Se A Systematic Review of School-Based Interventions to Prevent Bullying Rachel C. Vreeman, MD; Aaron E. Carroll, MD, MS Objective: To conduct a systematic review of rigor ously evaluated school-based interventions to decrease bullying. Data Sources: MEDLINE, PsycINFO, EMBASE, Edu- cational Resources Information Center, Cochrane Col- laboration, the Phy ogy: A SAGE Full terms bullying and bully ical Education Index, and Sociol- -xt Collection were searched for the Study Selection: We found 2090 article citations and reviewed the references of relevant articles. Two review- crs critically evaluated 56 articles and found 26 studies that met the inclusion criteria, Interventions: The types of interventions could be cat- egorized as curriculum (10 studies), multidiseiplinary or “whole-school” interventions (10 studies), social skills groups (4 studies), mentoring (1 study), and Social worker support (1 study), Main Outcome Measures: Dats were extracted regarding direct outcome measures of bullying (bully- ing, victimization, aggressive behavior, and school responses to violence) and outcomes indirectly related to bullying (school achievement, perceived school safety, self-esteem, and knowledge or attitudes toward bullying) Results: Only 4 of the 10 curriculum studies showed decreased bullying, but 3 of those 4 also showed no im- provement in some populations, Of the 10 studies eval ating the whole-school approach, 7 revealed decreased bullying, with younger children having fewer positive el fects, Three of the social skills raining studies showed no clear bullying reduction. The mentoring study found decreased bullying for mentored children. The study of increased school social workers found decreased bully- ing, truancy, theft, and drug use Conelusions: Many school-based interventions di- rectly reduce bullying, with better results for interven- tions that involve multiple disciplines. Curricular changes less often allect bullying behaviors. Outcomes indi- rectly related to bullying are not consistently improved by these interventions. Arch Pediatr Adolesc Med. 2007;161:78-88 ULLYING IS A FORM OF AG- {gression in which I or more children repeatedly and in- tentionally int rass, or physically harm a * Victims of bullying are perceived creased risk for depressive symptoms and suicidal ideation.*” Students who report victimization are 3 10 4 times more likely to report anxiety symptoms than unin- volved children.” The effects of bullying on emotional health may persist overtime; idate, ha Author Affiliations: Cailden's Health Services Research, Indiana University School of Medicine (Drs Vreeman and Caroll), and The Regenstie! Institute for Health Care col), Indianapolis, Ind. rc Downloaded From: https: by their peers as physically or psychologi- cally weaker than the aggressor(s),and vie~ ‘ums perceive themselves as unable to re- taliate* Although bullying, harassment, and vietimization can take many forms, the kkey elements of this behavior are aggres- sion, repetition, and the context of a re- lationship with an imbalance of power.” Bullying can impact the physical, emo tonal, and social health of the children involved. Vietims of bullying more often report sleep disturbances, enuresis, ab- dominal pain, headaches, and feeling sad than children who are not bullied." Bul- lies, their vietims, and those who are both bullies and victims have significantly in- 1 study" showed that children bullied re- peatedly through middle adolescence had lower self-esteem and more depressive symptoms as adults. Victims of bullying are more likely to feel socially rejected oF fsolated and to experience greater social ‘marginalization and lower social status." Bullying impacts child's experience of school on numerous levels. Bullying ere- ates problems with school adjustment and bonding, affecting the victims’ comple- tion of homework or desire to do well at school." In 1 study," 20% of grade- school children reported being fright- ened through much of the school day. Bul- lying seems to inerease school absenteeism, (eepnnvtED) SRCT PEDIATR ADOLESC MEDVVOL Te JANET —_WAFWARCHPEDIATRICS COM (©2007 American Medical Association. AI rights reserved, jamanetwork.com/ on 09/30/2019with victimized children becoming more school avoid- ant as the victimization increases."* Furthermore, in- volvement in bullying affects academic performance, al- though studies” show mixed results regarding which children are most affected. Most bullying takes place at school, particularly at times and places where supervi- sion is minimal.” Schools where adults tolerate more bul- lying may have more severe bullying problems." As school bullying increasingly becomes atopic of pub- licconcern and research efforts, a growing numberof stud- {es examines school-based interventions targeted to re- duce bullying, Although many of these interventions have been rigorously evaluated, the evaluations reveal mixed results.” For example, evaluations of the Olweus Bully- ing Prevention Program, a comprehensive “whole- school” intervention on which many subsequent pro- grams have been based, report reductions of 30% to 70% in the student reports of being bullied and bullying oth- ers. In contrast, evaltation ofa similar comprehen- sive prevention program implemented in Belgium did not show significant differences in victimization or bullying. scores among primary or secondary school students.” Al though some review articles have described several ofthese Interventions, to our knowledge, no systematic reviews of interventions to reduce bullying have been