Effect of a Digital Social Media Campaign on Young Adult Smoking Cessation

Abstract

Social Media (SM) may extend the reach and impact for smoking cessation among young adult smokers. To-date little research targeting young adults has been done on the use of social media to promote quitting smoking. We assessed the effect of an innovative multi-component web-based and SM approach known as Break-it-Off (BIO) on young adult smoking cessation.
The study employed a quasi-experimental design with baseline and 3-month follow-up data from 19 to 29 year old smokers exposed to BIO (n=102 at follow-up) and a comparison group of Smokers' Helpline (SHL) users (n=136 at follow-up). Logistic regression analysis assessed differences between groups on self-reported 7-day and 30-day point prevalence cessation rates, adjusting for ethnicity, education level, and cigarette use (daily or occasional) at baseline.
The campaign reached 37,325 unique visitors with a total of 44,172 visits. BIO users had significantly higher 7-day and 30-day quit rates compared with users of SHL. At three month follow-up, BIO participants (32.4%) were more likely than SHL participants (14%) to have quit smoking for 30 days (odds ratio = 2.95, 95% CI = 1.56- 5.57, p < .001)and BIO participants (91%) were more likely than SHL participants (79%) to have made a quit attempt (odds ratio = 2.69, 95% CI = 1.03-6.99, p = .04).
The reach of the campaign and findings on quitting success indicate that a digital/social media platform can complement the traditional SHL cessation service for young adult smokers seeking help to quit.
© The Author 2015. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Full text

© The Author 2015. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved.
For permissions, please e-mail: journals.permissions@oup.com.
351
Nicotine & Tobacco Research, 2016, 351–360
doi:10.1093/ntr/ntv119
Original investigation
Advance Access publication June 4, 2015
Introduction
In Canada, young adults (19–29) have the highest rate of smoking
at 24.4%,1 but their use of cessation services and products is low.2
This may reect the common belief among young adults that the
best way to quit is on their own.3 More than half of young adult
smokers want to quit, but few are successful,4,5 suggesting that there
may be a misalignment between the needs of young people and the
cessation methods available to them. However, those who are able to
quit before the age of 30 will have a life expectancy nearly as long as
those who never smoked,6 which holds hope for achieving popula-
tion health impact if effective, accessible and attractive interventions
are developed for this population. Innovative intervention solutions
for reaching young adult smokers are needed.
Web and mobile phone-based interventions have shown prom-
ise for encouraging smoking cessation in young adults7 and social
media (SM) tools and technologies have become a key resource for
young adults. SM is dened as “any electronic, networked informa-
tion resource that derives its principal value from user contributions,
Original investigation
Effect of a Digital Social Media Campaign on
Young Adult Smoking Cessation
Neill BruceBaskerville PhD1,2, SundayAzagba PhD1,2,
CameronNorman PhD3,4, KyleMcKeown BA5, K. StephenBrown PhD2,6
1Applied Health Sciences, University of Waterloo, Waterloo, ON, Canada; 2Propel Centre for Population Health
Impact, University of Waterloo, Waterloo, ON, Canada; 3Dalla Lana School of Public Health, University of Toronto,
Toronto, ON, Canada; 4CENSE Research + Design, Toronto, ON, Canada; 5Smokers’ Helpline, Canadian Cancer
Society, Toronto, ON, Canada; 6Statistics and Actuarial Sciences, University of Waterloo, Waterloo, ON, Canada
Corresponding Author: Neill Bruce Baskerville, PhD, Propel Centre for Population Health Impact, 200 University Ave.
West, University of Waterloo, ON N2L 3G1, Canada. Telephone: 519-888-4567 ext. 35236; Fax: 519-886-6424;
E-mail: nbbaskerville@uwaterloo.ca
Abstract
Introduction: Social media (SM) may extend the reach and impact for smoking cessation among
young adult smokers. To-date, little research targeting young adults has been done on the use of
SM to promote quitting smoking. We assessed the effect of an innovative multicomponent web-
based and SM approach known as Break-it-Off (BIO) on young adult smoking cessation.
Methods: The study employed a quasi-experimental design with baseline and 3-month follow-up
data from 19 to 29-year old smokers exposed to BIO (n=102 at follow-up) and a comparison group
of Smokers’ Helpline (SHL) users (n = 136 at follow-up). Logistic regression analysis assessed
differences between groups on self-reported 7-day and 30-day point prevalence cessation rates,
adjusting for ethnicity, education level, and cigarette use (daily or occasional) at baseline.
Results: The campaign reached 37 325 unique visitors with a total of 44 172 visits. BIO users had
significantly higher 7-day and 30-day quit rates compared with users of SHL. At 3-month follow-
up, BIO participants (32.4%) were more likely than SHL participants (14%) to have quit smoking for
30days (odds ratio=2.95, 95% CI=1.56 to 5.57, P < .001) and BIO participants (91%) were more
likely than SHL participants (79%) to have made a quit attempt (odds ratio=2.69, 95% CI=1.03 to
6.99, P=.04).
Conclusion: The reach of the campaign and findings on quitting success indicate that a digital/SM
platform can complement the traditional SHL cessation service for young adult smokers seeking
help to quit.
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from

352 Nicotine & Tobacco Research, 2016, Vol. 18, No. 3
engagement, and interaction”.8 Among Canadian Internet users aged
16 to 24, 91% use the SM platform Facebook, 73% use instant mes-
saging, and 33% participate in web based discussions on microblog-
ging sites such as Twitter or post images on sites such as YouTube9
suggesting that these tools may be a promising means to engage
young smokers.
