Bullying Depression and Suicidal It Yin Adolescents

7/29/2019 Bullying Depression and Suicidal It Yin Adolescents

1/11

Bullying, Depression, and Suicidality in AdolescentsANAT BRUNSTEIN KLOMEK, PH.D., FRANK MARROCCO, PH.D.,MARJORIE KLEINMAN, M.S., IRVIN S. SCHONFELD, PH.D., M.P.H,

AND MADELYN S. GOULD, PH.D., M.P.H.

ABSTRACTObjective: To assess the association between bullying behavior an d depression, suicidal ideation, and suicide attemptsamong adolescents. Method: A self-report survey wa s completed by 9th- through 12th-grade students (n = 2342) in sixNew York State high schools from 2002 through 2004. Regression analyses were conducted to examine the associationbetween being victimized and bullying others with depression, ideation, and attempts. Results: Approximately 9% of thesample reported being victimized frequently, an d 13% reported bullying others frequently. Frequent exposure tovictimization or bullying others was related to high risks of depression, ideation, and suicide attempts compared withadolescents not involved in bullying behavior. Infrequent involvement in bullying behavior also was related to increasedrisk of depression and suicidality, particularly among girls. Th e findings indicate that both victims and bullies are at high riskand that the most troubled adolescents are those who are both victims and bullies. Psychopathology was associated withbullying behavior both inand away from school. Conclusions: Victimization and bullying are potential risk factors foradolescent depression and suicidality. In evaluations of students involved in bullying behavior, it is important to assessdepression and suicidality. J. Am . Acad. ChildAdolesc. Psychiatry,2007;46(1):40-49. Key Words: bully, victimization,depression, suicide.

Nearly one third of 6th through 10th graders in theUnited States report moderate or frequent involve-ment in bullying, whether as a bully (13.0%), a victim(10.6%), or both (6.3%; Nansel et al., 2001). Althoughbullying behavior declines as children get older(Olweus, 1991), it is still a prevalent problem amonghigh school students (e.g., Kaltiala-Heino et al., 1999;Nansel et al., 2001). Harris (2005) found that 20% to30% of the students in grades 8 through 12 report

AcceptedJuly 3, 2006Drs.Brunstein KMomek, Marrocco,and Gould and Ms. K0einman are with

the Division of Childand Adolescent Psychiatry, Columbia University; Dr.Schonfeld is with City College, City University of Ne w York, anel the ChildPsychiatry Research TrainingProgram,Columbia University;Dr. Gould s alsowith the Ne w York State Psychiatric nstitute.

This project was supported by National Institute ofMental Health grantROI-MH64632.

Correspondence o Madelyn S. Gonlg Ph.D.,MP.H., Division of Child&AdolescentPsychiatry,DepartmentofPsychiatry, Columbia Universit'/NYSPI,1051 Riverside Drive, Unit 72, Ne w York, NY 10032; e-maik [email protected].

0890-8567/07/4601-0040@2006 by the American Academy of Childand Adolescent Psychiatry.DOI: 10.1097/01.chi.0000242237.84925.18

40

frequent involvement in bullying incidents as either avictim or a bully.The differentiation between bullies, victims, andbully-victims has been emphasized in the examination

of the association of bullying behavior and psycho-pathology (Forero et al., 1999; Kaltiala-Heino et al.,2000; Nansel et al., 2001). In studies examining therelationship between bullying and depression, victimswere found to manifest more depressive symptoms andpsychological distress than nonvictims (Hawker andBoulton, 2000; Kumpulainen and Rasanen, 2000;Kumpulainen et al., 1998; Mills et al., 2004; Neary andJoseph, 1994; Slee, 1995; van der Wal et al., 2003;Williams et al., 1996). Findings pertaining to bullies,however, are less consistent. Some studies did not findan association between being a bully and depression(Camodeca and Goossens, 2005; Fekkes et al., 2004;Juvonen et al., 2003), whereas other studies have foundthat bullies, not just victims, report high levels ofdepression (Forero et al., 1999; Kaltiala-Heino et al.,1999; Kaltiala-Heino et al., 2000; Kumpulainen et al.,2000; Roland, 2002; Salmon et al., 1998). Those whoare both victims and bullies are usually found to be at

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:1, JANUARY 2007


2/11

the highest risk for depression (Fekkes et al., 2004;Kaltiala-Heino et al., 1999; Kaltiala-Heino et al., 2000).

Studies reporting on the relationship betweenbullying and suicidal ideation usually find that victimsmanifest higher levels of suicidal ideation than non-victims (Rigby and Slee, 1999; van der Wal et al.,2003). Other studies have found an increased pre-valence of suicidal ideation among both victims andbullies (Kaltiala-Heino et al., 1999; Roland, 2002).Most studies assessing the relationship of bullyingand suicide attempts have found that victims weremore likely to attempt suicide than those no t involved(Cleary, 2000; Eisenberg et al., 2003; Mills et al.,2004). Kim and colleagues (2005) were the first to dem-onstrate an increased risk of suicide attempts amongboth victims and bullies.