published in peer-reviewed literature. The objective of this study was to review rigorously evaluated school-based inter- ventions to reduce or prevent bullying with the goal of determining whether these interventions worked. LES} We searched sever bibliographic databases ncuing MEDLINE {sms 11965, thr Ags 23,2001 PayINFO, EMBASE Educational Resores Information Center the Physical Ech cation Index, Sociology A SAGE Full-Text Collection, and the Cochrane Clinial Tras Rese alla of Aug 23,2004) \Werssed the search erms buying oe bully ax Medial Subject Headings or keywords: We used Keyword search because it twas more robust than searches using ony Medical subject Hend- ings. One of ts (RCV. reviewed the tiles fall etme r= ticks andthe bibliographies ofall relevat review atc to dtermine which stadies examined a school based interven tion to prevent or reduce balling. Article were immediate txeluded if they obviously didnot include an intervention oF didnot ocenr at school ‘Ate articles that clearly didnot meet the incision rite: sia were exiled, both of ts (RCV. and AEC), lined the Journal cation and atc text oer than the Methods section independently reviewed the atles The 2 reviewers independent decided on tal inchsion using standard form with predetermined eligilty criteria. Disagreements were re solved by consenmis reached ater discussion. Fr inlonon. 3 study needed odeseribe an experimental intervention with com troland intervention groups and toinclude follow-up evalu- tion with messured outcomes In adion, the intervention needed tobe school based and designed to educe or prevent Bullying Each anicle was analyzed to determine the study smchod intervention components, comes messed andre fall: There esto umesric of quality in chonsng ot eva ating study outcomes beyond the ncuston criteria, We didnot Tiehideo dicount sti led on belie ndaresemong treatment groups, sty power, tention rte, progam n= tensity becuse these characteristics are not associated dein= tively with the strength of treatment effects" Duplicate pub- lications or muluple articles that reported identical outcomes seasured over the same period on the same population were excluded, ‘We extracted data from the selected articles regarding di- rect outcome measures of bullying, including bullying, view cation, aggressive behavior, violence, school responses to vi lence, nd violent injuries. Data were also extracted forauteomes| thought to be indirectly related to bullying, such as school achievement, perception of school safety, self-esteem, or know!- fedge about of atitides toward bullying EE The systematic literature search identified 2090 articles, The online search of MEDLINE yielded 353 articles, and the search of EMBASE yielded 269 articles, 9 of which were not found by the MEDLINE search. The search of PsycINFO yielded 807 articles, Educational Re- sources Information Center yielded 552 articles, the Physi- cal Education Index yielded 16 articles, and Sociology: A. SAGE Full-Text Collection yielded 3 articles. An addi- ional 4 potential studies? were identified through searches of bibliographies and were also reviewed. Once articles that obviously did not address school-based in- terventions were excluded, 321 relevant articles r rained. Reviewing the abstracts of these articles allowed for the further exclusion of articles that did not address school-based interventions. Fifty-six articles were then as- sessed by both of the reviewers. Articles were most com- ‘monly excluded at this stage because they were not evalki- ations of interventions, they did not have control groups, ‘oF they did not address bullying. The 26 studies that met the selection criteria varied substantially in intervention type, study population, study design, and outcome measures. The detailed characte {stics of the studies ate reported in Fable ¥. The inte ventions could be divided into 5 eategories: curriculum interventions, multidiseiplinary or whole-school inte ventions, targeted social and behavioral sills groups, men- toring, and increased social work support. To maximize clarity and clinical usefulness, we present the subse: quent resus of the review based on the type of inte vention. All 26studies investigated interventions for some group of primary school students, but the primary grade levels varied from first to eighth grade. Six stud- 244 included secondary school students (older than eighth grade) in their interventions and outcomes. The selected studies reported a range of outcomes that were subsequently categorized into direct and indirect outcomes. The direct and indirect outcomes ofall of the studies are reported in Fable 2, (CURRICULUM INTERVENTIONS Tren studies®2*2* evaluated the implementation of new curriculum. The curriculum interventions included vid- cotapes, lectures, and written curriculum, and varied in intensity from a single videotape followed by classroom discussion to 15 weeks of classroom modules. The de- tails of the study designs, participants, intervention type, and important outcomes of the curriculum interven. (©2007 American Medical Association. AI rights reserved, jamanetwork.com/ on 09/30/2019Table 1. Study Characteristics = oe ey = ae ee nes ine aa GETS ionic or Fevdoins, [Sowios, Sale o<" eee a er Men Te, Lee tet eon ES "ean TER, at eee ae Fs “sa Kegdom fey ies gmptok “tek, ie, Geet eet " = Mize, Ni, mien een een yt, geet ag uel Kendom a a ' = ES Es meek, ep, me tages secglcovt est. np ‘900 ‘Sums inSclmces (1 css) pbk coend > " ce fret aon = Bessopeoe an = (ais)? ™ high ig, Sm rem ek? mem ta ea ne Sonnac eS "1996 slovakia in ‘ ces oe ce “ine sin age stein Ni te = ae i es ry aes: a a7 = Sa. Fi a ae = aS ea = = a a a wit, ind, sts tg Hagin gs hth SST See aoe = = ee iia eee Se sl : ae = = = ne ee coe Bae ae 3 St ve, Met "Bp, iim scent? feo meme, em, a, gr aS aera Se. Se Se Seeuy nea cl fan esa ats ig rks ca is Ea =. a ocd sos atcall dein Table he stills soils improvement in balging 0 A pretest, posttest, control group design: 6 of the 10 stud- tes randomized the assignment of the groups.22>2"25> The curriculum interventions did not consistently de- crease bullying, and several actually suggested that the bullying within the intervention group increased (Table) Of the 10 studies of curriculum interventions, 6 showed though bullying and victimization did not change si nificantly, Boulton and Flemington” did find that the stu- dents in the intervention group broadened their definition, of bullying slightly, and Englert found that the teach- cers reported a significant decrease in observed physical and verbal violence (P<.01). (©2007 American Medical Association. AI rights reserved, ‘Downloaded From: https:/jamanetwork.com/ on 09/30/2019Table 1. Study Characteristics (cont) a = Sar Se os Se ee Need sin tense es age Repos wakes tay ae cae es saan ier ee Sige Maatnes Mees” “ene Meta, e a = a = = vine, ie, item, wet ett Ont me nid ntti Sip TE, SURES ELE VS ae Seth y MERE Meat is, e ee, Six = a is ihn gar eee aa =o. See iored "fierertan ‘ernes see ee Aeiets erage gine age et wae EBay ii ante go amare “rarraion “els ag Es EE = ae ‘nflnscools panopa ie, ig enmeeg tei, tatty gn Ngee se a ih, REE BM Ey eee i i ee see, tanh sc tk ste nigga ye eg cae pian sae ahi PH enoing Stendhon, ES aoe ae aime, ae =e ong? ty gees kg a ‘at ye pinay sendy peach pao siee, coher pat co pores Ee ae ea eae Higdon om i, tn Oabed Pgs cee Ona gd a Wits ith, ‘te oo a a ae a. = EE Fees nage ty Tiger ag seit ET ty =] een iss = sie ed te te, Se, Oak Pe Wine eee OE. Ethan coger waa Sara “rat i : = a a ae = ao Sieger a Semeatier —ftenetion mre deem de, tana Pg Wen SE Age A ee e = aad aa = =. ae es ie ‘on (©2007 American Medical Assoc /ljamanetwork.comy on 09/30/2019 om. A rights reserved, Downloaded From: https:/Table 1. Study Characteristies (con!) Cie suey ataton Nehaaal asp sowee county __sutjcts__cntotaroup_"'Granp "erage rep ___Dengn___trenton Teor” Ataighment Soave Talangervenans Diese Unies 1Ssudknis 9-217 SGadeS Prat pestst-SS.GRN’ Social karing Random ot” "Ses “var Tandoas,” sacl hs ancora ssgnent Sete feomaké” —— guptaning beaver byte fone age eas ‘ies Faga® Unies — 12Tsudens n= 105 222012 Grae? Oungpaing Par edator Pe madaion Seated baat at rahe insane peuptaring mgr” on ayeson ‘a =e fila |S shear) aah ects Meraod Sou Stasis at n=18 ‘nena! Grade 6-7 Pretest poste, Bahra elle Sol Mahe ene “Aca SS scgae, ‘tavern Tanning gu inaesont “ages Ea ‘nee age anche as eiwtor ie ten eed as ine Band theo Fehon aa te aie a Helo, covet Tay and Unies a0 sus ih m= 15 raled Prat postast_ Socal sls Social ils Suen sed Dats "Kegon "snot sod ined eahot ~taing gow raniion easy. 2a itheal Combes” pa” Soe Sg cal Sian, Unie TOs ott pranyndpiaryand Pry Pte pte, aii by sar Regie 2 wendy ecndey "sstndry ‘coer graup "war “choo sae Nomen food sont ahaa "Sign feeiey ‘hone Negaai* Undated Meek” Gnd aho ay Maing Suds cad a er ae eg wie “aire —“‘aeon Weed ed) na saa"? Ran" ststten SS pulse” eat ‘uso hcalatve ‘Araviton NA, ant ppeabe, “formulary or hole-schal eterentons, components ae gen, Of the 4 studies that did show less bullying after a cur- riculum intervention, 3 also showed more bullying or vie- UUmization in certain populations or with certain mea- surement tools.®°* The study by Baldry and Farringion® showed a decrease in self-reported victimization among, older children (P<.05), but younger children actually re- ported more victimization (P<.01), and there were no significant differences in either victimization or bully. ing overall. Teglasi and Rothman’ found that teachers reported decreased antisocial behavior for children not identified as aggressive and increased aggressive behav- lor for the children previously identified as aggressive (P<.01 for both). The individual self-reports for aggres- sion did not reveal any significant effects [rom the inter- vention. A study by Rican et al” found significant de- creases in peer nominations of bullying (P=.02) and vietims (P=.03) using unspecified “broad criteria,” but no change in victimization using “narrower criteria." Only |Leurriculum intervention showed unequivocal improve- ments, and this was in an indirect outcome. The ran- domized trial of Elliott and Faupel" of a group problem- solving curriculum resulted in increased generation of responses to a simulated bullying situation by the inter- vention group. WHOLE-SCHOOL MULTIDISCIPLINARY INTERVENTIONS Ten studies" evaluated interventions using a mul Lidisciplinary whole-school approach that included some combination of schoolwide rules and sanctions, teacher training, classroom curriculum, conflict resolution tain- ‘ng. and individual counseling Table 1 describes the com- ponents ofeach of these multicisciplinary studies in de- tai. The whole-school studies involved more subjects than the curriculum interventions, with up to 42 schools ina single study. Only 2 of the studies”" evaluated inte ventions among secondary school sutdents, and the rest looked at primary schools. In contrast to the curricu- lum studies, only 2 of the whole-school studies incor- porated randomization in their study design. Two of the studies™° used a quasi-experimental