RESEARCH ARTICLE Open AccessAn examination of exposure and avoidancebehavior related to second-hand cigarette smokeamong adolescent girls in CanadaJennifer Schwartz1*, Raquel B Graham2, Chris G Richardson1, Chizimuzo T Okoli4, Laura L Struik2and Joan L Bottorff2,3AbstractBackground: Although rates of tobacco use and exposure to second-hand smoke (SHS) are declining in Canada,SHS exposure among non-smoking adolescents remains high. This study aimed to describe frequency, locations,and avoidance behavior related to SHS exposure among adolescent girls in British Columbia, Canada.Methods: Data were analyzed from 841 adolescent girls aged 13 to 15 years old who completed aninternet-delivered survey as part of a cohort study examining SHS exposure and substance use. Measures assesseddemographics, smoking behavior and intentions, frequency and locations of SHS exposure, and avoidance behaviorrelated to SHS.Results: Excluding their own smoking, 27% of girls reported exposure at least once a week and an additional 17%reported daily or almost daily exposure over the past month. Among girls who reported daily or almost dailyexposure, the locations of most frequent levels of high exposure were in the home, at or near school, inside avehicle, and outdoor public places. Avoidance behavior related to SHS exposure significantly differed by overall SHSexposure in the past month.Conclusions: Despite historically low smoking rates, many adolescent girls continue to report regular SHS exposurein multiple locations in British Columbia. Girls with the most frequent exposure were significantly less likely toreport habitual avoidance behavior related to SHS compared to those less frequently exposed. This study elucidatessettings of high SHS exposure among adolescent girls that could be targeted in future policy interventions.Additionally, future interventions could target adolescent girls who are frequently exposed to SHS and reportinfrequent avoidance behavior around their SHS exposure.Keywords: Second-hand smoke, Adolescents, Females, Tobacco, Risk reduction behaviorBackgroundAlthough rates of tobacco use and exposure to second-hand smoke (SHS) are declining in Canada, SHS ex-posure among non-smoking adolescents remains high[1]. According to the 2007–2009 Canadian Health Mea-sures Survey, a significantly greater percentage of non-smoking adolescents (12–19 years old) reported regularexposure to SHS compared to non-smoking children(6–11 years old) and adults (20–79 years old) [2]. Inaddition to poor academic performance and a greaternumber of school days missed due to poor health [3,4],SHS exposure within this demographic is associated witha number of negative health consequences (e.g., cancers,heart disease, ear infections, asthma, respiratory infec-tions, and decreased pulmonary function) [5]. Amongadolescents, girls are particularly vulnerable to the ad-verse health effects of tobacco smoke; recent researchdemonstrated that both active smoking and SHS ex-posure increase the risk for developing breast cancer,particularly when exposure occurs during puberty whenbreast cell proliferation is most rapid [6]. Results of ameta-analysis suggest a 60-70% increase for breast cancer* Correspondence: Schwarjennifer@gmail.com1School of Population & Public Health, University of British Columbia, 2206East Mall, Vancouver, BC V6T1Z3, CanadaFull list of author information is available at the end of the article© 2014 Schwartz et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.Schwartz et al. BMC Public Health 2014, 14:468http://www.biomedcentral.com/1471-2458/14/468risk in pre-menopausal women who report long-termSHS exposure [7]. As cigarette smoke exposure remainsone of the few modifiable risk factors for breast cancer, itis essential to develop harm reduction messages and strat-egies for adolescent girls who are at risk of regular SHSexposure.Emerging evidence indicates that targeted communica-tions (e.g., by gender and/or cancer-specific) are moreeffective than non-targeted alternatives such as gender-neutral messaging about non-specific forms of cancer[8-10]. Adolescence, or the transformation into woman-hood, is a period of marked awareness of physical changes(e.g., breast development) for girls [9]. These phases ofpronounced cognizance of health transitions may increasegirls’ risk perceptions, and have been recognized as teach-able moments for cancer prevention interventions [11].Therefore, targeting communications about the link be-tween tobacco exposure and an increased risk of breastcancer to adolescent girls takes advantage of this naturallyoccurring teachable moment to promote reductions inSHS exposure. Additionally, there is a growing body ofevidence that gender profoundly influences health behav-iors and responses to teachable