Several reviews of web-based research have noted the importance
of tailoring content to individuals and providing interactive compo-
nents to promote greater engagement with the site and greater success
in quitting.10–13 Other research suggests the importance of support
from peers, and the constant updating of website content.10,14,15
Arecent systematic review of mobile interventions found that text
messaging was the most commonly identied smoking cessation
intervention and pooled analysis showed that it doubled cessation
rates.16 However, there is little evidence concerning the effectiveness
of SM interventions for changing health behaviors using platforms
such as Facebook, Twitter, or YouTube.8,10,17–19 Furthermore, a recent
systematic review of SM-based interventions for health promotion
noted the scarcity of empirical evidence and the need for more inter-
ventions with participatory and user-generated features.20
Social marketing provides a means of raising awareness of an
issue or intervention and drawing attention to a specic interven-
tion aimed at large-scale social change issues.21 Implementing a
social marketing approach through SM to drive interest and engage-
ment for a web-based program is both practical and consistent with
the way young people use the media to share ideas and promote
themselves, thus the intervention was considered a central focus of a
larger campaign. The aim of this study was to examine the effect of
the Break-it-Off (BIO) campaign, a digital SM campaign on young
adult smoking cessation.
Methods
Interventions
The BIO campaign was developed and implemented by the
Canadian Cancer Society and funded by Health Canada. It started
January 2012 and was promoted through to the end of March 2012
and continues to be available—www.breakitoff.ca. The January to
March 2012 period represents the rst time the campaign ran. The
campaign was aimed at young adult smokers aged 19–29. The prin-
cipal aim of the campaign was to engage young adults in smoking
cessation through use of a web-site and SM. The campaign targeted
young adults in six provinces across Canada.
BIO used a “break-up” metaphor, comparing quitting smoking
with ending a romantic relationship, to provide support and encour-
age young adults to “break up” with their smoking addiction. The
campaign’s website guided users through the challenging stages of
ending an unhealthy relationship with smoking: getting it over with,
staying split up, and moving on with life (Figure1). Through the site,
users could learn about established quit methods, such as telephone
counseling, nicotine patches, and nicotine gum. Visitors to the site
could upload a video of their “break-up with smoking” experience
as well as announce their break-up status to friends via Facebook.
One of the central features of the campaign was a multi-platform
smartphone app that provided instant support during specic trigger
points (eg, when stressed, angry, tipsy, or bored). Afree BIO smart-
phone app was available for download (Figure2). The app provided
information that was designed to be time-sensitive in its response,
allowing the information to be accessed at the moment a smoker felt
an urge to smoke. BIO was promoted primarily through paid (eg,
Online Banner Ads through Facebook, Google, Yahoo, Microsoft,
etc.), and earned (eg, television, radio and print)media.
Smokers’ Helpline (SHL), the comparison intervention, is a quit-
line that provides a telephone-based smoking cessation service to
support people who would like to quit smoking, maintain their quit,
or help someone else to quit. Quitlines are an evidence-based smok-
ing cessation intervention22–24 offered in at least 53 countries25 and
usually at no cost to the user. SHL is an established intervention and
provides smokers who want to quit information, self-help materials,
referrals to other resources, tailored motivational counseling, as well
as proactive follow-up counseling to adult smokers, including young
adults 19–29, in six Canadian provinces. During the intervention
period, SHL was promoted through both paid and earned media as
well as referrals from health organizations and professionals.
Figure1. Break-it-off homepage.
352
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from

353Nicotine & Tobacco Research, 2016, Vol. 18, No. 3
Design and Participants
This study employed a quasi-experimental, pretest, and posttest
design with one experimental and one unmatched comparison
group26,27 to test the effect of the BIO intervention. The Transparent
Reporting of Evaluations with Non-randomized Designs guideline
statement was used to assist in the reporting of this study.28
Participants were English speaking smoking young adults
19–29 years of age in the Canadian provinces of Saskatchewan,
Manitoba, Ontario, New Brunswick, Nova Scotia, and Prince
Edward Island. BIO participants were recruited through a link to
the study registration form on http://Breakitoff.ca from February to
September 2012 and through ads placed in the general labor section
of http://Kijiji.ca (an online classied service) in various Canadian
cities from June 29 until September 20, 2012. Three months after
registration and completing an online baseline survey, participants
were emailed a link to an online follow-up survey. BIO Participants
received a $10 iTunes redemption code as incentive for register-
ing and another $15 iTunes redemption code upon completion of
the follow-up survey. Young adult SHL participants were recruited
through the SHL’s usual administrative and evaluation procedures
where participants complete baseline demographic and smoking sta-
tus questions over the telephone and a follow-up telephone survey
interview between October 1, 2010 and September 30, 2011. All
study participants provided informed consent at registration and
the study was approved by the University of Waterloo, Ofce of
Research Ethics.
Data Collection
The baseline and follow-up questionnaires were administered to
both the intervention and comparison groups. Questionnaires were
based on the questions contained within the minimal data set for
tobacco cessation29 with new questions added for the current study.
In addition to nancial incentives for completing the study, a modi-
ed Dillman method30 was employed by contacting participants that
registered at baseline up to ve times either by email or telephone
to increase response rates. The baseline questionnaire consisted of
demographic and baseline smoking behavior questions. The follow-
up questionnaire included questions on use of cessation services, sat-
isfaction with services, cessation support relationships, and smoking
behavior.
Measures
Seven-day and 30-day point prevalence abstinence (PPA) rates were
measured at 3-month follow-up for both groups. SHL participants
provided the date of last cigarette smoked to determine abstinence
at 3months based on a 7-month follow-up and according to SHL’s
administrative procedures. Quit rates were based on those partici-
pants who completed the follow-up surveys. For both treatment
groups, respondents who completed the follow-up survey, but did
not provide answers to the point prevalence questions were consid-
ered to be smokers. We followed the intention-to-treat principle in
that participants were analyzed in the groups to which they were
allocated, regardless of whether they received or adhered to the allo-
cated intervention.31
SmokingStatus
The self-reported 7-day point prevalence cessation rate was deter-
mined by asking, “Have you smoked any cigarettes, even a single
puff, in the last 7days?” The self-reported 30-day point prevalence
cessation rate was determined by asking, “Have you smoked any
cigarettes, even a single puff, in the last 30- days?”