design with ime- lagged age cohorts, Two studies, both evaluating the seminal Olweus Bul- lying Prevention Program, revealed disparate results. The Olweus Bullying Prevention Program pioneered the whole-school approach to preventing and reducing bul- lying with an intervention program in Bergen, Norway that included training for school personnel, materials for (©2007 American Medical Association. AI rights reserved, jamanetwork.com/ on 09/30/2019Downloaded From: https:/ (gepniytep) SRC PEDIATR ADOUSC MEDIVOL TANT (©2007 American Medical Assoc /ljamanetwork.comy on 09/30/2019 om. A rights reserved, Table 2. Study Outcomes by Intervention Type ermton team: hg opt ne, a, vag tea sou "a ae Sanaa ey Saga. tae Tipe nn ea Tana fatto.” “ime ests apo Ey ‘onset 225 BH our wha age spe sy eso nes ons te so, aed areeeeeer Seth: Sees os el Cminhin teams abingheacunb at; ene caren aerate roe sy fein tem mon gry ef hin em Sather "eaeaeenas he pale tin*10 can —_ Nea ton Cs stag Reet om ae a eee ee a eelereyee tans geariane Serena, eee eer ae en ocala ee ne ee ae leat ee SSnaaaen aan Semwereuaer mimesis — etiiataeyt lascon ee werse RNa Aagaueg® cara tecture RICA ot crs of hing Impresoci ile MANOA ‘ sir ee eee ed Sareea em § — Siemtmr etn apa cecesecaanace” Rivard Ge Sheet aoe SS A ae Soren an ee oy palermo, Fogg Cam mpeg enn casera ip er—enand "Bs rea er a re Te pine ee ele Wo Se uacponcutameatune forte fncpmamanmeents ee ae omeeetielier ieee et oat fase walutniaiepeeeeed ‘uli (P02) - peso Suse? Cat Neat suena» sgn ys "r pec esie fy np? Sarna enter, ee oe seas ertomme has simi, isin gael enepe nist wrmeticapasin wind ss a ee el ee eee i et et anereianae ome ine ecenmeeratert Ce ene eT aa seein st Sone Pay eenmersee Grea Ui eaenreryraer, Siacracekien Voges cuca neha algerie» ga br “ sceranacen ray re Si See ia mueel aay ae passe aera seerolre a ieam Bh gin, scary ocean sf: and eb nc oe ‘iit 7 Mec memes! Osa! ae w ieee suri cramteentena SENSES rer ctae heen peated See conmatecs: etieeesiie: Gein mene eene tan tate eso ah as a Sie nes crane eect, pene ees ean cn amaomais eae Magu —sutsipay Meyer wai acer, ies n ats unin ea ee “a reams aenaeeaea SPE Daiiny Sraciitees aeamas Rene mornaneraceimes Seariray nana anne See eens ee =aeeieeeeseen, Seniesa iepomoncae daha Sa ‘cntTable 2. Study Outcomes by Intervention Type (cont) =a wget ep pg ence ee rename tat Sac e ee a ear speaeetee bees eure Er Sena cbeyarceaerttel ieeemecnai ieenireienaeies te, atiny ieee eM: | mypeneina ay pi - on" Aenea kit BREE an a esivaiel salem” ee eee cin anes = sega eee Sheena Mnschney Simons ae eee Mar “a are Er) aoe a mat ieee eas ae at Susie cae Pcie ere ae Oger eins nena | gra aor ewe a cleat een ee Sete ae eae Serene Spleen tite mean Seeneeee meee Magda Wags trad pra tntac i Sf acco SNe Ne ge “HET n satan tan : Geer aa ay Meleenada ae gorti SRGiematemers” pemanmness sags aapney testenaigancascroma ines Moped “i ee cee a beferercegre niveau nainaes, Eeermen eee vnscpnny somo seston an der ‘cae my ss ore it} a epor eg blng com # Sree pe oe soeas Tent tiny inetd nters preset el an i es erent aarti ae nna a Grserenwwaamecs cerakiunss Ginter Situ einer Sucaeneccerceras, emer arene sree ea ae eeaemenrncar, eee caeataias een pr ea ogee sguam — GRaeeeenpeomy mote rote Te tan ecsceeiedaesk, Saemenenteanet Seen Tall 2S aes, es ace ee eee ee a ye epee ee mene eens Ege” Senn meee ee Sgegm aan atigee tna aaa sagem plies rc eed te sere vO Inca hrgem vince saa ‘resend and beta an acer pots bt Sqaeateniaageenr cat vemciad mgoenaigiy nd etoingommt, | aed Tae eae! aerate eee Eel area oaeee at ee er Seeunioe Mcecee Socannen coon, Sees raniveareatecnteh, rar mrat ates Nggt sn gece spose, men 2.8, ai te eeterrceeiceees «perma caateare, ivaiemeieemmncaes, pieuestons eae aaesr Skewes Seoremornaee . “ Prem eeeorry Bina nb progam, 01079 shone sh Resieer ate ‘Abravitor MANCOVA mpl ara of eovarnce Downloaded From: https:/ (gepnnytep) SRC PEDIATR ADOUSE MEDIVOL TAI [ANAT (©2007 American Medical Assoc /ljamanetwork.comy on 09/30/2019 om. A rights reserved,parents, a videotaped classroom curriculum, and evalu- ation through the bullying questionnaire developed by Olweus.” By using unspecified composite measures in- volving student questionnaires and teacher ratings, the follow-up evaluation found decreased bullying, de- creased victimization, decreased antisocial behavior, and improved school climate after the intervention."*” Evalue ation ofthe nationwide Olweus Bullying Prevention Pro- gram in Rogaland, Norway, revealed strikingly different results. Roland?’ reported increased vietimization and social exclusion for boys, and increased bullying for boys and girls based on student self-reports. Unlike in O- \weus' protocol, the schools inthis sample did not inter- act with the researchers during the intervention. The schools’ degece of involvement in the program was di- rectly related to positive effects from the antibullying pro- gram, particularly for girls. Although the evaluations ap- parently involved the same nationwide campaign and evaluation tools, Ohweus states that they “were com pletely different in terms of planning, data quality, times of measurement, and eontaet with the schools.”""” since the publication of the study by Olweus,"* ine terventions targeting the whole school have been im- plemented in several other countries. Overall, these whole-school studies had positive effects on bullying. OF the additional 8 studies, 7 revealed positive out- comes." Five of these studies!" reported decreases in bullying oF vietimization. Among kinder- garteners, Alsaker