suicidal ideation and behaviors among youths in juvenile detention

Copyright @ 2008 American Academy of Child and Adolescent Psychiatry. Unauthorized reproduction of this article is prohibited.

Suicidal Ideation and Behaviors Among Youthsin Juvenile Detention

KAREN M. ABRAM, PH.D., JEANNE Y. CHOE, B.A., JASON J. WASHBURN, PH.D., A.B.P.P.,LINDA A. TEPLIN, PH.D., DEVON C. KING, PH.D., AND MINA K. DULCAN, M.D.

ABSTRACT

Objective: To examine suicidal ideation, suicide attempts, lethality of suicide attempts, and the relationship between

psychiatric disorder and recent attempts in newly detained juveniles. Method: The sample included 1,829 juveniles, ages

10 to 18 years, sampled after intake to a detention center in Chicago. Interviewers administered the Diagnostic Interview

Schedule for Children to assess for thoughts of death, suicidal ideation, suicide plans, lifetime suicide attempts, number of

attempts, age at first attempt, attempts within the past 6 months, method of suicide attempts, and psychiatric disorder.

Results: More than one third of juvenile detainees and nearly half of females had felt hopeless or thought about death in

the 6 months before detention. Approximately 1 in 10 (10.3%, 95% confidence interval: 7.7%Y12.8%) juvenile detainees

had thought about committing suicide in the past 6 months, and 1 in 10 (11.0%, 95% confidence interval: 8.3%Y13.7%) had

ever attempted suicide. Recent suicide attempts were most prevalent in females and youths with major depression and

generalized anxiety disorder. Conclusions: Fewer than half of detainees with recent thoughts of suicide had told anyone

about their ideation. Identifying youths at risk for suicide, especially those suffering from depressive and anxiety disorders,

is a crucial step in preventing suicide. J. Am. Acad. Child Adolesc. Psychiatry, 2008;47(3):291Y300. Key Words: juvenile

detainee, suicidal ideation, suicide, psychiatric disorder.

Suicide is the third leading cause of death in youngpeople ages 15 to 24 years, affecting 9.5/100,000adolescents in 2003.1 Suicide among youths has nearlydoubled since 1950, increasing at a faster rate thanamong groups 25 years and older.2 Suicide is an evengreater risk in incarcerated youths; available nationaldata suggest that prevalence rates of completed suicideare between two and four times higher among youths in

custody than among youths in the community.3,4

Incarcerated youths have characteristics commonlyassociated with increased risk for suicide,5 such as highrates of psychiatric disorder6 and trauma.7,8 Conditionsassociated with confinement, such as separation fromloved ones,9 crowding,10 sleeping in locked rooms,4 andsolitary confinement,10,11 may also increase the risk forsuicide.

Accepted September 22, 2007.Drs. Abram, Washburn, Teplin, and Dulcan and Ms. Choe are with the

Department of Psychiatry and Behavioral Sciences, Psycho-Legal StudiesProgram, Northwestern University Feinberg School of Medicine; Dr. King is acontractor with NIMH.

This work was supported by National Institute of Mental Health grantsR01MH54197 and R01MH59463 (Division of Services and InterventionResearch and Center for Mental Health Research on AIDS) and grants 1999-JE-FX-1001 and 2005-JL-FX-0288 from the Office of Juvenile Justice andDelinquency Prevention. Major funding was also provided by the NationalInstitute on Drug Abuse, the Substance Abuse and Mental Health ServicesAdministration (Center for Mental Health Services, Center for SubstanceAbuse Prevention, Center for Substance Abuse Treatment), the NIH Center onMinority Health and Health Disparities, the Centers for Disease Control andPrevention (National Center on Injury Prevention and Control and NationalCenter for HIV, STD and TB Prevention), the National Institute on AlcoholAbuse and Alcoholism, the NIH Office of Research on Women`s Health, the

NIH Office on Rare Diseases, Department of Labor, the William T. GrantFoundation, and The Robert Wood Johnson Foundation. Additional funds wereprovided by The John D. and Catherine T. MacArthur Foundation, the Open SocietyInstitute, and the Chicago Community Trust. We thank Ann Hohmann, Ph.D.,Kimberly Hoagwood, Ph.D., and Heather Ringeisen, Ph.D., for indispensable advice,and Grayson Norquist, M.D., and Delores Parron, Ph.D., for their support. CeliaFisher, Ph.D., guided our human subject procedures. We thank project staff, especiallyAmy Mericle, Ph.D., Lynda Carey, M.A., and our field interviewers. Without thecooperation of the Cook County and State of Illinois systems, this study would not havebeen possible. Finally, we thank the participants for their time and willingness toparticipate.

Correspondence to Dr. Karen M. Abram, Northwestern University, FeinbergSchool of Medicine, 710 N. Lake Shore Drive, Suite 900, Chicago, IL 60611;e-mail: [email protected].

0890-8567/08/4703-0291�2008 by the American Academy of Child andAdolescent Psychiatry.

DOI: 10.1097/CHI.0b013e318160bce

WWW.JAACAP.COM 291J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 47:3, MARCH 2008


It is important to study youths immediately afterarrest and detention. Detention provides the firstopportunity in the juvenile justice system to system-atically screen youths for risk for suicide. Screening forcurrent and previous suicidal ideation and behavior iscritical for prevention. A recent national study of 79suicides among incarcerated and detained youths foundthat more than two thirds of the suicide victims hadmade previous attempts, reported suicidal ideation,made suicidal threats, or physically harmed themselves.5

Although studies have examined suicidal behaviors inyouths in long-term correctional facilities,12,13 youths inresidential facilities,14 and youths formally processed bythe juvenile justice system,15 few large-scale investiga-tions have examined detained youths in the UnitedStates. Findings vary widely. Prevalence rates of currentsuicidal ideation vary from 14.2% to 51%.8,16Y20 Racialand ethnic differences in suicidal ideation also varyacross studies; some reported higher prevalence ratesof ideation in non-Hispanic whites than AfricanAmericans and Hispanics,16,19 whereas others reportedno racial/ethnic differences.18,20,21 Similarly, somestudies reported higher prevalence rates of ideation infemales than males,16,18,19 whereas others found no sexdifferences.21,22

