UBC Faculty Research and Publications

The role of peer victimization in the physical activity and screen time of adolescents: a cross-sectional… Stearns, Jodie A; Carson, Valerie; Spence, John C; Faulkner, Guy; Leatherdale, Scott T Jul 19, 2017

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

Item Metadata


52383-12887_2017_Article_913.pdf [ 473.8kB ]
JSON: 52383-1.0348985.json
JSON-LD: 52383-1.0348985-ld.json
RDF/XML (Pretty): 52383-1.0348985-rdf.xml
RDF/JSON: 52383-1.0348985-rdf.json
Turtle: 52383-1.0348985-turtle.txt
N-Triples: 52383-1.0348985-rdf-ntriples.txt
Original Record: 52383-1.0348985-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessThe role of peer victimization in thephysical activity and screen time ofadolescents: a cross-sectional studyJodie A. Stearns1*, Valerie Carson1, John C. Spence1, Guy Faulkner2 and Scott T. Leatherdale3AbstractBackground: Negative peer experiences may lead adolescents with overweight and obesity to be less active andengage in more sitting-related behaviors. Our study is among the first to empirically test these associations andhypothesized that 1) peer victimization would mediate the negative association between body weight status andmoderate-to-vigorous physical activity (MVPA), and 2) peer victimization would mediate the positive associationbetween body weight status and screen time. Differences by gender were also explored.Methods: Participants were a part of the Year 1 data (2012–2013) from the COMPASS study, a prospective cohortstudy of high school students in Ontario and Alberta, Canada. The final sample consisted of 18,147 students ingrades 9 to 12 from 43 Ontario secondary schools. The predictor variable was weight status (non-overweight vs.overweight/obese), the mediator was peer victimization, and the outcome variables were screen time and MVPA.Multilevel path analysis was conducted, controlling for clustering within schools and covariates. A few differenceswere observed between males and females; therefore, the results are stratified by gender.Results: For both males and females peer victimization partially mediated the association between weight statusand screen time. Specifically, females with overweight/obesity reported 34 more minutes/day of screen time thandid females who were not overweight and 2 of these minutes could be attributed to experiencing peervictimization. Similarly, males who were overweight/obese reported 13 more minutes/day of screen time than themales who were not overweight and 1 of these minutes could be attributed to experiencing more victimization.Males and females who were overweight/obese also reported less MVPA compared to those who were notoverweight; however, peer victimization did not mediate these associations in the hypothesized direction.Conclusions: We found that higher rates of peer victimization experienced by adolescents with overweight andobesity partially explained why they engaged in more screen time than adolescents who were not overweight.However, the effects were small and may be of limited practical significance. Because this is one of the first studiesto investigate these associations, more research is needed before bully prevention or conflict resolution training areexplored as intervention strategies.Keywords: Negative peer experiences, Peer victimization, Mediation, Adolescents, Youth, Adolescents, Physicalactivity, Screen time, Sedentary behavior* Correspondence: jodie.stearns@ualberta.ca1Faculty of Physical Education and Recreation, University of Alberta, 1-113Van Vliet Complex, Edmonton, AB T6G 2H9, CanadaFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Stearns et al. BMC Pediatrics  (2017) 17:170 DOI 10.1186/s12887-017-0913-xBackgroundParticipating in regular physical activity (PA) is importantfor maintaining a healthy body weight, overall cardiovas-cular and psychological health, and motor skill develop-ment in children and adolescents [1, 2]. Limiting timespend sitting (i.e., sedentary behavior) is also importantfor the health of young people [3]. Screen-related behav-iors in particular, which are often done while sitting, areknown to be associated with poor health outcomes. Forinstance, a recent review found that overall screen timeand/or different screen-related behaviors (e.g., TV viewing,playing video games) were associated with unhealthy bodycomposition, cardiometabolic risk, and behavioral con-duct, and lower levels of fitness, pro-social behavior, andself-esteem [4]. Despite the known benefits of healthy ac-tive living, 95% of Canadian adolescents (aged 12–17 years)are insufficiently active and 76% engage in excessivescreen time [5]. Adolescents who are overweight or obesemay be particularly vulnerable as they tend to exhibit evenlower rates of PA and higher rates of screen time com-pared to their non-overweight counterparts [6, 7]. This isa particular concern because those who establish un-healthy habits early on in life tend to maintain them intoadulthood [8–10]. To inform interventions and healthpromotion programs, it is important to gain an under-standing of why adolescents who are overweight or obesetend to be less active and engage in higher levels of screentime.Extensive research demonstrates how low PA and ex-cessive screen time are risk factors for overweight/obes-ity [1, 4]. However, youth who are overweight or obesealso face unique barriers, including weight stigma anddiscrimination that increases their vulnerability to un-healthy behaviors, and perpetuates a “vicious cycle” forthese individuals [11–13]. Salvy and colleagues [11] re-cently proposed a theoretical framework describing theassociation between overweight/obesity (i.e., body weightstatus) and PA, and sedentary behavior, and the negativerole that peers can play on these associations in youngpeople. Specifically, peer social context, including thepresence or absence of peer adversity (e.g., peervictimization, peer rejection) and social isolation (e.g.,ostracism, loneliness), is proposed to mediate the nega-tive association between body weight status and PA andthe positive association between body weight status andsedentary behavior. Testing this model could provideimportant insights into interventions designed to getyouth who are overweight and obese moving more andaway from screens, such as school-level bully preventionprograms or conflict resolution training.Bullying is one aspect of the peer social context that isof particular concern. It is described as an “aggressivegoal directed behavior that harms another individualwithin the context of a power imbalance” [14]. Forms ofbullying include verbal (e.g., teasing), physical (e.g., hit-ting) and relational attacks (e.g., spreading rumors).Bullying can occur in person or through the internet orother computer technology (e.g., texting, emails, socialnetwork sites); the latter of which is described as “cyber-bulling” [15]. The experience of being bullied is called“peer victimization” and is the focus of this study.Research has shown that youth that are overweight orobese are more likely to experience peer victimization[16–21]. Specifically, their excess body weight is a phys-ical characteristic that makes them stand out from theirpeers, putting them at increased risk for being victimized[22]. For example, in a large sample of Canadian adoles-cents aged 11 to 16 years old, Janssen et al. [18] ob-served rates of peer victimization to be 10.7%, 14.4%,and 18.5% in healthy