Article summary

527Suicide and Life-Threatening Behavior 37(5) October 2007  2007 The American Association of Suicidology

Transgender Youth and Life-Threatening Behaviors Arnold H. Grossman, PhD., MSW, and Anthony R. D’Augelli, PhD

Sexual minority status is a key risk factor for suicide among lesbian, gay, and bisexual youth; however, it has not been studied among transgender youth. Fifty-five transgender youth reported on their life-threatening behaviors. Nearly half of the sample reported having seriously thought about taking their lives and one quarter reported suicide attempts. Factors significantly related to having made a suicide attempt included suicidal ideation related to transgender identity; experi- ences of past parental verbal and physical abuse; and lower body esteem, especially weight satisfaction and thoughts of how others evaluate the youths’ bodies. Sexual minority status is a key risk factor for life-threatening behaviors among transgen- der youth.

Suicide is the third leading cause of death for same-sex identity such as lesbian, gay, bisex- ual, and transgender) are uncertain (McDan-youth between the ages of 15 and 24 in the

United States. In 2002, 4,010 youth between iel, Purcell, & D’Augelli, 2001). While esti- mates for life-threatening behaviors (i.e.,these ages took their lives (Centers for Dis-

ease Control and Prevention [CDC], 2002). nonfatal acts where there is evidence that the individual had some intent to die) are avail-Estimates of completed suicide rates among

subgroups of the population, including sexual able for lesbian, gay, and bisexual (LGB) youth (Russell, 2003), comparable estimatesminority youth (i.e., youth who engage in

same-sex behaviors; have enduring same-sex for transgender youth do not exist. In one study, 19% of gay college stu-emotional or sexual attractions; or claim a

dents had made suicide attempts; many at- tempts were related to conflict about their

Arnold H. Grossman is with the Depart- sexual orientation (Savin-Williams & Cohen, ment of Applied Psychology at New York Univer- 1996). In a recent study, D’Augelli et al. sity; and Anthony R. D’Augelli is with the De- (2005) found that about one third of a com- partment of Human Development and Family

munity-based sample of 528 self-identifiedStudies at The Pennsylvania State University. LGB youth, ages 15–19, attempted suicide,We acknowledge the staff and volunteers

of the agencies who cooperated in recruiting par- which is generally comparable to findings in ticipants. Dr. Timothy S. O’Connell is thanked other studies of sexual minority youth (Re- for assistance with the research design and for his mafedi, 1994). That percentage is consider- research and management skills which helped the

ably higher than the 8.5% identified in a re-project begin. We also thank the study partici- cent survey of high school students (CDC,pants and the project’s research staff. This re-

search was supported the Research Challenge 2004). D’Augelli et al. also found that 15% Fund of New York University. of the suicides could be classified as lethal

Address correspondence to Arnold H. and that 17% of the attempts were attributed Grossman, PhD, MSW, Dept. of Applied Psy-

to the youths’ sexual orientation. In anotherchology, New York University, 239 Greene St., study, involving 77 college students recruitedSuite 400, New York, NY 10003; E-mail: arnold.

[email protected] from an undergraduate psychology pool and

528 Transgender Youth

gay, lesbian, and transgender student organi- When left unchecked, both the internal and external factors can lead to an array of adjust-zations, the majority between the ages of 18

and 24 (n = 69); Fitzpatrick, Euton, Jones, ment problems, including substance abuse and self-injurious behaviors (Burgess, 1999).and Schmidt (2005) found that cross-gender

role (i.e., personality traits associate with the Many helping professionals do not have ac- curate knowledge about the risks for suicidalopposite sex) was a predictor of suicide symp-

toms as determined on the Beck Suicide behavior among transgender youth (Lev, 2004). Our goal in this paper is to enhanceScale. Based on these and similar findings,

sexual minority status has been determined the knowledge of helping professionals about transgender youths’ and their risk for life-to be a key risk indicator for suicidal behav-

iors among adolescents (Russell, 2003). threatening behaviors so they will be prepared to provide effective intervention. In addition,Youth are predisposed to life-threaten-

ing behaviors under a variety of conditions, as most parents have control over their chil- dren’s medical care, helping professionals canincluding self-hatred, victimization by bully-

ing from peers, a history of family violence, play a vital role in helping parents under- stand the impact of their attitudes, behavior,substance abuse, and sexual identity conflicts.

“Research suggests that gay and lesbian and decisions on their transgender children as well as assist them in transitioning fromyouth are highly vulnerable on every count.