Cigarette Consumption
At registration, participants were asked whether their current ciga-
rette use was daily, occasional, or had quit and were trying to stay
quit.
Heaviness of SmokingIndex
Level of addiction was measured using the heaviness of smoking
index (HSI) that combines the number of cigarettes smoked per day
and the time to rst cigarette in the morning.32–34 High scores on the
Figure2. Break-it-off smartphone app page.
353
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from

354 Nicotine & Tobacco Research, 2016, Vol. 18, No. 3
HSI indicate higher levels of addiction and greater difculty in quit-
ting. HSI was categorized as low (scores of 0 to 2), medium (scores
of 3 and 4), and high (scores of 5 and 6).
Intention toQuit
Intention to quit smoking was measured by asking current smokers
if they intended to quit in the next 30days or not. Valid responses
were “yes” or “no.”
Used any CessationAid
At follow-up, participants were asked if they had used any of the fol-
lowing cessation aids other than SHL or BIO to help them quit and
included the options: Champix; Zyban; nicotine replacement therapy
gum, lozenges, inhaler, or nasal spray; advice from a physician, phar-
macist, nurse or other health professional; group cessation program;
self-help materials; quit contests; websites; and, smart phone apps.
At least One Support Person was Available
At follow-up, participants were asked if they had at least one person
they could count on for support in quitting smoking. Valid responses
were “yes” or “no.”
At least One Action Taken Toward Quitting
At follow-up, participants were asked if they had taken any of the
following actions toward quitting: cutting down the number of ciga-
rettes smoked; stopping smoking for 24 hours in an attempt to quit;
and set a quit date.
Larsen SatisfactionScore
User satisfaction was measured using the Larsen Satisfaction Score,35
which was computed by summing the responses for three questions
about satisfaction, each coded on a four-point scale with 1 being low.
The satisfaction questions were: To what extent has the program met
your needs? Overall, how satised were you with the service you
received from the program? If you were to seek help again, would
you use the program? The resulting Larsen Score ranges from 3 to
12, with 12 indicating “very high” satisfaction and 3“very low”.
Statistical Analyses
We compared baseline characteristics, use of cessation aids/supports,
service satisfaction, and smoking abstinence outcomes between BIO
and SHL young adult participants. Characteristics, smoking absti-
nence rates at 3 months after registration, use of cessation aids/
supports, satisfaction, and other cessation related outcomes were
summarized using the mean ± SD for continuous variables and fre-
quency percentages for categorical variables, and they were com-
pared between programs using the t test for independent groups or
the chi-square test, respectively. Further, to test for attrition bias,
baseline demographic characteristics and baseline smoking behav-
iors of participants that completed the follow-up survey and those
that did not were compared using the chi-square test. Finally, a sepa-
rate analysis of smoking abstinence rates was conducted using the
chi-square test that assumed participants lost to 3-month follow-up
were smokers.
First, logistic regression models were tted to examine the asso-
ciations of variables to the odds of having two primary outcomes—
7-day PPA and 30-day PPA. Variables signicantly associated with
quitting were included as covariates in the subsequent analysis to
examine the effect of BIO on smoking cessation. Second, regression
models were tted to examine the associations between the exposure
to the programs and the primary outcomes adjusting for possible
confounders identied at rst step. Avariable was considered to
enter the nal model if it produced a signicant association with the
outcome with a P value less than .05 for any level of the variable.
Assumptions such as sufcient sample size for single cell counts and
Goodness-of-t tests were used. Abackward elimination strategy
was employed in order to evaluate each covariate in the presence
of others. Potential confounders were removed one at a time start-
ing with the least signicant predictor and continuing until the P
value was less than 10% for all variables in the model. Findings
were summarized using unadjusted and adjusted odds ratios (ORs)
and corresponding 95% condence intervals (CIs). We used PROC
LOGISTIC in SAS 9.3.2 for analyses.
Results
Participant Characteristics
The ow of participants through the study is depicted in Figure3.
One-hundred twenty-two participants were excluded after baseline
given that they had already quit smoking for more than 30 days
prior to the commencement of the intervention. Atotal of 322 par-
ticipants were lost to follow-up (66% for BIO vs. 48% for SHL).
Atotal of 238 participants completed follow-up and contributed to
the analysis with overall follow-up rates for BIO and SHL of 34%
and 52%, respectively.
Important differences were found between BIO and SHL par-
ticipants at baseline (Table1). Users of SHL were more likely to be
female [63.2% vs. 49.0%, χ2 (1, N =238) = 4.81, P = .03], white
[89.1% vs. 76.2%, χ2 (1, N = 229)= 6.71, P= .01], have high
school education or less [50.7% vs. 36.3%, χ2 (1, N =238)=4.93,
P=.03], intended to quit in the next 30days [80.9% vs. 69.6%,
χ2 (1, N= 238) = 4.07, P = .04], and were much more likely to
be daily smokers [81.6% vs. 59.1%, χ2 (1, N = 224)=13.68, P <
.001]. Level of addiction in terms of cigarettes smoked per day and
time to rst cigarette in the morning was not signicantly different
between groups with 59.6% of SHL participants and 63.5% of BIO
participants reporting a low HSI. Signicant demographic, baseline
cigarette consumption and HSI differences were not found between
those that completed follow-up and those that did not for both BIO
and SHL participants (data not shown).
Program Reach and Participation
From January 2012 to March 2012, the total visits to http://
BreakitOff.ca were 44 172 with 37 325 unique visitors. The aver-
age user viewed 2.47 pages during their visit. Ontario was the most
prevalent audience with an average of 59% of visits, followed by
Saskatchewan with 22%. There were also 3937 installations of the
mobile app, 339 visitors interacted by posting content with the SM
components (Facebook and YouTube). Only 21 visitors connected
to SHL via the BIO campaign website. Larson satisfaction was
9.7 ± 1.97 and 78% of study participants visited the BIO website
two or more times with 18% visiting six or moretimes.