moments [10,11]. In fact,Richardson et al. found that gender-sensitive messagesabout the relationship between breast cancer and SHS in-creased adolescent girls’ awareness of the risks and stimu-lated information seeking about these risks [8].Previous studies have investigated SHS exposure in thegeneral population; however, few studies have identifiedspecific locations of greatest exposure among adolescents[12]. For example, non-smoking Canadians most com-monly report SHS exposure in public locations and thework place; however these findings may not extend to ad-olescents. Analyses from one study indicate that Canadianyouth are frequently exposed to SHS in the home or ve-hicle; but other locations such as around schools, busstops, and in public parks require further investigation [4].An understanding of where adolescent girls are mostexposed, as well as their efforts to avoid SHS, will pro-vide the foundation needed to develop targeted harm re-duction strategies that aim to reduce SHS exposure andultimately breast cancer risk. Given the heightened vul-nerability to carcinogens in cigarette smoke duringbreast development, there is a need to inform adolescentgirls of their potential risk. Utilizing data from a largecohort study of adolescents in British Columbia, thepresent study aimed to examine frequency, specific loca-tions, and avoidance behaviors related to SHS exposureamong adolescent girls.MethodsParticipantsParticipants were 841 adolescent girls aged 13 to 15 yearswho participated in an internet-based cohort study ofyouth in British Columbia, Canada (The BC AdolescentSubstance Use Survey [BASUS]) in the spring of 2011.Students were recruited from 48 participating publicsecondary schools in BC, and eligibility criteria includedbeing 13 years of age or older and the ability to read andcomplete the internet-based survey in English. All par-ticipants provided informed consent, as well as writtenparental consent in schools requiring participants toprovide parental consent. All students were recruited ina school environment – after viewing a brief presenta-tion during home room class, students were given an in-formational package that contained a unique login codeto set up an account on the survey website – and thesurvey was completed online during the students’ owntime or in some cases in school computer labs duringscheduled class time. The school-specific response rateranged from 2% to 100% with an average school re-sponse rate of 20%. The University of British ColumbiaBehavioural Research Ethics Board classified this studyas minimal risk and did not require us to obtain signedparental consent. Given the age of the students and theminimal risk of the study, we were approved to use apassive consent procedure in which the take home infor-mation package included a letter to parents about thestudy with our contact information should they wish usto block their son/daughter from participating in thestudy against their wishes (i.e., we could prevent an ac-count from being set up with their son/daughters' infor-mation package card, as well as their email address orhome mailing address). This procedure was approved bythe University of British Columbia Behavioral ResearchEthics Board and all School Districts, as well as individ-ual Secondary School Administrators except for oneschool district which has a standard policy requiringsigned parental consent for all research studies involvingtheir students. The signed parental consent procedureused in this school district was also approved by theUniversity of British Columbia Behavioural ResearchEthics Board. Data collected in the BASUS survey in-cluded participant characteristics such as age, ethnicity,family income, cigarette smoking behavior, intentions tosmoke, and information on SHS exposure.Measurement of exposureThe online questionnaire was used to collect detaileddata regarding adolescents’ exposure to SHS during thepast month. The following question was used to collectdata on overall SHS exposure: “Overall (excluding yourown smoking) in the past month were you exposed tosecond-hand smoke?”, with the following response op-tions: 1) never, 2) at least once in past month (low fre-quency), 3) at least once a week (medium frequency),and 4) every day or almost every day (high frequency).Additionally, the following multipart question was usedSchwartz et al. BMC Public Health 2014, 14:468 Page 2 of 9http://www.biomedcentral.com/1471-2458/14/468to examine SHS exposure in specific locations: “In thepast month (excluding your own smoking), how oftenwere you exposed to second-hand smoke: inside a car orother vehicle?; inside someone else’s home?; on an out-door patio of a restaurant or bar?; at a bus stop or shel-ter?; at an entrance to a building?; at your workplace?;at/near your school?; at any other public place such as ashopping mall, arena, concert, or sporting event?; and/oroutdoors such as on a sidewalk or in a park?”