. . . Conflicts about the disclosure of sexual their assigned birth sex and gender. orientation may influence young people to attempt suicide if they are otherwise predis- posed” (van Wormer, Wells, & Boes, 2000, DEFINING TRANSGENDER

AND RELATED TERMSp. 122). Recent research indicates that trans- gender youth experience victimization from their peers, negative parental reactions to Transgender is an umbrella term used

to describe people whose self-identificationtheir gender nonconforming expression and identity, substance abuse, and family violence or expression transgresses established gender

categories or boundaries (Green, 2004; Sears,that is similar to their LGB counterparts, who have higher rates of life-threatening 2005). People who identify as transgender or

“trans” live all or a substantial portion ofbehaviors than their heterosexual peers (D’Augelli et al., 2005; Grossman, D’Augelli, their lives expressing or presenting a gender

identity (i.e., an internal sense of gender) thatHowell, & Hubbard, 2005; Grossman, D’Augelli, & Salter, 2006). is other than their birth sex. The term trans-

gender is inclusive of individuals who identifyIn addition to the exogenous factors of rejection, maltreatment, and victimization, as transsexuals (i.e., identify with a gender dif-

ferent from their birth sex), cross-dressers (i.e.,youth who self-identify as transgender also experience personal distress and isolation. A wear clothes usually associated with a gender

other than their birth sex), and gender blendersprimary area of distress is a gender dysphoria; that is, a strong and persistent (long-term) (i.e., present ambiguous gender expressions).

Transgender communities include a varietydiscomfort and distress with one’s birth sex, gender, and anatomical body. Because of this of individuals who regularly transgress con-

ventional gender norms and role expecta-distress, body esteem—especially how one feels about one’s appearance and weight, and tions.

Most children learn about their birththe perception of how others think of one’s body—assumes importance. Gender dyspho- sex, gender, and gender role expectations

early in their development. Developmentalria also includes feelings arising from confu- sion about gender identity, for example, some research indicates that most two-year-olds

know whether they are girls or boys; by theyouth mistakenly believe that engaging is cross-gendered behavior means that they are time they reach the age of three, children use

the gender labels of “she” and “he” when re-lesbian or gay (Ryan & Futterman, 1998).

Grossman and D’Augelli 529

ferring to females and males. They also play tory of life-threatening behaviors among transgender youth (i.e., frequency of suicidewith toys associated with their own gender

and generally avoid toys associated with the ideation and attempts, methods used, and le- thality of attempts)? (2) Do parental reac-other gender (Marcus & Overton, 1978). By

ages four and five, they know that girls are tions to the youths’ gender nonconformity and transgender identity relate to youths’more likely to play with dolls and boys are

more likely to play aggressive sports (Connor life-threatening behaviors (i.e., suicide at- tempts)? (3) Do youths’ feelings about the& Serbin, 1977; Paley, 1984). While playing,

girls tend to assume the roles associated with appearance of their bodies, or body esteem, relate to life-threatening behaviors (i.e., sui-traditionally female professions (e.g., nurses,

teachers, and secretaries), while boys take cide attempts)? and, (4) What are the differ- ences between the transgender youth whoroles normally associated with traditionally

male professions (e.g., doctors, firefighters, had engaged in life-threatening behaviors (i.e., report attempted suicide) and those whoand truck drivers). Although some cultural

traditions have changed, Perrin (2002) ar- had not? Only the components of the assess- ment done for the larger project used in thegued that fundamental gender stereotypes

have remained in place and that most chil- analyses to address these questions are dis- cussed in this paper.dren express stereotypic ideas about what

each sex should wear, feel, and do. Further- The assessment procedure was an in- terview which focused on the experiences ofmore, children react in approving or disap-

proving ways toward each other according to transgender youth as well as a battery of stan- dard measures that assessed various aspects oftheir sex-appropriate behavior. Therefore,

youth who are gender nonconforming and adjustment and mental health. The protocol was based on a previous one used in a studyexpress identities that differ from their as-

signed birth sex receive varying responses of LGB youth (D’Augelli & Grossman, 2006). The earlier protocol was modifiedfrom others, many of which are disapproving

(Ryan & Futterman, 1998). Some gender based on findings from focus groups with transgender youth and on the advice of anonconforming youth will identify as trans-

gender and a subgroup of these youth experi- planning and evaluation group of transgen- der youth, adults, and professionals who hadence distress leading to life-threatening be-

haviors. The experiences of life-threatening worked with transgender youth (Grossman & D’Augelli, 2006). Because seeking parentalbehaviors among transgender youth have not

yet been explored in social science research. consent could put the youth at risk for expos- ing their gender identity or lead to harm, pa-Therefore, the investigators had no specific

hypotheses in designing and conducting this rental consent was waived. However, a youth advocate was available to discuss questionsresearch; instead, they aimed to generate in-

formation that would aid future research about the study or the youths’ participation in the study. The research procedures and(Way, 1998). protocols were approved by the institutional review boards on research with human sub- jects of New York University and Pennsylva-METHOD nia State University.