and Valkanover” found decreased vie~ timization on teacher and student reports, although there was no significant change in bullying on either student nominations or teacher ratings. An Italian schoolwide peer support intervention prevented some of the increased negative behaviors and attitudes reported in the control group on student reports.” Examining administrative of fice records, Metzler et aP* found decreased discipline referrals (P204) and harassment (P=.02) in seleet popu- lations after 2 years of implementing a schoolwide be- havioral management program. However, student re- ports of physical and verbal attacks did not significantly change. Twemlow etal" found decreased disciplinary re- ferral rates, decreased suspension rates, and increased achievement test scores after a schoolwide intervention. {In their evaluation ofa 12-week schoolwide program, Rac hey anel Criag! found mixed results. On peer and sell- reports, the students in grades 5 through 8 had de- creased victimization (P<05) and decreased peer isolation (P<01) compared with the controls." In contrast, the younger students reported increased levels of vietimiza- tion (P<.03) and increased exclusion (P<.01)." In ad- dition, neither student nor teacher reports showed asig- nificant decrease in bullying for any age group. Two of the studies” that revealed improvements af- tera multidisciplinary intervention only measured out- comes indirectly related to bullying. Some of the indirect outcomes were positively affected, but others remained un- changed. The evaluation by Sanches etal indicated no sig- nificant inerease in knowledge of bullying, but the inter- vention students were more likely to report seeing bullying and to express readiness to intervene personally (P<.05 forboth). A randomized controlled trial” evaluatingan in- tervention to develop “health-promoting schools” re- vealed an increased awareness of health-related polices and practices among school administrators ater the interven- tion (P=04). No significant changes in health-related poli- cies of practices occurred in the intervention group, in- cluding those related to bullying. The other evaluation ofa whole-school intervention that we identified revealed no significant decreasesin bullying, Melton etal implemented an intervention based on the (Olweus Bullying Prevention Program in 6 ruralschool dis- tticts in South Carolina. After 2 years, they found no sig- nificant differences in student cell-reports of bullying, vi Limization, generalantisocial behaviors, orattitudes toward, bullying between the intervention and control students, SOCIAL AND BEHAVIORAL SKILLS GROUP TRAINING INTERVENTIONS Fourstudies"” looked at targeted interventions involv- ing social and behavioral skills groups for children in- volved in bullying. Two of these interventions speci cally targeted children with high levels of aggression," while the other 2 targeted children who were them selves victims." Of the 4 studies, 3 focused on older students, in sixth through eighth grades," while the fourth looked at third-grade students.” The most posi- tive outcomes occurred for the youngest students." DeRosier* tested the efficacy of social skills group train- ing for third-grade students with peer relationship difficulties in 11 public primary schools in North Caro- lina. The intervention resulted in decreased aggression oon peer reports (P-<.001), deereased bullying on self- reports (P<.05), and fewer antisocial afiations on self reports (P<05) for the previously aggressive children. This ‘was the only social skills training intervention that showed clear reductions in bullying from the intervention. ‘The other social skills group interventions, all of which involved older children, did not result in clear changes. Meyer and Lesch'® evaluated a behavioral skills modifica: tion program for boys identified as bullies in South Alrica This intervention did not produce any statistically signifi cant decrease in bullying behaviors by peer report or sel report. Fast etl examined whether group training in peer ‘mediation for aggressive seventh graders would reduce thei level of aggression. The aggressive students in the inte vention group had a significant decrease in their problem Ibchaviors as measured by their teachers (P<.05): how- ever, no significant changes in disciplinary referrals forag- gressive behavior or impulsivity scores occurred. Tier and Dowd” used social skills roup training for eighth- grade girls with emotional and behavioral concerns. Al though the teacher data indicated statistically significant progress in the areas of friendships, behavior, interactions with peets, and level of teacher concer, the data and analy- sis were not given within the article. Data from the pupil questionnaires were not analyzed, but in student self reports, 8 reported no difference in vietimization and 7 re- ported improvement, OTHER INTERVENTIONS A single study, done by Bagley and Pritchard, ** exam- ined the ellects of an increase in the number of school (©2007 American Medical Association. AI rights reserved, jamanetwork.com/ on 09/30/2019social workers focused on problem behaviors, includ- ing bullying. Compared with matched control schools, they found a significant decrease in self-reported bully- ing within the primary school (P<.05), but worsening, bullying in the secondary school. For self-reports of the, truancy, fighting, and drug use, the primary and second- ary intervention schools had significant improvements (P<.05). A study by King etal” investigated the effects ofa mentoring