The variation in prevalence rates across studies ofdetained youths is likely due to differences in samplingand measurement. The largest study of detained youthsexamined 18,607 admissions to detention, not indivi-duals; youths may have been admitted more than once.16

The largest study of individual detainees sampled acombination of 451 youths held in detention and 1,350youths incarcerated in long-term facilities.19 However,findings were not reported by subsample, and combiningthe subsamples is problematic because youths in detentionand youths in prison have different patterns of suicidalbehavior.5 For example, in detained youths, 40% ofcompleted suicides occur within 3 days of admission. Incontrast, in youths in long-term facilities, more than 72%of completed suicides occur after 3 months.5

Variation in prevalence rates is also due to differencesinmeasurement. Although all of the previous studies usedquestionnaires to assess suicidal ideation and behavior,they used a variety of self-administered8,16,17,20Y22 andgroup-administered measures.19

There are several key omissions in the literature. First,many of the samples were too small or homogeneous toexamine differences by race/ethnicity and sex.17,18,21

Examining racial/ethnic and sex differences is importantbecause suicidal ideation and attempts vary by thesecharacteristics in the general population.23Y26 Under-standing demographic differences in suicide risk indetained youths also helps to identify culturally relevantand gender-specific interventions that are most needed.Second, only three studies examined prevalence rates

of suicide attempts in juvenile detainees18,19,22; pastsuicide attempts are powerful predictors of futureattempts.5 Moreover, due to methodological differencesin the studies, these rates vary widely. Rohde et al.27

reported a lifetime attempt rate of 19.4%, whereasMorris et al.19 reported a past-year attempt rate of15.5%. Esposito and Clum22 reported lifetime, past-year, and past-month attempt rates of 33%, 29%, and26%, respectively.Finally, to date, few studies of detained youths have

examined the relationship between psychiatric diagnosisand suicide risk. It is important to determine whichyouths are most at risk for suicide. Psychopathology isconsistently linked with risk for suicide in adolescents28

and is prevalent in detainees.6,29 Depression,15,27

anxiety,27 substance use,15,19 and behavioral distur-bances (in boys)27 have been identified as correlates ofpast suicide attempts in youths involved in the justicesystem. The few studies that examined psychiatricdisorders and suicide in detained youths relied on smallsamples27 or examined a limited range of disorders.18,19

To our knowledge, this is the first large-scaleepidemiological study of detained youths in the UnitedStates that uses a comprehensive, standardized interviewto assess suicidal ideation, suicide attempts, the lethalityof suicide attempts, and a wide range of psychiatricdisorders. Using data from the Northwestern JuvenileProject, we examine prevalence rates of suicidal ideationand behaviors, the relationship between recent suicideattempts and psychiatric disorders, and differences bysex and race/ethnicity.

METHOD

Participants and Sampling Procedures

Participants were part of the Northwestern Juvenile Project, alongitudinal study of 1,829 youths (ages 10Y18 years) detainedbetween 1995 and 1998 at the Cook County Juvenile TemporaryDetention Center (CCJTDC) in Chicago. The randomly selectedsample was stratified by sex, race/ethnicity (African American, non-Hispanic white, Hispanic), age (10Y13 years or 14 years and older),and legal status (processed as a juvenile or as an adult) to obtain

ABRAM ET AL.

292 WWW.JAACAP.COM J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 47:3, MARCH 2008


enough participants to examine key subgroups (e.g., females,Hispanics, younger children).During data collection, the CCJTDC received approximately

8,500 annual admissions (John Howard Association, unpublisheddata, 1992). The CCJTDC is used solely for pretrial detention andsentences <30 days. It houses detainees younger than 17 years,although youths up to age 21 years are detained at CCJTDC ifprosecuted for an arrest that occurred when they were younger than17 years. Like juvenile detainees nationwide,30 approximately 90%of the CCJTDC detainees are male, and most are racial/ethnicminorities (77.9% African American, 5.6% non-Hispanic white,16.0% Hispanic, and 0.5% from other racial/ethnic groups). Theage and offense distributions of the CCJTDC detainees are similar tothose of detained juveniles nationwide.31

We chose CCJTDC, which includes Chicago and surroundingsuburbs, for three reasons. First, nationwide, most juvenile detaineeslive in and are detained in urban areas.32 Second, Cook County isethnically diverse and has the third largest Hispanic population ofany county in the United States.33 Third, the CCJTDC`s sizeensured that enough participants would be available.No single site can represent the entire country because

jurisdictions differ in their options for diversion. Nevertheless,Illinois` criteria for detaining juveniles are similar to those of otherstates. Detainees were eligible to participate regardless of psychiatricmorbidity, alcohol or other drug intoxication, or fitness to stand trial.Within each stratum, names were selected using a random-numberstable. The final sampling fractions ranged from 0.018 to 0.689.Project staff explained the project to participants in their units and

assured them that anything they told us (except acute suicidal orhomicidal risk) would be confidential. Data are protected by aFederal Certificate of Confidentiality and Title 28 Code of FederalRegulations, Part 22. Participants signed an assent form or consentform, depending on their age. The Northwestern InstitutionalReview Board, the Centers for Disease Control and PreventionInstitutional Review Board, and the U.S. Office of Protection fromResearch Risks waived parental consent, consistent with federalregulations. We nevertheless tried to contact parents; however,despite repeated attempts, none could be found for 43.8% of theparticipants. In lieu of parental consent, youth assent was overseenby a participant advocate who represented the interests of theparticipants.Of the 2,275 names selected, 4.2% refused to participate. There

were no significant differences in refusal rates by sex, race/ethnicity,or age. Some youths processed as adults were counseled by theirlawyers to refuse participation; in this stratum, the refusal rate was7.1%. Twenty-seven youths left the detention center before wecould schedule an interview; 312 were not interviewed because theyleft while we were attempting to locate their caregivers. Eleven otherswere excluded: nine became physically ill and could not finish theinterview, one was too cognitively impaired to be interviewed, andone appeared to be lying. The final sample was 1,829. This samplesize allows us to reliably detect disorders that have a base rate in thepopulation of Q1.0% with a power of 0.80.34