Data are based on a convenience sam-Data for this report were taken from a larger exploratory study of the personal and ple of male-to-female (MTF; i.e., individuals

whose birth sex is male, whose gender iden-contextual factors influencing the develop- ment of transgender youth, aged 15 to 21. In tity is female, or who behave in ways tradi-

tionally associated with females) and female-the current report, the investigators focused on the history of life-threatening behaviors to-male (FTM; i.e., individuals whose birth

sex is female, whose gender identity is male,and their correlates. There were guided by four research questions: (1) What is the his- or who behave in ways traditionally associ-

530 Transgender Youth

ated with males) youth. Because transgender logical mothers, by their grandmothers (6), or by an adoptive mother (1). There were noyouth are a hidden population, it was not

possible to recruit a representative sample. differences between MTF and FTM youth regarding those raised by mothers andThe participants were recruited from pro-

grams of two social and recreation services grandmothers. Of those remaining, six youth were raised by their biological fathers, one byagencies providing services to LGBT youth

in New York City. Using a snowball sampling a stepfather, and two by other family mem- bers.technique, participants referred other youth

to the study. Youth were offered a $30 incen- tive to participate. The authors recognize Assessment that these recruitment techniques limit the generalizability of the results and that the Youth were assigned an interviewer who

was a master’s level clinician with experiencefindings may not be characteristic of trans- gender youth between the ages of 15 to 21. working with transgender youth. Interviews

took place in private rooms at the agenciesAdditionally, generalizability is not possible due to other research limitations, including or in nearby university offices. After giving

their informed consent, the youth completedthe fact that a convenience sample was used, that the youth self-identified as MTF and a questionnaire, and then participated in a

structured interview. The interviews wereFTM transgender youth, that the youth had access to a community-based organization conducted between 2001 and 2003.

Suicide Ideation. Suicidal ideation orserving lesbian, gay, bisexual, and transgen- der youth, or knew someone who did. Also, thoughts about life-threatening actions was

assessed with the following three questions:all data were based on youths’ self-reports, which have intrinsic limitations. (1) “How often have you seriously thought

about taking your own life?” (response op- tions: never, rarely, sometimes, often); (2)Participants “Within the last year, how often have you se- riously thought of taking your own life?”The investigators studied a sample of 31

MTF and 24 FTM youth between the ages of (same response options); and (3) “How much were these thoughts related to your being15 to 21. The respective mean ages of the two

groups were 17.5 (SD = 1.6) and 19.5 (SD = transgender?” (response options: not related, somewhat related, very related).1.6), a significant difference, t [53] = 4.55, p <

.001. As to ethnicity, 22 were of Hispanic Suicide Attempts. Past life-threatening behaviors defined as suicide attempts wereheritage and 33 were not. Regarding race, 41

identified as White, 7 as Black/African Amer- assessed with questions used in earlier studies of LGB youth suicide (D’Augelli et al., 2005;ican, 3 more than one race, 2 American In-

dian, 1 Asian, and 1 did not provide informa- D’Augelli & Hershberger, 1993; D’Augelli, Hershberger, & Pilkington, 2001). Additionaltion on race. Twenty MTF and 21 FTM youth

identified as White. Twenty-nine youth were questions were asked determine the serious- ness of reported suicide attempts as recom-attending school, with 22 in college and 7 in

high school. Three had graduated from high mended by O’Carroll et al. (1996). Youth were asked: “Have you ever actually tried toschool, 21 had completed various high school

grades, and 2 did not report their levels of kill yourself?” and “Was this attempt related to your being transgender?” Because foureducation as they could not identify the levels

because of repeated interruptions of their ed- youth reported multiple suicide attempts (ranging from 2 to 20) and detailed question-ucation due to periods of prolonged absences

from school. ing about each would have been prohibitive, focused inquiry for those youth was conductedFour fifths of the youth (79%) came

from two-person households. Approximately about the suicide attempt during which youth said they were most intent on takingthree fourths (42) were raised by their bio-

Grossman and D’Augelli 531

their own lives. With regard to this attempt, gency medical treatment), and 7 (very close to death upon discovery with intervention or fortu-they were asked, “What exactly did you do?”

