A School-Based Intervention for MentalIllness Stigma: A ClusterRandomized TrialBruce G. Link, PhD,a Melissa J. DuPont-Reyes, PhD,b Kay Barkin, BA,c Alice P. Villatoro, PhD,b Jo C. Phelan, PhD,d Kris Painter, PhDe

abstractOBJECTIVES: To determine the effectiveness of a school-based curriculum, Eliminating the Stigmaof Differences (ESD), in improving attitudinal and/or behavioral contexts regarding mentalillness in schools and increasing the likelihood that youth seek treatment for mental healthproblems when needed.

METHODS: We conducted a cluster randomized trial in sixth-grade classes from 14 schools in2011 and 2012 with follow-up at 6-month intervals through 24 months (2012–2015). Usinga fully crossed 2 3 2 3 2 factorial design, we compared ESD to a no-intervention controland to 2 comparator interventions: (1) contact with 2 young adults with a history of mentalillness and (2) exposure to antistigma printed materials. We implemented interventions inclassrooms in an ethnically and socioeconomically diverse school district. There were 416youth who participated in the follow-up, and 312 (75%) of these participated for the full2 years. Outcome measures were knowledge and positive attitudes, social distance from peerswith mental illness, and mental health treatment seeking.

RESULTS: Youth assigned to ESD reported greater knowledge and positive attitudes and reducedsocial distance (Cohen’s d = 0.35 and 0.16, respectively) than youth in the comparatorinterventions and no-intervention groups across the 2-year follow-up. Youth with high levelsof mental health symptoms were more likely (odds ratio = 3.51; confidence interval =1.08–11.31) to seek treatment during follow-up if assigned to ESD than if they were assignedto comparator interventions or no intervention.

CONCLUSIONS: ESD shows potential for improving the social climate related to mental illnesses inschools and increasing treatment seeking when needed. ESD and interventions like it showpromise as part of a public mental health response to youth with mental health needs in schools.

WHAT’S KNOWN ON THIS SUBJECT: Attitudes toward mentalillness develop early in life; youth with mental healthproblems encounter bullying, and only small proportions ofthose with mental health problems receive treatment. Limitedevidence suggests that short-term improvements inpreadolescents’ knowledge and attitudes are possible.

WHAT THIS STUDY ADDS: A cluster randomized, school-basedintervention implemented in a multiethnic community sampleshows that an antistigma curriculum intervention improvesknowledge and attitudes, reduces exclusionary tendencies,and leads youth with mental health problems to seektreatment in a 2-year longitudinal follow-up study.

To cite: Link BG, DuPont-Reyes MJ, Barkin K, et al. A School-Based Intervention for Mental Illness Stigma: A ClusterRandomized Trial. Pediatrics. 2020;145(6):e20190780

aDepartment of Sociology, School of Public Policy, University of California, Riverside, Riverside, California; bLatinoResearch Institute, The University of Texas at Austin, Austin, Texas; cMental Health Connection of Tarrant County,Fort Worth, Texas; dDepartment of Sociomedical Sciences, Mailman School of Public Health, Columbia University,New York, New York; and eSchool of Social Work, University of Texas Arlington, Arlington, Texas

Dr Link undertook the data analysis, produced the first draft of the manuscript, reviewed andrevised the manuscript after receiving feedback from coauthors, and contributed to thedevelopment of the design, measures, and analytic plan for the overall study as coprincipalinvestigator of the National Institute of Mental Health–funded study; Dr DuPont-Reyes contributed tothe development of the data collection instruments, participated in the implementation andmanagement of the study, contributed to the preparation of the data for analysis, and criticallyreviewed the analysis and writing of the current manuscript; Ms Barkin led the development of thecurriculum intervention, provided feedback on the analysis, and critically reviewed the manuscript;Dr Villatoro assisted with data merging, cleaning, and analysis and critically reviewed and revisedthe manuscript; Drs Phelan and Painter contributed to the conceptualization and design of theoriginal National Institute of Mental Health study, provided feedback on the analysis, and criticallyreviewed the current manuscript; and all authors approved the final manuscript as submitted andagree to be accountable for all aspects of the work.