These studies indicate a clear association betweenbullying and suicidality, but important questionsremain unanswered. Most of the studies did notexamine suicide attempts and focused on suicidalideation (Kaltiala-Heino et al., 1999; Rigby et al.,1999; Roland, 2002; van der Wal et al., 2003).Examination of attempts focused on victims and lesson bullies or bully-victims (Cleary, 2000; Davies andCunningham, 1999; Eisenberg et al., 2003; Mills et al.,2004). Moreover, most studies were not conducted inhigh schools (Ivarsson et al., 2005; Kim et al., 2005;Mills et al., 2004; Roland, 2002; van der Wal et al.,2003) and have not differentiated in-school from out-of-school bullying behaviors in their analysis (Kaltiala-Heino et al., 1999; Nansel et al., 2001; Rigby et k1.,1999). There is no study of U.S. adolescents focusingon bullying behavior and suicide attempts amongvictims, bullies, and bully-victims, nor is there anystudy differentiating in-school and out-of-school bully-ing behaviors.The purpose of this study was to examine theassociation between bullying behavior and depression,suicidal ideation, and suicide attempts among a largeAmerican sample of high school students. Specifically,we examined the prevalence of bullying behavior inand ou t of school; the association of bullying behaviorwith depression, suicidal ideation, and suicide attemptsby gender; and the impact of the co-occurrence ofbeing bully-victims. We hypothesized that greaterexposure to bullying behavior would increase the riskof depression, suicidal ideation, and suicide attempts.We also hypothesized that bullying behavior away


BULLYING, DEPRESSION, AND SUICIDALITY

from school would be less prevalent but that it wouldbe associated to the same extent with depression andsuicidality.

METHOD

SubjectsThis study targeted adolescents 13 through 19 years of age whowere enrolled in 9th through 12th grades in six high schools inNassau, Suffolk, and Westchester counties in New York. Fiveschools were public coeducational schools; one was a parochialall-boys school. These schools were part of a study examiningwhether asking about suicidality during a screening program createsdistress or increases suicidal ideation (Gould et al., 2005). Thisstudy included 2,341 of 3,635 students (64.4% participation rate)from fall 2002 through spring 2004. Reasons for nonparticipationincluded parental refusal (61.9%), student refusal (14.3%), andabsences (23.7%). Th e ethnic distribution of the participatingsample was 80.3% white, 5.1% black, 7.3% Hispanic, 3.8%Asian, and 3.5% other. A total of 58.1% of the students were boys(the inclusion of an all-male parochial school explains thehigh percentage 'of boys). The mean age of participatingstudents was 14.8 years (SD 1.2 years). There were no significantdifferences between participants and nonparticipants in sex, age,and race/ethnicity.Students were recruited with a waiver of parental consentfor parents and active written assent for youth. Th e recruitmentprocedures were based on those used in our earlier study (Gouldet al., 2004) and were developed in response to what the schoolsconsidered would best meet the needs of their community. Thepresent study received Institutional Review Board approval of awaiver of consent based on Federal Regulations [Tide 45; Part 46 ,Article 46.116(d)]. Tw o mailings with an information sheetdescribing -survey content and procedures, a response form, anda stamped response envelope were sent to parents 6 and 4 weeksbefore survey administration, providing parents with opportunitiesto refuse their children's participation and giving them amplepertinent information about the project. Students' written assentwas obtained immediately before the survey.Th e study procedures, consistent with the Family EducationalRights and Privacy Act and the Protection of Pupil RightAmendment, were approved by the Institutional Review, Boardof the New York State Psychiatric Institute/Columbia UniversityDepartment of Psychiatry.,MeasuresA self-report questionnaire assessed depression, suicidal ideationsuicide attempts, and involvement in bullying behavior both asa bully and as a victim. Th e assessment time frame was the pas4 weeks, with the exception of measuring lifetime suicide attempts.Demographic Questionnaire. The demographic questionnaireelicited information on age, grade, gender, racial/ethnic background, and household composition.Depression.Th e Beck Inventory (BDI-IA; Beck and Steer, 1993)

assessed cognitive, behavioral, affective, and somatic componentsof depression. Loss of libido was no t assessed. Th e BD I has beenused in >200 studies, including those with adolescent samples(Roberts et al., 1991; Strober et al., 1981; Teri, 1982). Each

41


3/11

KLOMEK ET AL .