Their responses were classified into 10 meth- itous circumstances saving child’s life). Following Cairns et al., life-threatening actions oc-ods: firearms, hanging, jumping, drowning,

stabbing, carbon monoxide, overdose, slash- curred when the actions were rated 3.0 or above (i.e., “not serious” or “serious”). In ad-ing/cutting, poisoning, and other. The youth

were also asked, “Where did you to it?” dition, youth were asked about their serious- ness in wanting to die. The question was:“When you did this had you been drinking

alcohol?” “What about any drugs?” and “Do you think you really wanted to die? Would you say definitely yes, yes, no, or defi-“When you made the attempt how likely did

you think it was that you would die?” The nitely no?” A review of the protocol material describing the suicide attempt was later con-youth were also asked if they wrote a suicide

note or if they did anything else because they ducted by another project staff member and given a second lethality rating. Discrepanciesbelieved that they might die, such as give

away possessions or say good bye to family between interviewer ratings and the second ratings were resolved by the first author, aor friends. This question would assist in jud-

ging the seriousness of youth’s intent to die. licensed master social worker with many years of clinical experience with youth; thereTo further assess the relation between

suicide attempts and the youths’ transgender was 64% agreement between the interviewer rating and the final rating. Youth were cate-identity status, the youth were asked to re-

spond to three statements from the Revised gorized as having engaged in life-threatening behavior or not.Homosexuality Attitude Inventory (RHAI;

Shidlo, 1994) related to life-threatening think- The youth who engaged in life-threat- ening behaviors were asked if they ever dis-ing and behavior: (1) “There have been times

when I’ve felt so rotten about being LGBT cussed any emotional problems with a coun- selor, psychologist, psychiatrist, social worker,that I wanted to be dead,” (2) “I have tried

killing myself because it seemed that my life or minister (yes/no); if yes, they were asked at what age they first saw this person and theas a LGBT person was too miserable to

bear,” and (3) “I have tried killing myself be- presenting problem(s). A question also as- sessed if they had ever been hospitalized be-cause I couldn’t accept my being transgen-

der.” The RHAI is answered on a 4-point cause of emotional problems, not including substance abuse (yes/no); if yes, they werescale from 0 (Strongly Disagree) to 3 (Strongly

Agree). These three items formed an index of asked their age, the length(s) of stay, and the presenting problem(s).transgender-related suicide negativity. Cron-

bach’s alpha for this index was .89. Childhood Gender Nonconformity. The participants completed a modified version ofThe lethality of the reported suicide

attempt was evaluated by the interviewer the Gender Conformity Scale (Hockenberry & Billingham, 1987) previously used byduring the interview using the lethality rating

scale developed by Cairns, Peterson, and D’Augelli, Hershberger, and Pilkington (2002) in their examination of gender atypicalityNeckerman (1988). The 7-point scale of sui-

cide attempts ranged from verbal threats to among lesbian, gay, and bisexual youth. The scale contains 16 items reflecting childhoodnear death. Specifically, the points were de-

fined as: 1 (verbal threat or ideation with no evi- frequency of thinking or acting in a manner typically associated with males and femalesdence of actual attempt), 2 (action leading to some

self-injury with suicidal intent), 3 (self-injurious (sample items: “I like rough-and-tumble play,” “I like dolls,” “I preferred boys’action with potentially serious physical consequences

but not life-threatening), 4 (potentially life threat- games”). Participants indicated the extent to which each item described them when theyening action), 5 (seriously life-threatening action,

often requiring medical treatment), 6 (critical were under 13 years of age, with response options ranging from 0 (never) to 6 (always).life-threatening action that required rapid emer-

532 Transgender Youth

This scale is a reliable measure of gender .85 for BE–Appearance, .87 for BE–Weight, and .79 for BE–Attribution.nonconformity; Hockenberry and Billingham

reported reliabilities of .89 to .91 for differ- ent versions of the measure.

Childhood Parental Abuse. Parental psy- RESULTS chological abuse was measured with seven verbal abuse and six physical abuse items Descriptive Findings from the Child and Adolescent Psychological Abuse Measure (Briere & Runtz, 1990). The Twenty-five of the transgender youth

(45% of the 55 youth in the study) seriouslyyouth were asked how often seven kinds of verbal abuse occurred when they were grow- thought about taking their lives, and 30

(55%) never had such thoughts. While 11ing up as a child under 13 years of age. The question was, “Verbal fights and arguments (20%) reported sometimes or often having seri-

ous thoughts of taking their lives, 14 (26%)can range from quiet disagreements to yell- ing, insulting, and more severe behaviors. reported that they had rarely such thoughts.

Approximately the same number of MTFWhen you were growing up, how often, if ever, did the following happen?” (Sample (n = 12) and FTM (n = 13) youth reported

that they sometimes or often having seriouslyitems: “yelled at,” “made you feel like a bad person”). With regard to the six kinds of thought of taking their lives. One-half of the

25 youth who seriously thought of takingphysical abuse, the question was, “Sometimes physical blows or violence occur between their lives (n = 12) said that the thoughts

were somewhat or very related to their beingparents and their children. When you were growing up, at the worst point, how often, transgender, with more MTF youth (n = 7)

than FTM youth (n = 5) relating theif ever, did the following happen?” (Sample items: “slapped you,” “beat you,” “kicked thoughts to their transgender identity. Of the

25 youth who ever thought seriously of tak-you”). For both sets of questions, youth an- swered with four options: 0 (never), 1 (rarely), ing their own lives, 8 (5 MTF and 3 FTM)

seriously thought of taking their lives within2 (sometimes), and 3 (often). Cronbach’s alphas were .92 for the verbal abuse subscale and .93 the last year.