PEDIATRICS Volume 145, number 6, June 2020:e20190780 ARTICLE by guest on October 27, 2021www.aappublications.org/newsDownloaded from

A long tradition of research hasexamined stigma toward people whodevelop mental illnesses,1 pointing tohow such stigma can harm socialrelations, lower self-esteem, andblock entry into mental healthtreatment.2–5 With respect to theexperience of young people, stigmapotentially contributes to problemssuch as large proportions of youthwith mental health problems notreceiving treatment for thoseproblems,6,7 experiencing bullyingand exclusion,8,9 and rising suiciderates among young people in theUnited States.10 The stigmatizingattitudes that may contribute to theseproblems begin early in life, andschools are important contexts inwhich mental health problems areexperienced and stigma isenacted.11–14 Schools are alsoa powerful socializing institution andconfer knowledge, attitudes, andbeliefs about a host of circumstancesyouth will confront as they grow anddevelop. It follows that a response toproblems of mental illness stigmamight be usefully implemented inschool settings and that sucha response might increase knowledge,improve attitudes, reduce exclusion,and help youth with mental healthproblems find their way to treatment.The current study assesses theeffectiveness of mental illness stigmainterventions in a 2-year longitudinalfollow-up study of sixth-grade youthattending school in a multiethnicschool district in Texas.

Few studies have sought to interveneearly, address stigma, and advanceknowledge and attitudes that mightfacilitate treatment seeking. Sixstudies of preadolescents15–20

implemented curricula designed toreduce stigma. Three included controlgroups,16,18,20 and 3 included follow-ups of 6 weeks to 4 monthspostintervention.16,19,20 Noneassessed treatment-seekingbehaviors, and none assessedstudents beyond 4 months.Comprehensive reviews of the larger

number of studies of high schoolstudents point to the need for studieswith longer follow-up, measures oftreatment-seeking behaviors, andcomparison of multipleinterventions.21,22

We conducted a multiprongedintervention with randomization ofintervention arms to sixth-gradeclasses. Intervention modalitiesincluded the following: (1)a curriculum intervention,Eliminating the Stigma of Differences(ESD), which was our primaryintervention of interest; (2) a contactintervention; and (3) saturation ofclassrooms with antistigma printedmaterials. These interventions werefully crossed with a no-interventioncontrol. In a previous report, weexamined short-term effectiveness onyouth attitudes and beliefs and foundthat the curriculum interventionimproved mental health knowledgeand attitudes.23

Although useful, previous studiesleave critical questions unanswered.First, evidence about thesustainability of beneficial changes islacking because previous studies hadrelatively short or no follow-up.Second, no evidence exists concerningchanges in treatment seeking amongyouth in need. The current analysisprovides such evidence by assessingknowledge and attitude change at 6,12, 18, and 24 monthspostintervention. This longitudinaldesign also allows for an evaluationof whether the interventions changedtreatment-seeking behaviors amongyouth who experienced mental healthproblems. The racial, ethnic, andsocioeconomic diversity of the studysample also allowed us to determineif intervention effects differed acrosssocial groupings.

METHODS

Study Design

The study was conducted in an urbanschool district in Texas with

randomization of interventions toschools. All interventions wereadministered by physical educationteachers, who are responsible for thedistrict-wide sixth-grade healthcurriculum. The superintendent ofschools sent letters to principalswithin the district introducing thestudy; 16 principals agreed toparticipate. Schools were rankedaccording to performance onstatewide standardized examinationsand were combined to createrelatively equal scoring pairs ofschools that were then randomlyassigned in blocks of 2 to the 8 cellsof the fully crossed design by thestudy team (Supplemental Table 3).Before the study began, 2 schoolsdropped out for nonstudy-relatedreasons. Because the 2 schools werelarge in size, the study was repeatedduring a second academic year witha new set of sixth-grade students in 5of the smaller participating schoolsbecause they were demographicallysimilar to the lost schools.23 Theinterventions assigned to the lostschools were randomly allocated tothese replacement schools. No one atthe schools knew of their assignedintervention arm before deciding toparticipate. All sixth-graders inparticipating schools were eligible,and students were excluded if theydid not receive parental consent andgive their assent for participation.The institutional review boards ofColumbia University Herbert andFlorence Irving Medical Center andMy Health My Resources of TarrantCounty, the county’s local mentalhealth authority, approved the study.

Interventions were clusterrandomized to schools becauseindividual randomization was notpractical and would not reflect theway the interventions would bedelivered if adopted for widescaleuse. Additionally, clusterrandomization reduced the possibilityof contamination that might occur ifparticipants were individuallyrandomized and interacted with other

2 LINK et al by guest on October 27, 2021www.aappublications.org/newsDownloaded from

students in the same school about theinterventions they received.

