tions in the home and secondhand smoke exposure
among primary schoolchildren before and after introduc-
tion of the Scottish smoke-free legislation. Tob Control.
2009;18:409-415.
55. Tsai Y-W, Chang L-C, Sung H-Y, et al. e impact of
smoke-free legislation on reducing exposure to second-
hand smoke: dierences across gender and socioeconomic
groups. Tob Control. 2013;2009:78-79.
56. Lee JT, Glantz SA, Millett C. Eect of smoke-free leg-
islation on adult smoking behaviour in England in
the 18 months following implementation. PLoS One.
2011;6:e20933.
57. Haw SJ, Gruer L. Changes in exposure of adult non-
smokers to secondhand smoke after implementation of
smoke-free legislation in Scotland: national cross section-
al survey. BMJ. 2007;335:549.
58. Li FX, Robson PJ, Ashbury FD, et al. Smoking frequen-
cy, prevalence and trends, and their socio-demographic
associations in Alberta, Canada. Can J Public Health.
2009;100:453-458.
59. Singh A, Sahoo N. Urban-rural dierentials in the factors
associated with exposure to second-hand smoke in India.
BMJ Open. 2013;3:e003542.
60. Harris JK, Geremakis C, Moreland-Russell S, et al. De-
mographic and geographic dierences in exposure to
secondhand smoke in Missouri workplaces, 2007-2008.
Prev Chronic Dis. 2011;8:A135.
61. Borland R, Yong H-H, Cummings KM, et al. Determi-
nants and consequences of smoke-free homes: ndings
from the International Tobacco Control (ITC) Four
Country Survey. Tob Control. 2006;15(Suppl 3):iii42-
iii50.
62. Callinan JE, Clarke A, Doherty K, Kelleher C. Legislative
smoking bans for reducing secondhand smoke exposure,
smoking prevalence and tobacco consumption. Cochrane
Database Syst Rev. 2010;CD005992.
63. Frazer K, Callinan JE, McHugh J, et al. Legislative smok-
ing bans for reducing harms from secondhand smoke
exposure, smoking prevalence and tobacco consumption.
Cochrane Database Syst Rev. 2016;2:CD005992.
64. Government of Canada. Bill C-45: An Act Respecting Can-
nabis and to amend the Controlled Drugs and Substances
Act, the Criminal Code and other Acts. 2017. Available at:
http://www.parl.ca/Content/Bills/421/Government/C-
45/C-45_1/C-45_1.pdf. Accessed March 10, 2019.
65. Okoli CTC, Kelly T, Hahn EJ. Secondhand smoke
and nicotine exposure: a brief review. Addict Behav.
2007;32:1977-1988.
66. Avila-Tang E, Elf JL, Cummings KM, et al. Assessing
secondhand smoke exposure with reported measures. Tob
Control. 2013;22:156-163.
67. Frazer K, Callinan JE, McHugh J, et al. Legislative smok-
ing bans for reducing harms from secondhand smoke
exposure, smoking prevalence and tobacco consumption.
Cochrane Database Syst Rev. 2016;2:CD005992.
68. Prochaska JJ, Grossman W, Young-Wol KC, Benowitz
NL. Validity of self-reported adult secondhand smoke ex-
posure. Tob Control. 2015;24:48-53.
69. Connor Gorber S, Schoeld-Hurwitz S, Hardt J, et al. e
accuracy of self-reported smoking: a systematic review of
the relationship between self-reported and cotinine-as-
sessed smoking status. Nicotine Tob Res. 2009;11:12-24.
70. Cummings KM, Proctor RN. e changing public im-
age of smoking in the United States: 1964-2014. Cancer
Epidemiol Biomarkers Prev. 2014;23:32-36.