SHL reached 488 young adult smokers (19 to 29years of age)
from six provinces during the January 2012 to March 2012 inter-
vention period. Ontario represented 75% of users. Similar to BIO,
Larson satisfaction was 9.5 ± 2.19 and 68% of SHL participants 19
to 29years of age received two or more telephone calls for coun-
seling with 10% receiving six or more calls.
354
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from

355Nicotine & Tobacco Research, 2016, Vol. 18, No. 3
BIO—Regression Analysis
Table2 provides the results of association between participant char-
acteristics and the primary study outcomes. As expected, having
post-secondary education or higher is signicantly associated with
increased odds of quitting smoking for 7-day PPA, adjusted odds
ratio (AOR) = 2.16, 95% CI= 1.06 to 4.41, P= .03. In addition,
occasional cigarette use is signicantly associated with increased
odds of quitting for both 7-day PPA and 30-day PPA (AO R=6.26,
95% CI=3.10 to 12.67, P < .001 and AOR=5.93, 95% CI=2.82
to 12.46, P < .001, respectively). Other characteristics that were
positively associated but not signicant were being 25 to 30years
of age, non-white, intending to quit in the next 30days and having
social support (Table2).
BIO users had signicantly higher 7-day and 30-day quit rates
compared with users of SHL (Table3). The 7-day quit rate for BIO
(47.1%) was more than double that of SHL (15.4%), OR= 4.87,
95% CI=2.66 to 8.93, P < .001 and AOR=3.89, 95% CI=1.98 to
7.67, P < .001, controlling for education, ethnicity and daily or occa-
sional cigarette use. Although it was less prominent, the same pattern
was seen with 30-day quit rates, BIO (32.4%) and SHL (14.0%),
OR=2.95, 95% CI=1.56 to 5.57, P < .001 and AOR=2.14, 95%
CI=1.05 to 4.38, P=.04, controlling for covariates. For second-
ary outcomes, 91% of BIO participants made a quit attempt dur-
ing the 3-month intervention period compared to 79.1% of SHL
participants, OR=2.69, 95% CI=1.03 to 6.99, P=.04. Although
not statistically signicant, 89.4% of BIO participants versus 79.4%
of SHL participants cut-down amount smoked, OR= 2.18, 95%
CI = 0.88 to 5.44, P= .09. Both BIO and SHL participants were
equally likely to have set a quit date (Table3).
Treating all those who did not complete follow-up as smokers,
there was a signicant difference [χ2 (1, N=560)=7.55, P=.004]
for 7-day PPA between users of BIO (16.1%) and SHL (8.0%) and
the 30-day point prevalence was also higher for BIO participants
(11.0%) compared to SHL (7.3%); however, the difference was not
statistically signicant [χ2 (1, N=560)=1.91, P=.15].
Discussion
The current study found that the BIO campaign had a signicant
effect on young adult cessation rates. The results compare favorably
to other digital SM interventions36 such as “Happy Ending” in the
Netherlands which reported a 44.7% 7-day PPA rate at 3months.37
Further, a randomized trial of SHL had similar results to those
reported here. Participants in the SHL arm of the trial had a 7-day
quit rate at 3-month follow-up of 19.8%38 as compared to 15.4% in
this study. The doubling of quit rates for the digital SM intervention
group (AOR=2.14, 95% CI=1.05 to 4.38) was similar to the nd-
ings of a recent systematic review of internet-based and text messag-
ing interventions that also reported a doubling of quit rates.16,36 In
addition, BIO participants were signicantly more likely to make a
quit attempt as compared to SHL participants (91% vs. 79%). BIO
is helping smokers to try to quit and it is argued that increasing quit
attempts is key to improving the overall population cessation rate.39
Study Limitations
There are important limitations of the current study that are worth
noting. First, some of the limitations of this study are inherent in
the nature of the quasi-experimental design27 and internet-based
research in general such as the low study recruitment and high attri-
tion rate.40,41 However, attrition bias was not a limitation as the
baseline characteristics between those who completed follow-up
and those that did not were not signicantly different for both com-
parison groups. Second, threats to internal validity, such as selection
effects, were mediated by adjusting for observed group differences.
Break It Off
Smokers’ Helpline
BIO registrants
consented and
completed baseline
(N=342)
SHL registrants
consented and
completed baseline
(N=340)
Ineligible:
Already quit
smoking
(n=79)
Lost to
follow-up
(n=125)
Completed 3
month follow-up
(n=136)
Assessed for smoking
cessation outcomes
(n = 136)
Assessed for smoking
cessation outcomes
(n = 102)
Completed 3
month follow-up
(n=102)
Ineligible:
Already quit
smoking
(n = 43)
Lost to
follow-up
(n= 197)
Break-it-off
3 month intervention: n = 299
Smokers’ Helpline
3 month intervention: n = 261
Final N = 238
Figure3. Participant flow-diagram of recruitment and data collection.
355
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from

356 Nicotine & Tobacco Research, 2016, Vol. 18, No. 3
However, there is possibility of bias due to unobserved characteris-
tics. In addition, BIO participants may have been more motivated to
participate due to receiving an incentive of up to CaD $25. Third,
in terms of generalizability, the BIO campaign results may only be
applicable to those young adults who are motivated to use SM.