Participants and Procedures

The study included a previouslypublished,23 in-school pre-postassessment (Phase I) and longitudinalfollow-up (Phase II) with in-homeassessments at 6, 12, 18, and24 months postintervention(2012–2015). On initial contactwith research staff and beforerandomization, parents andparticipants were able to consent andassent to Phase I only or to Phase Iand Phase II. In Phase I, 751 of the1252 invited (60%) students agreedto participate, and 721 completedsuch participation. Of the 751, 484(64.4%) agreed to participate in bothPhase I and Phase II (SupplementalTable 4 includes Phase IIparticipation rates by demographicand other variables). Of thoseagreeing to Phase II, 416 (87%)completed at least 1 assessment inPhase II; 99% of Phase II participants(n = 412) participated at 6 months,89% (n = 370) participated at 12months, 81% (n = 338) participatedat 18 months, and 75% (n = 312)participated at 24 months. Because ofthe nature of the interventions,neither teachers nor participantswere blinded to the interventionsreceived. Follow-up assessmentswere self-administered on laptopcomputers in each participant’s home.The Phase II sample (N = 416) was,on average, 11.5 years old, and 56%were girls. Forty-four percent self-identified as Latino, 22.4% as AfricanAmerican, 26.3% as white, and 7.4%as other, results that correspondrelatively closely to the ethnicdistribution in the population of theschools we studied at the time ofrecruitment (49.7% Latino, 19.9%white, 21.7% African American, and8.7% other race and/or ethnicgroups).24 Parent or guardianeducational attainment was 17% lessthan high school, 59% high schoolgraduate or some college, and 23%college graduate, figures that

correspond relatively closely topopulation estimates of the citywhere our study was conducted(15.8% less than high school, 55.5%high school or some college, and28.8% college graduate).25

We assessed the possibility ofselective attrition after interventionassignment among participants whoagreed to Phase II and found nostatistically significant differences byage, sex, race and/or ethnicity, socialdesirability bias, intervention groupassignment, preinterventionattitudes, primary language spoken athome, or parent or guardianeducational attainment betweenthose who eventually participated(n = 416) and those who did not (n =68; Supplemental Tables 5 and 6).

Interventions

We tested ESD against comparatorinterventions and a no-interventiongroup control. Each intervention wasdeveloped to correspond with how itmight actually be implemented inschool settings rather than to achievean equal balance of time devotedto each.

Curriculum

ESD is a 3-module, 3-hour curriculumdelivered within 1 week, with eachmodule involving a didacticcomponent, group discussion, andhomework exercises. Module 1addresses the bases on which othersare judged to be different; thedefinition, causes, and consequencesof stigma; ways to end stigma; anoverview on the definition,description, causes, and treatments ofmental illness; and barriers totreatment seeking. Modules 2 and 3address specific mental disorders,including attention-deficit/hyperactivity disorder, anxietydisorders, depression, schizophrenia,and bipolar disorder and includecontent that stimulates empathy(Supplemental Table 7). A video withan explanation of the curriculum’spurpose and a walkthrough to help

teachers with its presentation wasprovided. Fidelity was reliablyassessed by study staff and found tobe generally high when usinga 60–item measure based on 2existing tools with good psychometricproperties.23

Contact

Two young adults, 1 man and 1woman, with histories ofhospitalization for bipolar disordereach prepared and delivered a 10-minute in-class presentation abouttheir experiences, which was thenfollowed by a group discussionmoderated by teachers. Guided byprevious research, the presentationswere constructed to moderatelydisconfirm stereotypes of mentalillness.26

Printed Materials

Teachers displayed posters inclassrooms for 2 weeks and providedstudents with bookmarks thatreferred to people with mentalillnesses in terms of the individuals’personal traits and abilities ratherthan language that labels people as“mentally ill.”

Measures

All outcome measures wereadministered at each wave of datacollection (preintervention, 3 weekspostintervention, and then 6, 12, 18,and 24 months postintervention).Question wording and descriptivefrequencies are found inSupplemental Tables 8 through 10.

Knowledge and Positive Attitudes(a = .78) is a 21-item measureadapted from Wahl et al16 assessingyouth knowledge about and attitudestoward mental illness. High scoresindicate greater knowledge and morepositive attitudes.

Children’s social distance (a = .89) isa 6-item measure gauging the extentto which youth are unwilling tointeract with someone who isidentified as having a mental illness,including as a lunchmate, friend, or

PEDIATRICS Volume 145, number 6, June 2020 3 by guest on October 27, 2021www.aappublications.org/newsDownloaded from

collaborator on a school project.16

High scores indicate a greater desirefor social distance.

Mental health treatment seeking wasassessed by asking whether youthhad taken medicine for a mentalhealth problem or talked toa therapist or counselor abouta mental health problem (coded “1” ifeither or both, coded “0” if neither).

A self-reported mental healthsymptoms checklist (a = .87) uses 21stem questions from the NationalInstitute of Mental Health DiagnosticInterview Schedule for Children(DISC), Version IV27 to identifyyouth with a high probability ofneeding professional mental healthtreatment (Supplemental Table 11).Youth endorsing two-thirds ormore of the symptoms either atthe beginning of the study (averageof pretest and post-test scores)or at the end (average of the 18-and 24-month scores) werecategorized as having highprobability of need; 18.5% of youthmet these criteria.