response ranged from 0 ("symptoms not present") to 3 ("symptomis severe"), with a maximum total score of 60. A cutoff point of 16was used to dichotomize BD I scores. This cutoff has correctlyclassified 81% of adolescent psychiatric patients with majordepressive disorder (Strober et al., 1981) and has been recom-mended to detect possible depression in normal populations (Becket al., 1993).Suicidal deation.Th e Suicide Ideation Questionnaire (SIQ-JR)assesses suicidal thoughts and is designed for large-scale, school-based screening of adolescents (Reynolds, 1988). The 15-item SIQ-JR uses a 7-point Likert-type scale, ranging from 0 ("I never had thisthought") to 6 ("This thought was in my mind almost every day"),assessing the frequency of specific suicidal thoughts during thepast month. It assesses thoughts related to death and dying,passive and active suicidal ideation, and suicidal intent. Reliabilityof the SIQ-JR is high, ranging from 0.91 to 0.96 for internalconsistency (Keane et al., 1996; Reynolds, 1988; Reynolds andMazza, 1999) and from 0.87 to 0.93 for test-retest reliability(Reynolds and Mazza, 1999). Th e SIQ-JR has demonstratedcriterion validity (King et aL, 1993; Reynolds, 1988; Reynolds,1990; Reynolds and Mazza., 1999), construct validity in clinicalsamples (King et al., 1993; King et al., 1996; King et al., 1997;Sibthorpe et al., 1995; Siemen et al., 1994), an d predictive validity(Keane et al., 1996). Suicidal ideation was considered serious ifthe adolescent scored 31 or higher on the SIQ-JR, scored 5 or 6on two or more of the six "critical" SIQ-JR items (Reynolds, 1988),or responded with either of the two most serious response optionsof the BD I suicide item.Suicide Attemnpt History. Seven questions asking about lifetimeand recent suicide attempts were derived from the depressionmodule of the Diagnostic Interview Schedule for Children(Shaffer et al., 2000) and an earlier suicide screen (Shaffer er al.,2004). These items have demonstrated good construct validity(Gould et al., 1998; Shaffer et al., 2004). Th e assessment of anattempt included questions about occurrences, injuries sustained,medical care sought, and hospitalization (Meehan er al., 1992). Theadolescent was considered to have a history of attempt if he or shereported any past attempt, regardless of timing, injury, or medicalattention, because there is no evidence that injury orneed for medicalattention is a clear indication of severity of attempts amongadolescents. Th e availability of a lethal agent is the most significantdetermination of the lethality of impulsive attempts, whereas intentand severity of psychopathology may make the most importantcontributions to the lethality of attempts by hopeless, dysphoricadolescents (Brent et al., 1987).Bullying Behavior.Several questions regarding bullying behaviorwere derived from the World Health Organization study on youthhealth (Nansel et al., 2001). Questions about bullying werepreceded with the following explanation: "W e say a student isbeing bullied when another student or group of students says ordoes nasty and unpleasant things to him or her. It is also bullyingwhen a pupil is teased repeatedlyin awayhe or she doesn't like. Butit is not bullying when two students of about the same strengthquarrel or fight." Victimization was assessed by two parallelquestions asking respondents, "How often have you been bullied inschool in the past four weeks?" and "How often have you beenbullied away from school in the past four weeks?" Similarly,bullying was assessed by additional parallel questions. Thefrequency items were coded on a five-point scale ranging fromno t at all to most days. Respondents were classified as "nevervictimized," "victimized less than weekly," or "victimized fre-quently" (at least three to four times in the past 4 weeks). Similarly,

42

respondents were classified as "never bullying," "bullying less thanweekly," or "bullying frequently." Th e reliability (Cronbach's a)of the questionnaire is .63.Statistical Analysis

A series of logistic regression models was used to determinewhether depression, ideation, and attempts (as dichotomizedoutcomes) were associated with bullying behavior (less than weeklyand frequently). Th e categories "never victimized" and "neverbully" were the reference groups in these analyses. Th e analyseswere conducted separately for events in and away from schooland were adjusted for schools, grade, and gender.Another series of logistic regression analyses was conducted toexamine gender differences in the association between bullyingbehavior with depression, ideation, and attempts. These regressionanalyses were first conducted separately for boys and girls andwere adjusted for schools and grade. To avoid small numbers, wecombined having been victimized in and away from school andcombined bullying in the two locations. This was justified by therelationship between being victimized/bullying others in and awayfrom school (rw,timijd inandawyiFm schooli 0.42,p 0.80) to detect the interactions.Finally, the relationship of the three outcome measures to theco-occurrence of being victimized and bullying others wasexamined. Respondents were classified into five mutually exclusivecategories: "never bullies nor victims," "infrequent bullies orvictims," "frequently victims only, .... frequently bullies only," or"frequently both bullies and victims (bully-victims)." Respondentswho were neither victims nor bullies served as the reference group. Ifa student was a bully and a victim, we used the maximum score todecide to which group he or she belonged. Interactions betweengender and bullying behaviors were examined in additional models.We chose not to include random effects for school or class inthe regression analyses because the sample dusters (school) andrandomization unit (class within school) had little impact on theoutcomes or correlates (gender, depression, serious suicide ideation/behavior, bullying behavior), as indicated by the intraclass co-efficients, which were close to zero (intraclass coefficients


4/11


Bullying BehaviorVictim in school

NeverLess than weeklyFrequently

Victim away from schoolNeverLess than weeklyFrequently

Bully others in schoolNeverLess than weeklyFrequently

Bully others away from schoolNeverLess than weeklyFrequently

TABLE 1Prevalence of Bullying Behavior In and Out of School

Boys (n = 1,272), Girls (n = 908), Total (N= 2,180),% (no.) % (no.) % (no.)77.3 (983)14.9 (189)

7.9 (100)89.9 (1,143)7.6 (97)

2.5 (32)68.7 (872)20.6 (262)10.7 (136)81.9 (1,042)11.7 (149)

6.4 (82)

83.1 (755)12.3 (112)

4.5 (41)89.3 (812)

8.7 (79)2.0 (18)

84.5 (768)11.3 (103)4.2 (38)89.1 (810)

8.0 (73)2.9 (26)

79.7 (1,738)13.8 (301)6.5 (141)

89.6 (1,955)8.1 (176)2.3 (50)

75.3 (1,640)16.8 (365)

8.0 (174)84.9 (1,852)10.2 (222)

4.9 (108)

Boys were significantly more likely than girls to bevictims in school and to be bullies in and away fromschool. There were no significant differences betweenboys and girls in the frequency of being victims awayfrom school.Association of Bullying Behavior With Depression, SeriousSuicidal Ideation, and Suicide Attempts