. The pres-ence of home cigarette smoking restrictions was assessedwith the following yes/no question: “Are there any re-strictions against smoking cigarettes in your home?”Stage of change related to avoidance of SHSA brief measure, which was developed based on the Pro-chaska’s Stage of Change model on an adolescent sample[13,14], was used to examine avoidance behavior relatedto SHS exposure. Specifically, the measure assessed ado-lescents’ stage of change (i.e., maintenance, action, prepar-ation, contemplation and pre-contemplation) related toreducing exposure to SHS with the following question:When you are exposed to second-hand cigarette smokedo you consistently do things to reduce your exposureto the smoke? (Please check only one)1) Yes, I have been for more than 6 months(MAINTENANCE).2) Yes, I have been, but for less than 6 months(ACTION).3) No, but I intend to in the next 30 days(PREPARATION).4) No, but I intend to in the next 6 months(CONTEMPLATION).5) No, and I do NOT intend to in the next 6 months(PRE-CONTEMPLATION).Based on responses to this question, adolescents werecategorized as being in a particular stage with regard toavoidance behavior related to SHS exposure.Statistical analysisDescriptive statistics of the sample are provided, and abivariate analysis (Pearson Chi-square test) was performedto examine the relationship between overall SHS exposureand risk reduction behavior around SHS in the pastmonth. An alpha level of p < 0.05 (2-tailed) was used to in-dicate statistical significance. All statistical analyses wereconducted using IBM SPSS Statistics 19.0. A series ofVenn diagrams were created to evaluate the extent towhich environmental and social variables are related tohigh levels of SHS exposure in three of the most com-monly reported exposure settings. Among adolescentswho reported high SHS exposure in the home (n = 130),we were specifically interested in understanding howmany also had parents who smoke and reported homesmoking restrictions. Among those who reported highSHS exposure inside a vehicle (n = 96), we were interestedin examining how many also had friends or parents whosmoke. Finally, among adolescents who reported highSHS exposure at school and other outdoor locations(n = 88), we evaluated how many adolescents also hadfriends who smoke. A 95% confidence interval wasused to provide a range of plausible values for these pa-rameters, and to investigate the likelihood of specific ex-posure groups overlapping with the larger exposure groupof interest.ResultsTable 1 displays the adolescent girls’ demographics,smoking behavior and intentions, and reported levels ofSHS exposure. Fifty-seven percent of girls were 14 yearsold, 69% were in grade nine, 50% were Caucasian, and78% reported an average family income. Most of the girlshad never tried smoking (88%) and reported that theydefinitely did not intend on smoking in the future (75%);of those girls who had tried smoking, the majority werenot current smokers (60%) and were 13 years of age orolder when they tried their first cigarette. Moreover, amajority of girls reported that they did not have parents(68%) or friends (80%) who smoke, and had home smok-ing restrictions (87%).As shown in Table 2, overall SHS exposure varied,with 17% (139/841) of the sample reporting exposureevery day or almost every day, and 27% (223/841) re-porting exposure at least once a week. Reports of anySHS exposure in various smoking locations among girlswith at least weekly overall exposure were as follows:32% (115/358) in the home, 44% (159/360) in a vehicleor car, 48% (172/359) in someone else’s home, 86% (311/360) at/near school, 75% (267/358) on an outdoor patioof a bar or restaurant, 77% (277/359) at a bus stop orshelter, 93% (331/357) outdoors on sidewalk or park,89% (319/359) at an entrance to a building, and 91%(323/357) at any other public place such as a shoppingmall, arena, concert, or sporting event. Patterns of ex-posure varied by location, particularly among high SHSexposure groups. For example, among adolescent girlswho reported exposure every day or almost every day,39% (95%CI: 30.8-47.5) reported in home exposure everyday or almost every day. Similarly, among those who re-ported exposure every day or almost every day, 34%(95%CI: 26.2-42.4) reported being exposed ‘a lot’ at ornear school.Figure 1 illustrates the adolescent girls’ stage of changewith regard to avoidance behavior around SHS by overallSHS exposure in the past month. For example, 18% andSchwartz et al. BMC Public Health 2014, 14:468 Page 3 of 9http://www.biomedcentral.com/1471-2458/14/46814% of adolescent girls who were exposed to SHS everyday or