weight, overweight, and obese par-ticipants, respectively. Further, a recent meta-analysisconfirmed this association does not differ by gender[21]. Adolescents perceive that weight-related stigma isthe primary reason that peer victimization occurs, andverbal attacks are the most common type ofvictimization (e.g., made fun of, called names, teased)[23]. Among a sample of adolescents seeking weight-losstreatment, 64% had experienced weight-basedvictimization and, of these participants, 78% had en-dured the teasing/bulling for one year, and 36% had ex-perienced the attacks for five years, with peers (92%) andfriends (70%) being the most common perpetrators [24].Bullying often occurs in PA settings. For instance, Puhland others [23] found that 85% of participants in theirstudy had witnessed weight-based teasing during PA and58% had observed this behavior at least sometimes, often,or very often. Observational studies reveal that higherrates of peer victimization is associated with lower phys-ical education (PE) attendance, and less PA [25–27], andweight criticism during sports and PA is associated withlower sport enjoyment and lower participation in mild-intensity PA [28]. Further, a recent systematic review of 15qualitative studies found that adolescents with overweightor obesity reported peer victimization, including social ex-clusion, stereotyping, verbal bullying, and physical bully-ing, as barriers to PA participation [13].Two studies from the Youth Risk Behavior Survey sug-gest that negative peer experiences can lead to higherlevels of screen time in adolescents in grades 9–12 [29,30]. One found that being bullied in the last 12 monthswas associated with reporting ≥3 h of TV viewing perday in males, and ≥3 h per day of computer use in bothmales and females in grades 9–12 [29]. The otherobserved that females who were bullied on school prop-erty in the last 12 months had an increased odds of ac-cumulating ≥3 h/day of video game/computer use,although no associations were found for males. Thus, itseems plausible that higher levels of peer victimizationStearns et al. BMC Pediatrics  (2017) 17:170 Page 2 of 11experienced by adolescents with overweight and obesitymay help explain why they tend to shy away from activ-ity and stray towards screen-based behaviors.There are several potential reasons why greater peervictimization may lead to less PA and greater time insitting-related behaviors. Salvy and authors [11] pro-posed that negative peer interactions elicit psychological“pain” which impairs executive function and induces ap-athy. As the individual tries to cope with the pain, theymay be more likely to choose sedentary activities such asscreen-based behaviors. Those who experience peervictimization may also avoid PA settings due to fear ofbeing bullied, reduced enjoyment of PA, and/or becausethey are socially excluded and/or not invited to partici-pate in PA activities [13, 27, 28].To our knowledge, the framework proposed by Salvyet al. (2012) has yet to be empirically tested. Though thecausal pathways cannot be rigorously tested in cross-sectional designs [31], such studies can be useful as afirst step in obtaining a snapshot of concurrent associa-tions and to justify the need for conducting longitudinalstudies [32]. The first aim of the study was to examinewhether peer victimization mediates the negative associ-ation between body weight status and PA in adolescents.The second aim was to investigate whether peervictimization mediates the positive association betweenbody weight status and screen time in adolescents.Consistent with the theoretical framework by Salvy et al.[11], it was hypothesized that peer victimization wouldmediate the associations between body weight status andboth PA and screen time. Because some differences existbetween males and females in the literature, differencesby gender were also explored.MethodsDesign and procedureThis cross-sectional study uses data from Year 1 (2012–2013 school year) of the COMPASS study. COMPASS is aprospective cohort study designed to annually collect hier-archical longitudinal data from a convenience sample of24,173 grade 9 to 12 students attending 43 secondaryschools in Ontario, Canada. Eligible students were re-cruited via an active-information passive-consent proced-ure. Parents were mailed an information letter, and weretold to contact the COMPASS research coordinator if theydid not want their child to participate. This procedureallowed us to obtain robust data, achieve higher participa-tion rates (82.1% participation rate among eligible stu-dents), and maintain student confidentiality. Eligiblestudents willing to participate provided their assent andcompleted surveys during class time. Eligible studentscould withdraw or decline to participate at any time, andwere assured that their answers would be kept confiden-tial, and that no one at their school or home would knowhow they responded. Honest responses to the questionswere also encouraged. All procedures were approved bythe University of Waterloo Office of Research Ethics andparticipating School Boards. More information on theCOMPASS study methods and procedures can be foundin print [33] or online [34].MeasuresModerate-to-vigorous physical activity (MVPA) wasassessed with two questions including time per dayspent doing moderate (e.g., walking, biking to school,recreational swimming) and hard (e.g., jogging, teamsports, fast dancing, jump-rope) physical activities oneach of the last 7 days. The scores for moderate andhard physical activities from each day were summed anddivided by 7 to create an average minutes of MVPA/dayscore. Hours per day of MVPA was then calculated bydividing minutes/day by 60. Screen time was assessedwith three questions including usual time per day spentwatching/streaming TV shows or movies, playing video/computer games, and surfing the internet. The responsesto the three questions were summed to create the screentime variable. Hours of screen time/day was then calcu-lated by dividing minutes per day by 60. 1-week test-retest reliability intraclass correlation coefficients (ICC)for this scale have been reported as 0.75 for MVPA, 0.54for watching TV shows/movies, 0.65 for video/computergames, and 0.71 for surfing the internet [35]. Whencompared to accelerometer-measured PA, the criterionvalidity ICCs were 0.22 for moderate PA, 0.18 for hardPA, and 0.25 for MVPA. These findings are comparableto other studies that examined the association betweenself-report PA measures and accelerometers [36].Bullying was defined as physical attacks (e.g., gettingbeaten up, pushed, or kicked), verbal attacks (e.g., get-ting teased, threatened, or having rumors spread aboutyou), cyber-attacks (e.g., being sent mean text messagesor having rumors spread about you on the internet), andtheft or damage of property. Frequency of peervictimization was assessed with one question: “In the last30 days, how often have you been bullied by other stu-dents?” Response options included a) I have not beenbullied by other students in the last 30 days, b) less thanonce a week, c) about once a week, d) 2 or 3 times aweek, or e) daily or almost daily. For ease of interpret-ation peer victimization was collapsed into 2 categoriesincluding 1) was not bullied in the last 30 days and 2)was bullied in the last 30 days. These questions are simi-lar to the “global” measure of peer victimization fromthe Olweus Bully/Victim Questionnaire [37] and toother