Students who were involved in bullyin'g behavior inor ou t of school, whether as a victim or a bully, were atsignificantly higher risk for depression, serious suicidalideation (SSI), and suicide attempts compared withstudents who were never victims or bullies (Table 2).The more frequent the involvement in bullyingbehavior was (whether as a victim or a bully), themore likely the student was depressed, ha d SS1, or hadattempted suicide. Fo r example, students who werefrequently victims in or away from school were seventimes more likely to be depressed compared withstudents who were never victims. Students who werevictims infrequently in or away from school were two tothree times more likely to be depressed. Students whofrequently bullied others in or away from school werethree times more likely to be depressed compared withstudents wh o never bullied others. Students who bulliedothers infrequently were approximately two times morelikely to be depressed. A similar pattern was foiund forSSI and suicide attempts. The only exception to thepattern of greater risk being associated with greater


bullying involvement was for victimization away fromschool and suicide attempts; however, the number ofparticipants who were frequently victimized away fromschool and made an attempt wa s too small (n = 3) toyield adequate statistical power.Gender Differences in Associations BetweenPsychopathology and Bullying Behavior

Frequently victirAized boys were more likely to bedepressed, have SSI, and attempt suicide than boys whow6re never victimized (p = .000, p = .000, p = .058,respectively; Table 3). Boys who were infrequentlyvictimized were more likely to attempt suicide but werenot more likely to be depressed or have SSI than boywho were never victimized. Boys who bullied othersfrequently were more likely to be depressed and haveSSI' but were not more likely to attempt suicidecompared with boys who never bullied others. Boywho bullied others infrequently were not at a higher riskfor depression, SSL, or suicide attempts.

Among girls, any involvement in bullying behaviorswas associated with a higher risk for depression, SS1,and suicide attempts. Girls who were victimized,whether infrequently or frequently, were more likelyto be depressed, have SSI, and attempt suicide than girlwho were never victims. Girls who bullied otherswhether infrequently or frequently, were more likely tobe depressed, have SSI, and attempt suicide comparedwith girls who never bullied others.

43

X2 GenderDifferences; p13.91; .001

1.43; .49

73.26;


5/11

KLOMEK ET AL.

00u

0 v* c1l

UN

::I

if

g C

0,0

V -

a) Co

>.ZJ-

\0 * Co c'4Cc).. , C .*cq

N N~

orl'a,

UN -o c

06 C--i

0.2

0.z

C'cl

0PQ2-

0

0

'3Co

Z- 4

c4 14?

.0

002 -

0 z o

0Q2


oo

00-ZU

00

0

00C/)Ei

00

'3j

\2 AC-i c

,-

=

Co

u-Oc~li-

'- V

CoCO4

--- *ClCo* i

Co

Cot N 'i-'- *1- CI'3 Cl-C0

Co -4* Co-IN

CI\N

Co

ONI'll- coc\I

- ccI

cien

.. :

- u C:

U

0o0

0 .

zU-fia'u-Ca'C)

- 'cz aCd)

44


6/11


Significant interactions between gender and thebullying-related variables indicated that girls whobullied others frequently were at significantly higherrisk for depression and suicide attempts than compar-able boys (odds ratio [OR] = 3.93, 95% confidenceinterval [CI] = 1.6-9.4, p = .002; OR 4.08, 95% CI1.2-13.8, p = .024, respectively). Interactions betweenbullying behavior and gender were no t significantwith regard to suicidal ideation.Impact of Co-occurrence of Bullying and Victimization

Among boys, 2.5 % were frequently both bullies andvictims (bully-victims), 6.1% were frequently victimsonly, 10.4% were freqiiently bullies only, and 28.1%were infrequently bullies or victims (Table 4). Amonggirls, nearly 1% were frequently bully-victims, 4.5%were frequently victims only, 4.4% were frequentlybullies only, and 22.1% were infrequently bullies orvictims. A significantly greater proportion of girls

(nearly 70%) than boys (52.8%) were neither bulliesno r victims (X2 = 62.55; df= 5,p < .001).Boys who were frequently bully-victims were morelikely to be depressed and have SSI than boys who wereno t inv'lved in bullying behavior. Th e small number of

boys who were bully-victims and made a suicideattempt ,(n = 1) precluded a meaningful examinationof the risk of attempts. Boys who frequently bulliedothers only were not more likely to be depressed, haveSSI, or attempt suicide compared with boys who neverbullied others. Boys who were frequently victims onlywere more likely to be depressed, have SSI, and attemptsuicide than boys who were never victimized. Boys whowere infrequently involved were more likely to attemptsuicide but not more likely to be depressed or have SSIthan boys who were never involved.

Among girls, any involvement in bullying behaviorincreased the risk for depression, SSI, and suicideattempts. Girls who were frequently bully-victims were

TABLE 3Association of Bullying Behavior With Depression, Suicidal Ideation, and Suicide Attempts by Gender

Victim In and Out of School Bully Others In and Out of SchoolNever Less Than Weekly Frequently Never Less Than Weekly Frequently

(n M =935;" (n M=225;" (n M = 109;a (n M 815;" (n M =291;' (n M = 163;"nF =716) nF= 139) nF =49) nF =725), nF 131) nF=48)Depression

Male, % (no.) 4.1 (38) 6.7 (15) 21.1 (23) 4.8 (39) 7.2 (21) 9.8 (16)OR5 1.74 6.45*** 1.55 2.10*95% CI 0.94-3.23 3.65-11.42 0.89-2.70 1.12-3.94Female, % (no.) 10.5 (75) 27.3 (38) 51.0 (25) 11.3 (82) 23.7 (31) 52.1 (25)ORb 3.25*** 9.73*** 2.43*** 8.39***95% CI 2.08-5.08 5.23-18.10 1.52-3.88 4.53-15.53