almost every day, and at least once a week, re-spectively, were in the pre-contemplation stage with re-gard to avoidance behavior around SHS. Additionally,15% of girls who were exposed at least once in the pastmonth were in the pre-contemplation stage. Based onchi-square analysis, stage of change with regard to SHSrisk reduction behavior significantly differed by overallSHS exposure in the past month among adolescent girls(p < 0.001).A series of Venn diagrams were generated to explorepatterns of SHS exposure in multiple locations, and theextent to which other factors (e.g., having parents whosmoke) might overlap with reporting SHS exposure inparticular locations. We were specifically interested ingaining insight into the extent that high SHS exposurein the home overlapped with reporting home smokingrestrictions and/or having parents who smoke. We werealso interested in examining the extent to which highSHS exposure in a vehicle coincided with having parentswho smoke and/or friends who smoke. Finally, we exam-ined the overlap between those who reported havingfriends who smoke and those who reported high SHSexposure at/near school and outdoors (on a sidewalk orin a park). As displayed in Panel 1 of the Venn diagrams(Figure 2), among girls with both high (every day or al-most every day) overall SHS exposure in the past monthand complete data on in-home exposure (n = 130), 42%(95% CI: 33.1-50.5) also had high SHS exposure in thehome; 59% (95% CI: 49.5-66.9) had parents who smoke;and 74% (95% CI: 66.1-81.7) had home smoking restric-tions. Furthermore, among these highly exposed girlswith complete data on in-home exposure, 4% (95% CI:1.4-9.2) had high SHS exposure in the home and re-ported home smoking restrictions; 16% (95% CI: 10.5-23.9) had high SHS exposure in the home and had par-ents who smoke; 19% (95% CI: 13.1-27.3) had parentswho smoke and home smoking restrictions; and an add-itional 18% (95% CI: 13.1-27.3) had high SHS exposurein the home, home smoking restrictions, and parentswho smoke.As displayed in Panel 2 of the Venn diagrams (Figure 2),among girls with both high (every day or almost everyday) overall SHS exposure in the past month and com-plete data on SHS exposure in cars (n = 96), 34% (95% CI:25.2-44.9) also had high SHS exposure inside a car orother vehicle; 79% (95% CI: 69.4-86.5) had parents whosmoke; and 47% (95% CI: 36.7-57.3) had friends whosmoke. Furthermore, among these highly exposed girls,15% (95% CI: 8.5-23.6) had high SHS exposure inside acar or other vehicle and had parents who smoke; 4%(95% CI: 1.3-10.9) had high SHS exposure inside acar or other vehicle and had friends who smoke; andan additional 14% (95% CI: 7.7-22.4) had high SHS expos-ure inside a car or other vehicle, friends who smoke, andparents who smoke.As displayed in Panel 3 of the Venn diagrams (Figure 2),among girls with both high (every day or almost everyTable 1 Characteristics of adolescent girls (N = 841) in astudy on second-hand smoke (SHS) exposureDemographicsAge13 years 144 (17.1%)14 years 479 (57.0%)15 years 218 (25.9%)Grade8 251 (30.0%)9 578 (69.0%)10 8 (1.0%)EthnicityCaucasian 419 (50.1%)Aboriginal 111 (13.3%)Asian 286 (34.2%)Other 21 (2.5%)Family Income (self-reported)Below average 46 (5.9%)Average 611 (78.0%)Above average 126 (16.1%)Smoking behavior and intentionsEver tried smokingYes 97 (11.5%)Current smoker (smoked ≥once in past 30days);among those who indicated they had ever triedsmoking (N=97)Yes 38 (40%)Age at first cigarette; among those who indicatedthey had ever tried smoking (N=97)≤12 years 39 (41.5%)13 years 29 (30.9%)≥14 years 26 (27.7%)Smoking intentionsDefinitely/probably yes 28 (3.9%)Probably not 153 (21.3%)Definitely not 539 (74.9%)SHS exposureParents smokeYes 236 (31.7%)Friends smokeYes 128 (20.3%)Home smoking restrictionsYes 692 (87.0%)Schwartz et al. BMC Public Health 2014, 14:468 Page 4 of 9http://www.biomedcentral.com/1471-2458/14/468Table 2 Frequency of second-hand smoke exposure at specific locations among adolescent girls with at least weeklyoverall exposure in the past month (N = 362)Overall (excluding your own smoking) in the past month were you exposed tosecond hand smoke?At least once a week(N = 223) % (95% CI)Every day or almost everyday (N = 139) % (95% CI)In the past month (excluding your own smoking), how often were you exposed to SHS:In your own home Never (n = 243) 80% (73.8-84.8) 48% (38.3-55.4)At least once in past month (n = 24) 7% (4.0-11.1) 7% (3.2-12.3)At least once a week (n = 29) 10% (6.1-14.2) 6% (2.7-11.4)Every day or almost every day (n =62)4% (1.7-7.2) 39% (30.8-47.5)Inside a car or other vehicle Never (n = 201) 67% (59.7-72.5) 38% (30.1-46.8)Once (n = 39) 12% (7.9-16.8) 9% (5.3-15.8)A few times (n = 54) 15% (10.9-20.8) 14% (9.2-21.6)More than a few times (n = 28) 