program for “at-risk” children, The men- tored students were significantly les likely than their non- mentored age-matched peers to report bullying (P<.002) physically fighting (P=.001), and feeling depressed (P2.006) in the past 30 days. Le} As governments, schools, and educators invest increas- ing amounts of money and time into anuibullying inter- ventions, the findings ofthis review provide evidence for how to best achieve the desired outcome of decreased bul- lying within schools. By systematically gathering and com- piling the growing number of studies evaluating these in- terventions, it becomes clear that some of the antibullying, interventions actually decrease bullying, while others have no ellect or even seem to increase the amount of bully- Ing, Grouping the studies by the typeof intervention seems tw olfer the most insight into what leads to suceess. The curriculum interventions were generally de- signed to promote an anuibullying attitude within the elass- room and to help children develop prosocial confliet reso- lution sills. Most of these interventions drew on the social cognitive principles of behavioral change, with fo- cus on changing students attitudes, altering group norms, and increasing sell-llicacy. Curriculum changes are often attractive because they usually require a smaller com- mitment of resources, personnel, and effort. Neverthe- less, the interventions that consisted only of classroom- level curriculum seldom improved bullying, The basis in social, cognitive, behavioral change may explain part of the problem: previous work” suggests that younger children benelit less from these techniques. However, the failure of classroom-level interventions for older and younger students points to the systemic nature of bul- lying and supports the theory of bullying as a socio- cultural phenomenon, If bullying is a systemic group process involving bullies, victims, peers, adults, par- ents, school environments, and home environments, an intervention on only 1 level is unlikely to have a signifi- cant consistent impact, Furthermore, ifbullying is, as some propose, a sociocultural phenomenon springing from the existence of specified social groups with different levels of power,” then curriculum aimed at altering the atti- tudes and behaviors of only a small subset of those groups. fs unlikely to have an elfect. Similarly, the targeted interventions providing train- ing in social skills did not clearly improve bullying or vie- mization.’ The failure of these interventions, also based largely on social, cognitive, behavioral changes, points again to the inability of « single-level interven” tion to combat bullying effectively. Interestingly, the older children had worse outcomes from the social skills trait- {ng groups than the younger children." The 1 study look- {ngat younger children found decreased aggression, bul- lying, and antisocial affiliations. Although one cannot generalize from asingle study, itis possible that address- {ng social skills changes in the context of a small tar- geted group during a particular developmental window could be effective. Overall, the studies of social skills group interventions suggest again that failing toaddress the sys- temic issues and social environment related to bullying undermines success. ‘The whole-school interventions, which included rule tiple disciplines and complementary components di- rected at different levels ofthe school organization, more often reduced victimization and bullying than the inter- ventions that only included classroom-level curricula oF social skills groups. The whole-school interventions ad- dress bullying as a systemic problem meriting a sys- temic solution. They seek to alter the school’s entire en- Vironment and to involve individuals, peer groups, classrooms, teachers, and administration. The success of the whole-school interventions suggests that bullying does, indeed, spring from factors external to individual children’s psychosocial problems, including a complex process of social interactions. An evaluation of whele- school approaches by Smith etal" in 2004 suggests that these interventions may “reflect reasonable rate of return on the investment inherent in low-cost, nonstig- ‘matizing primary prevention programs.” Our review of fers further support for using whole-school interven- tions to reduce or prevent bullying. Despite the evidence pointing toward the value of whole-school approaches, significant barriers may still limit their effectiveness. The implementation of the in- tervention can vary significantly, and this clearly alters the results, The original antibullying whole-school ap- proach studied in Bergen by Olweus” and the evalua- tion ofthe same program in Rogaland by Roland?" pro- duced the most strikingly disparate results. The contrast ray have been the result of decreased school staf! par- ticipation at the Rogaland schools.” In addition, the Ok ‘weuls program does not include detailed instructions for replicating an identical program in another school set- ting, Dificulty in replicating this program may conteib- ute to the lack of stceess when used in other settings, suchas South Carolina.” Although the adaptation of the interventions in different settings may ereate more cule turaly appropriate interventions, these modifications may produce some ofthe varianee in suecess. Unfortunately the specific components of a given intervention are gen~ erally not described sufficiently to enable faithful repl cation. The specific school environment could also sig- nificantly impact