Control variables are sex, age, a 6-item scale of social desirabilitybias,28,29 self-identified race and/orethnicity, parent or guardianeducational attainment, and theprimary language spoken at home.

Analysis

We assessed clustering of (1) youthwithin classrooms and (2) follow-upassessments within youth usingintraclass correlation coefficients.Clustering by classroom was modestwith intraclass correlationcoefficients for each outcome ,10%(knowledge and attitudes 0.094,social distance 0.081, and treatmentseeking 0.019). Clustering ofoccasions within youth wassubstantial and is addressed by usinggeneralized estimating equations(GEEs).30 Because we foundpreintervention differences betweencluster randomized interventiongroups on some variables(Supplemental Table 12), we controlfor preintervention values of thedependent variable and otherpotentially important covariates.

In analyses of knowledge andattitudes and social distance, weshow effects for each of theintervention groups separately andtest for interactions betweeninterventions in our fully crosseddesign. If no significant interactionsbetween interventions were found,only the main effects of eachintervention with dummy codes (1 =receiving intervention; 0 = notreceiving intervention) are presented.For mental health treatment seeking,for which the number of cases in the

relevant high-symptom group isrelatively small (n = 77), we simplycompare youth receiving and notreceiving the curriculum.

We also assess whether interventioneffects persist by examininginteractions between interventionand time period. Finally, weexplore whether interventions aresignificantly more or less effective forgroups, defined by preinterventionattitudes and sociodemographicfactors.

Missing data are addressed by usingmultiple imputation via chainedequations in Stata 15.1 (Stata Corp,College Station, TX).31 Twenty-fivedata sets were imputed andrecombined by using Rubin’s rules.32

Results presented imputed values forcovariates but not dependentvariables. Sensitivity analyses thatused (1) complete cases and (2) animputation of all variables, includingdependent variables, resulted insimilar conclusions (SupplementalTables 13 through 15).

RESULTS

Does the Curriculum InterventionImprove Knowledge and Attitudesand Reduce Social Distance?

We found no significant 3- or 2-wayinteractions between interventionsand therefore present results

TABLE 1 Multiple Linear Regression (GEEs) Showing the Effect of Preintervention Levels of the Outcome Variable, Time Period and Curriculum, Contact,and Materials Interventions on Knowledge and Attitudes and Social Distance: Regression Coefficients and Robust SEs (n = 416 Youth; n = 1432Observations for Knowledge and Attitudes; n = 1428 Observations for Social Distance)

Variables Knowledge and/or Attitudes Social Distance

Equation 1 Unadjusted Equation 2 Adjusteda Equation 1 Unadjusted Equation 2 Adjusteda

Preintervention knowledge and attitudes 0.514*** (0.037) 0.502*** (0.037) — —

Preintervention social distance — — 0.483*** (0.037) 0.473*** (0.037)Time period (6 mo is reference)12 mo 0.006 (0.015) 0.006 (0.015) 20.041 (0.023) 20.041 (0.023)18 mo 20.002 (0.016) 20.002 (0.017) 20.003 (0.029) 20.003 (0.029)24 mo 0.044* (0.020) 0.044* (0.020) 20.103 (0.033) 20.014 (0.033)

Curriculum (1), all other interventions and no interventions (0) 0.139*** (0.030) 0.103*** (0.031) 20.123* (0.049) 20.103* (0.051)Contact (1), all other interventions and no interventions (0) 0.002 (0.029) 0.005 (0.028) 0.052 (0.048) 0.049 (0.048)Material (1), all other interventions and no interventions 20.001 (028) 0.016 (0.028) 20.029 (0.050) 20.044 (0.052)

—, not applicable.a Adjusted for child age, sex, race and/or ethnicity, social desirability score, caregiver education, and whether English was the main language spoken in the home.* P , .05.*** P , .001.

4 LINK et al by guest on October 27, 2021www.aappublications.org/newsDownloaded from

showing the main effects. Table 1,Equation 1 shows the effect ofeach intervention whilecontrolling for time period andpreintervention values of thedependent variable. Neitherthe contact nor the materialsintervention was significant foreither dependent variable.Exposure to the curriculumintervention is associated witha significant increase inknowledge and attitudes (b = 0.139;P , .001; Cohen’s d 0.35) anda significant decrease in socialdistance (b = 20.123; P , .05;Cohen’s d 0.16). Equation 2shows that the curriculumremains significantly associatedwith both knowledge andattitudes and social distance whencontrolling for covariates. Figure 1shows means adjusted forcovariates calculated for eachdependent variable at each timeperiod and indicates that thecurriculum intervention is bothbeneficial (higher knowledgeand attitudes and lower socialdistance) and enduring in itseffect over the study period.