adolescent population health surveys such as theOntario Student Drug Use and Health Survey [38] andthe Health Behavior in School-aged Children studywhich was conducted in 33 countries [16, 18].Stearns et al. BMC Pediatrics  (2017) 17:170 Page 3 of 11Weight status was assessed using two self-reportedheight and weight questions [39] that are consistent withother large-scale surveys [40, 41]. Body mass index(BMI) was calculated as kg/m2 and age- and sex- specificnon-overweight (coded as 0), overweight/obese weightstatus (coded as 1) categories were calculated based onWorld Health Organization standards [42]. In a valid-ation study, the 1-week test-retest reliability ICCs were0.96 for height, 0.99 for weight, and 0.95 for BMI [39].Concurrent validity ICCs of self-reported and objectivelymeasured values were 0.88 for height, 0.84 for weight,and 0.84 for BMI.Covariates included grade, ethnicity/race, weeklyspending money, and future education plans. Ethnicity/race was assessed with the question “How would youdescribe yourself?” (mark all that apply). Responses werecollapsed into White, Black, Asian, Aboriginal (FirstNations, Metis, Inuit), Latin American/Hispanic, andmixed/other.Because adolescents are not necessarily aware of theirhousehold income and the education levels of their par-ents [43], weekly spending money and future educationplans were used as indicators of personal economic sta-tus. Adolescents whose parents have attained a highereducation tend to have a higher disposable income interms of weekly allowance and job income [44], andweekly spending money has been shown to be positivelyassociated with vigorous exercise and watching TVamong adolescents [43]. Research in Norway found thatplans for higher education were highly stable across ado-lescence, and the participants’ educational plans tendedto correspond well with their parents education [45].Weekly spending money was assessed with the question“About how much money do you usually get each weekto spend on yourself or to save?”, and included moneyfrom allowances and jobs like babysitting and deliveringpapers. To be consistent with other COMPASS studies,[46–49] and in order to retain as many cases as possible,this variable was collapsed into “zero”, “$1–20”, “$21–100”, “> $100”, and “don’t know”. Future education planswas assessed with the question “What is the highest levelof education you think you will get?” with six responseoptions including completed high school or less; college/trade/vocational certificate; university bachelor’s degree;university master’s/PhD/law school/medical school/teachers’ college degree; and I don’t know.AnalysisPreliminary analyses were completed using IBM SPSSVersion 22. Univariate outliers for the dependent variableswith a z-score above 3 or below −3 (screen time = 463cases, MVPA = 335 cases) were coded as missing. Afurther 481 multivariate outliers (all standardized residual>3) for screen time and 191 multivariate outliers (allstandardized residuals >3) for MVPA were detected andcoded as missing. Coding the outliers as missing allowedthese values to be estimated in the main analysis. The as-sumptions of homoscedasticity and multivariate normalitywere met. The error variance also appeared to be similaracross schools. An inspection of the bivariate correlationsshowed no evidence of multicollinearity (i.e., r’s < .70 andVIF < 10).Multilevel path analysis, controlling for clustering byschools, was used to test the multiple meditation model.Mediation was examined using the product of coefficientmethod (Cerin & MacKinnon, 2009). It involved estimat-ing 1) the associations between weight status and peervictimization (α path coefficient), 2) the associationbetween peer victimization and the outcome variableswhile controlling for weight status (β path coefficient),and 3) the mediated effect (αβ path coefficient). Thoughprevious methods required a significant pathway betweenthe predictor and outcome variables to proceed with me-diation analysis, new procedures do not require this step[50]. However, both the total effects (i.e., association be-tween the predictor and outcome variables) and direct ef-fects (i.e., the association between the predictor andoutcome variables with the indirect effect removed) willstill be presented. The mediated effect (or indirect effect)is the estimated effect of weight status on MVPA andweight status on screen time through peer victimization.Because weight status is dichotomous, the indirect effectcan be interpreted as the mean difference between groups(non-overweight vs overweight/obese) in units of the out-come (MVPA, screen time) attributable to the pathwaythrough peer victimization [51]. The significance of themediation effect p < .05 and the 95% confidence intervalsprovided evidence of mediation [50].The path analysis was computed in Mplus Version 7.1using the WLSMV estimator, which employs “weightedleast square parameter estimates using a diagonal matrixwith standard errors and a mean- and variance-adjustedchi-square test statistic that use a full weight matrix”[52]. Probit regression was used to test associations be-tween the control variables and peer victimization, andbody weight status and peer victimization. Linear regres-sion was used to test all associations with screen timeand MVPA. This resulted in a fully saturated model andtherefore model fit statistics were not available. Grade,ethnicity/race, weekly spending money, and future edu-cation plans were added as control variables by includingthem as exogenous variables predicting all outcome vari-ables in the model. In a preliminary analysis, differencesby gender were explored within the proposed model.When comparing differences by subgroups, formal testsof moderation are recommended [53]. If significant dif-ferences exist, stratification by groups is justified. Thus,to test for gender moderation on each pathway,Stearns et al. BMC Pediatrics  (2017) 17:170 Page 4 of 11interaction terms were created between weight statusand gender, and weight status and peer victimization.The interaction terms were then tested for their effecton the outcome variables one by one within the model,with gender included as a main effect. Significant differ-ences were found on two pathways; therefore, the modelis presented separately for males and females.All of the outcome variables were missing on less than5% of the cases, and 2.5% of the total cases in the datasetwere missing. Missingness on the outcome variables waspredicted by multiple variables including variables fromthe larger dataset that are not part of the main analysis.We therefore assumed that the data was missing at ran-dom and estimated the missing cases using full-information maximum likelihood. The variables predict-ive of missingness but not included in the analysis (i.e.,participation in school and non-school sports, whetherthe last week was a typical week for PA, perceivedsupport for bullying from the school) were added as aux-iliary variables. Cases missing on all variables (n = 13) orone of the x-variables (i.e., weight status and all covari-ates; n = 6142) were excluded from the analysis. Thisresulted in a final sample size of 18,147 participants.ResultsTable 1 presents the sociodemographic information. Ap-proximately half of the sample was female (49%) and 73-75% were white. Table 2 presents the descriptive infor-mation for the model variables. Specifically, 19% of fe-males and 32% of males were overweight