The final sample comprised 1,172 males (64.1%) and 657 females(35.9%), 1,005 African Americans (54.9%), 296 non-Hispanicwhites (16.2%), 524 Hispanics (28.7%), and 4 from other racial/ethnic groups (0.2%). The mean age of participants was 14.9 years,and the median age was 15 years.Participants were interviewed for 2 to 3 hours in a private area,

almost always within 2 days of intake. Female participants wereinterviewed only by female interviewers. Interviewers were trainedfor at least 1 month; most had a master`s degree in psychology or an

associated field and experience interviewing high-risk youths. Onethird of our interviewers were fluent in Spanish. We maintainedinterviewer consistency throughout the study by monitoring scriptedinterviews with mock participants. Additional information on ourmethods can be found in the article by Teplin et al.6

Measures

We used version 2.3 of the Diagnostic Interview Schedule forChildren (DISC),35,36 the most recent English and Spanish versionsthen available. The DISC 2.3 assesses the presence of the followingDSM-III-R disorders in the past 6 months: major depressive episode,dysthymia, mania, hypomania, panic, separation anxiety, over-anxious, generalized anxiety, obsessive-compulsive, schizophrenia,attention-deficit/hyperactivity, oppositional defiant, conduct, mar-ijuana use, alcohol use, and other drug use. The DISC 2.3 hasspecific questions that assess thoughts of death, suicidal ideation,suicide plans, lifetime suicide attempts, number of suicide attempts,age at first suicide attempt, suicide attempts within the past 6months, and method of suicide attempts. For analyses includingpsychiatric diagnoses, we excluded items related to suicidal ideationand behavior from the diagnostic algorithms for major depression(major depression, modified) and dysthymia (dysthymia, modified)to avoid inflating the relationship between them.The DISC 2.3 is highly structured, contains detailed symptom

probes, and has acceptable reliability and validity.35,37Y40 Addi-tional information on our use of the DISC 2.3 has been publishedelsewhere.6,41

Statistical Analysis

Because selected strata were oversampled, we used sample weights,based on the CCJTDC population, to estimate descriptive statisticsand model parameters that reflect the CCJTDC population.Weighted analyses were conducted using Stata, version 9.0. Taylorseries linearization was used to estimate SEs.42,43 Logistic regressionwas used to assess demographic differences in the prevalence rates ofsuicidal ideation and behaviors and for the predictive models.

RESULTS

Table 1 shows the prevalence rates of suicidal ideationand behavior by sex and race/ethnicity. Results aresummarized by type of suicidal ideation and behavior.

Hopelessness

More than one third of the sample had ever felt thatlife was hopeless. Significantly more females than malesever felt hopeless (odds ratio [OR] 1.43, 95%confidence interval [CI] 1.10Y1.86).

Thoughts About Death and Dying

More than one third of the sample had thought morethan usual in the past 6 months about death and dying.Among males, significantly more Hispanics (OR 1.97,95% CI 1.26Y3.09) and African Americans (OR 1.64,95% CI 1.09Y2.47) than non-Hispanic whites had


SUICIDALITY AMONG JUVENILE DETAINEES

Copyright @

2008 Am

erican Academ

y of Child and A

dolescent Psychiatry. U

nauthorized reproduction of this article is prohibited.

TABLE 1Prevalence of Suicidal Ideation and Behavior by Sex and Race/Ethnicity for Juvenile Detainees

SuicidalIdeationand Behavior

Totala

(N = 1,826)(%)

Females Males

TestsofSex

Totalb

(n = 656)(%)

AA(n = 430)

(%)

W(n = 89)(%)

H(n = 136)

(%)

Tests ofRace/

Ethnicity

SpecificTests ofRace/

Ethnicity

Totalc

(n = 1,170)(%)

AA(n = 574)

(%)

W(n = 207)

(%)

H(n = 386)

(%)

Tests ofRace/

Ethnicity


Ethnicity

Ever felt lifewas hopeless

36.2 44.2 40.7 45.6 52.2 ns 35.6 34.8 42.6 38.2 ns p < .01,F 9 M

Thought a lotabout deathor dying inpast 6 mo

35.4 31.5 32.0 26.7 35.4 ns 35.7 35.5 25.1 39.8 p < .01 AA 9 W;H 9 W

ns

Thought alot aboutdeath forQ2 wkd

20.2 18.8 18.5 13.4 25.8 ns 20.3 20.3 16.3 22.1 ns ns

Thought alot about killingyourself in thepast 6 mo

10.3 19.3 17.4 20.0 28.7 p < .05 H 9 AA 9.5 8.5 18.1 12.0 p < .01 W 9 AA p < .001,F 9 M

Thought aboutkilling yourselfa lot of thetime forQ2 wke

3.7 8.3 7.0 4.4 17.2 p < .001 H 9 W 3.3 2.6 9.3 5.2 p < .01 W 9 AA p < .001,F 9 M

Had specificsuicide plane

5.5 10.5 10.0 8.9 14.8 ns 5.1 4.7 6.9 6.8 ns p < .01,F 9 M

Told anyoneaboutsuicidal wishe

4.7 9.5 8.112.2 15.0 ns 4.3 4.3 11.2 2.5 p < .001 W 9 H;W 9 AA

p < .01,F 9 M

Ever attemptedsuicide

11.0 27.1 22.0 42.8 31.7 p < .001 W 9 AA;H 9 AA

9.8 9.4 18.0 9.2 p < .05 W 9 AA;W 9 H

p < .001,F 9 M

Attemptedsuicide inpast 6 mo

3.0 8.4 8.0 7.9 11.5 ns 2.5 1.8 5.7 4.9 ns p < .001,F 9 M

Note: AA = African American; W = non-Hispanic white; H = Hispanic; F = female; M = male.a The original sample included 1,829 participants, but 3 did not receive the Diagnostic Interview Schedule for Children, version 2.3.b One female of BOther^ race is included in the Total column but excluded from all analyses of race/ethnicity.c Three males of BOther^ race are included in the Total column but are excluded from all analyses of race/ethnicity.d This variable refers to thoughts of suicide lasting for at least 2 weeks in the past 6 months.e In the past 6 months.

ABRAM

ET

AL.

294

WWW.JA

ACAP.C

OM

J.AM.ACAD.CHIL

DADOLESC.PSYCHIA

TRY,47:3,MARCH

2008


thoughts of death and dying. During the past 6 months,approximately one fifth of youths had thought aboutdeath and dying for at least 2 weeks.