4% (1.7-7.2) 14% (9.2-21.6)A lot (n = 38) 2% (0.8-5.4) 24% (17.1-31.8)Inside someone else’s home Never (n = 187) 59% (51.5-64.8) 41% (32.8-49.7)Once (n = 57) 17% (12.1-22.3) 15% (9.2-21.6)A few times (n = 70) 19% (13.7-24.2) 21% (14.6-28.7)More than a few times (n = 22) 4% (1.7-7.2) 10% (5.8-16.6)A lot (n = 23) 2% (0.8-5.4) 13% (8.1-20.0)At/near your school Never (n = 49) 15% (10.9-20.8) 11% (6.4-17.5)Once (n = 40) 14% (9.4-18.8) 7% (3.7-13.2)A few times (n = 131) 40% (33.1-46.2) 31% (23.5-39.4)More than a few times (n = 59) 16% (11.7-21.8) 17% (11.0-24.0)A lot (n = 81) 15% (10.9-20.8) 34% (26.2-42.4)On an outdoor patio of a restaurant or bar Never (n = 91) 26% (20.1-31.9) 25% (17.8-32.6)Once (n = 46) 17% (12.1-22.3) 7% (3.2-12.3)A few times (n = 136) 40% (33.5-46.7) 34% (26.2-42.4)More than a few times (n = 48) 12% (7.9-16.8) 16% (10.4-23.2)A lot (n = 37) 5% (2.9-9.4) 18% (12.2-25.6)At a bus stop or shelter Never (n = 82) 20% (15.2-26.2) 27% (19.7-34.9)Once (n = 38) 11% (7.5-16.3) 10% (5.3-15.8)A few times (n = 133) 44% (36.9-50.3) 27% (19.0-34.1)More than a few times (n = 57) 18% (13.3-23.8) 13% (7.5-19.1)A lot (n = 49) 7% (4.3-11.6) 24% (17.1-31.8)Outdoors such as on a sidewalk or in a park Never (n = 26) 5% (2.6-8.9) 11% (6.4-17.5)Once (n = 35) 15% (10.2-19.8) 2% (0.6-6.7)A few times (n = 136) 44% (36.5-49.8) 29% (21.6-37.2)More than a few times (n = 87) 22% (16.8-28.1) 28% (20.3-35.7)A lot (n = 73) 15% (10.2-19.8) 30% (22.2-37.9)At an entrance to a building Never (n = 40) 11% (7.2-15.8) 12% (6.9-18.3)Once (n = 38) 11% (7.2-15.8) 10% (5.8-16.6)A few times (n = 148) 47% (40.0-53.4) 32% (24.2-40.2)More than a few times (n = 80) 23% (18.1-29.5) 21% (14.0-28.0)A lot (n = 53) 9% (5.3-13.2) 25% (17.8-32.6)Never (n = 34) 9% (5.3-13.2) 11% (6.4-17.5)Schwartz et al. BMC Public Health 2014, 14:468 Page 5 of 9http://www.biomedcentral.com/1471-2458/14/468day) overall SHS exposure in the past month and com-plete data on SHS exposure outdoors (at/near school andon sidewalks or in parks) (n = 88), 53% (95% CI: 42.5-64.0)also had high SHS exposure at or near school; 51% (95%CI: 40.3-61.9) had friends who smoke; and 47% (95% CI:36.0-57.5) had high SHS exposure outdoors. Furthermore,among these highly exposed girls, 11% (95% CI: 5.9-20.3)had high SHS exposure at or near school and had friendswho smoke; 10% (95% CI: 5.1-19.0) had high SHS expos-ure at or near school and had high SHS exposure out-doors; and an additional 15% (95% CI: 8.4-24.3) had highSHS exposure at or near school, friends who smoke, andhigh SHS exposure outdoors. Relationships displayed inthe Venn diagrams are also shown in a table for ease of in-terpretation (see Additional file 1: Table S1).DiscussionTo our knowledge, this is the first study to describe thereported frequency, locations, and avoidance behaviorsrelated to SHS exposure among adolescent girls inCanada. We found that although the majority of adoles-cent girls in our sample had never tried smoking, a con-cerning proportion reported regular exposure to SHS.Of the 841 girls, 139 (17%) reported exposure to SHSevery day or almost every day in the past month, and anadditional 223 (27%) reported exposure at least once aweek. Within this group, regular exposure frequently oc-curred in the home, at/near school, inside a vehicle, atbus stops, and other public locations. Of note, the in-home SHS exposure rate (19%) reported by this cohortof adolescent girls is consistent with recent research byHealey et al. on youth in New Zealand aged 14 to15 years, who reported that in-home SHS exposure ratesvaried from 12% among youth with no parents whosmoked, to 85% among youth with both parents whosmoked [15].Avoidance behavior around SHS was examined using abrief measure adapted from the stage of change model[13]. Compared to girls who reported less SHS exposure,girls who were exposed every day or almost every daywere more likely to be in the pre-contemplation, con-templation, or preparation stages. However, 60% of girlswho were exposed every day or almost every day were inthe maintenance stage, suggesting that the majority ofgirls with regular SHS exposure consistently take actionto reduce their exposure. It is possible that greater SHSexposure prompts avoidance behaviors, although mo-tivational factors underlying efforts to avoid SHS werenot assessed in this study. Future studies may explorestrategies girls implement to reduce SHS exposure andTable 2 Frequency of second-hand smoke exposure at specific locations among adolescent girls with at least weeklyoverall exposure in the past month (N = 362) (Continued)At any other public place such as a shopping mall, arena,concert, or sporting eventOnce (n = 23) 9% (5.7-13.7) 2% (0.6-6.7)A few times (n = 156) 49% (41.7-55.2) 35% (26.8-43.1)More than a few times (n = 80) 