Portal de Periódicos da CAPES

Gender Identity and Psychosexual Disorders

2005; Volume: 3; Issue: 4 Linguagem: Inglês

10.1176/foc.3.4.598

1541-4108

Kenneth J. Zucker, Susan J. Bradley,

Gender Studies in Language

Back to table of contents Previous article Next article INFLUENTIAL PUBLICATIONSFull AccessGender Identity and Psychosexual DisordersKenneth J. Zucker, Ph.D., and Susan J. Bradley, M.D.Kenneth J. ZuckerSearch for more papers by this author, Ph.D., and Susan J. BradleySearch for more papers by this author, M.D.Published Online:1 Oct 2005https://doi.org/10.1176/foc.3.4.598AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack Citations ShareShare onFacebookTwitterLinked InEmail This chapter provides an overview of gender identity and psychosexual problems in children and adolescents. Three terms—gender identity, gender role, and sexual orientation—are useful in organizing a conceptual framework in thinking about these issues.Gender identity refers to a person’s basic sense of self as male or female. It includes both the awareness that one is male or female and an affective appraisal of such knowledge. In the clinical literature, the term gender dysphoria has been used to characterize a person’s sense of discomfort or unease about his or her status as male or female. Based on clinical work with children born with physical intersexual conditions, Money et al. (1957) concluded that gender identity typically appears in its nascent form between ages 2 and 3 years. During the past four decades, this original clinical observation has been buttressed by normative empirical studies, which have demonstrated that children in this age range—typically by about 30–36 months—are able to categorize people by sex on the basis of phenotypic social cues, such as clothing and hairstyle (for reviews, see Ruble and Martin 1998; Zucker et al. 1999). According to some researchers, this ability likely precedes the ability to self-categorize oneself as a boy or a girl.Gender role refers to a person’s behavioral adoption of cultural markers of masculinity and femininity. In children, gender role preference can be measured in a variety of ways, including playmate affiliations, toy interests, role and fantasy play, and endorsement of various personality attributes. Conceptually, one can consider a child to be primarily masculine or feminine based on the pattern of gender role behavior. Alternatively, one can view a child as both masculine and feminine (androgynous) or as neither masculine nor feminine (undifferentiated). Over the past 50 years, many studies have been conducted on gender role behavior in children (Ruble and Martin 1998). Almost without exception, these studies have shown that boys and girls differ significantly with regard to several sex-typed attributes, including toy and fantasy play (Fagot 1977), peer affiliation preference (Maccoby 1998; Maccoby and Jacklin 1987), aggression (Maccoby and Jacklin 1974, 1980), activity level (Benenson et al. 1997; Campbell and Eaton 1999; Eaton and Enns 1986), and rough-and-tumble play (DiPietro 1981).Sexual orientation refers to the pattern of a person’s erotic responsiveness. Heterosexual, bisexual, and homosexual are the three sexual orientations most commonly described by contemporary nosologists in sexology, although it is also important to consider the age of the sexual partner (either in fantasy or in behavior), as in heterosexual or homosexual pedophilia. Gender identity and gender role are typically viewed as developing before the emergence of sexual orientation, although this view is not held universally (Isay 1989).Diagnostic criteriaIn DSM-IV-TR (American Psychiatric Association 2000), four diagnoses are of relevance with regard to gender identity and psychosexual problems during childhood and adolescence:1. Gender identity disorder (GID) (Table 1)2. Gender identity disorder not otherwise specified (GIDNOS) (Table 2)3. Transvestic fetishism (Table 3)4. Sexual disorder not otherwise specified (Table 4)During childhood, only the first two of these four diagnoses are of relevance. In adolescence, however, all of these diagnoses may be utilized as gender identity and psychosexual concerns become more differentiated, in part because of the increased salience of erotic behavior. Therefore, in addition to concerns about gender identity proper (which can be diagnosed with either GID or its residual diagnosis, GIDNOS), the clinician must also be attentive to paraphilic behavior (transvestic fetishism) or distress regarding one’s sexual orientation, most typically in homosexuality (sexual disorder not otherwise specified).Only one study has formally evaluated the reliability of the GID diagnosis in children, which was based on the criteria from DSM-III (American Psychiatric Association 1980). Based on chart data from 36 consecutive referrals to a child and adolescent