Do Curriculum Effects on Knowledgeand Attitudes and Social DistanceVary by Youth Attributes?

We assessed interactions betweenassignment to the curriculum andpreexisting attitudes, socialdesirability bias, sex, age, race and/orethnicity, and parent or guardianeducational attainment on knowledgeand attitudes and social distance. Ofthe 16 interactions examined, 1 wassignificant: Latino youth did notexperience curriculum-associatedreductions in social distance(Supplemental Fig 4). We found nosignificant interaction between thecurriculum and race and/or ethnicityon knowledge and attitudes ortreatment seeking and no significantinteraction between the curriculumand any of the other variablesexamined.

Does the Curriculum InterventionLead Youth With High SymptomLevels to Seek Treatment?

Table 2 compares youth with highsymptom levels assigned to thecurriculum with youth with highsymptom levels who were notassigned to the curriculum. Asexpected, preintervention mentalhealth treatment seeking issignificantly associated withtreatment seeking postintervention.Importantly, whether before addingcovariates (odds ratio [OR] = 3.51;confidence interval [CI] = 1.08–11.39)

or after doing so (OR = 3.90; CI =1.09–13.87), assignment to thecurriculum is associated withincreased odds of mental healthtreatment seeking. Becausepostintervention treatment seekingcould represent first-ever treatmentseeking, a continuation of treatmentseeking, or return to treatmentseeking, supplementary analysesexamined curriculum effects for theseoutcomes separately (SupplementalTable 16) and found ORs that weresimilar in magnitude to thosereported above but that were notstatistically significant because of

FIGURE 1Mental illness stigma outcome variables: adjusted means for intervention groups at 6-, 12-, 18-, and24-month follow-up postintervention in the Texas Stigma Study (n = 416). Means are adjusted forpreintervention knowledge and attitudes, child age, sex, race and/or ethnicity, social desirabilityscore, parent or guardian educational attainment, and English as the primary language spoken inthe home. A, Knowledge and positive attitudes outcome. B, Social distance outcome.

PEDIATRICS Volume 145, number 6, June 2020 5 by guest on October 27, 2021www.aappublications.org/newsDownloaded from

smaller sample sizes. We found noevidence of interaction betweencurriculum assignment and timeperiod. Figure 2 shows predictedprobabilities of treatment seeking,with youth with high symptom levelsassigned to curriculum being muchmore likely to experience treatmentseeking than youth with highsymptom levels assigned tocomparator interventions or nointervention.

In contrast, we found no significanteffect of the curriculum interventionon treatment seeking in youth withfewer symptoms. A test of theinteraction (not shown) revealed thatthe effect of curriculum assignmentwas significantly different in the high-as opposed to the low-symptomgroup. Figure 2 shows predictedprobabilities in the lower-symptomgroup and indicates no effect of thecurriculum in relation to comparatorinterventions and the no-interventioncontrol.

DISCUSSION

We examined whether a school-basedantistigma curriculum interventiontargeted early in the life course mightimprove the context in which mentalhealth problems are experienced,increasing knowledge, improving

attitudes, and inducing treatmentseeking when mental health problemsare present. The results generallysupport an impact of the ESDcurriculum in achieving these goals.Youth assigned to the curriculumintervention experienced significantlyincreased knowledge and morepositive attitudes toward mentalillness and reported significantlyreduced social distance from youthwith mental illness. Because ourstudy is unique in that it is a clusterrandomized design amongpreadolescent youth with 2-yearfollow-up, we cannot make directcomparisons to effect sizes achievedin other studies. However, a meta-analysis of mental health stigmainterventions among adults andadolescents (not preadolescents)reported average effect sizes forrandomized studies to be smallerthan ours (0.21 for attitudes andknowledge and 0.10 for socialdistance).33 Additionally,participants in the ESDintervention were substantiallymore likely to seek treatment formental health problems if theyexperienced them than were youthin other intervention groups orcontrol group, a finding that, toour knowledge, has not previouslybeen addressed in stigma-intervention research.

A unique aspect of our design was the2-year follow-up, which revealed thatintervention effects did notappreciably dissipate over thatperiod. Additionally, with 1 exception(Latino youth with respect to socialdistance), there were no significantdifferences in the effectiveness of thecurriculum intervention by race and/or ethnicity, sex, educational level ofcaregivers, or youth preinterventionattitudes.