effectiveness. The small class size, excellent teacher training, and tradition of social wel fare intervention in some settings could enable better e fects. The suggestion that whole-school interventions may not work as well for younger children, seen in 2 of the °¥ also merits consideration. This, albeit lim evidence may support a developmental theory whereby bullying begins in early childhood as individuc als assert themselves to gain dominance and then gradu ally evolves s children useless socially repre to dominate others." Schoolwide rules and changes in (©2007 American Medical Association. AI rights reserved, jamanetwork.com/ on 09/30/2019the school’s overall responses to bullying may not be as effective in the younger population before they follow their natural developmental progression into conformity with social norms. There are several limitations to this systematic re- view that warrant consideration, We only included std- fesin the English language. Although we may have, there- fore, overlooked some relevant studies, we located few nnon-English-language studies that required exclusion. Some of the inchided studies did not have ideal meth- odological strength; however, many ofthe studies were reasonably well done and offered important counterbal- ances to the findings. The study results may be overes- timated because, in many cases, schools or districts were randomized to treatment conditions, but the students were evaluated as the unit of analysis. The unit of analysis prob- lem could result ina higher type I error ifintraclass cor- relation is not taken into account.” Even so, many of the studies that did this still found no treatment elfects, The use of variable outcome measures may further limit the ability to measure accurately the effects of these inter- ventions. The most common outcome measures were self reports of victimization and bullying that may not wholly correspond with information obtained from peers oF teachers or from observations." Sull, self-reportsare the standard measure used in most studies evaluating be- hhavioral interventions. Despite the diversity of the evi- dence reviewed, the studies were primarily performed in Europe and the United States, which may limit the gen- eralizability of the conclusions. In addition, several in- terventions with positive results, including interven- tions using mentoring, increased social workers by 2.5 workers, and social skills groups for younger children, were only studied on a single occasion, thus limiting their gen- cralizability. Finally, while we atlempted to separate out the most elfective components of intervention strategies, many of the studies involved numerous complementary components that were not evaluated individually. In conclusion, fairly consistent evidence suggests that children’s bullying behavior can be significantly reduced by well-planned interventions. The chance of success is greater the intervention incorporates a whole-school ap- proach involving muliple disciplines and the whole school community. The school stas commitment to implement- ing the intervention also may play a crucial role in its sue- cess, The use of curriculum oF targeted social skill groups: alone less often results in any decrease in bullying and some- limes worsens bullying and victimization. Caution should be exercised in supposing that antibullying interventions invariably produce the intended results. This review re- veals that not all programs have proved effective. Most re- ductions in bullying tend to be relatively small and related ‘more to the proportion of children being victimized rather than the proportion engaging in bullying. Additional re- search to evaluate bullying behaviors and antibullying in- terventions is clearly needed, Accepted for Publication: August 3, 2006, Correspondence: Rachel C. Vrceman, MD, Children's Health Services Research, Indiana University School of Medicine, 699 West Dr, Riley Research Room 330, In- dianapolis, IN 46202 (rvreeman@iupui.edu), (nepnnyrep) ARCH PEDIATR ADOLISC MEDVOL TAI ANI Author Contributions: Dr Vreeman had full access to all the data in the study and takes responsibility for the in- tegrity of the data and the accuracy of the data analysis, Study concept and design: Vreeman and Carroll. Acquisi- tion of data: Vreeman. Analysis and interpretation of data: Vreeman and Carroll. Drafting of the manuscript: Vree- man. Critical revision of the manuscript for important in- tellectual content: Vrceman and Carrol, Statistical anal sis: Veeeman and Carroll. Administrative, technical, and ‘material support: Vreeman and Carroll. Study supervi- sion: Veeeman and Carroll Financial Disclosure: None reported, Disclaimer: The views expressed herein ate those of the authors and do not necessarily represent the views of In- diana University School of Medicine —_ EES} 1. lew Gira Feuer uljing: ein hung cle, Paar Re 00024-18810, ‘hu Dlg sch at and an tt ieenton gegen, Promot ue WRK 27 31 ‘hes Sar ary in Sh PAY, JngeTas, Olas Cana Se, Tere o Schoo Bling ACos-ational Perspective New orc MY Rowtes 10007-8 \Wtans Chars, Logan RoinonDAssoction crs heath sympa wih bling in pinay stool chen BU 1863131710. 