Serious suicidal ideationMale, % (no.) 1.9 (18) 3.1 (7) 10.1 (11) 2.1 (17) 3.8 (11) 4.9 (8)

ORb 1.69 5.88*** 1.92 2.57*95% CI 0.70-4.12 2.67-12.92 0.88-4.18 1.07-6.20

Female, % (no.) 3.3 (24) 13.7 (19) 14.3 (7) 3.7 (27) 10.0 (13) 20.8 (10)ORb 4.75*** 5.26*** 3.13** 7,18***95% CI 2.50-9.00 2.10-13.17 1.54-6.34 3.17-16.25

Suicide attemptsMale, 0/o (no.) 1.8 (17) 4.5 (10) 4.6 (5) 2.0 (16) 3.4 (10) 3.7 (6)ORb 2.64* 2.70 1.86 2.0795% CI 1.18-5.90 0.97-7.55 0.83-4.18 0.78-5.51Female, % (no.) 5.0 (36) 10.9 (15) 20.0 (10) 4.5 (33) 10.9 (14) 29.2 (14)OR 5 2.30* 4.98*** 2.72** 8.91***95% CI 1.21-4.38 2.24-11.08 1.38-5.34 4.23-18.77

Note: M = male; F = female; OR odds ratio; CI = confidence interval." Numbers vary slightly because of missing data.bAdjusted for schools attended and grade.*p < .05; **p < .01; **'p< .001.

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 46:1, JANUARY 2007 45


7/11

KLOMEK ET AL .

TABLE 4Co-occurrence of Bullying and Victimization: Association With Depression, Suicidal Ideation, and Suicide Attempts

Neither Bully Infrequently Frequently Frequently Frequently Bothnor Victim' Bully or Victim Victim Only Bully Only Bully and Victim(n M=671;b (n M =357;b (n M =78;' (n M = 132 ;b (n M = 32;bn F=615) n F=200) n F =41) n F = 40) n F = 9)

DepressionMale, % (no.) 3.6 (24) 5.6 (20) 20.5 (16) 6.8 (9) 22.6 (7)ORC 1.34 6.00*** 1.63 6.36***95% CI 0.75-2.40 3.08-11.70 0.74-3.57 2.50-16.17Female, % (no.) 9.1 (56) 19.0 (38) 46.3 (19) 47.5 (19) 75.0 (6)ORW 2.38*** 9.29*** 8.81*** 32.15***95% CI 1.52-3.72 4.71-18.35 4.46-17.43 6.29-164.30

Serious suicidal ideationMale, % (no.) 1.8 (12) 2.8 (10) 7.7 (6) 2.3 (3) 16.1 (5)

ORc 1.56 4.54** 1.29 9.93***95% CI 0.67-3.61 1.66-12.44 0.36-4.68 3.23-30.54Female, % (no.) 2.9 (18) 8.5 (17) 12.5 (5) 20.5 (8) 22.2 (2)

ORc 3.29** 5.26** 9.04*** 10.00"*95% CI 1.65-6.57 1.81-15.29 3.56-22.97 1.86-53.64Suicide attempts

Male, % (no.) 1.3 (9) 3.7 (13) 5.1 (4) 3.8 (5) 6.5 (1)ORW 2.71" 3.88* 2.96 2.4595% CI 1.17-6.28 1.17-12.85 0.97-9.05 0.30-20.14Female, % (no.) 3.9 (24) 8.1 (16) 17.1 (7) 28.2 (11) 33.3 (3)OR 2.25* 5.43*** 9.89 ***b 12.66**95% CI 1.16-4.37 2.11-13.92 4.26-22.95 2.74-58.48Note: M = male; F = female; OR = odds ratio; CI = confidence interval.a Reference group for the indicated significance.bNumbers vary slightly because of missing data.'Adjusted for schools attended and grade.*P .05; **P < .01; *P < .001.

32 times more likely to be depressed and 10 to 12 timesmore likely to have SSI or to attempt suicide comparedwith girls who were not involved in bullying behavior.Girls who were frequently bullies only or werefrequently victims only were more likely to bedepressed, have SSI, or attempt suicide comparedwith girls who never bullied others or who were nevervictimized, respectively. Even girls who were infre-quently bullies or victims were two to three times morelikely to be depressed, have SSI, or attempt suicide.

Significant interactions between gender and thebullying-related variables indicated that girls whofrequently bullied others only were at significantlyhigher risk for depression and SSI than comparableboys (OR 5.29, 95 % CI 1.9-1 4 .9,p = .002; OR 6.83,95% CI 1.4-33.2, p = .017, respectively). Interactionsbetween bullying behavior an d gender were notsignificant with regard to suicide attempts.

46

DISCUSSIONThis study found an association between bullyingbehavior and depression, SSI, and suicide attemptsamong high school students. Approximately 9% of the

sample reported being frequently victimized, and 13%reported bullying others frequently. These rates aresimilar to previous reports (Kaltiala-Heino et al., 1999;Nansel et al., 2001).Most bullying behavior occurred in school comparedwith away from school, as found in other studies(Williams et al., 1996; Wolke et al., 2001). Bullyingbehavior away from school, however, was still prevalentamong high school students (25.5%). Bullying beha-vior was more prevalent among boys than girls,consistent with previous reports (Kumpulainen et al.,1998; Wolke.et al., 2001). The decrease in theprevalence of bullying behavior from childhood toadolescence (Olweus, 1991) seems to indicate that



8/11


bullying behaviors are less normative in high schoolsand that those adolescents still engaged in bullyingbehavior in high school may be more disturbed. Assuch, our findings can be generalized only to highschool students.