We found no effect of the contactintervention, although contact hasoften been shown to be the mosteffective way to change attitudes. Apotential explanation is that contact isnot as effective in youth, a possibilitythat is supported by a meta-analysisshowing diminished effects ofcontact compared with educationalinterventions in adolescents.33

Our study is limited by its use of self-reported data for both attitudes andtreatment-seeking behaviors. Our useof self- rather than interviewer-administered survey instruments andour ability to control for socialdesirability bias somewhat mitigateour concerns about reporting bias.Because treatment seeking is self-reported by the youth participating,we were not able to acquire detailsabout the type or adequacy oftreatment. Loss to follow-up, which isa common problem for longitudinalstudies, is another limitation.Somewhat reassuring was thefinding of no significant differencesbetween those successfullyfollowed and those lost to follow-up, the correspondence betweenour achieved sample and thepopulation from which they weredrawn in terms of race and/orethnicity and socioeconomic status,and our multiple sensitivityanalyses implemented to addressmissing values. Despite matchingschools on test scores, our clusterrandomization of classroomsresulted in preinterventiondifferences between groups onsome baseline characteristics,

TABLE 2 Logistic Regression (GEE) Showing the Effect of the ESD Curriculum Intervention onTreatment Seeking for Mental Health Problems Among Youth With High Symptom Levels(n = 77 Youth; n = 275 Observations)

Variables Mental Health Help Seeking (Therapist or TakingMedication = 1, All Others = 0), OR (95% CI)

Equation 1Unadjusted

Equation 2 Adjusted forCovariatesa

Preintervention mental health help seeking 7.49*** (2.20–25.43) 10.28*** (1.13–1.50)Time period (6 mo is reference)12 mo 1.64 (0.93–2.88) 1.77 (0.91–3.44)18 mo 1.10 (0.55–2.18) 1.12 (0.51–2.45)24 mo 1.27 (0.60–2.69) 1.34 (0.57–3.16)

Curriculum (1), all other interventions and nointervention (0)

3.51* (1.08–11.39) 3.90* (1.09–13.87)

a Adjusted for child age, sex, race and/or ethnicity, social desirability score, caregiver education, and whether English wasthe main language spoken in the home.* P , .05.** P , .01.*** P , .001.

6 LINK et al by guest on October 27, 2021www.aappublications.org/newsDownloaded from

leading us to adjust for covariatesand preintervention measures ofdependent variables. Althoughdiverse in terms of race and/orethnicity and parental education, ourstudy is limited in generalizability

because it was conducted in 1 schooldistrict in Texas.

Our study joins a handful of othersthat have intervened withpreadolescent students and shown

positive effects on knowledge andattitudes. We added a substantiallylonger follow-up and includedmeasures of treatment-seekingbehaviors in an intervention designedto be relatively easily disseminated.Our study, in combination with otherstudies, suggests strongly that youthcan be positively influenced ata relatively young age, fosteringchanges in mental health attitudesand behaviors that last, as our studyhas shown, for at least 2 years.Although we cannot know from thedata in hand, we might expect that ifinterventions were delivered annuallyin every sixth-grade class, the contextexperienced by people with mentalillness could improve even more,a possibility that future researchmight examine. We do know thatnegative attitudes toward mentalillnesses and the exceptionally largepercentage of people who experiencebut do not receive treatment for suchillnesses are problems that have beenwith us for a long time. Interventionssuch as ESD represent a partial butpositive response to this publicmental health challenge.

ABBREVIATIONS

CI: confidence intervalDISC: Diagnostic Interview

Schedule for ChildrenESD: Eliminating the Stigma of

DifferencesGEE: generalized estimating

equationOR: odds ratio

This trial has been registered at www.clinicaltrials.gov (identifier NCT03597048).

Deidentified individuals’ participant data will not be made available.

DOI: https://doi.org/10.1542/peds.2019-0780

Accepted for publication Dec 3, 2019

Address correspondence to Bruce G. Link, PhD, School of Public Policy, University of California, Riverside, Interdisciplinary South 4133, 900 University Ave, Riverside,

CA 92521. E-mail: bruce.link@ucr.edu

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2020 by the American Academy of Pediatrics

FIGURE 2Predicted probabilities of mental health treatment seeking at 6-, 12-, 18- and 24-month follow-upamong youth with high and low symptoms in the Texas Stigma Study. Predicted probabilitiesadjusted for preintervention mental health treatment, child age, sex, race/ethnicity, social de-sirability score, parent/guardian educational attainment and English as primary language spoken inthe home. Results show probabilities with covariates set at mean values. Curriculum versuscomparator and no intervention P , .05. A, Youth with high symptom levels (n = 77). B, Youth withlow symptom levels (n = 336).

PEDIATRICS Volume 145, number 6, June 2020 7 by guest on October 27, 2021www.aappublications.org/newsDownloaded from

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

FUNDING: Supported by the National Institute of Mental Health (grant 1 R01 MH095254-01; to Drs Link and Painter [coprincipal investigators]). Funded by the

National Institutes of Health (NIH).