or obese and21% of females and 15% of males had been victimized atleast once during the last 30 days. On average, femalesreported 4.5 h per day of screen time and 1.8 h per dayof MVPA and males reported 5.2 h per day of screentime and 2.2 h per day of MVPA.Gender differencesSignificant gender differences were found for two path-ways. Specifically, gender moderated the associationbetween peer victimization and screen time(B = 0.380 ± 0.073, p < .001), with females having astronger association than males. The associationbetween weight status and screen time was also moder-ated by gender (B = −0.368 ± 0.098, p < .001), withfemales having a stronger association than males.Path analysis - femalesThe full model for females including unstandardizedcoefficients and standard errors is presented in Fig. 1.All analyses adjusted for grade, ethnicity/race, weeklyspending money, and future education plans. Weightstatus was positively associated with peer victimization(α path coefficient; B = 0.139 ± 0.041, p = 0.001). Peervictimization was positively associated to screen time (βTable 1 Sociodemographic informationCharacteristic Females(n = 8904)Males(n = 9243)Grade – count (%)9 2112 (23.7) 2233 (24.2)10 2336 (26.2) 2305 (24.9)11 2255 (25.3) 2333 (25.2)12 2201 (24.7) 2372 (25.7)Ethnicity/race – count (%)White 6653 (74.7) 6701 (72.5)*Black 269 (3.0) 420 (4.5)*Asian 522 (5.9) 503 (5.4)Aboriginal 210 (2.4) 257 (2.8)Latino/Hispanic 164 (1.8) 223 (2.4)*Other/Mixed 1086 (12.2) 1139 (12.3)Anticipated education level – count (%)High school diploma or graduationequivalency or less328 (3.7) 471 (5.1)*College/trade/vocational certificate 1754 (19.7) 2874 (31.1)*University Bachelor’s degree 2317 (26.0) 2289 (24.8)Master’s/PhD/law school/medicaldegree/teachers’ college degree3212 (36.1) 2410 (26.1)*I don’t know 1293 (14.5) 1199 (13.0)*Weekly spending money – count (%)Zero 1229 (13.8) 1443 (15.6)*$1–20 2688 (30.2) 2758 (29.8)$21–100 2743 (30.8) 2431 (26.3)*$100+ 1205 (13.5) 1572 (17.0)*I don’t know 1039 (11.7) 1039 (11.2)Differences by gender tested via chi-square tests of independence*indicates significant differences (p < .05) between males and femalesTable 2 Descriptive statistics for the main model variablesCharacteristic Females(n = 8904)Males(n = 9243)Weight status – count (%)Non-Overweight 7189 (80.7) 6312 (68.3)*Overweight/Obese 1715 (19.3) 2931 (31.7)Peer victimization – count (%)None 7012 (79.2) 7781 (85.0)*At least once in the past 30 days 1841 (20.8) 1371 (15.0)Daily screen time – mean hours/day (SD) 4.500 (2.675) 5.231 (2.729)*Daily MVPA – mean hours/day (SD) 1.758 (1.146) 2.175 (1.262)*MVPA moderate-to-vigorous physical activityDifferences by gender tested via chi-square tests of independence (weightstatus, peer victimization), and independent samples t-tests (screen time, MVPA)Numbers in the table may not tally to the total N due to missing data*indicates significant differences (p < .05) between females and malesStearns et al. BMC Pediatrics  (2017) 17:170 Page 5 of 11coefficient; B = 0.291 ± 0.039, p < .001). Further, 8%(R2 = .083) of the variance was explained for screentime; however, was reduced to 2% (R2 = .024) when thecovariates were removed.The total, direct and indirect effects are presented inTable 3. The total effect of weight status on screen timewas significant and indicates that the females with over-weight/obesity participated in 0.562 more hours per day(or 34 min per day) of screen time than the females whowere not overweight (± 0.074, p < .001). When the indirecteffects were taken into account, the direct pathway fromweight status to screen time remained significant(B = 0.521 ± 0.075, p < .001). The indirect effect of weightstatus on screen time through peer victimization was sig-nificant. Specifically, there was 0.040 additional hours perday (or 2 min per day) of screen time in the females withoverweight/obesity compared to females who were notoverweight that could be attributed to increased peervictimization (± 0.014, p = .004). Therefore, our first hy-pothesis that peer victimization mediates the positive as-sociation between weight status and screen time waspartially supported. Greater peer victimization may par-tially explain why adolescents with overweight and obesityengage in more minutes of screen time than those whoare not overweight. However, the effects are very smalland the practical significance of such findings arequestionable.As mentioned previously, weight status was positively as-sociated with peer victimization (α path coefficient;B = 0.139 ± 0.041, p = 0.001). Unexpectedly, peervictimization was positively associated with MVPA (β coef-ficient; B = 0.105 ± 0.015, p < .001). Further, 5% (R2 = .049)of the variance was explained for MVPA; however, theseproportions were reduced to 1% (R2 = .013) for MVPAWeight StatusMVPAScreen TimePeerVictimization0.291 (0.039)***-0.088 (0.027)**0.105 (0.015)***0.521 (0.075)***0.139 (0.041)**Fig. 1 The final model for females with unstandardized beta values and standard errors. Non-significant pathways are indicated by a dotted line.Weight status is coded as “non-overweight” = 0, “overweight/obese” = 1. Peer victimization is coded as “has not been bullied in the last 30 days” = 0,“has been bullied at least once in the last 30 days” = 1. MVPA = moderate-to-vigorous physical activity. p < .05; **p < .01; ***p < .001. Model wasadjusted for grade, ethnicity/race, weekly spending money, and future education plansTable 3 Unstandardized path coefficients for direct, total indirect, specific indirect, and total effects (N = 18,147)Model OutcomesScreen Time (hours/day) MVPA (hours/day)Coefficient (SE) p-value 95% CI Coefficient (SE) p-value 95% CIFemalesDirect Effects 0.521 (0.075) <.001 0.375, 0.668 −0.088 (0.027) .001 −0.141, −0.036Indirect Effects 0.040 (0.014) .004 0.013, 0.068 0.015 (0.004) .001 0.006, 0.023Total Effects 0.562 (0.074) <.001 0.419, 0.708 −0.074 (0.027) .006 −0.127, −0.021MalesDirect Effects 0.194 (0.053) <.001 0.090, 0.298 −0.058 (0.027) .034 −0.112, −0.004Indirect Effects 0.019 (0.008) .012 0.004, 0.035 0.003 (0.002) .197 −0.001, 0.007Total Effects 0.214 (0.053) <.001 0.109, 0.319 −0.056 (0.027) .043 −0.109, −0.002Unstandardized path coefficients are presentedMVPA moderate-to-vigorous physical activityWeight status is coded as 0 = non-overweight, 1 = overweight/obeseModel was adjusted for grade, ethnicity/race, weekly spending money, and future education plansStearns et al. BMC Pediatrics  (2017) 17:170 Page 6 of 11when the covariates were removed. The pathway betweenweight status and MVPA (i.e., total effect) was significantfor females. This indicates that females with overweight/obesity participated in 0.074 less hours per day (or 4 minper day) of MVPA than the females who were not over-weight (± 0.027, p = .006). When the indirect effects wereaccounted for, the direct effect between MVPA and weightstatus remained significant (p = .001). The indirect effectthrough peer victimization was also significant. Specifically,there was 0.015 additional hours per day (or 1 min per day)of MVPA in the females who were overweight/obese com-pared to the females that were not overweight that couldbe attributed to increased peer victimization (± 0.004,p = .001). Therefore, our second hypothesis was not sup-ported for females. Females with overweight and obesitydid engage in less MVPA and were more likely to have beenvictimized compared to the adolescents who were not