Fourteen (26%) youth reported a his-for the physical abuse subscale. Body Esteem. Many transgender peo- tory of life-threatening behaviors (i.e., a sui-

cide attempt), 6 MTF and 8 FTM. Ten youthple have feelings of estrangement from the bodies of their birth sex; commonly expressed reported one attempt, three reported two at-

tempts, and one reported 20 attempts. Theas feeling trapped in the wrong body. Body esteem was assessed using the Body-Esteem ages of the youths’ suicide attempts ranged

from 10 to 17; with half of the youth first(BE) Scale for Adolescents and Adults (Men- delson, Mendelson, & White, 2001). The BE attempting suicide at ages 15 or 16. All youth

reporting a suicide attempt said that at leastScale has three subscales: BE–Appearance (i.e., general feelings about appearance), BE– one of those attempts related to their being

transgender. Ten of the 14 youth reportedWeight (i.e., satisfaction with one’s weight), and BE–Attribution (i.e., others’ evaluation that the first attempt related to their gender

identity. These findings were supported byof one’s body and appearance). The BE– Appearance contains 10 items (sample item: their responses to the statements of transgen-

der-related suicide negativity index. Six (3“I like what I see when I look in the mirror”), the BE–Weight has 8 items (sample item: “I each MTF and FTM) youth reported that

they mainly agree or strongly agree with theam satisfied with my weight”), and the BE– Attribution contains 4 items (sample item: statement that they tried to kill themselves

because they could not accept their being“People my own age like my looks”). The scales are scored on a 5-point scale from 1 transgender, while 10 (5 each MTF and

FTM) of the youth gave the same responses(never) to 4 (always). Cronbach’s alphas were

Grossman and D’Augelli 533

saying that there had been times that they 1.5) and they stayed an average of 21 days. Not surprisingly, the presenting problemsfelt so badly about being LGBT that they

wanted to be dead. Finally, 8 youth (5 MTF were “suicidal thinking” and “self-injury.” With regard to the likelihood of deathand 3 FTM) indicated that they mainly agree

or strongly agree with the statement that they as a result of the suicide attempt, three said it was likely, seven were unsure, and four re-tried killing themselves because they felt that

their life as an LGBT person was difficult. ported death to be unlikely. This finding cor- responds to their reports about writing sui-Regarding the methods used for the

suicide attempts in which the youth were cide notes, giving away possessions, or saying good-bye to family or friends, as only threemost set on taking their lives, the most fre-

quent methods were drug overdose (n = 6), youth responded affirmatively to this ques- tion. There were no differences betweenand slashing/cutting (n = 5). Two reported

hanging (n = 2) and one poisoning (n = 1). MTF and FTM youth with regard to plan- ning the suicide attempt, drinking or usingThe MTF youth reported using all four at-

tempt methods, while FTM reported only drugs when making the attempt, intending to die, writing a suicide note, or being admittedusing overdosing and slashing/cutting. Ten

youth reported that these attempts took place for psychiatric treatment. In response to the question, “Do youat home; three said on the street or in their

neighborhoods; and one indicated an “other” think you really wanted to die?” three youth said definitely yes, three said yes, five said no,setting. Five youth made the attempt when

they knew someone was in the vicinity that and three said definitely no. However, the le- thality ratings indicated that attempts of sixcould stop them, while nine indicated no one

was nearby. When asked how long they had youth were “not serious” and eight were “se- rious.” Consequently, the lethality of the at-been planning the suicide attempt, three

youth reported no prior planning, seven re- tempts of two more youth than those who indicated that they really wanted to die wasported 1 day, three reported 7 days, and one

did not respond. Thirteen youth reported “serious” or “very serious.” There were no significant differences between MTF andthat they had not been drinking or using

drugs when they made the attempt, while FTM with regard to the lethality of their sui- cide attempts.one reported using drugs and being intoxi-

cated on alcohol or drugs. Although the lethality ratings indi- cated eight youth were “serious” about theirWhen asked if they ever discussed

emotional problems with a counselor, psy- attempts being life-threatening and six were “not serious,” comparisons between the twochologist, social worker, or minister, 12 of

the youth who engaged in life-threatening groups on the variables of interest were non- significant. Therefore, they were combinedbehaviors answered in the affirmative; two

youth indicated that they did not. Those who as “attempters” (n = 14), and they were com- pared to the “non-attempters” (n = 41) to de-did were a mean age of 13 years (SD = 4/6)

when they first saw this person. The most termine if those variables could distinguish those two groups.frequent presenting problems were “sexual

orientation concerns” and “family problems.” Other problems identified by the youth were: Comparisons of Youth Who Did

and Did Not Attempt Suicidedepression, mood disorder, and bipolar disor- der; one half of the youth listed two prob- lems. Additionally, 5 of the 12 youth indi- A multivariate analysis of variance