gender identity clinic, Zucker et al. (1984) found high agreement with DSM-III criteria for GID. Validity evidence was also found: the children who met the complete diagnostic criteria were, on average, more cross-gendered in their behavior than were the children who did not meet the complete diagnostic criteria. Subsequent analyses with larger numbers of children have verified this distinction (Zucker and Bradley 1995; Zucker et al. 1992, 1993b).Among demographic variables, age of the child at assessment has been most consistently associated with the GID diagnosis. Based on 488 consecutive referrals from two specialized gender identity clinics (one in Toronto, Ontario, Canada, and the other in Utrecht, the Netherlands), Cohen-Kettenis et al. (2003) found that the children who met the complete criteria for GID (mean age, 6.8 years) were significantly younger than the children who were subthreshold (mean age, 8.6 years). Elsewhere it was shown that children who met the complete diagnostic criteria were from a higher social class and were more likely to come from an “intact” two-parent family. Sex of the child and IQ were not associated with the presence or absence of the complete diagnostic criteria for GID (Zucker and Bradley 1995).To test which variables contributed to the correct classification of the children in the two diagnostic groups, a discriminant-function analysis was performed (Zucker and Bradley 1995). Age, sex, IQ, and parents’ marital status contributed to the discriminant function, with age showing the greatest power. In the DSM-III group, 82.6% were correctly classified, and in the non-DSM-III group, 68.8% were correctly classified.The data regarding age appear to be related to older children’s tendency not to verbalize the wish to be of the opposite sex, which was a distinct criterion for GID in both DSM-III and DSM-III-R (American Psychiatric Association 1987). Clinical evidence may suggest continued discomfort with gender identity issues, but the older child may be more aware of social convention and thus may not verbalize concerns, at least during an initial diagnostic assessment.These findings led to the question of whether it was appropriate to retain the wish to be the opposite sex as a distinct criterion in DSM-IV (Bradley et al. 1991; Zucker et al. 1998). The DSM-IV Subcommittee on Gender Identity Disorders recommended that the verbalized wish to be the opposite sex should be collapsed with several other behavioral criteria to index the child’s cross-gender identification (see Criterion A in Table 1) (Bradley et al. 1991). In a reanalysis of an existing data set, Zucker et al. (1998) showed that this change somewhat reduced the influence of age on the likelihood of a child meeting the requirements in Criterion A.Table 1 also shows Criterion B for the diagnosis of GID: the child’s sense of discomfort about his or her sex and the sense of inappropriateness in the gender role of that sex. At present, formal studies to fully evaluate the reliability and validity of the modified GID diagnosis for children have not been conducted.Perhaps the most substantive change in the diagnostic criteria for GID in DSM-IV and DSM-IV-TR pertains to the inclusion of age-related criteria so that the diagnosis can be applied to all phases of the life cycle. This change resulted in the elimination of two diagnoses from DSM-III-R: 1) transsexualism and 2) gender identity disorder of adolescence or adulthood, nontranssexual type.From a clinical point of view, the establishment of the GID diagnosis in adolescents requires that one pay close attention to the descriptors strong and persistent (Criterion A) and persistent discomfort (Criterion B). Unfortunately, no gold standard is available to make this kind of clinical judgment and, to date, there has been little in the way of systematic empirical research regarding the reliability and validity of the diagnosis in adolescents (see, however, Smith et al. 2001; Zucker and Bradley 1995).No formal empirical studies have been conducted to examine the reliability and validity of the diagnosis of transvestic fetishism in adolescents; however, the DSM-IV-TR diagnostic criteria for transvestic fetishism (see Table 3) are sufficiently loose to make their use with adolescents not particularly difficult (Zucker and Blanchard 1997; Zucker and Bradley 1995). Clinically, one encounters adolescents who display a range of fetishistic cross-dressing (e.g., from wearing women’s underwear while masturbating to complete cross-dressing accompanied by erotic arousal). Conceptually, clinicians should be interested in whether such behavior represents an erotic preference (Zucker and Blanchard 1997). With adolescents, this is not always clear, because many adolescent boys also display considerable heterosexual arousal and interaction without the use of feminine apparel. However, DSM-IV-TR