Given the novelty of the BIO concept, the campaign may not work
as effectively at another time and setting. Fourth, we were not able
to obtain corresponding SHL data for the same time periods as BIO;
however, using the older data may have prevented contamination of
our ndings given that a SHL specic promotion campaign happened
during the fall of 2012. Finally, while biochemical validation was
not conducted, the recommendation of the Society for Research on
Nicotine and Tobacco is that biochemical validation is not required
for large population health trials.42 In fact, a Cochrane Review of
Internet-based interventions for smoking cessation found that very
few studies used this method11 and accurate estimates of the preva-
lence of cigarette smoking among Canadians can be derived from
self-reported smoking status data.43
Implications for Practice
There are important policy and practice considerations given the
reach of a campaign such as BIO compared to traditional quitline
services using telephone-based counseling. Most young adults who
live on their own do not have landlines, prefer other modes of
communication such as texting, and the percentages are growing
for other population groups.44 Older technology is quickly being
replaced by mobile smartphone technology which act as computers,
are projected to be used by 5 billion people world-wide by 2025, and
have almost limitless functionality for accessing information and
facilitating interaction.45 Our ndings support the need for deter-
mining the role of SM and mobile technology interventions within
a tobacco cessation system, given the effectiveness and potential for
greater reach into population groups not served by or motivated to
use traditional cessation services.46 For example, building on the
results of the initial BIO campaign ndings, Health Canada has com-
mitted to extending the BIO campaign for 5years Canada-wide,47
expanding the reach and the potential depth of information tools
and resources available through BIO.
Implications for Research
Given that research on the effectiveness of SM for health pro-
motion and behavior change is still new, there are a number of
areas for further research. More formative evaluation is needed to
answer such questions as how SM are used by different audiences.
Table1. Sociodemographic Characteristics, Smoking Behaviors at Baseline, and Use of Cessation Aids and Supports by Young Adult
Smokers in Break-It-Off (BIO) and Smokers’ Helpline (SHL) Programs
BIO user (N=102), (n, %) SHL user (N=136), (n, %) Pa
Sociodemographics
Gender
Male 52 (51.0) 50 (36.8) .028
Female 50 (49.0) 86 (63.2)
Ageb
19 to 24 54 (54.5) 64 (47.1) .257
25 to 29 45 (45.5) 72 (52.9)
Highest level of education
High school or less 37 (36.3) 69 (50.7) .026
Post-secondary or higher 65 (63.7) 67 (49.3)
Ethnicityb
White 77 (76.2) 114 (89.1) .010
Other ethnicity 24 (23.8) 14 (10.9)
Working statusb
Not working 49 (49.5) 62 (45.6) .554
Working (full or part-time) 50 (50.5) 74 (54.4)
Smoking behavior at baseline
Cigarette consumptionb
Smoking daily 52 (59.1) 111 (81.6) <.001
Not smoking daily 36 (40.9) 25 (18.4)
Heaviness of smoking indexb
Medium or high 31 (36.5) 55 (40.4)
Low 54 (63.5) 81 (59.6) .556
Intent to quit in next 30days
No 31 (30.4) 26 (19.1) .044
Yes 71 (69.6) 110 (80.9)
Use of cessation aids and supports
Have used cessation aids
No 40 (39.2) 48 (35.3) .535
Yes 62 (60.8) 88 (64.7)
Social support: at least one support person was available
No 11 (12.0) 22 (16.2) .374
Yes 81 (88.0) 114 (83.8)
aP value for chi-square test of association.
bSome missing values for this variable in the study population.
356
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from

357Nicotine & Tobacco Research, 2016, Vol. 18, No. 3
Table2. Multiple Logistic Regression for the Association Between Participant Characteristics With 7-day and 30-day Smoking Abstinence at 3 Months
7-day point prevalence at 3months 30-day point prevalence at 3months
Abstinenta n (%) OR crude (95% CI) OR adjustedb (95% CI) Abstinent, n (%) OR crude (95% CI) OR adjustedc (95% CI)a
Sociodemographics
Gender
Maled31 (30.4) 1.00 1.00 23 (22.6) 1.00 1.00
Female 38 (27.9) 0.89 (0.51 to 1.56) 0.90 (0.45 to 1.82) 29 (21.3) 0.93 (0.50 to 1.73) 1.12 (0.53 to 2.39)
Age
19 to 24d31 (26.3) 1.00 1.00 21 (17.8) 1.00 1.00
25 to 29 31 (31.6) 1.30 (0.74 to 2.29) 1.09 (0.54 to 2.21) 31 (26.5) 1.67 (0.89 to 3.11) 1.62 (0.76 to 3.45)
Highest level of education
High school or lessd21 (19.8) 1.00 1.00 18 (17.0) 1.00 1.00
Post-secondary or higher 48 (36.4) 2.31 (1.28 to 4.19) 2.16 (1.06 to 4.41) 34 (25.6) 1.70 (0.89 to 3.22) 1.38 (0.64 to 2.97)
Ethnicity
Whited53 (27.8) 1.00 1.00 40 (20.9) 1.00 1.00
Other ethnicity 15 (39.5) 1.70 (0.82 to 3.50) 1.86 (0.74 to 4.69) 11 (29.0) 1.54 (0.70 to 3.37) 2.06 (0.77 to 5.52)
Working status
Not workingd29 (26.1) 1.00 1.00 20 (18.0) 1.00 1.00
Working (full or part time) 39 (31.4) 1.30 (0.74 to 2.29) 1.15 (0.57 to 2.29) 31 (25.0) 1.52 (0.81 to 2.85) 0.94 (0.44 to 2.00)
Smoking behavior at baseline
Cigarette consumption
Smoking dailyd29 (17.8) 1.00 1.00 21 (12.9) 1.00 1.00
Not smoking daily 36 (59.0) 6.65 (3.48 to 12.74) 6.26 (3.10 to 12.67) 28 (45.9) 5.74 (2.90 to 11.34) 5.93 (2.82 to 12.46)
Heaviness of smoking index
Medium or high 16 (18.6) 1.00 1.00 13 (15.1) 1.00 1.00
Low 49 (36.3) 2.49 (1.31 to 4.76) 1.00 (0.42 to 2.34) 36 (26.7) 2.04 (1.01 to 4.12) 0.90 (0.35 to 2.32)
Intent to quit in next 30days
Nod10 (17.5) 1.00 1.00 8 (14.0) 1.00 1.00
Yes 59 (32.6) 2.27 (1.07 to 4.81) 2.19 (0.79 to 6.12) 44 (24.3) 1.97 (0.87 to 4.47) 2.56 (0.79 to 8.28)
Use of cessation aids and supports
Have used cessation aids
Nod27 (30.7) 1.00 1.00 20 (22.7) 1.00 1.00
Yes 42 (28.0) 0.88 (0.49 to 1.56) 0.70 (0.34 to 1.45) 32 (21.3) 0.92 (0.48 to 1.74) 0.72 (0.33 to 1.56)
Social support: at least one support person was available
Nod5 (15.2) 1.00 1.00 4 (12.1) 1.00 1.00
Yes 59 (30.3) 2.43 (0.89 to 6.60) 2.45 (0.79 to 7.62) 45 (23.1) 2.18 (0.73 to 6.52) 2.22 (0.64 to 7.70)
CI=condence interval; OR=odds ratio.