Depression, SSI, and suicide attempts were signifi-cantly associated with victimization and with bullyingothers both in and away from school. Higher exposuresto being victimized or bullying others generally wererelated to higher risk of depression, suicidal ideation,and suicide attempts, yet infrequent involvement inbullying behavior also was related to increased risk ofdepression and suicidality, particularly among girls.

Our findings are consistent with reports that bullies,not just victims, are at higher risk of depression,ideation, and attempts (Forero et al., 1999; Kaltiala-Heino et al., 1999; Kaltiala-Heino et al., 2000;Kumpulainen et al., 2000; Roland, 2002; Seals andYoung, 2003; Wolke et al., 2001). Moreover, ourfindings replicate reports that the most disturbed groupare the bully-victims (Austin and Joseph, 1996;Juvonen et al., 2003; Kaltiala-Heino et al., 1999; Kimet al., 2005). Although our findings are not consistentwith reports that depression is rare among adolescentbullies (Camodeca et al., 2005; Fekkes et al., 2004;Juvonen et al., 2003), these studies did not account forgender differences. The inconsistent results may beexplained by the gender differences we found. Ourresults suggest that there is a different threshold inwhich bullying is associated with depression andsuicidality among girls and boys. Girls who bulliedothers were at risk for depression, ideation, andattempts even when the bullying wa s infrequent.Among boys, however, only frequent bullying wasassociated with depression, ideation, and attempts.Infrequent bullying among boys may be a morenormative behavior, consistent with reports that thelevel of aggression is higher among boys compared withgirls (Achenbach and Edelbrock, 1981). The genderdifferences in the association with bullying may beunderstood by the findings of Wasserman et al. (2005),indicating that female adolescents who demonstrateconduct problems are at higher risk of affectivedisorders than comparable male adolescents. Thesegirls were found to be more aggressive but also moredepressed than boys with conduct problems. Bullyingbehavior could be another example of a "genderparadox" (Tiet et al., 2001; Wasserman et al., 2005)


in which girls are less likely to be bullies, but when theyare they have a more severe impairment than their malecounterparts.

The results suggest a different gender threshold invictimization as well. Among girls, victimization at anyfrequency increased the risk of depression, ideation, andattempts. Among boys, only frequent victimizationincreased the risk of depression and ideation, althoughinfrequent victimization was associated with an increasedrisk of attempts. These findings contradict reports thatonly female victims are at greater risk fo r depression andideation (Kim et al., 2005; van derWal et al., 2003).Limitations

One limitation was the use of a convenient, ratherthan a random, sample of schools. The schools weresuburban and predominantly white, limiting thegeneralizability to urban and more ethnically orsocioeconomically diverse settings. However, studiesreporting on ethnicity and socioeconomic status asfactors in bullying behavior have shown inconsistentresults. Research in elementary school students suggeststhat socioeconomic status is inversely related to bullyingand victimization (Veenstra et al., 2005; Wolke et al.,2001). Olweus (1999), in contrast, found that bullyingwa s unrelated to social class. Seals and Young (2003)found no statistically significant differences in bullyinginvolvement based on ethnicity. However, Nansel et al.(2001) found that Hispanic youth reported higherinvolvement in bullying of others, whereas black youthreported being bullied with less frequency. We had a'similar finding: Hispanics were slightly more likely tobully others in school compared with other ethnicgroups. The small number of minority students in thepresent study precludes the examination of theassociation of bullying behaviors with depression, SSI,and suicide attempts by race. Further studies shouldinclude these analyses. Second, only tw o thirds ofeligible subjects participated in the study. Despite nosignificant differences between participants and non-participants in demographic factors (e.g., sex, gradelevel, ethnicity), the same cannot be said about clinicalfactors (e.g., BDI and SIQ-JR scores). Third, data onbullying were based on self-reports. Information aboutbullying also can be obtained from peers, parents, andteachers. Fourth, we found a cross-sectional associationbetween bullying and depression/suicidality, but caus-ality cannot be determined.

47


9/11

KLOMEK ET AL .

Clinical ImplicationsOur findings suggest that bullying behavior in and

away from high school is a prevalent problem amongadolescents. Being a victim or a perpetrator appears toheighten the risk for depression, suicidal ideation, andsuicide attempts. Our results emphasize that bullyingmay be a marker of suicidal behavior and that routineprevention of bullying should be considered part of anysuicide prevention strategy. Both bullying behavioraway from school and in-school bullying should bescrutinized. Mental health practitioners evaluatingsuicidal tendencies should consider bullying one ofthe potential risk factors. Conversely, in evaluations ofstudents involved in bullying behavior, it is importantto assess depression and suicidality.

Disclosure:The authorshave no financialrelationships o disclose.