criteria for transvestic fetishism do not address the issue of erotic preference.In DSM-IV-TR, the term autogynephilia was introduced as a new descriptive construct, giving recognition to its increased clinical use in understanding the linkage between transvestic fetishism and GID. The term autogynephilia was constructed from Greek roots meaning “love of oneself as a woman” and has been defined as a male’s propensity to be sexually aroused by the thought or image of himself as a female (for review, see Zucker and Blanchard 1997). It is most commonly the case that transvestic adolescents who experience autogynephilic feelings and fantasies are the ones who will present with the request for sex-reassignment surgery.Clinical findingsPhenomenologyChildren with GID present with a rather coherent set of behavioral signs. In a boy, these signs include making verbal statements that he is, or would like to be, a girl; cross-dressing in girls’ or women’s clothing; having a preference for culturally stereotypical feminine toys and activities; emulating females in fantasy play; having a preference for girls as playmates; expressing dislike of his sexual anatomy (e.g., concealing his penis, sitting to urinate in order to embellish the fantasy of having female genitalia); and being averse to rough-and-tumble play and group sports. In a girl, the inverse is observed. Of particular note is the intense aversion to culturally stereotypical girls’ clothing and the desire to have her hairstyle look like that of a boy, such that a naive observer would perceive her as male. Taken together, these characteristics point to the child’s very strong cross-gender identification, and several research studies have established their discriminant validity or relative uniqueness (Green 1974, 1987; Zucker 1985, 1992; Zucker and Bradley 1995).Apart from the features of GID described in the formal DSM diagnostic criteria, there are other aspects that have been noted clinically and empirically. For example, many clinicians have observed how intense and rigid cross-gender identification can be in these youngsters (Coates 1990, 1997; Coates and Wolfe 1995, 1997), and sometimes this extends into domains that surprise even the most sophisticated practitioner. One example of this pertains to sex-typed color preferences, which some gender theorists might characterize as an overelaborated gender schema (Ruble and Martin 1998). Clinically, one is often impressed by the rigidity of color choices by children with GID. For example, the boys invariably prefer pink and purple, as exemplified, perhaps, in the critically acclaimed film Ma Vie en Rose (My Life in Pink), about a boy with GID (Kline 1998). One can observe this in the type of clothing that they request and in their drawings, which are often of idealized, beautiful, and benevolent females. Girls with GID invariably reject colors like pink and purple and will often prefer dark colors, such as blue and black. About 10% of boys with GID appear preoccupied not with benevolent females but with malevolent ones: they endlessly draw pictures of angry, hostile females, such as the Wicked Witch of the West, Cruella De Vil from 101 Dalmatians, or Ursula from The Little Mermaid.The study of the child’s actual sense of gender identity has been somewhat more complicated. Most children with GID “know” that they are male or female; that is, if they are asked the question “Are you a boy or a girl?” they answer correctly (Zucker et al. 1993b). Some of these youngsters, however, do not seem to know the answer or else appear to be confused about their gender status (i.e., whether they are male or female). In part, this may be a developmental phenomenon, but it also may be a sign of the severity of the overall condition (Stoller 1968a). Indeed, Zucker et al. (1999) reported evidence for a developmental lag in acquisition of gender constancy among children with GID.The child’s internal representation of gender has been more difficult to study, although clinical experience suggests that the stability of the gender sense can be quite labile in some children. Zucker et al. (1993b) reported the results of a gender identity interview schedule that identified two factors, tentatively labeled cognitive gender confusion and affective gender confusion, both of which discriminated children referred for gender identity problems from clinical and nonclinical control subjects. The following responses were given by an 8-year-old boy (IQ, 133) who, by parent report, met DSM-IV criteria for GID:Interviewer (I): Are you a boy or a girl?Child (C): Both.I: Are you a girl?C: Kind of.I: When you grow up, will you be a mom or a dad?C: Don’t know.I: Could you ever grow up to be a mom?C: Yes.I: Are there any good things about being a boy?C: Yes.I: Tell me some of the good things about being a boy.C: Boys don’t have to have