over-weight; however, those who were victimization tended toperform more MVPA.Path analysis – MalesThe full model for males including unstandardized coef-ficients and standard errors is presented in Fig. 2. Allanalyses adjusted for grade, ethnicity/race, weekly spend-ing money, and future education plans. Weight statuswas positively associated with peer victimization (α pathcoefficient; B = 0.094 ± 0.035, p = 0.008). Controlling forweight status, peer victimization was positively associ-ated with screen time (β coefficients; B = 0.208 ± 0.043,p < .001). Further, 4% (R2 = .041) of the variance was ex-plained for screen time, however these proportions werereduced to 0.9% (R2 = .009) for screen time when the co-variates were removed.The total, direct and indirect effects for the male modelare presented in Table 3. The pathway between weightstatus and screen time (i.e., total effect) was significant formales. This indicates that the males with overweight/obes-ity participated in 0.214 more hours per day (or 13 minper day) of screen time than the males who were not over-weight (± 0.053, p < .001). When the indirect effects wereaccounted for, the direct effect from weight status andscreen time remained significant (p < .001). The indirecteffect through peer victimization was also significant(B = 0.019 ± 0.008, p = .012). Specifically, there was 0.019additional hours per day (or 1 min per day) of screen timein the males with overweight/obesity compared to themales who were not overweight that could be attributedto increased peer victimization. Thus, our first hypothesisthat peer victimization mediates the positive associationbetween weight status and screen time was partially sup-ported in males. Greater peer victimization may partiallyexplain why males who are overweight or obese engage inmore minutes of screen time than do males who are notoverweight. However, the effects are very small and thusmay not have practical significance.As mentioned, weight status was positively associatedwith peer victimization (α path coefficient;B = 0.094 ± 0.035, p = 0.008). Controlling for weight sta-tus, peer victimization was not associated with MVPA(B = 0.029 ± 0.020, p = 0.149). Further, the modelexplained 4% (R2 = .035) of the variance in MVPA; how-ever, these proportions were reduced to 0.1% (R2 = .001)for MVPA when the covariates were removed. The path-way between weight status and MVPA (i.e., total effect)was significant for males. This indicates that males withoverweight/obesity participated in 0.056 less hours per dayof MVPA (or 3 min per day) than males who were notoverweight (± 0.027, p = .043). When the indirect effectswere accounted for, the association between weight statusand MVPA (i.e., direct effect) remained significantWeight StatusMVPAScreen TimePeerVictimization0.208 (0.043)***-0.058 (0.027)*0.029 (0.020)0.194 (0.053)***0.094 (0.035)**Fig. 2 The final model for males with unstandardized beta values and standard errors. Non-significant pathways are indicated by a dotted line. Weightstatus is coded as “non-overweight” = 0, “overweight/obese” = 1. Peer victimization is coded as “has not been bullied in the last 30 days” = 0, “hasbeen bullied at least once in the last 30 days” = 1. MVPA = moderate-to-vigorous physical activity. *p < .05; ** p < .01;***p < .001. Model was adjustedfor grade, ethnicity/race, weekly spending money, and future education plansStearns et al. BMC Pediatrics  (2017) 17:170 Page 7 of 11(p = .034). The indirect effect through peer victimizationwas not significant (p = .197). Thus, our second hypothesisthat peer victimization mediates the positive associationbetween weight status and MVPA was not supported.DiscussionOur study is among the first to empirically examinewhether negative peer experiences help explain why ado-lescents who are overweight and obese tend to engage inmore screen time and be less active than those who arenot overweight. Previous research has shown that over-weight and obese adolescents tend to have fewer friend-ship nominations and to be on the periphery offriendship networks [54]. These individuals are often vic-tims of bullying [16–19], and this type of conflict isknown to have negative effects on psychological health[55]. Consistent with the model proposed by Salvy et al.[12], our findings suggest that in both males and femalesthe association between weight status and higher levelsof screen time is partially explained by peervictimization. Specifically, we found evidence of partialmediation whereby females with overweight/obesity re-ported 34 more minutes of screen time per day than didfemales who were not overweight and 2 of these minutescould be attributed to experiencing peer victimization.Similarly, males who were overweight/obese reported 13more minutes of screen time per day than the maleswho were not overweight and 1 of these minutes couldbe attributed to experiencing more victimization. Malesand females who were overweight/obese also reportedless MVPA compared to those who were not overweight;however, peer victimization did not mediate theseassociations in the hypothesized direction.Similarly, Vanderwater et al. [56] found that adoles-cents with overweight/obesity spent less time withfriends, which resulted in them being less active, andsubsequently led to spending more time watching TV.This study along with ours highlights some of the peerdifficulties (i.e., peer victimization, less time spent withfriends) adolescents with overweight/obesity face andhow these issues can influence their screen-related activ-ities. It is also consistent with a growing body of litera-ture suggesting that weight-related stigma and peerdifficulties experienced by individuals who areoverweight negatively impact health behaviors [12].Though the overall findings were similar for males andfemales, a few differences were observed. Specifically,stronger associations were observed between weight sta-tus and screen time, and peer victimization and screentime for females (yet still significant for males). Higherrates of peer victimization were also observed forfemales (21%) compared to males (15%). It is possiblethe psychological impacts of peer victimization areslightly more harmful to females compared to males, orthat the type of victimization experienced by femalestends to be more detrimental. For example, across stud-ies females who are overweight/obese have a lower qual-ity of life than do males who are overweight/obese [57].Further, females are more often teased about theirweight, and report being more bothered by these experi-ences compared to males [58]. Females may also feelmore pressure to conform to societal body ideals, andthus peer victimization could have greater impacts onbody image and consequently their health and health be-haviors [59]. Dating may be one source of this pressureas obese females are less likely to date than their peers,yet this difference is not seen in males [17]. Further, fe-males tend to report greater bullying when they believetheir body is too fat; whereas, boys report greater bully-ing when their believe their body is too thin [16].We do acknowledge that the mediation effect of peervictimization on the weight status-screen time associ-ation is small. Some studies have found that adolescentswith overweight and obesity underreport their weightcompared to healthy weight adolescents, which can re-sult in a misclassification of adolescents with overweightas healthy weight [60]. Therefore, it is possible that someof the overweight participants were misclassified as non-overweight thereby reducing the magnitude of theassociations. Indeed, we found the rates of overweight/obesity in this sample (25.6%) to be lower than nationalrates in Canada (33.2%) [61]. However, the format of theCOMPASS height and weight questions are slightlydifferent than previous surveys, which may influencefindings. For example, in the review by Sherry et al. [60]females were found to underreport their weight morethan males, yet this bias was not found in the COM-PASS survey [39]. Also, a recent meta-analysis foundthat the strength of association between weight statusand peer victimization did not differ between studiesthat used self-report vs. objective measures of heightand weight, suggesting that self-reported weight statusdoes not bias this association [21].Further, it should be mentioned that when peervictimization was taken into account, there was still adirect association between weight status and screen timein both males and females. This suggests that there areother unmeasured mechanism(s) that explain theseassociations. As previously mentioned, Vanderwater andco-authors [56] found that time spent with friends wasan important mediator of the weight status-TV time as-sociation; therefore, future research will benefit fromexamining the individual and combined impacts offriendship (e.g., presence of a friend, number of friends,time spent with friends) and negative peer experiences.Other research has found that having a best friendbuffers a child from the negative psychosocial conse-quences of peer victimization [62]; therefore, anotherStearns et al. BMC Pediatrics  (2017) 17:170 Page 8 of 11possible avenue to explore is whether having one ormore friends negates or reduces the negative impacts ofpeer victimization on screen time.The findings of this study were not consistent betweenscreen time and MVPA. Although weight status wasnegatively associated with MVPA as hypothesized, peervictimization was positively associated with MVPA in fe-males, and unassociated in males. Consequently, it didnot mediate the weight status-MVPA association in thehypothesized direction. This is surprising consideringpeer victimization is associated with lower PA and PE at-tendance [25–27], and adolescents with overweight/obesity describe victimization as a barrier to PA partici-pation [13]. Again, the potential underreporting ofweight (and subsequently BMI) [60] in the participantswith overweight and obesity could have attenuated thefindings. Another potential explanation is that partici-pants with overweight/obesity overreported their PAcompared to the participants who were not overweight,however this phenomenon is less consistent in the litera-ture [63, 64]. Finally, it is possible that some adolescentswith overweight/obesity may be more resilient and lessaffected by victimization from their peers [65]. For in-stance, Faith and colleagues [28] found that childrenwho were criticized for their weight participated in lessmild-intensity PA but this association was moderated byproblem-focused coping skills, such that weight criticismdid not lead to lower PA in children who could copewith the criticisms. As this is one of the first studies totest the model proposed by Salvy et al. [11], and becausethe associations between weight status, peervictimization, and MVPA are supported by previous re-search and theory, we suggest that researchers continueto examine these associations in greater detail amongboth children and adolescents.Future research will be important for advancing the the-ory around weight status, negative peer experiences, andPA, sedentary behavior, and screen time. Studies should in-vestigate whether the impact of peer victimization on PAand screen time is specific to weight-based teasing (ratherthan general peer victimization), and examine the full rangeof negative peer experiences that young people encounter(e.g., peer rejection, lack of friends, ostracism, loneliness).Ecological momentary assessment and natural observationsin PE classes and playgrounds could be used to investigatewhether PA tends to decrease, and/or sedentary behaviorincreases immediately after a peer victimization experience.Further, studies should examine these associations using anobjective measure of sedentary behavior (e.g., inclinometer,accelerometer) in addition to a screen time measure, andideally adopt longitudinal designs. When the data collectionis complete, the COMPASS study will have four waves ofdata, and we will be able to examine these associationsacross multiple time points.The strengths of our study include the large sampleand the wide age range of participants (grades 9–12).The multilevel multiple path analysis allowed us to con-trol for clustering within schools, and to examine mul-tiple outcome variables simultaneously. Additionally,some limitations should also be noted. First, the study iscross-sectional, and thus we cannot be certain that thepaths in the model are specified accurately. In the ab-sence of temporal precedence, it is recommended thatthe causal sequence be informed by theory [66]. Futureresearch using additional waves of the COMPASS studywill allow us to establish the temporal sequence of theassociations. Second, all of the measures were self-reported, which are known to have associated biases,and this could have impacted study findings. Further,though we recognize that a multiple item measure ofpeer victimization would have been ideal, the one-itemmeasure used in this study is consistent with mostpopulation-based studies in the health literature [18, 67].In addition, the use of self-report measures allowed theCOMPASS study to collect information from a verylarge sample of students, where objective measures arenot feasible within the passive consent protocol requiredfor collecting substance use data. Also, our model onlyexamined peer victimization, and did not assess the fullrange of potential negative peer experiences that youngpeople encounter. Similarly, we only assessed screentime and thus the findings cannot be generalized to allsitting-related behaviors or sedentary behavior. Finally,despite the large sample size, this was a convenient sam-ple of schools in Ontario and thus the findings may notgeneralize to all schools and students in Ontario.ConclusionPeer victimization partially explains why adolescentswith overweight and obesity engage in higher levels ofscreen time than adolescents who are not overweight.This is one of the first studies to investigate the impactsof peer victimization on the health behaviors of adoles-cents, and thus more research is needed before bullyprevention or conflict resolution training are explored asintervention strategies. The use of objective measuresand longitudinal designs, and examining the immediateimpact of peer victimization within specific contexts willbe useful for progressing this topic area.AbbreviationsBMI: Body mass index; MVPA: Moderate-to-vigorous physical activity;PA: Physical activityAcknowledgementsWe would like to acknowledge the contributions of the project manager,Chad Bredin, along with the entire COMPASS team for their importantcontributions to data collection and the management of the dataset.Stearns et al. BMC Pediatrics  (2017) 17:170 Page 9 of 11FundingThe COMPASS study was supported by a bridge grant from the CanadianInstitutes of Health Research (CIHR) Institute of