Thoughts About Suicide

Approximately 10% of our sample had thought aboutsuicide in the past 6 months. Significantly more femalesthan males had thought about suicide (OR 2.27, 95%CI 1.56Y3.29). Among females, more Hispanics thanAfrican Americans had thought about suicide (OR 1.91,95% CI 1.22Y2.99). Among males, significantly morenon-Hispanic whites than African Americans endorsedthoughts of suicide (OR 2.37, 95% CI 1.36Y4.12).

Nearly 4% of our sample had thought Ba lot^ aboutsuicide for at least 2 weeks in the past 6 months;significantly more females than males had such ideation(OR 2.63, 95% CI 1.53Y4.53). Among females,significantly more Hispanics than non-Hispanic whiteshad thought about killing themselves for Q2 weeks (OR4.47, 95% CI 1.48Y13.47). Among males, significantlymore non-Hispanic whites than African Americans hadthought about suicide for Q2 weeks (OR 3.84, 95% CI1.66Y8.87).

Suicide Plan

Nearly 6% of the sample developed a specific plan forsuicide in the past 6 months. Significantly more femalesthan males had a plan (OR 2.17, 95% CI 1.32Y3.57).

Telling Someone About Suicidal Thoughts

Nearly 5% of the sample told someone in the past 6months about having suicidal ideation. Among thosewho endorsed suicidal ideation in the past 6 months,46.1% had told someone about their suicidal thoughts.Significantly more females than males had told someoneabout their suicidal thoughts (OR 2.32, 95% CI1.35Y3.97). Among males, significantly more non-Hispanic whites than African Americans (OR 2.80,95% CI 1.32Y5.95) or Hispanics (OR 4.88, 95% CI2.21Y10.77) told someone that they were thinkingabout suicide.

Suicide Attempts

Eleven percent of the sample had made at least onesuicide attempt. Participants who had ever attemptedsuicide had made, on average, two attempts (range1Y11, SD 0.15). The average age at first suicide attemptwas 12.7 years (range 5Y17, SD 0.24). Significantly

more females than males had attempted suicide (OR3.44, 95% CI 2.33Y5.08). Among females, significantlymore non-Hispanic whites (OR 2.65, 95% CI1.63Y4.28) and Hispanics (OR 1.64, 95% CI1.10Y2.53) had attempted suicide than African Amer-icans. Among males, significantly more non-Hispanicwhites than African Americans (OR 2.12, 95% CI1.21Y3.71) and Hispanics (OR 2.17, 95% CI1.10Y4.28) had attempted suicide.Approximately 3% of the sample had attempted

suicide in the past 6 months. Significantly more femalesmade suicide attempts in the past 6 months than males(OR 3.54, 95% CI 1.91Y6.57).Table 2 shows that 283 participants had ever

attempted suicide. Most common methods were cutting(26.9%), drug overdose (23.8%), and jumping(20.7%). Less common methods included hanging(9.5%), firearms (3.7%), and ingestion other than drugs(1.8%). Approximately 14% used methods other thanthose specifically listed by the DISC 2.3 (e.g., runninginto traffic). Significantly more males than femalesattempted suicide by jumping (OR 5.62, 95% CI1.97Y16.04). Among males, significantly more Hispa-nics attempted suicide with firearms than AfricanAmericans (OR 18.06, 95% CI 3.90Y83.53).

Psychiatric Correlates of Recent Suicide Attempts

Table 3 shows that, controlling for sex, age, and race/ethnicity in separate analyses of individual disorders,nearly all of the disorders significantly increased the oddsof a recent suicide attempt. We tested for interactionsbetween sex and each of the disorders associated with arecent suicide attempt. Overanxious disorder increasedthe odds of making a suicide attempt, but less so forfemales than for males (OR 0.19, 95% CI 0.47Y0.77; p< .05). We also tested for interactions between race/ethnicity and each of the disorders associated with arecent suicide attempt. Generalized anxiety disordersignificantly increased the odds for a recent attempt forboth Hispanics (OR 58.83, 95%CI 13.60Y247.52) andAfrican Americans (OR 4.96, 95% CI 1.02Y24.00);however, it increased the odds significantly more forHispanics than African Americans (OR 10.09, 95% CI1.64Y73.00).In a regression analysis containing sex, age, and

race/ethnicity along with all of the disorders that wereindividually associated with suicide attempt, majordepressive episode modified (OR 3.21, 95% CI



Copyright @

2008 Am

erican Academ

y of Child and A

dolescent Psychiatry. U

nauthorized reproduction of this article is prohibited.

TABLE 2Method of Most Recent Suicide Attempt by Sex and Race/Ethnicity for 283 Juvenile Detainees

Method ofMost RecentSuicideAttempt

Totala

(N = 283)(%)

Females Males

Totalb

(n = 177)(%)

AfricanAmerican(n = 95)(%)

Non-HispanicWhite(n = 38)(%)

Hispanic(n = 43)(%)

Tests ofRace/

Ethnicity

SpecificTests

of Race/Ethnicity

Total(n = 106)

(%)

AfricanAmerican(n = 40)(%)

Non-HispanicWhite (n = 35)

(%)

Hispanic(n = 31)(%)

Tests ofRace/

Ethnicity


Ethnicity

TestsofSex

Cutting 26.9 39.8 31.8 50.7 34.3 ns 24.0 23.6 33.7 20.0 ns nsDrug overdose 23.8 34.7 37.5 28.5 43.1 ns 21.4 23.1 24.8 10.8 ns nsJumping 20.7 5.3 8.5 2.6 0.0 N/A 24.0 22.7 12.9 37.9 ns p < .01,

M 9 FHanging 9.5 4.7 7.4 0.0 2.1 N/Ac 10.5 11.6 10.9 5.2 ns nsFirearms 3.7 1.7 0.0 5.2 4.3 N/Ad 4.1 1.3 2.9 18.9 p < .01 H 9 AA nsOther ingestion 1.8 2.7 2.1 2.6 5.4 ns 1.6 1.4 0.0 3.6 N/Ae nsOther 13.8 11.1 12.7 10.4 10.7 ns 14.3 16.4 14.7 3.6 ns ns

a Of 1,826 participants who received the Diagnostic Interview Schedule for Children, version 2.3, 283 reported a history of a suicide attempt.b One female of BOther[ race is included in the Total column but excluded from all of the analyses of race/ethnicity.c No non-Hispanic white females endorsed hanging as their most recent method of suicide attempt, so racial/ethnic comparison could not be made among females for this variable.d No African American females endorsed firearms as their most recent method of suicide attempt, so racial/ethnic comparison could not be made among females for this variable.e No non-Hispanic white males endorsed other ingestion as their most recent method of suicide attempt, so racial/ethnic comparisons could not be made among males for this

variable.