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

REFERENCES

1. Link BG, Stuart H. On Revisiting SomeOrigins of the Stigma Concept as ItApplies to Mental Illnesses. In: Gaebel W,Rössler W, Sartorius N, eds. The Stigmaof Mental Illness-End of the Story? NewYork, NY: Springer International; 2017:3–28

2. Corrigan PW, Watson AC. Understandingthe impact of stigma on people withmental illness. World Psychiatry. 2002;1(1):16–20

3. Link BG. Understanding labeling effectsin the area of mental disorders: anassessment of the effects ofexpectations of rejection. Am Sociol Rev.1987;52(1):96–112

4. Pescosolido BA, Martin JK, Long JS,Medina TR, Phelan JC, Link BG. “Adisease like any other”? A decade ofchange in public reactions toschizophrenia, depression, and alcoholdependence. Am J Psychiatry. 2010;167(11):1321–1330

5. Yanos PT. Written-Off: Mental HealthStigma and the Loss of HumanPotential. Cambridge, United Kingdom:Cambridge University Press; 2018

6. Avenevoli S, Swendsen J, He JP, BursteinM, Merikangas KR. Major depression inthe national comorbidity survey-adolescent supplement: prevalence,correlates, and treatment. J Am AcadChild Adolesc Psychiatry. 2015;54(1):37–44.e2

7. Merikangas KR, He JP, Burstein M, et al.Service utilization for lifetime mentaldisorders in US adolescents: results ofthe National Comorbidity Survey-Adolescent Supplement (NCS-A). J AmAcad Child Adolesc Psychiatry. 2011;50(1):32–45

8. DuPont-Reyes MJ, Villatoro AP, PhelanJC, Painter K, Link BG. Patterns ofaggressive behaviors across mentalhealth in sixth graders. J InterpersViolence. 2018;886260518793991

9. Leary MR, Kowalski RM, Smith L, PhillipsS. Teasing, rejection, and violence: case

studies of the school shootings.Aggress Behav. 2003;29(3):202–214

10. Stone DM, Simon TR, Fowler KA, et al.Vital signs: trends in state suicide rates- United States, 1999–2016 andcircumstances contributing to suicide -27 states, 2015. MMWR Morb MortalWkly Rep. 2018;67(22):617–624

11. Green JG, McLaughlin KA, Alegría M,et al. School mental health resourcesand adolescent mental health serviceuse. J Am Acad Child AdolescPsychiatry. 2013;52(5):501–510

12. Ndetei DM, Mutiso V, Maraj A, AndersonKK, Musyimi C, McKenzie K. Stigmatizingattitudes toward mental illness amongprimary school children in Kenya. SocPsychiatry Psychiatr Epidemiol. 2016;51(1):73–80

13. Wahl OE. Children’s views of mentalillness: a review of the literature.Psychiatr Rehabil Skills. 2002;6(2):134–158

14. Wahl O, Susin J, Lax A, Kaplan L, ZatinaD. Knowledge and attitudes aboutmental illness: a survey of middleschool students. Psychiatr Serv. 2012;63(7):649–654

15. Watson AC, Otey E, Westbrook AL, et al.Changing middle schoolers’ attitudesabout mental illness through education.Schizophr Bull. 2004;30(3):563–572

16. Wahl OF, Susin J, Kaplan L, Lax A, ZatinaD. Changing knowledge and attitudeswith a middle school mental healtheducation curriculum. Stigma ResAction. 2011;1(1):44–53

17. Desocio J, Stember L, Schrinsky J.Teaching children about mental healthand illness: a school nurse healtheducation program. J Sch Nurs. 2006;22(2):81–86

18. Pitre N, Stewart S, Adams S, et al. Theuse of puppets with elementary schoolchildren in reducing stigmatizingattitudes toward mental illness. J MentHealth. 2007;16(3):415–429

19. Spagnolo AB. Examining the Effects ofAnti-stigma Messages on the Attitudesof Early Adolescents [doctoraldissertation]. Newark, NJ: University ofMedicine and Dentistry of New Jersey;2009

20. Ventieri D, Clarke DM, Hay M. Theeffects of a school-based educationalintervention on preadolescents’knowledge of and attitudes towardmental illness. Adv Sch Ment HealthPromot. 2011;4(3):5–17

21. Mellor C. School-based interventionstargeting stigma of mental illness:systematic review. Psychiatr Bull(2014). 2014;38(4):164–171

22. Salerno JP. Effectiveness of universalschool‐based mental health awarenessprograms among youth in the UnitedStates: a systematic review. J SchHealth. 2016;86(12):922–931