Nutrition, Metabolism andDiabetes (INMD) through the "Obesity - Interventions to Prevent or Treat"priority funding awards (OOP-110788; grant awarded to S. Leatherdale) andan operating grant from the Canadian Institutes of Health Research (CIHR)Institute of Population and Public Health (IPPH) (MOP-114875; grant awardedto S. Leatherdale). Dr. Carson is supported by a Canadian Institutes of HealthResearch (CIHR) New Investigator salary award. Drs. Faulkner and Leatherdaleare both Chairs in Applied Public Health Research funded by the PublicHealth Agency of Canada (PHAC) in partnership with Canadian Institutes ofHealth Research (CIHR).Availability of data and materialsThis is an ongoing study, therefore the data is not publically available at thistime. Access to the data supporting the findings of the study can berequested from.https://uwaterloo.ca/compass-system/compass-system-projects/compass-study.Authors’ contributionsJAS conceived and designed this particular study, performed the statisticalanalysis, and drafted the manuscript. VC assisted with the conception anddesign of the study, made substantial contributions to the analysis andinterpretation of data, and revised the manuscript critically for importantintellectual content. JCS made substantial contributions to the analysis andinterpretation of data, and revised the manuscript critically for importantintellectual content. GF made substantial contributions to the analysis andinterpretation of data, and revised the manuscript critically for importantintellectual content. STL conceived of the COMPASS study and wrote thefunding proposal, developed the tools, and is leading study implementationand coordination. For this particular study, STL made substantialcontributions to the analysis and interpretation of data, and revised themanuscript critically for important intellectual content. All authors read andapproved the final manuscript.Ethics approval and consent to participateEthics approval was granted by the University of Waterloo Office of ResearchEthics (ORE #17264). Participating school board and school ethicscommittees approved all procedures. Parents provided consent through anactive-information passive-consent procedure. Participants could decline toparticipate or withdrawal at any time.Consent for publicationNot applicable.Competing interestsThe authors declare that we have no competing interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Faculty of Physical Education and Recreation, University of Alberta, 1-113Van Vliet Complex, Edmonton, AB T6G 2H9, Canada. 2School of Kinesiology,Faculty of Education, University of British Columbia, Vancouver, Canada.3School of Public Health and Health Systems, University of Waterloo,Waterloo, Canada.Received: 10 August 2016 Accepted: 29 June 2017References1. Poitras VJ, Gray CE, Borghese MM, Carson V, Chaput JP, Janssen I, et al.Systematic review of the relationships between objectively measuredphysical activity and health indicators in school-aged children and youth.Appl Physiol Nutr Metab. 2016;41:S197–239.2. Janssen I, LeBlanc AG. Systematic review of the health benefits of physicalactivity and fitness in school-aged children and youth. IJBNPA. 2010;7:1–16.3. Tremblay MS, Carson V, Chaput J-P, Connor Gorber S, Dinh T, Duggan M, etal. Canadian 24-hour movement guidelines for children and youth: anintegration of physical activity, sedentary behaviour, and sleep. Appl PhysiolNutr Metab. 2016;41:S311–27.4. Carson V, Hunter S, Kuzik N, Gray CE, Poitras V J, Chaput J-P, et al.Systematic review of sedentary behaviour and health indicators in school-aged children and youth: an update 1. Appl Physiol, Nutr, and Metab.2016;41:S240-S265.5. ParticipACTION. PartcipACTION report on physical activity of children andyouth. 2016. http://stage.participaction.com/sites/default/files/downloads/2016%20ParticipACTION%20Report%20Card%20-%20Full%20Report.pdf.Accessed 7 June 2016.6. Janssen I, Katzmarzyk PT, Boyce WF, Vereecken C, Mulvihill C, Roberts C, etal. Comparison of overweight and obesity prevalence in school-aged youthfrom 34 countries and their relationships with physical activity and dietarypatterns. Obes Rev. 2005;6:123–32.7. Janssen I, Katzmarzyk PT, Boyce WF, King MA, Pickett W. Overweight andobesity in Canadian adolescents and their associations with dietary habitsand physical activity patterns. J of Adolesc Health. 2004;35:360–7.8. Biddle SJ, Pearson N, Ross GM, Braithwaite R. Tracking of sedentarybehaviours of young people: a systematic review. Prev Med. 2010;51:345–51.9. Craigie AM, Lake AA, Kelly SA, Adamson AJ, Mathers JC. Tracking of obesity-related behaviours from childhood to adulthood: a systematic review.Maturitas. 2011;70:266–84.10. Telama R. Tracking of physical activity from childhood to adulthood: areview. Obes Facts. 2009;2:187–95.11. Salvy S-J, Bowker JC, Germeroth L, Barkley J. Influence of peers and friendson overweight/obese youths’ physical activity. Exerc Sport Sci Rev.2012;40:127–32.12. Puhl R, Suh Y. Health consequences of weight stigma: implications forobesity prevention and treatment. Curr Obes Rep. 2015;4:182–90.13. Stankov I, Olds T, Cargo M. Overweight and obese adolescents: what turnsthem off physical activity? IJBNPA. 2012;9:53.14. Volk AA, Dane AV, Marini ZA. What is bullying? A theoretical redefinition.Dev Rev. 2014;34:327–43.15. Kowalski RM, Limber SP. Psychological, physical, and academic correlates ofcyberbullying and traditional bullying. J Adolesc Health. 2013;53:S13–20.16. Brixval CS, Rayce SL, Rasmussen M, Holstein BE, Due P. Overweight, bodyimage and bullying—an epidemiological study of 11-to 15-years olds. Eur JPub Health. 2011;22:126–30.17. Pearce MJ, Boergers J, Prinstein MJ. Adolescent obesity, overt and relationalpeer victimization, and romantic relationships. Obes Res. 2002;10:386–93.18. Janssen I, Craig WM, Boyce WF, Pickett W. Associations between overweightand obesity with bullying behaviors in school-aged children. Pediatrics.2004;113:1187–94.19. Lumeng JC, Forrest P, Appugliese DP, Kaciroti N, Corwyn RF, Bradley RH.Weight status as a predictor of being bullied in third through sixth grades.Pediatrics. 2010;125:e1301–7.20. Ottova V, Erhart M, Rajmil L, Dettenborn-Betz L, Ravens-Sieberer U.Overweight and its impact on the health-related quality of life in childrenand adolescents: results from the European KIDSCREEN survey. Qual LifeRes. 2012;21:59–69.21. Van Geel M, Vedder P, Tanilon J. Are overweight and obese youths moreoften bullied by their peers? A meta-analysis on the relation betweenweight status and bullying. Int J Obes. 2014;38:1263–7.22. Juvonen J, Graham S. Bullying in schools: the power of bullies and theplight of victims. Annu Rev Psychol. 2014;65:159–85.23. Puhl RM, Luedicke J, Heuer C. Weight-based victimization towardoverweight adolescents: observations and reactions of peers. J Sch Health.2011;81:696–703.24. Puhl RM, Peterson JL, Luedicke J. Weight-based victimization: bullyingexperiences of weight loss treatment–seeking youth. Pediatrics. 2013;131:e1–9.25. Henriksen P, Rayce S, Melkevik O, Due P, Holstein B. Social background,bullying, and physical inactivity: national study of 11-to 15-year-olds. ScandJ Med Sci Spor. 2015; doi:10.1111/sms.12574.26. Roman CG, Taylor CJ. A multilevel assessment of school climate, bullyingvictimization, and physical activity. J Sch Health. 