ABRAM

ET

AL.

296

WWW.JA

ACAP.C

OM

J.AM.ACAD.CHIL

DADOLESC.PSYCHIA

TRY,47:3,MARCH

2008


1.05Y9.81) and generalized anxiety disorder (OR 3.40,95% CI 1.51Y7.67) significantly increased the odds ofhaving made a recent suicide attempt.

DISCUSSION

One of every 10 newly detained youths has a historyof attempted suicide. Because suicide attempts are apowerful predictor of future attempts,5 detained youthsare at greater risk than youths in the generalpopulation.25,26,28,44,45

Risk varied by demographic characteristics. Femaleshave a higher risk for suicide than males, a finding con-sistent with previous studies of detained youths16,18,19

and a study of delinquency among youths in the generalpopulation.46 The association with recent suicideattempts persists even after controlling for currentpsychiatric disorders, which also tend to be moreprevalent among girls.6

Our study confirms and extends what is known aboutracial/ethnic differences among detained youths16,19,22,27;non-Hispanic whites generally have higher risk for suicidethan youths of color. We did, however, find a few ex-ceptions. Among females, Hispanics were the most likelyto have suicidal ideation. Studies of youths in the generalpopulation have also found that Hispanic females havehigher prevalence rates of suicidal ideation and behaviorthan non-Hispanic white47Y49 and African American50,51

females. Suicidal ideation and behavior appears to behighest in U.S.-born Hispanic females from traditionalHispanic families who may have difficulty coping withcontrasting social role expectations at home and amongpeers.52,53

We also found that significantly more AfricanAmerican and Hispanic males had thoughts aboutdeath in the past 6 months than non-Hispanic whites. Itis unclear whether and how concern about death amongAfrican American and Hispanic males is related to riskfor suicide. Such concern may result from a greaterlikelihood of having lost siblings and peers to violentdeath compared with non-Hispanic white males.54,55

These findings also may reflect an awareness of theirown heightened risk for mortality. In our sample,African American and Hispanic males are at substan-tially greater risk for an early violent death than non-Hispanic whites.56

The most common methods for recent suicideattemptsVcutting and drug overdoseVare also themost common in the general population.28 A strikingfinding was that Hispanic males who attempted suicideweremore likely to use a firearm than African American ornon-Hispanic white males. This finding is of particularconcern because half of all of the completed suicides byyoung men in the general population involve firearms.57

Many psychiatric disorders were associated withhaving made a recent suicide attempt. At minimum,detainees who are in any type of distress must beconsidered at risk for self-harm. When accounting forcomorbidity in a multivariate model, however, onlymajor depression (modified) and generalized anxietydisorder remained significant predictors. These inter-nalizing disorders are often the most difficult forcorrectional staff to identify; affected youths tend tobe compliant and cause little trouble. Anxiety disorderswere also more strongly associated with a recent suicideattempt for males and Hispanics, groups that are lesslikely to be detected in detention as needing services.

TABLE 3Odds Ratios (ORs) and Confidence Intervals (CIs) for the

Association Between Specific Psychiatric Disorders and RecentSuicide Attempt

Diagnosis

Recent Suicide Attempt (Past 6 Mo)

ORa 95% CI F df p

Generalizedanxiety disorder

9.89 3.34Y29.27 17.17 1, 1,798 .001

Overanxious disorder 8.80 3.03Y25.55 16.00 1, 1,808 .001Major depression 6.88 2.73Y17.30 16.81 1, 1,804 .001Hypomania 6.78 2.66Y17.26 16.12 1, 1,786 .001Oppositionaldefiant disorder

5.19 1.98Y13.59 11.22 1, 1,807 .001

Panic disorder 4.86 1.43Y16.44 6.46 1, 1,804 .011Obsessive-compulsivedisorder

4.80 2.07Y11.17 13.30 1, 1,802 .001

Psychotic disorder 3.92 1.03Y14.92 4.00 1, 1,810 .046Separation anxietydisorder

3.56 1.46Y8.71 7.79 1, 1,810 .005

Alcohol use disorder 3.51 1.47Y8.41 7.95 1, 1,783 .005Conduct disorder 3.34 1.31Y8.52 6.38 1, 1,803 .012Dysthymia 3.22 1.27Y8.22 6.03 1, 1,808 .014Other substanceuse disorder

2.44 1.05Y5.71 4.27 1, 1,808 .039

Marijuana usedisorder

1.30 0.52Y3.27 0.31 1, 1,783 .580

Attention-deficit/hyperactivitydisorder

1.26 0.59Y2.69 0.37 1, 1,789 .545

Mania 0.95 0.23Y3.96 0.01 1, 1,797 .939

a Logistic regression analyses control for sex, race/ethnicity,and age.




Although preliminary, these findings suggest that suiciderisk is manifested differently in males and females and byrace/ethnicity and that we need gender-specific and cul-turally relevant interventions for detained youths.18,46

A few limitations to the study are noteworthy.Because our measure of suicidal ideation and behaviorwas part of a larger diagnostic module, our data were notas comprehensive and detailed as clinical measures ofsuicidal ideation and behavior (e.g., Suicidal BehaviorsInterview,58 the Suicidal Ideation Questionnaire,59 theScale for Suicidal Ideation60,61). Using an interviewinstead of a self-report questionnaire may underestimatethe prevalence of suicidal ideation. In contrast, theturmoil of recent detention may increase participants`suicidal symptoms or their awareness of symptoms.Although we had a large and diverse sample, ourstatistical power limited analyses of racial/ethnic differ-ences for uncommon behaviors, such as the method ofsuicide attempts. Furthermore, correlational analyses donot infer causality. Finally, findings may generalize onlyto juvenile detainees living in urban areas. Despite theselimitations, our findings have implications for researchand for clinical services.