babies and get their spine ripped open . . .I: Are there any things that you don’t like about being a boy?C: Yes.I: Tell me some of the things that you don’t like about being a boy.C: . . . Everything. Please don’t make me tell.I: Do you think it is better to be a boy or a girl?C: Girl.I: Why?C: There are so many reasons. Girls are more mature . . . positive . . . better . . .I: In your mind, do you ever think that you would like to be a girl?C: Yes.I: Can you tell me why?C: Because the guys don’t accept me for what I am. . . . Girls have better bands, like the Spice Girls.I: In your mind, do you ever get mixed up and you’re not really sure if you are a boy or a girl?C: Yes.I: Tell me more about that.C: Practically all the time. I think the doctor was totally wrong and deformed me. I am a woman in a man’s body. He gave me a girl’s mind and switched brains or bodies.I: Do you ever feel more like a girl than like a boy?C: Yes.I: Tell me more about that.C: Well, I do lots of times, especially when I’m in the bath. . . . This is going way too deep. . . . Well, usually girls want me to play their games. . . . Then I get mixed up being a girl or a boy.I: You know what dreams are, right? Well, when you dream at night, are you ever in the dream?C: Yes.I: In your dreams, are you a boy, a girl, or sometimes a boy and sometimes a girl?C: Both. . . . it’s just me as a girl, me and my friends . . . find out the secret of being trapped in a boy’s body. We go to the doctor’s to find out, look into his files and find nothing about me. Then we destroy him.I: Do you ever think that you really are a girl?C: Yes.I: Tell me more about that.C: . . . Practically all the time . . . because of my feelings. . . . I want to be a lead singer of a girls’ band when I grow up.Age at onsetGreen (1976) reported that the age at onset of cross-gender behaviors in GID is typically during the preschool years. In his sample of boys, for example, 55% were cross-dressing by their third birthday, 80% were cross-dressing by their fourth birthday, and 90% were cross-dressing by their fifth birthday. Many experienced clinicians have observed that repetitive, intense cross-gender behaviors appear even before a child’s second birthday. Clinical data on girls reveal a similar age at onset (Zucker and Bradley 1995). It is important to note that among more typical children, a display of various gender role behaviors can also be observed during this period in the life cycle. This similarity suggests that the underlying mechanisms for both patterns may be the same, albeit mirror images.Associated psychopathologyIn DSM-IV-TR, it is noted that children with GID may have co-occurring behavioral difficulties, including social isolation, low self-esteem, separation anxiety, and depression. What are the data regarding the presence of other types of psychopathology in children with GID? If other forms of psychopathology are present, how is the association with GID to be understood?Unfortunately, omnibus structured interview schedules that cover the gamut of childhood psychopathology have not been utilized in this clinical population. However, several more narrowly focused empirical studies have reported on the presence of general psychopathology in both boys and girls with GID (Coates and Person 1985; Cohen-Kettenis et al. 2003; Zucker and Bradley 1995). These studies have shown that both boys and girls with GID display levels of general psychopathology similar to those of demographically matched psychiatric control subjects and levels greater than those of control subjects without psychiatric disorders. For example, children with GID have been shown to display behavior problems at a level comparable to the clinic-referred standardization sample (Coates and Person 1985; Cohen-Kettenis et al. 2003; Zucker and Bradley 1995), as measured with the Child Behavior Checklist (CBCL) (Achenbach and Edelbrock 1981), a parent-report instrument of behavior problems. In boys with GID, internalizing problems were somewhat more common than externalizing problems, a finding that is consistent with clinical observations that many of these boys experience anxiety, depression, and social withdrawal (see also Zucker et al. 1996a). More generally, it should be noted that the overall functioning of children with GID varies considerably. Some of these youngsters show pervasive behavioral difficulties and often require intensive intervention for these problems in their own right; on the other hand, some youngsters show minimal behavioral psychopathology and function quite well in the different environments of daily life.How might these associated behavioral difficulties be best understood? Zucker and Bradley (1995) provided data that showed that in boys with GID, behavioral difficulties based on CBCL measures increased with age. This was interpreted as being consistent with the