23% (17.6-29.1) 21% (14.6-28.7)A lot (n = 64) 10% (6.1-14.2) 31% (23.5-39.4)Figure 1 Stage of change related to avoidance behavior around SHS by frequency of SHS exposure in the past month in adolescentgirls (n = 841). *Based on chi-square test, p < 0.001. *Based on chi-square test, p < 0.001.Schwartz et al. BMC Public Health 2014, 14:468 Page 6 of 9http://www.biomedcentral.com/1471-2458/14/468evaluate the effectiveness of such strategies. Further-more, social norms may influence girls’ perceived abilityto avoid SHS. Therefore, interventions that address so-cial pressures that undermine avoidance of SHS couldbe created.The Venn diagrams (Figure 2), which display SHS ex-posure locations, sources, and related variables amongadolescent girls with high overall SHS exposure in thepast month, provide interesting findings. For example,despite reporting home smoking restrictions, 22% ofadolescent girls reported being exposed in their homeevery day or almost every day. Meanwhile, as displayedin the second Venn diagram (Panel 2), only 28% of girlswho reported having parents who smoke also reportedhigh SHS exposure inside a car or other vehicle. Asshown in the third Venn diagram (Panel 3), the majorityof girls who reported being regularly exposed at/nearschool did not actually have friends who smoke, suggest-ing they were exposed in areas where smoking occursamong a larger group of peers. These findings have im-portant implications for informing the development ofrisk-reduction messages and/or interventions, and de-monstrate the importance of targeting such messages/interventions to specific populations and/or locations.For example, messages could target locations where ado-lescent girls are frequently exposed, such as bus stopsand at/near schools. Moreover, interventions couldtarget individuals in the pre-contemplation and/or con-templation stages with the goal of shifting them intothe action or maintenance stages related to avoidanceof SHS.The current findings also highlight the need to educateparents about the importance of enforcing completehome smoking bans and protecting girls from SHS ex-posure. With increasing prohibitions on smoking inpublic places, the home is becoming a primary source ofSHS exposure [16]. This is reflected in our findings –39% of the girls with high overall SHS exposure reportedFigure 2 Venn diagrams displaying SHS exposure locations, sources, and related variables among adolescent girls with high overallSHS exposure in the past month (N = 139). Panel a. In-home Exposure (N=130). Panel b. Car Exposure (N=96). Panel c. Outdoors – At/nearschool and on sidewalks or in parks (N=88).Schwartz et al. BMC Public Health 2014, 14:468 Page 7 of 9http://www.biomedcentral.com/1471-2458/14/468being exposed every day or almost every day in theirhome. This is especially concerning in light of evidencethat particulate matter from tobacco smoke in homeswhere smoking occurred were at unhealthy levels evenin remote areas of the home where smoking does nottypically occur [17]. Given that adolescents may be lim-ited in influencing home smoking restrictions, it is im-portant that parents are educated about the dangers ofSHS exposure in the home. Information about SHS andbreast cancer could provide the basis for a renewed mes-sage to parents about the importance of smoke-freehomes, especially for adolescent girls.Our findings offer important implications for policymakers. Despite prohibitions on smoking in most publicplaces since 2008 in British Columbia [16], the girls inour study reported regular SHS exposure in several ofthe locations where these bans are in place. For example,of those most exposed to SHS, 34% reported regular ex-posure at or near school, 25% at an entrance to a build-ing, 24% inside a car or other vehicle, and 18% on anoutdoor patio of a restaurant or bar. These findingshighlight the importance of ensuring the development,implementation, and enforcement of well-defined pol-icies in relation to smoking prohibitions in these places,in order to create environments that protect adolescentsrather than placing the burden on them to avoid SHS.This study is not without limitations. British Columbiahas a strong history of tobacco control, which has re-sulted in decreased smoking rates and policies restrictingsmoking in public spaces. Levels of SHS exposure andefforts to avoid SHS are likely to be higher in otherregions where smoking rates remain high. It is alsopossible that the order in which the questions were pre-sented may have influenced individual responses. Par-ticipants were first asked about their efforts to reduceexposure, followed by questions about specific locationsof exposure, and finally about their overall smoke expos-ure during the past month. Priming individuals to thinkabout exposure in specific