(MANOVA) was performed to test for differ-cated that they had been hospitalized because of emotional problems (not including sub- ences across youth suicide attempter groups

(i.e., attempted suicide yes or no) on the fol-stance abuse); 3 of the 5 youth were admitted to a psychiatric facility after a suicide at- lowing seven psychological variables: child-

hood gender nonconformity, transgender-tempt. They were a mean age of 15 (SD =

534 Transgender Youth

related suicide negativity, parental verbal sulted, made to feel guilty, and embarrassed in front of others. From 13% to 36% of theabuse, parental physical abuse, body es-

teem—appearance, body—esteem weight, youth reported “sometimes” or “often” being physically abused by their parents related toand body esteem—attribution. The signifi-

cant MANOVA, λ = .55, F = 5.20 (1, 53) p < their gender expression on each of the six items. More than 25% reported being.001, was followed up with seven univariate F

tests. Study results indicated that five of these slapped, beat, or hit very hard, and from 13% to 20% reported being punched, kicked, andpsychological variables were related to sui-

cide attempt status. The results of these anal- pushed very hard. Significant differences were found between those who attemptedyses are presented in Table 1.

Suicide Attempts and Childhood Gender suicide and those who did not with regard to verbal abuse and physical abuse; attemptersConformity. We examined whether or not

the childhood gender conformity differed be- reporting more verbal and physical abuse by their parents than non-attempters.tween suicide attempters and nonsuicide at-

tempter. A significant difference was not found. Body Esteem. The means for the three body esteem scales were all mid-range on theTransgender-Related Suicide Negativity.

As can be deduced from the results discussed 5-point scale: BE–Appearance (M = 3.3, SD = .82), BE–Weight (M = 3.4, SD = .91), andearlier, the suicide attempters, on average,

“mainly agreed” or “strongly agreed” with BE–Attribution (M = 3.4, SD = .86). Signifi- cant differences were found between suicidethe three items on the transgender-related

suicide negativity index. A significant differ- attempters and non-attempters with regard to BE–Weight and BE–Attribution, but notence was found on the index of transgender-

related suicide negativity between those in relation to BE–Appearance: those youth who attempted suicide were less satisfied withyouth who attempted suicide and those who

had not attempted suicide. their weight and others disliked their bodies more than those who had not attempted sui-Parents’ Reactions to Atypical Gender Ex-

pression. From 35% to 73% of the youth re- cide. ported “sometimes” or “often” being verbally abused by their parents related to their gen- der expression on each of the seven items. DISCUSSION The largest percentages reported being yelled at or criticized; however, approximately 50% The findings of this study provide evi-

dence that transgender youth, whether MTFreported “sometimes” or “often” being in-

TABLE 1 Means and Standard Deviations for Psychological Variables for Transgender Suicide Attempters and Nonattempters (N = 55)

Attempters Nonattempters (N = 14) (N = 41)

Variable M SD M SD F

Childhood gender nonconformity 3.58 1.47 3.94 ns Transgender-related suicide negativity 1.10 .99 .20 .53 18.13*** Parental verbal abuse 1.60 .85 1.09 .71 4.86* Parental physical abuse 1.42 .99 .73 .63 8.90* Body esteem—appearance 3.01 .88 3.34 .79 ns Body esteem—weight 2.92 1.11 3.54 .80 5.05* Body esteem—attribution 3.11 .95 3.68 .77 4.89*

*p < .05; ***p < .001.

Grossman and D’Augelli 535

or FTM, are at risk for suicidal ideation and damental difference between males and fe- males, there are also systems of social ruleslife-threatening behaviors. Almost half of the

transgender youth in the study thought seri- and customs concerning what males and fe- males are supposed to be and say (Perrin,ously of taking their lives, and half of those

related these thoughts to their transgender 2002). When youth transition from their as- signed birth sex, they contradict many ofidentity. One quarter reported a suicide at-

tempt, with almost three quarters of those these rules and customs. They may also indi- cate that the physiology which led to desig-youth relating their first or only suicide at-

tempt to their transgender identity, while the nating them one gender or the other was not accurate. Consequently, it is important toremaining youth attributed subsequent at-

tempts to their being transgender. This pro- learn about transgender youths’ feelings about their bodies and how they think othersportion of sexual minority youth is larger

than the proportion of LGB youth found by evaluate their bodies. The findings indicate significant associations between suicide at-D’Augelli et al. (2005) who attempted suicide

and attributed their attempts to their sexual tempts and two aspects of body esteem, weight satisfaction and others’ evaluation oforientation.