Directions for Future Research

We suggest several directions for future research:

1. Investigate factors that underlie sex and racial/ethnicdifferences in suicidal ideation and behavior. Studiesare needed, for example, to investigate why suicidalideation was most common in Hispanic females,whereas non-Hispanic white females had thehighest prevalence rates of suicide attempts. Sui-cidal ideation may play a different role in the riskfor suicide for different groups. Further research isalso needed to examine whether the dispropor-tionate violent deaths among African American andHispanic males,56 in part, reflect their underlyingsuicidal ideation. Suicidal behavior in these youthsmay manifest as self-destructive, reckless, ordangerous behavior, often referred to as victim-precipitated homicide or Bsuicide-by-cop.^62

2. Study the relationship between adverse life eventsand thoughts of death. Although Bthoughts of death^is a common risk factor for suicide in the generalpopulation, it may reflect the greater exposureto violence, loss, and trauma experienced by detainedyouths.7,55 Studies are needed to examine whether

Bthoughts of death[ remains a useful marker forsuicidal ideation and behavior in a population thatroutinely experiences loss and violence.

3. Develop methods to assess suicidal ideation. Self-report questionnaires may yield higher prevalencerates of suicidal ideation than face-to-face inter-views.63Y65 It is unclear, however, which methodsproduce the most valid estimates. A multimethodcross-validational approach using both interviewsand self-report may produce the most accurateinformation.64 More research is needed toidentify which methods of suicide assessment aremost accurate, especially for high-risk youths.

Our findings highlight two clinical implications:First, juvenile detention facilities must systematicallyscreen for suicide risk. Juvenile detention centers oftenprovide the first opportunity to systematically screenyouths for risk for suicide and to provide interventions,yet the majority of facilities do not perform soundscreening for Bemergent risk.^66 A recent study foundthat facilities that screen all juveniles within 24 hours ofarrival had lower prevalence rates of serious suicideattempts than those that screen only juveniles con-sidered at risk for suicide.67 Less than half of thedetainees with recent suicidal thoughts had shared thiswith someone else. Juvenile justice facilities cannot relyon juvenile detainees to inform staff that they arecontemplating suicide.67 Identifying youths at riskfor suicide is a crucial step to preventing suicide, bothin detention centers and after youths return to theircommunities.Second, psychiatric services in detention must be

increased. Youths with psychiatric disorders, especiallydepression and anxiety, may be at particular risk forsuicide attempts. Detention center staff should betrained to recognize depressive and anxiety disordersin detainees and refer affected youths for psychiatricservices. The competent assessment and treatment ofpsychiatric disorders in detained youths will preventuntimely deaths.

Disclosure: Dr. Dulcan is a consultant to the Strattera Global AdvisoryBoard for Eli Lilly. The other authors report no conflicts of interest.

REFERENCES

1. Hoyert DL, Kung H-C, Smith BL. Deaths: Preliminary Data for 2003(PHS 2005-1120). Hyattsville, MD: National Center for HealthStatistics, 2005.

ABRAM ET AL.



2. National Center for Health Statistics. Health, United States, 2004 WithChartbook on Trends in the Health of Americans. Hyattsville, MD: USDepartment of Health and Human Services; 2004.

3. Memory JM. Juvenile suicides in secure detention facilities: correction ofpublished rates. Death Stud. 1989;13:455Y463.

4. Gallagher CA, Dobrin A. Deaths in juvenile justice residential facilities.J Adolesc Health. 2006;38:662Y668.

5. Hayes LM. Juvenile Suicide in Confinement: A National Survey.Mansfield, MA: National Center on Institutions and Alternatives,2004.

6. Teplin LA, Abram KM, McClelland GM, Dulcan MK, Mericle AA.Psychiatric disorders in youth in juvenile detention. Arch Gen Psychiatry.2002;59:1133Y1143.

7. Abram KM, Teplin LA, Charles DR, Longworth SL, McClelland GM,Dulcan MK. Posttraumatic stress disorder and trauma in youth injuvenile detention. Arch Gen Psychiatry. 2004;61:403Y410.

8. Shelton D. Health status of young offenders and their families. J NursScholarsh. 2000;32:173Y178.

9. Pogrebin M. Jail and the mentally disordered: the need for mental healthservices. J Prison Jail Health. 1985;5:13Y19.

10. Parent DG, Leiter V, Kennedy S, Livens L, Wentworth D, Wilcox S.Conditions of Confinement: Juvenile Detention and Corrections Facilities.Washington, DC: Office of Juvenile Justice and Delinquency Preven-tion, U.S. Department of Justice; 1994.

11. Marcus P, Alcabes P. Characteristics of suicides by inmates in an urbanjail. Hosp Community Psychiatry. 1993;44:256Y261.

12. Evans W, Albers E, Macari D, Mason A. Suicide ideation, attempts andabuse among incarcerated gang and nongang delinquents. Child AdolescSoc Work J. 1996;13:115Y126.

13. Penn JV, Esposito CL, Schaeffer LE, Fritz GK, Spirito A. Suicideattempts and self-mutilative behavior in a juvenile correctional facility.J Am Acad Child Adolesc Psychiatry. 2003;42:762Y769.

14. Holsinger K, Holsinger AM. Differential pathways to violence and self-injurious behavior: African American and white girls in the juvenilejustice system. J Res Crime Delinquency. 2005;42:211Y242.

15. Wasserman GA, McReynolds LS. Suicide risk at juvenile justice intake.Suicide Life Threat Behav. 2006;36:239Y249.

16. Cauffman E. A statewide screening of mental health symptoms amongjuvenile offenders in detention. J Am Acad Child Adolesc Psychiatry.2004;43:430Y439.

17. Goldstein NE, Arnold DH, Weil J, et al. Comorbid symptom patterns infemale juvenile offenders. Int J Law Psychiatry. 2003;26:565Y582.

18. Rohde P, Seeley JR, Mace DE. Correlates of suicidal behavior in ajuvenile detention population. Suicide Life Threat Behav. 1997;27:164Y175.

19. Morris RE, Harrison EA, Knox GW, et al. Health risk behavioral surveyfrom 39 juvenile correctional facilities in the United States. J AdolescHealth. 1995;17:334Y344.