5. Rg. Persian schol and tht aan sho stds. ‘Br auc schol 190 Sp 195-108 6, Date JA Pie Tallhan SK. The ri and eet of bling at school ‘Sch eth 200872172 120 iatitrHeino Ringel, Macunen Rena Rantanen Bullying esion and aul esionn Fish stent chot ere. UL 1280 Frees Salman Jas A, Sih OM. uljing in schoo: presen and sl esto ecnday eel chi ts, alia Hino impala, artanen P. Ringel A Buh at eho a in ‘stor false tik for mental ord J Ades 200023684 er 10. ObeusD. Vetination by pester andlongem ous: A ‘ink Aeenert 8 ee Sor WithranalIiton and Shyness Chen ‘ase: Lawrnce A Esau Aspe: 10075 21 11, Bonn, Bech OW fanharlV uyng perpen: te ecammendaons lt 13 ea etn he inthe Uned Sate So Heath, 2005,75:84.20 12 Eenbarg ME, Murr Sane D, Par CL. Per arassmant, schoo co etna, nd aadeiaheenent J Soh Heth 20057837 216 18 Har ower Over Whathrn say abot bug. Ect ava. sanz4i0.22 14, Keshentr Bl, Lad GU. Par iin: cae or consqunc of choo rlaustan? Ch Dv 1095671905411. 15: Nyrar Joseph. Buc pti ant asscsion vith Een’ esoral) dnesonsind 13 years, Br Jive set 100757 1-54 16 Glew GM Fan MY KalonW, Raf, Remi MA ulling,pychoseil ad [use an aaa permanent Schoa Arch Pear Ao. ese Med 0051581026108 1. dv, Wihina& Graham Pe rasan, eyo ausment a school furctning i ently aescenen. J EacPsyhol 200092340 0 18 Nace TR.Ovarpec Pla RS, Fan Wi, Simons Monon B, Suc Bu ying behaviors ong US you evans and association wth psyetosocil juste. JAN 200 28520882100. 8. Pery Lets of pear agrssin Dev Payal 108824807. repo iy de ae ses7808- 20 Sin Arado Cai erenons ta ec shoo uljing. Can J Pajhuty- 2008 4891-98. 2. ObeusD lina etot Wha We eo and Wha We an Do. Cambri Mase Blaha Pier L193 22 Obes Buljing at scoot antes fa schol basa itenen- ian rota. Chis Payal Psych TOD435 171-110. (©2007 American Medical Association. AI rights reserved, ‘Downloaded From: https:/jamanetwork.com/ on 09/30/2019a. Py Steven Oe our, Van ot Buln in amish chal eve ‘ono at-ulying nanan n primary and scandy schon. J abe Psyto 20070195 210, KEM, BoP, Moskos tl. Const ot uy sass wih ‘ies vse tect inst anaes tani convled as ANA ooze7 207-282 6 ‘od Gtr, Long M, Young, ing ei: Curent Stes Reseach, and itaremons V2 NewYork, NY Havoc Maen {Tran Pree 200157160, Twelon 5, Fay P, Satz FO, Ges ML Eins R, Ean Craog 2 scl schellering nine conto of anelemeny shel Ironton reduce lene. Am Psy 200 1820880 25. Askar FD, Valaover SExy gnosis and prevention of vtmztnin 44, Deo ME ulin rans and combating bujing: laces of Kine Jnosn ran Se. aor Haraesment Scho! Ne Schoo asedsocil sis oop intention. JC Chi Aas: Poca 2004 ork MY Guts Press 200-175-105 sian, 2% Cowie, SmihPK Bouton Laver Cooperation inet Clssmam. 45, Fat Fan, Sen. Hou becoming meditate ages students Landon, Erg Gai For 0 ‘hi Se 20082556117, 2a Blog elpg easton such ang manpenat. Tat 4, jr Lech An aa ins ot ah programe tun Po nartiing ad Meaping Bung Oso gan Hi balyngtopams teonomianiannt SAP JONG Adsl eth tn avira 16091890. soe 28 Wotan 8 Liner SF. Cannan, cdma fur Youth Was 4, Ty TD Te a of al sls as a san ite th > non, Oto vant nd segue Pr: 188. Tear Me wgn tn ll ru en ue 20, Bay A, frogs Osho an era ream athec- yg Stel ies tenn cho Super aang 200 S42 Sonet hing and iio sho Ages Be 200130.-15. Sonim sk ot: an eprint dy of cel ail wa th at 20 Soil Fog hota on eh xn Smetana sree eel a steal papi ion in tion ofan eon BAR, 4a ig Vou De HeCalan Weng sas and hel Pac conecnes tush mucosa mer ean ea 1. Bul gl Che's solos baling ies, ee Phe Pat coma soars 50, Owes, Vision song schulhientrvenon nd preein. 52 Et ese ters foes Meron Pinay Old Venn aor shaman nd rein. cee ncaa ano Ragone sserzan Unvesiy ‘enon. Thousand Oaks, Clit: Sage Pubcabons;192:270-205. Fe ma sea ASueyo!BEST—Auhin Enz 5: Obes D- Bulg ang chon nen an revetin. Pe 4 ra 2. a. Py 2. a 4 2 ate Fram Schools Teter At ibing Progam for Seveth Gade ‘Shen Tlahsse- Fis Ste Unierty Calle of cation, 203 Rican? nsrove Sits Tota shor, mani retin upon ina Gch own, dan WY Acad Se. 1996774300 Tels ethan STORIES: aseacn- based programto race apps sive bhai. J et Peal 2001397134 Warden, Maan Gis J Maes 6, Maced L.A erausintacitrar's sy traning popes, Eau ajo 197-740-409, Menai Coca Bel 8 Conia Enhancing chen epensiilty ‘ota ton apanstbulyeg:evuaton ta een itarvenon in aan idl sbal- ess Bei 20082814 Mati, gan A shy J, Sprague Evin af acomprhesive aha lormaraamentpregamtsinpoe sche idepostnebekavr spp. ate Trt Oa 001 2048-478, chal Palmer, ooh Pov Dies 6. randomised lo ani veronéelp heath promoting schools in Ausra: thesauth mst Sy ey study. Aust 12 Pl Heat 2000242226 a Cay W. Elution ef clog progam reucebujing ins, an Gouri, 200236281256 Rand Bujng nana tae aenaliooatns inRoneagn schol in 13 yas. Agrss Boor. 200026135148, Santer Auber Lew sent ohms T, Cay 0. Prevent ing buljg an sol hres n lean shal te Egat Respest 2, tere RD tlaon Ona VL de. Aggession nde Thought ‘te Le pan Loni, Ena: ape Pubes; 1982 10-125. Comer Norman. Pret es Behaviour Bckoghan, Egan: Open nersy Pres 105 58. Senna, Murphy. Paycalagy and Heath roman Buckingham, nga 54 Wel FW aliry AC An appt of he Sere Sai ini in 7, Oper Univers rss: 1987 ty inarentn programing: vee sts on acivaing the be sperspeciveln eadlscants perceptions af asteping-nebehour sus Cmiaogt lag Pcl 1007 253-15, ale AC Wn Pay evento ot nguyen, grok eds Payctlogy and Lavina Changing Word London, Engl Rowe, 2001550. Sih JD. Schneider BH, Smith PK Arado K. Teles of whale Schoa anulhng programs: ase of aluaton seach, Schoo Psych Fav 200423547 96 iy KAresig bling ins hoy and prac, Tend sues rine ‘mia sts 208 2501-6 58 Holey PH, The onan sx! dominic: sata aed eon ny spc. Dy Re. 1808 1997-122 58, Roo Mura OM. Astana of smoking preenton rags ar juste for eerste unto aa, Reat ave @.OHE.2848-58 60, Pele A, ann M, Anemia conpuison of meso sargtg ages slonad imestonnschoo setings ate Payal 200092.950-388. (nepniytep) SRC PEDIATR ADOUSE MEDIVOL TAI ANAT (©2007 American Medical Association. AI rights reserved, ‘Downloaded From: https:/jamanetwork.com/ on 09/30/2019