Five youth said that they had been ad- one’s body and appearance. In his discussion of transgender people and their bodies,mitted to psychiatric hospitals after a suicide

attempt. In relation to the lethality of the sui- Green (2004, p. 90) stated, “Most of us are not seeking perfection when measuredcide attempts in which the youth were most

intent on ending their lives, six were rated as against external stereotypes; rather, most of us are seeking an internal sense of comfortnot serious, and eight were rated as serious

or very serious. These results support find- when measured against our own sense of our- selves.” In others words, transgender peopleings from a qualitative study using focus

groups in which transgender youth described endeavor to change their bodies so that they can be perceived by others as the males andthemselves as vulnerable (i.e., having no

comfort or safety zones), a situation which females they consider themselves to be. Most transgender youth do not have the access andput them at risk for suicide (Grossman &

D’Augelli, 2006). resources needed to change their bodies so that they are pleasing to themselves and toNonheterosexual youth are stigma-

tized in most cultures. While there are differ- others. The findings indicate that there is aent responses to gay, lesbian, bisexual, and

transgender people, all of these groups expe- need for a variety of interventions with both transgender youth and their parents. Amongrience negative judgments and discrimination

(Perrin, 2002), and the most vulnerable lack these should be: (1) educational programs for parents and other guardians about theirfamily and peer support systems (Grossman

& Kerner, 1998; van Wormer et al., 2000). transgender children and the negative out- comes of psychological abuse, verbal or phys-In comparing the suicide attempters with the

non-attempters in this study, the investiga- ical; (2) psycho-educational programs for transgender youth about approaches to chang-tors found that the youth who attempted sui-

cide experienced more verbal and physical ing their bodies incrementally so that they gain higher body esteem by knowing thatabuse from their parents that those who did

not. These findings are consistent with those they will be able to facilitate changes over time; (3) intervention programs for transgen-of Proctor and Groze (1994) who studied 221

self-identified LGB youth. In comparing der youth who have personal conflict and dis- tress related to their transgender identity tothose who attempted suicide with those who

did not, Proctor and Groze found significant help them cope with the stress of living as a transgender person; (4) training programs fordifferences in family relations, peer relations,

school performance, and self-perception to mental health professionals (e.g., counselors, social workers, psychologists, psychiatrists)be the most salient.

Although physiology is the most fun- so that they can work with transgender

536 Transgender Youth

youth, and their parents and guardians to re- nize and treat mental health disorders, such as bipolar disorder and major depression,duce the youths’ psychological distress and

decrease the likelihood that youth will en- specifically associated with individuals who attempt or complete suicide (Moscicki, 1997).gage in life-threatening behaviors; and (5)

training mental health professions to recog-

REFERENCES

Burgess, C. (1999). Internal and external of lesbian, gay, and bisexual youths in high school. School Psychology Quarterly, 17, 148–167.stress factors associated with the identity develop-

ment of transgender youth. In G. Mallon (Ed.), Fitzpatrick, K. K., Euton, S. J., Jones, J. N., & Schmidt, N. B. (2005). Gender role, sex-Social services with transgendered youth (pp. 35–47).

New York: Harrington Park Press. ual orientation and suicide risk. Journal of Affective Disorders, 87, 35–42.Briere, J., & Runtz, M. (1990). Differen-

tial adult symptomatology associated with three Green, J. (2004). Becoming a visible man. Nashville: Vanderbilt University Press.types of child abuse histories. Child Abuse and Ne-

glect, 14, 357–364. Grossman, A. H., & D’Augelli, A. R. (2006). Transgender youth: Invisible and vulnera-Cairns, R. B., Peterson, G., & Necker-

man, H. J. (1988). Suicidal behavior in aggressive ble. Journal of Homosexuality, 51, 111–128. Grossman, A. H., D’Augelli, A. R., How-adolescents. Journal of Clinical Child Psychology, 17,

298–309. ell, T. J., & Hubbard, S. (2005). Parents’ reac- tions to transgender youths’ gender nonconform-Centers for Disease Control and Pre-

vention. (2002). 10 Leading Cause of Death by Age ing expression and identity. Journal of Gay & Lesbian Social Services, 18, 3–16.Group—2002. Atlanta, GA: Office of Statistics and

Programming, National Center for Injury Pre- Grossman, A. H., D’Augelli, A. R., & Salter, N. P. (2006). Male-to-female transgendervention and Control, CDC.

Centers for Disease Control and Pre- youth: Gender expression milestones, gender atypicality, victimization, and parents’ responses.vention. (2004, May 21). Surveillance summaries:

Youth Risk Behavior Surveillance—United States, Journal of GLBT Family Studies, 2, 71–92. Grossman, A. H., & Kerner, M. S. (1998).2003, MMWR, 53(No. SS-2).