20. Esposito CL, Clum GA. Specificity of depression symptoms andsuicidality in a juvenile delinquent population. J Psychopathol Behav.1999;21:171Y182.

21. Sanislow CA, Grilo CM, Fehon DC, Axelrod SR, McGlashan TH.Correlates of suicide risk in juvenile detainees and adolescent inpatients.J Am Acad Child Adolesc Psychiatry. 2003;42:234Y240.

22. Esposito CL, Clum GA. Social support and problem-solving asmoderators of the relationship between childhood abuse and suicidality:applications to a delinquent population. J Trauma Stress. 2002;15:137Y146.

23. Flannery DJ, Singer MI, Wester K. Violence exposure, psychologicaltrauma, and suicide risk in a community sample of dangerouslyviolent adolescents. J Am Acad Child Adolesc Psychiatry. 2001;40:435Y442.

24. Gould MS, Greenberg T, Velting DM, Shaffer D. Youth suicide risk andpreventive interventions: a review of the past 10 years. J Am Acad ChildAdolesc Psychiatry. 2003;42:386Y405.

25. Lewinsohn PM, Rohde P, Seeley JR. Adolescent suicidal ideation andattempts: prevalence, risk factors, and clinical implications. Clin Psychol.1996;3:25Y46.

26. Kessler RC, Borges G, Walters EE. Prevalence of and risk factors for

lifetime suicide attempts in the National Comorbidity Survey. Arch GenPsychiatry. 1999;56:617Y626.

27. Rohde P, Mace DE, Seeley JR. The association of psychiatric disorderswith suicide attempts in a juvenile delinquent sample. Crim Behav MentHealth. 1997;7:187Y200.

28. Gould MS, King R, Greenwald S, et al. Psychopathology associated withsuicidal ideation and attempts among children and adolescents. J AmAcad Child Adolesc Psychiatry. 1998;37:915Y923.

29. Wasserman GA, McReynolds LS, Lucas CP, Fisher P, Santos L. Thevoice DISC-IV with incarcerated male youths: prevalence of disorder.J Am Acad Child Adolesc Psychiatry. 2002;41:314Y321.

30. Snyder HN, Sickmund M. Juvenile Offenders and Victims: 1999 NationalReport. Washington, DC: Office of Juvenile Justice and DelinquencyPrevention; 1999.

31. Sickmund M, Sladky TJ, Kang W. Census of Juveniles in ResidentialPlacement Databook, 2005. Available at: http://www.ojjdp.ncjrs.org/ojstatbb/cjrp . AccessedMarch 1, 2007.

32. Pastore AL, Maguire K. Sourcebook of Criminal Justice StatisticsY1999.Washington, DC: U.S. Department of Justice, 2000.

33. U.S. Bureau of the Census. The Hispanic Population. Washington, DC:U.S. Department of Commerce; 2001.

34. Cohen J. Statistical Power Analysis for the Behavioral Sciences, 2nd ed.Hillsdale, NJ: Erlbaum; 1988.

35. Shaffer D, Fisher P, Dulcan MK, Davies M. The NIMH DiagnosticInterview Schedule for Children Version 2.3 (DISC-2.3): description,acceptability, prevalence rates, and performance in the MECA study.J Am Acad Child Adolesc Psychiatry. 1996;35:865Y877.

36. Bravo M, Woodbury-Farina M, Canino GJ, Rubio-Stipec M. TheSpanish translation and cultural adaptation of the Diagnostic InterviewSchedule for Children (DISC) in Puerto Rico. Cult Med Psychiatry.1993;17:329Y344.

37. Shaffer D, Schwab-Stone M, Fisher P, et al. The Diagnostic InterviewSchedule for ChildrenYRevised version (DISC-R): I. Preparation, fieldtesting, interrater reliability, and acceptability. J Am Acad Child AdolescPsychiatry. 1993;32:643Y650.

38. Fisher PW, Shaffer D, Piacentini J, et al. Sensitivity of the DiagnosticInterview Schedule for Children, 2nd edition (DISC-2.1) for specificdiagnoses of children and adolescents. J Am Acad Child Adolesc Psychiatry.1993;32:666Y673.

39. Piacentini J, Shaffer D, Fisher P, Schwab-Stone M, Davies M, Gioia P.The Diagnostic Interview Schedule for ChildrenYRevised Version(DISC-R): III. Concurrent criterion validity. J Am Acad Child AdolescPsychiatry. 1993;32:658Y665.

40. Schwab-Stone M, Fisher PW, Piacentini J, Shaffer D, Davies M, BriggsM. The Diagnostic Interview Schedule for ChildrenYRevised version(DISCYR): II. Test-retest reliability. J Am Acad Child Adolesc Psychiatry.1993;32:651Y657.

41. Abram KM, Teplin LA, McClelland GM, Dulcan MK. Comorbidpsychiatric disorders in youth in juvenile detention. Arch Gen Psychiatry.2003;60:1097Y1108.

42. Cochran W. Sampling Techniques, 3rd ed. New York: Wiley; 1977.43. Levy P, Lemeshow S. Sampling of Populations: Methods and Applications,

3rd ed. New York: Wiley; 1999.44. Johnson JG, Cohen P, Gould MS, Kasen S, Brown J, Brook JS.

Childhood adversities, interpersonal difficulties, and risk for suicideattempts during late adolescence and early adulthood. Arch GenPsychiatry. 2002;59:741Y749.

45. Foley DL, Goldston DB, Costello EJ, Angold A. Proximal psychiatric riskfactors for suicidality in youth: the Great Smoky Mountains Study. ArchGen Psychiatry. 2006;63:1017Y1024.

46. Thompson MP, Kingree JB, Ho C. Associations between delinquencyand suicidal behaviors in a nationally representative sample ofadolescents. Suicide Life Threat Behav. 2006;36:57Y64.

47. Roberts RE, Chen Y-W. Depressive symptoms and suicidal ideationamong Mexican-origin and Anglo adolescents. J Am Acad Child AdolescPsychiatry. 1995;34:81Y90.

48. Roberts RE, Chen R, Roberts CR. Ethnocultural differences inprevalence of adolescent suicidal behaviors. Suicide Life Threat Behav.1997;27:208Y217.