REFERENCESAchenbach TM , Edelbrock CS (1981), Behavioral problems and compe-tencies reported by parents of normal and disturbed children aged fourthrough sixteen. Monogr Soc Res ChildDev 46:1-82Austin S,Joseph S (1996), Assessment of bully/victim problems in 8 to 11year-olds. BrJEducPsycho! 66:447-456Beck AT , Steer RA (1993), Manualfor he Beck Depression Inventory. SanAntonio, TX: The Psychological CorporationBrent D (1987), Correlates of the medical lethality of suicide attempts inchildren and adolescents. JAm Acad ChildAdolZec Psqrhiatrq26:87-91Camodeca M, Goossens FA (2005), Aggression, social cognitions, anger andsadness in bullies an d victims. J ChildPycholPVchiat7y46:186-197Cleary SD (2000), Adolescent victimization and associated suicidal and

violent behaviors. Adolescence 35:671-682Davies M, Cunningham G (1999),Adolescent parasuicide in the Foyle area.IrishjPsycholMed 16:9-12Eisenberg ME , Neumark-Sztainer D, Story M (2003), Associations ofweight-based teasing and emotional well-being among adolescents.Arch PediatrAdolesc Med 157:733-738Fekkes M, Pijpecrs FI, Verloove-Vanhorick SP (2004), Bullying behavior andassociations with psychosomatic complaints and depression in victims.JPediatr144:17-22Forero R, McLellan L, Rissel C, Bauman A (1999), Bullying behavior andpsychosocial health among school students in New South 'Wales,Australia: cross sectional survey. BMJ319:344-348Gould MS, King R, Greenwald S (1998), Psychopathology associated withsuicidal ideation and attempts among children and adolescents. JAmAcad ChildAdolesc Psychiatry37:915-923Gould MS, Marrocco FA, Kleinman M (2005), Evaluating iatrogenic riskof youth suicide screening programs: a randomized controlled trial.JAMA 293:1635-1643GouldMS, Velting D, KMeinman M, Lucas C, ThomasJG, ChungM (2004),Teenagers' attitudes about coping strategies an d help-seeking behaviorfor suicidality. JAm Acad ChildAdolesc Psychiat7)43:1124-1133Harris S (2005), Bullying at school among older adolescents. SchoolNursRev22:18-21Hawker DSJ, Boulton MJ (2000), Twenty years' research on peervictimization and psychosocial maladjustment- a meta-analytic reviewof cross-sectional studies.j ChildPsycholPsychiatry41:441-455

Ivarsson T, Broberg AG , Arvidsson T, Gillberg C (2005), Bullying inadolescents: psychiatric problems in victims and bullies as measured by

48

the Youth Self Report (YSR) and the Depression Self-Rating Scale(DSRS). NordjPschiatry59:365-373Juvonen J, Graham S, Schuster MA (2003), Bullying among youngadolescents: the strong, the wveak, and the troubled. Pediatrics112:1231-1237Kaltiala-Heino R, Rimpela M, Marttunen M, Rimpela A, Rantanen P(1999), Bullying, depression, an d suicidal ideation in Finnishadolescents. school survey. BMJ319:348-351Kaltiala-Heino I, Rimpela M, Rantanen P, Rimpela A (2000), Bullying atschool: an indicator of adolescents at risk for mental disorders.JAdolesc

23:661-674Keane EM , Dick RW, Bechtold DW, Manson SM (1996), Predictive andconcurrent validity of the Suicidal Ideation Questionnaire amongAmerican Indian adolescents. JAbnonn ChildP*yho! 24:735-747Kim YS, Koh YJ, Leventhal B (2005), School bullying and suicidal risk inKorean middle school students. Pediatrics115:357-363King CA, Ghaziuddin N, McGovern L, Brand E, Hill E,Naylor M (1996),Predictors of comorbid alcohol and substance abuse in depressedadolescents. JAm Acad ChildAdolesc Psychiatry35:743-751King CA, Hill EM, Naylor M, Evans T, Shain B (1993), Alcoholconsumption in relation to other predictors of suicidality amongadolescent inpatient girls. JAm Acad ChildAdolesc Psychiatry32:82-88King CA, Katz SH, Ghaziuddin N, Brand E, Hill E, McGovern L (1997),Diagnosis an d assessment ofdepression and suicidality using the NIMHDiagnostic Interviesw Schedule for Children (DISC-2.3).JAbnorm Child

Pscho!25:173-181Kumpulainen K, Rasanen E (2000), Children involved in bullying at elementaryschool ag e their psychiatric symptoms and deviance in adolescence: anepidemiological sample. ChildAbuseNegl24:1567-1577Kumpulainen K, Rasanen E, Henttonen 1 (1998), Bullying an d psychiatricsymptoms among elementary school-age children. ChildAbuse Negl22:705-717Meehan PJ, Lamb JA, Saltzman LE, O'Carroll PW (1992), Attemptedsuicide among young adults: progress toward a meaningful estimate ofprevalence. AmjPyVhiatry 149:41-44Mills C, Guerin S, Lynch F, Daly I, Fitzpatrick C (2004), The relationshipbetween bullying, depression and suicidal thoughts/behaviour in Irishadolescents. IrJ PsycholMed 21:112-116Murray D (1998), Design and Analysis of Group-RandomizedTrials. NewYork: Oxford University PressNansel TR, Overpeck M, Pilla RS, Ruan WJ, Simons-Morton B, Scheidt P(2001), Bullying behaviors among US youth: prevalence and associationwith psychosocial adjustment.JAWA 285:2094-2100Neary A, Joseph S (1994), Peer victimization an d its relationship to self-concept and depression among schoolgirls. Personaliy IndividualDifferences 16:183-186Olweus D (1991), Bully/victim problems among school children: Basic factsand effects of a school based intervention program. In: The developmentand reatmentof childhoodaggression,Pepler D, Rubin K, eds. Hillsdale,NJ : Lawrence Erlbaum Associates, pp 411-448Olweus D (1999), Sweden. In: The Natureof School Bullying: A Cross-National Perspective,Smith PK , Morita Y, Junger-Tas J, Olweus D,Caralano R, Slee PT, eds. London: Roudedge, pp 7-27Reynolds W (1990), Development ofa semi-structured clinical interview forsuicidal behaviors in adolescents. Pychol Assess J Consult Clin Psychol2:382-390Reynolds W (1988), SIQ ProfessionalManual Odessa, FL: PsychologicalAssessment ResourcesReynolds WM, Mazza JJ (1999), Assessment of suicidal ideation in inner-city children and young adolescents: reliability and validity of theSuicidal Ideation Questionnaire-JR. SchoolPsychol Rev 28:17-30Rigby K, Slee P (1999), Suicidal ideation among adolescent school children,involvement in bully-victim problems, and perceived social support.Suicide Life Threat Behav 29:119-130Roberts RE, Lewinsohn PM , Seeley JR (1991), Screening fbr adolescentdepression: a comparison ofdepression scales.JAm Acad ChildAdolescPqchiatry30:58-66Roland E (2002), Bullying, depressive symptoms an d suicidal thoughts. EducRes 44:55-67