2013;83:400–7.27. Storch EA, Milsom VA, DeBraganza N, Lewin AB, Geffken GR, Silverstein JH. Peervictimization, psychosocial adjustment, and physical activity in overweight andat-risk-for-overweight youth. J Pediatr Psychol. 2007;32:80–9.28. Faith MS, Leone MA, Ayers TS, Heo M, Pietrobelli A. Weight criticism duringphysical activity, coping skills, and reported physical activity in children.Pediatrics. 2002;110:e23–3.Stearns et al. BMC Pediatrics  (2017) 17:170 Page 10 of 1129. Hertz MF, Everett Jones S, Barrios L, David-Ferdon C, Holt M. Associationbetween bullying victimization and health risk behaviors among highschool students in the United States. J Sch Health. 2015;85:833–42.30. Demissie Z, Lowry R, Eaton DK, Hertz MF, Lee SM. Associations of schoolviolence with physical activity among US high school students. J Phys ActHealth. 2014;1131. Rose BM, Holmbeck GN, Coakley RM, Franks EA. Mediator and moderatoreffects in developmental and behavioral pediatric research. J Dev BehavPediatr. 2004;25:58–67.32. Dishman RK, Hales DP, Pfeiffer KA, et al. Physical self-concept and self-esteem mediate cross-sectional relations of physical activity and sportparticipation with depression symptoms among adolescent girls. HealthPsychol. 2006;25:396.33. Leatherdale ST, Brown KS, Carson V, et al. The COMPASS study: alongitudinal hierarchical research platform for evaluating naturalexperiments related to changes in school-level programs, policies and builtenvironment resources. BMC Public Health. 2014;14:1.34. COMPASS System. https://uwaterloo.ca/compass-system/compass-system-projects/compass-study. Accessed 7 July 2016.35. Leatherdale ST, Laxer RE, Faulkner G. Reliability and validity of the physicalactivity and sedentary behaviour measures in the COMPASS questionnaire.2014. https://uwaterloo.ca/compass-system/publications/reliability-and-validity-physical-activity-and-sedentary. Accessed 7 July 2016.36. Chinapaw MJM, Mokkink LB, van Poppel MNM, van Mechelen W, TerweeCB. Physical activity questionnaires for youth. Sports Med. 2010;40:539–63.37. Solberg ME, Olweus D. Prevalence estimation of school bullying with theOlweus bully/victim questionnaire. Aggress Behav. 2003;29:239–68.38. Sampasa-Kanyinga H, Willmore J. Relationships between bullyingvictimization psychological distress and breakfast skipping among boys andgirls. Appetite. 2015;89:41–6.39. Leatherdale ST, Laxer RE. Reliability and validity of the weight status anddietary intake measures in the COMPASS questionnaire: are the self-reported measures of body mass index (BMI) and Canada’s food guideservings robust. IJBNPA. 2013;10:42.40. Brener ND, Mcmanus T, Galuska DA, Lowry R, Wechsler H. Reliability andvalidity of self-reported height and weight among high school students. JAdolesc Health. 2003;32:281–7.41. Wong SL, Leatherdale ST, Manske SR. Reliability and validity of a school-based physical activity questionnaire. Med Sci Sports Exerc.2006;38:1593–600.42. Onis Md, Onyango AW, Borghi E, Siyam A, Nishida C, Siekmann J.Development of a WHO growth reference for school-aged children andadolescents. Bull World Health Organization. 2007;85:660–7.43. Currie CE, Elton RA, Todd J, Platt S. Indicators of socioeconomic status foradolescents: the WHO health behaviour in school-aged children survey.Health Educ Res. 1997;12:385–97.44. Soteriades ES, DiFranza JR. Parent's socioeconomic status, adolescents'disposable income, and adolescents' smoking status in Massachusetts. Am JPublic Health. 2003;93:1155–60.45. Friestad C, Lien N, Klepp K-I. Educational plans-when are they establis hed?Implications for the measurement of socio-economic status in youth.Young. 2001;9:18–32.46. Herciu AC, Laxer RE, Cole A, Leatherdale ST. A cross-sectional studyexamining factors associated with youth binge drinking in the COMPASSstudy: year 1 data. J Alcohol Drug Depend. 2014;17247. Reid JL, Hammond D, McCrory C, Dubin JA, Leatherdale ST. Use ofcaffeinated energy drinks among secondary school students in Ontario:prevalence and correlates of using energy drinks and mixing with alcohol.Can J Public Health. 2015;106:101–8.48. Leatherdale ST. An examination of the co-occurrence of modifiable riskfactors associated with chronic disease among youth in the COMPASSstudy. Cancer Causes Control. 2015;26:519–28.49. Leatherdale ST, Harvey A. Examining communication-and media-basedrecreational sedentary behaviors among Canadian youth: results from theCOMPASS study. Prev Med. 2015;74:74–80.50. Cerin E, MacKinnon DP. A commentary on current practice in mediatingvariable analyses in behavioural nutrition and physical activity. Public HealthNutr. 2009;12:1182–8.51. Hayes AF. Beyond baron and Kenny: statistical mediation analysis in thenew millennium. Commun Monogr. 2009;76:408–20.52. Muthen LK, Muthen BO. Mplus user’s guide. 7th ed. Muthen & Muthen: LosAngeles, CA; 1998-2012.53. Atkin AJ, van Sluijs EMF, Dollman J, Taylor WC, Stanley RM. Identifyingcorrelates and determinants of physical activity in youth: how can weadvance the field? Prev Med. 2016;87:167–9.54. Strauss RS, Pollack HA. Social marginalization of overweight children. ArchPediatr Adolesc Med. 2003;157:746–52.55. Reijntjes A, Kamphuis JH, Prinzie P, Telch MJ. Peer victimization andinternalizing problems in children: a meta-analysis of longitudinal studies.Child Abuse Negl. 2010;34:244–52.56. Vandewater EA, Park SE, Hébert ET, Cummings HM. Time with friends andphysical activity as mechanisms linking obesity and television viewingamong youth. IJBNPA. 2015;12 Suppl 1:S6.57. Buttitta M. Lliescu C, Rousseau a, Guerrien a. Quality of life in overweightand obese children and adolescents: a literature review. Qual Life Res.2014;23:1117–39.58. Neumark-Sztainer D, Falkner N, Storey M, Perry C. Hannan, PJ. Mulert SWeight-teasing among adolescents: correlations with weight status anddisordered eating behaviors Int J Obes. 2002;26:123.59. Voelker DK, Reel JJ, Greenleaf C. Weight status and body image perceptionsin adolescents: current perspectives. Adoles Health Med Ther. 2015;6:149.60. Sherry B, Jefferds ME, Grummer-Strawn LM. Accuracy of adolescent self-report of height and weight in assessing overweight status: a literaturereview. Arch Pediatr Adolesc Med. 2007;161:1154–61.61. Shields M, Tremblay MS. Canadian childhood obesity estimates based onWHO, IOTF and CDC cut-points. IJPO. 2010;5:265–273.62.62. Hodges EVE, Boivin M, Vitaro F, Bukowski WM. The power of friendship:protection against an escalating cycle of peer victimization. Dev Psychol.1999;35:94.63. McMurray RG, Ward DS, Elder JP, Lytle LA, Strikmiller PK, Bagget CD, et al.Do overweight girls overreport physical activity? Am J Health Behav.2008;32:538–46.64. Slootmaker SM, Schuit AJ, Chinapaw MJ, Seidell JC, Van Mechelen W.Disagreement in physical activity assessed by accelerometer and self-reportin subgroups of age, gender, education and weight status. IJBNPA. 2009;6:1.65. Russell-Mayhew S, McVey G, Bardick A, Ireland A. Mental health, wellness,and childhood overweight/obesity. J Obes. 2012; doi:10.1155/2012/281801.66. MacKinnon DP. Introduction to statistical mediation analysis. Chapter 3.Taylor & Francis Group: New York, USA; 2008.67. Craig W, Harel-Fisch Y, Fogel-Grinvald H, Dostaler S, Hetland J, Simons-Morton B, et al. A cross-national profile of bullying and victimization amongadolescents in 40 countries. Int J Public Health. 2009;54:216–24.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Stearns et al. BMC Pediatrics  (2017) 17:170 Page 11 of 11


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items