aNumber and percent of participants who were abstinent at 7days or 30days at 3months follow-up in each group.
bConfounders included in all nal models: education, ethnicity, cigarette consumption, intent to quit in next 30days, and social support.
cConfounders included in all nal models: age, education, ethnicity, cigarette consumption, intent to quit in next 30days, and social support.
dReferent group.
357
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from

358 Nicotine & Tobacco Research, 2016, Vol. 18, No. 3
Other questions relate to implementation science; for example,
what are the design features that encourage deeper engagement
with SM to support behavior change and the dose needed to
produce a behavior change response.17 BIO is a multicompo-
nent intervention; thus, it was not possible to determine to what
extent each component of BIO contributed to the overall cam-
paign’s impact. For example, it is not known if the smartphone
app is more effective than the website itself. However, the BIO
website “Get it over with” section was the most popular aspect
of the campaign among BIO users and the “Break-up methods”
section of the website received the most page views. Research to
disentangle which elements of a multicomponent intervention are
accounting for change is needed as is further research to explore
the cumulative effect of the intervention and how it can connect
with other SM or technological resources.
Overall, there is very little evidence on the effectiveness and cost-
effectiveness of SM interventions available to decision makers. The
World Health Organization recently made a call to action for more
research on the effectiveness of SM and mHealth interventions for
behavior change.48 In addition, further research is needed to ensure
that smartphone and SM interventions incorporate evidence-based
practices22 rather than nonevidence-based practices such as hypno-
sis49,50 and that this knowledge is translated for policy-makers.50–53
Conclusion
A large number of young adults prefer a forum such as BIO for
help to quit smoking in comparison to traditional quitline services.
The reach of the campaign and ndings on quitting success indicate
that a multicomponent digital and SM campaign offers a promis-
ing opportunity to promote smoking cessation. While there is no
one-size-ts-all policy for smoking cessation, an integrated approach
that combines traditional quitline cessation services and a BIO type
SM campaign may be more effective, particularly in reaching young
adult smokers.
Funding
This work was supported by research grants from the Canadian Cancer
Society Research Institute (grant numbers 2011-70099 and 2011-701019).
Declaration of Interests
None declared.
Acknowledgments
NBB led the conceptualization and design of the study and CN, KM and KSB
contributed to the design of the study. NBB and SA drafted the manuscript.
NBB, SA, CN, KM, and KSB critically revised the manuscript for important
intellectual content. NBB is principal investigator and CN and KSB are co-
investigators on the research funding application. KM provided administra-
tive, technical, and material support. NBB is the guarantor. We thank Barbara
Zupko, Nghia C.Nguyen, Larry Squire, Lynda Hayward, and Matt Grey of
the University of Waterloo Propel Centre for Population Health Impact for
assistance in conducting the research, data analysis and helpful comments.
References
1. Statistics Canada. Canadian Community Health Survey, 2010. Ottawa,
ON: Statistics Canada; 2011.
2. Minian N, Schwartz R, Di Sante E, Philipneri A. Impact of the Smoking
Cessation System on Young Male Smokers. Special Report. Toronto, ON:
Ontario Tobacco Research Unit; 2010.
3. Bader P, Travis HE, Skinner HA. Knowledge synthesis of smoking cessa-
tion among employed and unemployed young adults. Am J Public Health.
2007;97(8):1434–1443. doi:10.2105/AJPH.2006.100909.
4. Curry SJ, Sporer AK, Pugach O, Campbell RT, Emery S. Use of tobacco
cessation treatments among young adult smokers: 2005 National
Health Interview Survey. Am J Public Health. 2007;97(8):1464–1469.
doi:10.2105/AJPH.2006.103788.
5. Fagan P, Augustson E, Backinger CL, et al. Quit attempts and intention
to quit cigarette smoking among young adults in the United States. Am J
Public Health. 2007;97(8):1412–1420. doi:10.2105/AJPH.2006.103697.
Table3. Multiple Logistic Regression for the Associations Between the Social Media Campaigns and 3-Month Smoking Cessation
Outcomes
Outcomes Favorable outcomesa, n (%) OR crude (95% CI) OR adjustedb (95% CI)
Primary outcomes
7-day PPA
Smokers’ helplinec21 (15.4) 1.00 1.00
Break-it-off 48 (47.1) 4.87 (2.66 to 8.93) 3.89 (1.98 to 7.67)
30-day PPA
Smokers’ helplinec19 (14.0) 1.00 1.00
Break-it-off 33 (32.4) 2.95 (1.56 to 5.57) 2.14 (1.05 to 4.38)
Secondary outcomes
Made a quit attempts (stopped for ≥24 hours)
Smokers’ helplinec87 (79.1) 1.00 1.00
Break-it-off 61 (91.0) 2.69 (1.03 to 6.99) 2.05 (0.77 to 5.49)
Cut-down amount smoked
Smokers’ helplinec85 (79.4) 1.00 1.00
Break-it-off 59 (89.4) 2.18 (0.88 to 5.44) 2.18 (0.76 to 6.24)
Set a quite date
Smokers’ helplinec69 (63.3) 1.00 1.00
Break-it-off 42 (65.6) 1.11 (0.58 to 2.11) 1.30 (0.61 to 2.73)
CI=condence interval; OR=odds ratio; PPA=point prevalence abstinence.
aNumber and percent of participants who obtained favorable outcomes at 3months follow-up in each group.
bConfounders included in all nal models: education, ethnicity, and cigarette consumption at baseline.
cReferent group.