influence of social ostracism, which becomes more pronounced over time. Two subsequent studies, utilizing a more direct measure of poor peer relations, confirmed this inference empirically; indeed, in a multiple regression analysis, it was the measure of poor peer relations, not age, that accounted for the most variance in predicting CBCL behavior problems in both boys and girls with GID (Cohen-Kettenis et al. 2003; Zucker et al. 2002).A study by S.R. Fridell (unpublished doctoral dissertation, 2001) provided further evidence that the cross-sex-typed behavior of boys with GID may well be related to how well they are liked by other children. Fridell created 15 age-matched experimental play groups consisting of one boy with GID and two nonreferred boys and two nonreferred girls (age range, 3–8 years). After two 60-minute play sessions, conducted a week apart, each child was asked to select their favorite playmate from the group. The nonreferred boys most often chose the other nonreferred boy as their favorite playmate, thus indicating a distinct preference over the boy with GID. The nonreferred girls chose the other girl as their favorite playmate, thus showing a relative disinterest in either the boy with GID or the two nonreferred boys.In another line of research, Zucker and Bradley (1995) showed that a composite measure of maternal psychopathology also predicted the extent of behavioral difficulties detected with the CBCL, which suggests that generic familial emotional and psychiatric factors also contribute to the degree of general psychopathology (see also Cohen-Kettenis et al. 2003; Zucker et al. 2002).Another perspective on the nature of the associated psychopathology has been advanced by Coates and Person (1985), who provided data on a high rate of separation anxiety disorder in boys with GID. These researchers argued that the high rate of separation anxiety could be accounted for by a great deal of familial psychopathology, which rendered the mothers of these boys unpredictably available. The authors claimed that the emergence of the separation anxiety actually preceded the first appearance of the feminine behavior, which was understood to serve a representational coping function of recapturing an emotionally unavailable mother.As noted elsewhere (Zucker and Green 1992), Coates and Person (1985) did not have empirical evidence available to document the putative temporal relation between separation anxiety and GID, because both diagnoses were made concurrently at the time of assessment. Rather, the temporal relation was inferred on the basis of clinical evidence. Subsequently, Zucker et al. (1996a) confirmed the high rate of co-occurring traits of separation anxiety disorder in boys with GID, and A.S. Birkenfeld-Adams (unpublished doctoral dissertation, 1999) has shown a rate of insecure attachment to the mother, but the temporal aspect of the hypothesis of Coates and Person (1985) remains to be tested.Biophysical markersBy observation, it is apparent that the vast majority of children with GID do not have abnormalities in physical sex differentiation—such as in the various physical intersexual disorders—that might, on theoretical grounds, contribute to the condition (Meyer-Bahlburg 1994; Zucker 1999c). Because sex hormone levels are so low during childhood (Sizonenko 1980), it is unlikely that a standard endocrine assessment would detect abnormalities. Green (1976) and Rekers et al. (1979) reported normal XY karyotypes in boys with GID. Green also found that the feminine boys he studied did not differ in height and weight from nonfeminine boys at the time of assessment (Roberts et al. 1987), although they were hospitalized more often before their participation in the study.However, if one starts with a sample of children or adolescents with certain physical intersexual conditions, such as genetic females with congenital adrenal hyperplasia (CAH) raised as girls, genetic males with partial androgen insensitivity syndrome raised as girls, and genetic males with cloacal exstrophy raised as girls, gender identity problems or gender dysphoria appears to be present in a subgroup of these youngsters (for reviews, see Zucker 1999c, 2002). However, these conditions are invariably already known to the clinician at the time of diagnostic assessment.Differential diagnosisDiagnostic issues in childrenSeveral diagnostic issues require consideration in relation to GID. A small number of boys engage in a type of cross-dressing that appears to be quite different from the type of cross-dressing that is part of the clinical picture in GID. In the latter, cross-dressing encompasses a range of behaviors, including the wearing of dresses, women’s shoes, and jewelry, all of which enhance the fantasy or desire to be like the opposite sex. In the former, cross-dressing is limited to the use of undergarments, such as panties