locations may have influencedresponses to the overall exposure question. Current smo-kers, although more likely to associate with other smokersand less likely to engage in SHS avoidance behavior, wereincluded in analyses to reduce the likelihood of biasingthe sample towards less SHS exposure than expected in atypical Canadian school. Although the average participa-tion rate within schools was relatively low, the represen-tativeness of our sample is supported by findings thatindicate a smoking rate of 4.5% among adolescents ingrades 8–10, which is similar to results based on the na-tionally representative Youth Smoking Survey (2010–11),which indicated a 2% and 10% prevalence of current smo-kers among youth in grades 6–9 and 10–12, respectively[18]. Furthermore, the finding that 7.4% of girls in thisstudy reported every day or almost every day exposure inthe home is similar to findings from the 2011 CanadianTobacco Use Monitoring Survey which found that 7.1% ofCanadian children aged 12–17 years were regularly ex-posed to tobacco smoke in their homes [19]. It should alsobe noted that 78% of our sample reported an average fam-ily income, which may have resulted from social desirabil-ity bias. Nonetheless, we acknowledge that this relativelyhigh proportion potentially limits generalizability of ourfindings to individuals in other socioeconomic groups.Lastly, the descriptive data displayed in the Venn diagramsshould be interpreted with consideration for the smallsample sizes.Future researchThe findings from this study suggest the need for furtherresearch among adolescent girls in other regions to as-sess the influence of smoking rates and policies restrict-ing smoking in public spaces on girls’ SHS exposure.From a methodological perspective, although researchershave validated adolescents’ self-reported smoking status[20], research is needed to validate the self-reported SHSexposure measures used in this study. Since youth maybe in situations where they have limited influence on thesmoking behaviors of adults around them, it is recom-mended to evaluate approaches to inform parents andothers who smoke regarding the specific health risks ofSHS exposure for adolescent girls. In addition, an ex-ploration of adolescent girls’ efforts to avoid SHS expos-ure and the effectiveness of their strategies could informinterventions to promote and support girls and youngwomen in their efforts to avoid SHS.ConclusionsSHS exposure among adolescent girls in British Columbiaremains high despite many smoking restrictions in publicspaces. This level of exposure is especially concerninggiven the recent evidence demonstrating an associated in-creased risk of breast cancer [7]. Compared to girls whoreported less frequent exposure, those who reported themost exposure were significantly less likely to report avoi-dance behavior around SHS. Developing approaches to in-form both smokers and those exposed to SHS about thespecific SHS exposure risk to young women is important.Interventions targeting adolescent girls with frequent SHSexposure who are in the pre-contemplation and/or con-templation stages in relation to SHS risk reduction behav-ior are needed to reduce exposure and ultimately breastcancer risk.Additional fileAdditional file 1: Table S1. Sources and related variables of SHSexposure by specific exposure locations among adolescent girls withhigh overall SHS exposure (n = 139).Schwartz et al. BMC Public Health 2014, 14:468 Page 8 of 9http://www.biomedcentral.com/1471-2458/14/468Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsJS was involved in conceptualizing and designing the study, conducting thestatistical analyses, and drafting the manuscript. CG (Co-PI of the STARTstudy) was involved in conceptualizing and designing the study, participatedin the statistical analyses, and revised the manuscript for intellectual content.RG was involved in drafting the manuscript, creating figures, and revising themanuscript. CO was involved in conceptualization of the study. LS wasinvolved in drafting and revising the manuscript. JB (PI of the START study)was involved in conceptualization, design, and revising the manuscript.All authors read and approved the final manuscript.AcknowledgementsThis research was supported by funding from the Canadian BreastCancer Research Alliance and the Canadian Breast Cancer Foundation(grant # 020659), and from the Canadian Institutes of Health Research toChris G. Richardson, PhD.Author details1School of Population & Public Health, University of British Columbia, 2206East Mall, Vancouver, BC V6T1Z3, Canada. 