49. Tortolero SR, Roberts RE. Differences in nonfatal suicide behaviorsamong Mexican and European American middle school children. SuicideLife Threat Behav. 2001;31:214Y223.

50. Grunbaum JA, Kann L, Kinchen S, et al. Youth Risk BehaviorSurveillance, United States, 2003. MMWR Morb Mortal Wkly Rep.2004;53:1Y96.

51. Sen B. Adolescent propensity for depressed mood and help seeking:race and gender differences. J Ment Health Policy Econ. 2004;7:133Y145.

52. Zayas LH, Lester RJ, Cabassa LJ, Fortuna LR. Why do so many Latinateens attempt suicide? A conceptual model for research. Am JOrthopsychiatry. 2005;75:275Y287.

53. Zimmerman JK. Crossing the desert alone: an etiological model of femaleadolescent suicidality. In: Gilligan C, Rogers AG, Tolman DL , eds.Women, Girls, and Psychotherapy. New York: Haworth; 1991:223Y240.

54. Crouch JL, Hanson RF, Saunders BE, Kilpatrick DG, Resnick HS.Income, race/ethnicity, and exposure to violence in youth: results fromthe national survey of adolescents. J Community Psychol. 2000;28:625Y641.

55. Viboch M. Childhood Loss and Behavioral Problems: Loosening the Links.New York: Vera Institute of Justice; 2005.

56. Teplin LA, McClelland GM, Abram KM, Mileusnic D. Early violentdeath among delinquent youth: a prospective longitudinal study.Pediatrics. 2005;115:1586Y1593.

57. Centers for Disease Control and Prevention, 2005. WISQARS Fatal

Injuries: Leading Causes of Death Reports. Available at: http://webappa.cdc.gov/sasweb/ncipc/leadcaus10.html. Accessed June 13, 2005.

58. Reynolds WM. Development of a semistructured clinical interview forsuicidal behaviors in adolescents. Psychol Assess. 1990;2:382Y390.

59. Reynolds WM. Suicidal Ideation Questionnaire: Professional Manual.Odessa, FL: Psychological Assessment Resources; 1988.

60. ClumGA, Curtin L. Validity and reactivity of a system of self-monitoringsuicide ideation. J Psychopathol Behav. 1993;15:375Y385.

61. Beck AT, Kovacs M, Weissman A. Assessment of suicidal intention: theScale for Suicide Ideation. J Consult Clin Psychol. 1979;47:343Y352.

62. Lindsay M, Lester D. Suicide-by-Cop: Committing Suicide by ProvokingPolice to Shoot You. Amityville, NY: Baywood; 2004.

63. Kaplan ML, Asnis GM, Sanderson WC, et al. Suicide assessment: clinicalinterview vs. self-report. J Clin Psychol. 1994;50:294Y298.

64. Klimes-Dougan B. Screening for suicidal ideation in children andadolescents: methodological considerations. J Adolesc. 1998;21:435Y444.

65. Levine S, Ancill RJ, Roberts AP. Assessment of suicide risk by computer-delivered self-rating questionnaire: preliminary findings. Acta PsychiatrScand. 1989;80:216Y220.

66. Wasserman GA, Jensen PS, Ko SJ, et al. Mental health assessments injuvenile justice: report on the Consensus Conference. J Am Acad ChildAdolesc Psychiatry. 2003;42:751Y761.

67. Gallagher CA, Dobrin A. The association between suicide screeningpractices and attempts requiring emergency care in juvenile justicefacilities. J Am Acad Child Adolesc Psychiatry. 2005;44:477Y484.

Healthy Steps for Young Children: Sustained Results at 5.5 Years Minkovitz CS, Strobino D, Mistry KB, Scharfstein DO,Grason H, Hou W, Ialongo N, Guyer B

Objective: We sought to determine whether Healthy Steps for Young Children has sustained treatment effects at 5.5 years, given earlyfindings demonstrating enhanced quality of care and improvements in selected parenting practices.Methods:Healthy Steps was a clinicaltrial that incorporated developmental specialists and enhanced developmental services into pediatric care in the first 3 years of life. Atotal of 5565 children were enrolled at birth and followed through 5.5 years. Healthy Steps was evaluated at 6 randomization and 9quasi-experimental sites. Computer-assisted telephone interviews were conducted with mothers when Healthy Steps children were 5.5years of age. Outcomes included experiences seeking care, parent response to child misbehavior, perception of child`s behavior, andparenting practices to promote development and safety. Logistic regression was used to estimate overall effects of Healthy Steps,adjusting for site and baseline demographic characteristics. Results: A total of 3165 (56.9%) families responded to interviews (usual care:n = 1441; Healthy Steps: n = 1724). Families that had received Healthy Steps services were more satisfied with care (agreed thatpediatrician/nurse practitioner provided support, 82.0% vs 79.0%; odds ratio: 1.25 [95% confidence interval: 1.02Y1.53]) and morelikely to receive needed anticipatory guidance (54.9% vs 49.2%; odds ratio: 1.33 [95% confidence interval: 1.13Y1.57]) (all P < .05).They also had increased odds of remaining at the original practice (65.1% vs 61.4%; odds ratio: 1.19 [95% confidence interval:1.01Y1.39]). Healthy Steps families reported reduced odds of using severe discipline (slap in face/spank with object, 10.1% vs 14.1%;odds ratio: 0.68 [95% confidence interval: 0.54Y0.86]) and increased odds of often/almost always negotiating with their child (59.8%vs 56.3%; odds ratio: 1.20 [95% confidence interval: 1.03Y1.39]). They had greater odds of reporting a clinical or borderline concernregarding their child`s behavior (18.1% vs 14.8%; odds ratio: 1.35 [95% confidence interval: 1.10Y1.64]) and their child reading books(59.4% vs 53.6%; odds ratio: 1.16 [95% confidence interval: 1.00Y1.35]). There were no effects on safety practices. Conclusions:Sustained treatment effects, albeit modest, are consistent with early findings. Universal, practice-based interventions can enhance qualityof care for families with young children and can improve selected parenting practices beyond the duration of the intervention. Reprintedwith permission from Pediatrics 2007;120(3):e658Y668 by the AAP.

ABRAM ET AL.


suicidal ideation and behaviors among youths in juvenile detention

Documents