358
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from

359Nicotine & Tobacco Research, 2016, Vol. 18, No. 3
6. Jha P, Ramasundarahettige C, Landsman V, et al. 21st-century hazards
of smoking and benets of cessation in the United States. N Engl J Med.
2013;368(4):341–350. doi:10.1056/NEJMsa1211128.
7. Villanti AC, McKay HS, Abrams DB, Holtgrave DR, Bowie JV. Smoking-
cessation interventions for U.S.young adults: a systematic review. Am J
Prev Med. 2010;39(6):564–574. doi:10.1016/j.amepre.2010.08.009.
8. Norman CD. Social Media for Health Promotion With Youth and Young
Adult Substance Use: A Resource and Evidence Review. Ottawa, ON:
Health Canada, Tobacco and Drugs Initiative; 2012.
9. Statistics Canada. Canadian Internet Use Survey, Internet Use, by Age
Group and Internet Activity. Table 358-0153. Ottawa, ON: Statistics
Canada; 2011.
10. Bennett GG, Glasgow RE. The delivery of public health interventions
via the Internet: actualizing their potential. Annu Rev Public Health.
2009;30:273–292. doi:10.1146/annurev.publhealth.031308.100235.
11. Civljak M, Sheikh A, Stead LF, Car J. Internet-based interven-
tions for smoking cessation. Cochrane Database Syst Rev. 2010;9.
doi:10.1002/14651858.CD007078.pub3.
12. Crutzen R, de Nooijer J, Brouwer W, Oenema A, Brug J, de Vries NK.
Strategies to facilitate exposure to internet-delivered health behavior change
interventions aimed at adolescents or young adults: a systematic review.
Health Educ Behav. 2011;38(3):49–62. doi:10.1177/1090198110372878.
13. Shahab L, McEwen A. Online support for smoking cessation: a sys-
tematic review of the literature. Addiction. 2009;104(11):1792–1804.
doi:10.1111/j.1360-0443.2009.02710.x.
14. Brouwer W, Kroeze W, Crutzen R, etal. Which intervention characteristics
are related to more exposure to internet-delivered healthy lifestyle promo-
tion interventions? Asystematic review. J Med Internet Res. 2011;13(1):e2.
doi:10.2196/jmir.1639.
15. Graham AL, Cha S, Papandonatos GD, etal. Improving adherence to web-
based cessation programs: a randomized controlled trial study protocol.
Trials. 2013;14:48. doi:10.1186/1745-6215-14-48.
16. Free C, Phillips G, Watson L, etal. The effectiveness of mobile-health tech-
nologies to improve health care service delivery processes: a systematic
review and meta-analysis. PLoS Med. 2013;10(1):e1001363. doi:10.1371/
journal.pmed.1001363.
17. Cobb NK, Graham AL, Byron MJ, Niaura RS, Abrams DB; Workshop
Participants. Online social networks and smoking cessation: a scientic
research agenda. J Med Internet Res. 2011;13(4):e119. doi:10.2196/
jmir.1911.
18. Gold J, Pedrana AE, Sacks-Davis R, etal. A systematic examination of the
use of online social networking sites for sexual health promotion. BMC
Public Health. 2011;11:583. doi:10.1186/1471-2458-11-583.
19. Struik LL, Baskerville NB. The role of Facebook in Crush the Crave,
a mobile- and social media-based smoking cessation interven-
tion: qualitative framework analysis of posts. J Med Internet Res.
2014;16(7):e170.
20. Chou WY, Prestin A, Lyons C, Wen KY. Web 2.0 for health promotion:
reviewing the current evidence. Am J Public Health. 2013;103(1):e9–18.
doi:10.2105/AJPH.2012.301071; 10.2105/AJPH.2012.301071.
21. Andreasen AR. Marketing social marketing in the social change
marketplace. J Public Policy Mark. 2002;21(1):3–13. doi:10.1509/
jppm.21.1.3.17602.
22. Fiore M, Jaen C, Baker TB, etal. Treating Tobacco Use and Dependence:
2008 Update. Clinical Practice Guidline. Rockville, MD: U.S. Department
of Health and Human Services, Public Health Service; 2008.
23. Stead LF, Perera R, Lancaster T. A systematic review of interventions for
smokers who contact quitlines. Tob Control. 2007;16:i3–8. doi:10.1136/
tc.2006.019737.
24. Stead LF, Hartmann-Boyce J, Perera R, Lancaster T. Telephone
counselling for smoking cessation. Cochrane Database Syst Rev.
2013;8:CD002850. doi:10.1002/14651858.CD002850.pub3
25. World Health Organization. Developing and Improving Toll-Free Tobacco
Quitlines: AWorld Health Organization Manual. Geneva, Switzerland:
World Health Organization; 2011.
26. Campbell DT, Stanley JC. Experimental and Quasi-Experimental Designs
for Research. Boston, MA: Houghton Mifin Company; 1963.
27. Shadish WR, Cook TD, Campbell DT. Experimental and Quasi-
Experimental Designs for Generalized Causal Inference. New York, NY:
Houghton Mifin Company; 2002.
28. Des Jarlais DC, Lyles C, Crepaz N; TREND Group. Improving the
reporting quality of nonrandomized evaluations of behavioral and pub-
lic health interventions: the TREND statement. Am J Public Health.
2004;94(3):361–366.