2School of Nursing, University ofBritish Columbia, 3333 University Way, Kelowna, BC V6T 1Z3, Canada.3Professorial Fellow, Faculty of Health Sciences, Australian Catholic University,St. Patrick’s Campus, 115 Victoria Parade, Fitzroy, Victoria 3065, Australia.4Tobacco Treatment and Prevention Division, University of Kentucky Collegeof Nursing, 315 College of Nursing Building, Lexington, KY 40536, USA.Received: 2 February 2014 Accepted: 9 May 2014Published: 17 May 2014References1. Canadian Partnership Agains Cancer: Cancer Control Snapshot: Second-hand smoke and cancer. 2012.2. Wong SL, Malaison E, Hammond D, Leatherdale ST: Secondhand SmokeExposure Among Canadians: Cotinine and Self-Report Measures Fromthe Canadian Health Measures Survey 2007–2009. Nicotine Tob Res 2012,15:693–700.3. Ho S-Y, Lai H-K, Wang M-P, Lam T-H: Exposure to Secondhand Smoke andAcademic Performance in Non-Smoking Adolescents. J Pediatr 2010,157:1012–1017. e1.4. Leatherdale ST, Ahmed R: Second-hand smoke exposure in homes and incars among Canadian youth: current prevalence, beliefs about exposure,and changes between 2004 and 2006. Cancer Causes Control 2009,20:855–865.5. U.S. Department of Health and Human Services: Children and SecondhandSmoke Expoure. Excerpts from The Health Consequences of Involuntary Expoureto Tobacco Smoke: A Report of the Surgeon General. Atlanta, GA: U.S.Department of Health and Human Services, Centers for Disease Control andPrevention, Coordinating Center for Health Promotion, National Center forChronic Disease Prevention and Health Promotion, Office on Smoking andHealth; 2007.6. Johnson KC: Tobacco Smoke and Breast Cancer Risk: Rapid Evolution ofEvidence and Understanding in the Early 21st Century. In CigaretteConsumption and Health Effects. Hauppauge, NY: Nova Science Publishers;2013:1–20.7. Collishaw N, Boyd N, Cantor K, Hammond K, Johnson K, Miller J, Miller A,Miller M, Palmer J, Salmon A, Turcotte F: Canadian Expert Panel onTobacco Smoke and Breast Cancer Risk. 2009.8. Richardson CG, Struik LL, Johnson KC, Ratner PA, Gotay C, Memetovic J,Okoli CT, Bottorff JL: Initial Impact of Tailored Web-Based MessagesAbout Cigarette Smoke and Breast Cancer Risk on Boys’ and Girls’ RiskPerceptions and Information Seeking: Randomized Controlled Trial.JMIR Res Protoc 2013, 2:e53.9. Bottorff JL, McKeown SB, Carey J, Haines R, Okoli C, Johnson KC, Easley J,Ferrence R, Baillie L, Ptolemy E: Young women’s responses to smokingand breast cancer risk information. Health Educ Res 2010, 25:668–677.10. The Chief Public Health Officer’s Report on the State of Public Health inCanada, 2012 - Public Health Agency of Canada. [http://www.phac-aspc.gc.ca/cphorsphc-respcacsp/2012/index-eng.php].11. McBride CM, Emmons KM, Lipkus IM: Understanding the potential ofteachable moments: the case of smoking cessation. Health Educ Res 2003,18:156–170.12. Vozoris N, Lougheed MD: Second-hand smoke exposure in Canada:prevalence, risk factors, and association with respiratory andcardiovascular diseases. Can Respir J 2008, 15:263.13. Prochaska JO, Velicer WF, Rossi JS, Goldstein MG, Marcus BH, Rakowski W,Fiore C, Harlow LL, Redding CA, Rosenbloom D: Stages of change anddecisional balance for 12 problem behaviors. Health Psychol 1994,13:39–39.14. Richardson CG, Schwartz J, Struik LL, Bottorff JL: Adapting the Stage ofChange model to investigate adolescent behavior related to reducingsecond hand smoke exposure. Open J Prev Med 2013, 03:160–164.15. Healey B, Hoek J, Wilson N, Thomson G, Taylor S, Edwards R: Youthexposure to in-vehicle second-hand smoke and their smokingbehaviours: trends and associations in repeated national surveys(2006–2012). Tob Control 2013.16. Provincial and Territorial Smoke-Free Legislation. [www.nsra-adnf.ca/cms/file/files/all_jurisdictions_Mar_2012(1).pdf]17. Van Deusen A, Hyland A, Travers MJ, Wang C, Higbee C, King BA, Alford T,Cummings KM: Secondhand smoke and particulate matter exposure inthe home. Nicotine Tob Res 2009, 11:635–641.18. Summary of Results of the 2010–11 Youth Smoking Survey. [http://www.hc-sc.gc.ca/hc-ps/tobac-tabac/research-recherche/stat/_survey-sondage_2010-2011/result-eng.php].19. Table Nine: Exposure of Children at Home to Environmental TobaccoSmoke (ETS), by Province and Age Group, Canada 2008 - SupplementaryTables - Annual (February-December 2008) - Canadian Tobacco UseMonitoring Survey (CTUMS) 2008. [http://www.hc-sc.gc.ca/hc-ps/tobac-tabac/research-recherche/stat/_ctums-esutc_2008/ann-table 9-eng.php].20. Wong SL, Shields M, Leatherdale S, Malaison E, Hammond D: Assessmentof validity of self-reported smoking status. Health Rep 2012, 23:47–53.doi:10.1186/1471-2458-14-468Cite this article as: Schwartz et al.: An examination of exposure andavoidance behavior related to second-hand cigarette smoke amongadolescent girls in Canada. BMC Public Health 2014 14:468.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitSchwartz et al. BMC Public Health 2014, 14:468 Page 9 of 9http://www.biomedcentral.com/1471-2458/14/468