10/11

Salmon G, James A, Smith DM (1998), Bullying in schools: self reportedanxiety, depression, and self esteem in secondary school children. BM J317:924-925

Seals D, Young J (2003), Bullying and victimization: prevalence andrelationship to gender, grade level, ethnicity, self-esteem, and depression.Adolescence 38:735-747Shaffer D, Fisher P, Lucas CP , Dulcan MK, Schwab-Stone ME (2000),NIMH Diagnostic Interview Schedule for Children Version IV (NIMHDISC-IV): description, differences from previous versions, andreliability of some common diagnoses. J Am Acad Child AdolescPsychiatry39:28-38Shaffer D, Scott M,Wilcox H (2004), Th e Columbia Suicide Screen: validityand reliability of a screen for youth suicide and depression.JAm AcadChildAdolesc Psychiatry43:71-79Sibthorpe B, DrinkwaterJ, Gardner K, Bammer G (1995), Drug use, bingedrinking and attempted suicide among homeless and potentiallyhomeless youth. AustNZJ Psychiatry29:248-256SiemenJR,Warrington CA, Mangano EL (1994), Comparison of the MillonAdolescent Personality Inventory and the Suicide Ideation Questionnaire-Junior with an adolescent inpatient sample. PsycholRep 75:947-950Slee PT (1995), Bullying: health concerns of Australian secondary schoolstudents. IntJAdolescYouth 5:215-224Strober M, Green J, Carlson G (1981), Utility of the Beck Depression


Inventory with psychiatrically hospitalized adolescents.J Consult ClinPsychol49:482-483Teri L (1982), Th e use of the Beck Depression Inventory with adolescents.JAbnorm ChildPsychol 10:277-284Tier QQ , Wasserman GA, Loeber R, McReynolds LS , Miller LS (2001),Developmental and sex differences in types of conduct problems.J Child

FamStud 10:181-197van der Wal MF, de Wit CA, Hirasing RA (2003), Psychosocial healthamong young victims and offenders of direct and indirect bullying.Pediatrics 111:1312-1317Veenstra R, Lindenberg S, Oldehinkel AJ, De Winter AF, Verhulst FC ,Ormel J (2005), Bullying and victimization in elementary schools: acomparison of bullies, -victims, bullylvictims, and uninvolved preado-lescents. Dev Psychol 41:672-682Wasserman GA, McReynolds LS, Ko SJ, Katz LM, Carpenter JR (2005),Gender differences in psychiatric disorders at juvenile probation intake.AmJ PublicHealth95:131-137Williams K, Chambers M, Logan S, Robinson D (1996), Association ofcommon health symptoms with bullying in primary school children.BMJ 313:17-19Wolke D, Woods S, Stanford K, Schulz H (2001), Bullying andvictimization of primary school children in Englaind and Germany.prevalence and school factors. Br J Psychol 92:673-696

Variation in the Diagnosis of Child Abuse in Severely Injured Infants Matthew Trokel, MD, Anthony Wadimmba, MB , ChB, JohnGriffith, PhD, Robert Sege, MD, PhDObjective: Diagnosis of child abuse is difficult and may reflect patient, practitioner, and system factors. Previous studies have demonstratedpotential lethal consequences if cases of abuse are missed and suggested a role for continuing medical education in improving the accuracy ofdiagnosis ofsuspected abuse. Although the majority of injured American children are treated at general hospitals, most published studies ofsevere injury resulting from child abuse have been conducted at children's hospitals. The objective of this study was to evaluate the role ofhospital type in observed variations in the frequency of diagnosis of child physical abuse among children with high-risk injuries. Methods:Hospital discharge data were evaluated, and adjusted rates of abuse diagnosis were reported according to hospital type. A regression modelestimated the number of cases of abuse that would have been diagnosed if all hospitals identified abuse as frequently as observed at pediatricspecialty hospitals. This study consisted ofchildren wh o were


11/11

COPYRIGHT INFORMATION

TITLE: Bullying, Depression, and Suicidality in Adolescents

SOURCE: Journal of the American Academy of Child and

Adolescent Psychiatry 46 no1 Ja 2007

PAGE(S): 40-9

WN: 0700102114010

The magazine publisher is the copyright holder of this article and it

is reproduced with permission. Further reproduction of this article in

violation of the copyright is prohibited.

Copyright 1982-2007 The H.W. Wilson Company. All rights reserved.

Bullying Depression and Suicidal It Yin Adolescents

Documents

Transcript of Bullying Depression and Suicidal It Yin Adolescents