23. Painter K, Phelan JC, DuPont-Reyes MJ,Barkin KF, Villatoro AP, Link BG.Evaluation of antistigma interventionswith sixth-grade students: a school-based field experiment. Psychiatr Serv.2017;68(4):345–352

24. Arlington Independent School District.Student data reports. Available at:https://www.aisd.net/district/about/student-data/. Accessed July 8, 2019

25. US Census Bureau. Selectedcharacteristics of the native andforeign-born populations, ArlingtonTexas. American Community Survey2012: 5-year Estimates. Available at:https://data.census.gov/cedsci/table?q=Arlington%20Texas%202012&g=1600000US4804000&hidePreview=false&table=S0501&tid=ACSST5Y2012.S0501&y=2012&vintage=2012&layer=place&cid=DP05_0001E&lastDisplayedRow=60&mode=. Accessed January20, 2020

26. Reinke RR, Corrigan PW, Leonhard C,Lundin RK, Kubiak MA. Examining twoaspects of contact on the stigma ofmental illness. J Soc Clin Psychol. 2004;23(3):377–389

8 LINK et al by guest on October 27, 2021www.aappublications.org/newsDownloaded from

27. Shaffer D, Fisher P, Lucas CP, DulcanMK, Schwab-Stone ME. NIMHDiagnostic Interview Schedule forChildren Version IV (NIMH DISC-IV):description, differences fromprevious versions, and reliability ofsome common diagnoses. J Am AcadChild Adolesc Psychiatry. 2000;39(1):28–38

28. Baxter SD, Smith AF, Litaker MS, et al.Recency affects reporting accuracy of

children’s dietary recalls. AnnEpidemiol. 2004;14(6):385–390

29. Crandell VC, Crandell VJ, Kratkovsky W.A child’s social desirability scale.J Consult Clin Psychol. 1965;29(1):27–36

30. Liang KY, Zeger SL. Longitudinal dataanalysis using generalized linearmodels. Biometrika. 1986;73(1):13–22

31. Azur MJ, Stuart EA, Frangakis C, Leaf PJ.Multiple imputation by chained

equations: what is it and how does itwork? Int J Methods Psychiatr Res.2011;20(1):40–49

32. Rubin DB. Multiple Imputation forNonresponse in Surveys. New York, NY:John Wiley & Sons; 2004

33. Corrigan PW, Morris SB, Michaels PJ,Rafacz JD, Rüsch N. Challenging thepublic stigma of mental illness: a meta-analysis of outcome studies. PsychiatrServ. 2012;63(10):963–973

PEDIATRICS Volume 145, number 6, June 2020 9 by guest on October 27, 2021www.aappublications.org/newsDownloaded from

DOI: 10.1542/peds.2019-0780 originally published online May 20, 2020; 2020;145;Pediatrics 

and Kris PainterBruce G. Link, Melissa J. DuPont-Reyes, Kay Barkin, Alice P. Villatoro, Jo C. Phelan

TrialA School-Based Intervention for Mental Illness Stigma: A Cluster Randomized

ServicesUpdated Information &

http://pediatrics.aappublications.org/content/145/6/e20190780including high resolution figures, can be found at:

Referenceshttp://pediatrics.aappublications.org/content/145/6/e20190780#BIBLThis article cites 26 articles, 1 of which you can access for free at:

Subspecialty Collections

al_issues_subhttp://www.aappublications.org/cgi/collection/development:behaviorDevelopmental/Behavioral Pediatricsfollowing collection(s): This article, along with others on similar topics, appears in the

Permissions & Licensing

http://www.aappublications.org/site/misc/Permissions.xhtmlin its entirety can be found online at: Information about reproducing this article in parts (figures, tables) or

Reprintshttp://www.aappublications.org/site/misc/reprints.xhtmlInformation about ordering reprints can be found online:

by guest on October 27, 2021www.aappublications.org/newsDownloaded from

DOI: 10.1542/peds.2019-0780 originally published online May 20, 2020; 2020;145;Pediatrics 

and Kris PainterBruce G. Link, Melissa J. DuPont-Reyes, Kay Barkin, Alice P. Villatoro, Jo C. Phelan

TrialA School-Based Intervention for Mental Illness Stigma: A Cluster Randomized

http://pediatrics.aappublications.org/content/145/6/e20190780located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://pediatrics.aappublications.org/content/suppl/2020/05/19/peds.2019-0780.DCSupplementalData Supplement at:

by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397. the American Academy of Pediatrics, 345 Park Avenue, Itasca, Illinois, 60143. Copyright © 2020has been published continuously since 1948. Pediatrics is owned, published, and trademarked by Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it

by guest on October 27, 2021www.aappublications.org/newsDownloaded from