Connor, J. M., & Serbin, L. A. (1977). Self-esteems and supportiveness as predictors of emotional distress in gay males and lesbian youth.Behaviorally-based masculine and feminine activ-

ity-preference scales for preschoolers. Child Devel- Journal of Homosexuality, 35, 25–29. Hockenberry, S. L., & Billingham, R. E.opment, 48, 1411–1416.

D’Augelli, A. R., & Grossman, A. H. (1987). Sexual orientation and boyhood gender conformity: Development of the Boyhood Gender(2006). Researching lesbian, gay, and bisexual

youth: Conceptual, practical and ethical consider- Conformity Scale. Archives of Sexual Behavior, 16, 475–492.ations. Journal of Gay & Lesbian Issues in Education,

3(2/3), 35–56. Lev, A. I. (2004). Transgender emergence: Therapeutic guidelines for working with gender-vari-D’Augelli, A. R., Grossman, A. H.,

Salter, N. P., Vasey, J. J., Starks, M. S., & Sin- ant people and their families. New York: Haworth Clinical Practice Press.clair, K. O. (2005). Predicting the suicide at-

tempts of lesbian, gay, and bisexual youth. Suicide Marcus, D., & Overton, W. F. (1978). The development of cognitive gender constancyand Life-Threatening Behavior, 35, 646–660.

D’Augelli, A. R., & Hershberger, S. L. and sex role preferences. Child Development, 49, 434–444.(1993). Lesbian, gay, and bisexual youth in com-

munity settings: Personal challenges and mental McDaniels, J. S., Purcell, J., & D’Aug- elli, A. R. (2001). The relationship betweenhealth problems. American Journal of Community

Psychology, 21, 421–448. sexual orientation and risk for suicide: Research findings and future directions for research andD’Augelli, A. R., Hershberger, S. L.,

Pilkington, N. W. (2001). Suicidality patterns prevention. Suicide and Life-Threatening Behavior, 31, 84–105.and sexual orientation-related factors among les-

bian, gay, and bisexual youths. Suicide and Life- Mendelson, B. K., Mendelson, M. J., & White, D. R. (2004). Body-Esteem Scale for Ad-Threatening Behavior, 31, 250–264.

D’Augelli, A. R., Hershberger, S. L., & olescents and Adults. Journal of Personality Assess- ment, 76, 90–106.Pilkington, N. W. (2002). Incidence and mental

health impact of sexual orientation victimization Moscick, E. K. (1997). Identification of

Grossman and D’Augelli 537

suicide risk factors using epidemiology studies. bian and gay youth: Care and counseling. New York: Columbia University Press.Psychiatric Clinics of North America, 20, 499–517.

O’Carroll, P. W., Berman, A. L., Maris, Savin-Williams, R. C., & Cohen, K. M. (1996). The lives of lesbians, gays, and bisexuals: Chil-R. W., Moscicki, E. K., Tanney, B.L., & Silver-

man, M. (1996). Beyond the Tower of Babel: A dren to adults. Forth Worth, TX: Harcourt Brace. Sears. J. T. (Ed.). (2005). Gay, lesbian, andnomenclature for suicidology. Suicide and Life-

Threatening Behavior, 26, 237–252. transgender issues in education: Programs, practices, policies, and practices. New York: Harrington ParkPaley, V. G. (1984). Boys and girls: Super-

heroes in the doll corner. Chicago: University of Press. Shidlo, A. (1994). Internalized homopho-Chicago Press.

Perrin, E. C. (2002). Sexual orientation in bia: Conceptual and empirical issues in measure- ment. In B. Greene & G. M. Herek (Eds.), Lesbianchild and adolescent health care. New York: Kluwer

Academic/Plenum. and gay psychology: Theory, research, and clinical ap- plications (pp. 176–205). Thousand Oaks, CA:Proctor, C., & Groze, V. (1994). Risk

factors for suicide among gay, lesbian, and bisex- Sage. van Wormer, K., Wells, J., & Boes, M.ual youths. Social Work, 39, 504–512.

Remafedi, G. (Ed.). (1994). Death by denial: (2000). Social work with lesbians, gays, and bisexuals: A strengths perspective. Boston, MA: Allyn andStudies of suicide in gay and lesbian teenagers. Boston:

Alyson Publications. Bacon. Way, N. (1998). Everyday courage: The livesRussell, S. T. (1993). Sexual minority

youth and suicide risk. American Behavioral Scien- and stories of urban teenagers. New York: New York University Press.tist, 46, 1241–1257.

Russell, S. T. (2003). Sexual minority youth and suicide risk. American Behavioral Scien- tist, 46, 1241–1257. Manuscript Received: July 3, 2006

Revision Accepted: December 30, 2006Ryan, C., & Futterman, D. (1998). Les-