29. Campbell HS, Ossip-Klein D, Bailey L, Saul J; North American Quitline
Consortium. Minimal dataset for quitlines: a best practice. Tob Control.
2007;16:i16–i20. doi:10.1136/tc.2007.019976.
30. Dillman D, Smyth J, Christian L. Internet, Mail, and Mixed-Mode Surveys:
The Tailored Design Method. New York, NY: Wiley; 2009.
31. Gupta SK. Intention-to-treat concept: a review. Perspect Clin Res.
2011;2(3):109–112. doi:10.4103/2229–3485.83221.
32. Heatherton TF, Kozlowski LT, Frecker RC, Rickert W, Robinson J.
Measuring the heaviness of smoking: using self-reported time to the rst
cigarette of the day and number of cigarettes smoked per day. Br J Addict.
1989;84(7):791–799.
33. John U, Meyer C, Schumann A, etal. A short form of the Fagerström
Test for Nicotine Dependence and the Heaviness of Smoking Index in
two adult population samples. Addict Behav. 2004;29(6):1207–1212.
doi:10.1016/j.addbeh.2004.03.019.
34. Pérez-Ríos M, Santiago-Pérez MI, Alonso B, Malvar A, Hervada X,
de Leon J. Fagerstrom test for nicotine dependence vs heavy smoking
index in a general population survey. BMC Public Health. 2009;9:493.
doi:10.1186/1471-2458-9-493.
35. Larsen DL, Attkisson CC, Hargreaves WA, Nguyen TD. Assessment of
client/patient satisfaction: development of a general scale. Eval Program
Plann. 1979;2(3):197–207. doi:10.1016/0149-7189(79)90094-6.
36. Civljak M, Stead LF, Hartmann-Boyce J, Sheikh A, Car J. Internet-based
interventions for smoking cessation. Cochrane Database Syst Rev.
2013;7:CD007078. doi:10.1002/14651858.CD007078.pub4.
37. Brendryen H, Kraft P. Happy ending: a randomized controlled trial
of a digital multi-media smoking cessation intervention. Addiction.
2008;103(3):478–84. doi:10.1111/j.1360-0443.2007.02119.x.
38. Smith PM, Cameron R, McDonald PW, Kawash B, Madill C, Brown KS.
Telephone counseling for population-based smoking cessation. Am J
Health Behav. 2004;28(3):231–241.
39. Zhu SH, Lee M, Zhuang YL, Gamst A, Wolfson T. Interventions to
increase smoking cessation at the population level: how much progress
has been made in the last two decades? Tob Control. 2012;21(2):110–118.
40. Bull SS, Levine D, Schmiege S, Santelli J. Recruitment and retention of
youth for research using social media: experiences from the just/us study.
Vulnerable Child Youth Stud. 2013;8(2):171–181.
41. Maher CA, Lewis LK, Ferrar K, Marshall S, De Bourdeaudhuij I, Vandelanotte
C. Are health behavior change interventions that use online social networks
effective? Asystematic review. J Med Internet Res. 2014;16(2):e40.
42. Benowitz NL, Jacob P, Ahijevych K, et al. Biochemical verication
of tobacco use and cessation. Nicotine Tob Res. 2002;4(2):149–159.
doi:10.1080/14622200210123581.
43. Wong SL, Shields M, Leatherdale S, Malaison E, Hammond D.
Assessment of validity of self-reported smoking status. Health Rep.
2012;23(1):47–53.
44. Canadian Radio-Television and Telecommunications Commission.
Navigating Convergence: Charting Canadian Communications Change
and Regulatory Implications. (No. Convergence Consulting Group’s fore-
cast). Ottawa, ON: Canadian Radio-Television and Telecommunications
Commission; 2010.
45. Miller G. The smartphone psychology manifesto. Perspect Psychol Sci.
2012;7(3):221–237.
46. Abrams D. Comprehensive smoking cessation policy for all smokers:
systems integration to save lives and money. In Bonnie RJ, Stratton K,
Wallace RB eds. Ending the Tobacco Problem: Blueprint for the Nation.
Washington, DC: National Academies Press; 2010:A-1–A-50.
47. Health Canada. Break it off smoking cessation campaign. http://news.
gc.ca/web/article-en.do?nid=886789&tp=1 September 23, 2014 Accessed
October 30, 2014.
359
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from

360 Nicotine & Tobacco Research, 2016, Vol. 18, No. 3
48. World Health Organization. World Health Organization. mhealth: New
Horizons for Health Through Mobile Technologies. Global Observatory
for eHealth Series (vol. 3). Geneva, Switzerland: World Health
Organization; 2011.
49. Abroms LC, Padmanabhan N, Thaweethai L, Phillips T. iPhone apps for
smoking cessation: a content analysis. Am J Prev Med. 2011;40(3):279–
285. doi:10.1016/j.amepre.2010.10.032; 10.1016/j.amepre.2010.10.032.
50. Richardson CG, Vettese L, Sussman S, Small SP, Selby P. An investiga-
tion of smoking cessation video content on YouTube. Subst Use Misuse.
2011;46(7):893–897. doi:10.3109/10826084.2011.570628.
51. Abroms LC, Lee Westmaas J, Bontemps-Jones J, Ramani R, Mellerson J. A
content analysis of popular smartphone apps for smoking cessation. Am J
Prev Med. 2013;45(6):732–736.
52. Backinger CL, Pilsner AM, Augustson EM, Frydl A, Phillips T, Rowden
J. YouTube as a source of quitting smoking information. Tob Control.
2011;20(2):119–122. doi:10.1136/tc.2009.035550.
53. Prochaska JJ, Pechmann C, Kim R, Leonhardt JM. Twitter=quitter?
An analysis of Twitter quit smoking social networks. Tob Control.
2012;21(4):447–449.
360
at University of Waterloo on June 28, 2016http://ntr.oxfordjournals.org/Downloaded from