and nylons. As with boys with GID, the cross-dressing has a compulsive and self-soothing flavor to it. However, it is not accompanied by other signs of cross-gender identification; in fact, apart from the cross-dressing, these boys are conventionally masculine (Zucker and Blanchard 1997). Many male adolescents and adults who have a diagnosis of transvestic fetishism recall such cross-dressing during childhood (Zucker and Bradley 1995); however, no prospective studies of prepubertal boys engaging in this form of cross-dressing have been conducted to determine what proportion, if any, of these boys develop transvestic fetishism.When all of the clinical signs of GID are present, there is little difficulty in making the diagnosis. If one accepts the idea of a spectrum of cross-gender identification, then there is more room for ambiguity, and one must be prepared to identify what Meyer-Bahlburg (1985) referred to as the “zone of transition between clinically significant cross-gender behavior and mere statistical deviation from the gender norm” (p. 682).Friedman (1988), for example, suggested that there is a subgroup of boys who are “unmasculine” but not feminine. Based on clinical experience, Friedman argued that these boys have a “persistent, profound feeling of masculine inadequacy which leads to negative valuing of the self” (p. 199). Although it is not described in the formal clinical literature, there is also probably a subgroup of girls who are “unfeminine” but not masculine and may have similar feelings. Such youngsters would not meet DSM-IV-TR criteria for GID, but the residual diagnosis of GIDNOS could be used in such cases.For girls, the main differential diagnostic issue concerns the distinction between GID and what is known in popular culture as tomboyism. According to Webster’s Ninth New Collegiate Dictionary, a tomboy is “a girl of boyish behavior.” Green et al. (1982) studied a community sample of tomboys and found that, compared with a control group of nontomboys, tomboys displayed a greater number of masculine traits, such as a preference for boys as playmates, interest in rough-and-tumble play, and play with guns and trucks. In many respects, the cross-gender behavior of such tomboys is similar to the masculine gender role preferences of girls who are referred clinically for gender identity concerns (Bailey et al. 2002; Zucker and Bradley 1995).Based on critiques of DSM-III criteria for GID in girls (Zucker 1982), DSM-III-R and DSM-IV (American Psychiatric Association 1994) criteria for girls were modified in the hope of better differentiating these two groups. Clinical experience suggests that at least three characteristics are useful in making a differential diagnosis. First, girls with GID express a profound unhappiness with their female gender status; in contrast, Green (1980) noted that his sample of tomboys were “generally content being female” (p. 262). Second, girls with GID display a marked aversion to culturally defined feminine clothing and will do their utmost to avoid having to wear it. Their refusal to wear “girls’ clothes” under any circumstance often precipitates clinical referral. Although tomboys prefer functional and casual clothing (Green et al. 1982), they do not display the same type of rigid rejection of feminine clothing. Third, girls with GID, unlike tomboys, often express discomfort with or dislike of their sexual anatomy.Diagnostic issues in adolescentsThe clinician will encounter at least four types of psychosexual problems among adolescents (Bradley and Zucker 1990; Zucker and Bradley 1995). First, clinical experience suggests that persistent cross-gender identification throughout childhood is a risk factor for the continuation of GID into adolescence and adulthood. As noted earlier, it is important to evaluate the fixedness of the desire to change sex, because therapeutic decisions will be influenced, at least to some extent, by the adolescent’s openness to consider alternatives to sex reassignment (Newman 1970). From a differential diagnostic standpoint, the residual diagnosis of GIDNOS can be used for individuals whose desire to change sex does not quite fit the criteria for GID.The second psychosexual problem that can be observed in adolescents involves individuals who have a history of GID or a subclinical variant. These adolescents show various signs of cross-gender identification but do not voice a desire to change sex. They are circumspect about their sexual orientation, so it is not possible to classify them as homosexuals. These youngsters often are referred because of continued social ostracism. Many of these adolescents are able to acknowledge distress about not “fitting in” because of their cross-gender behavior. In these cases, the residual diagnosis of GIDNOS could be used to indicate that the adolescent continues to struggle with gender identity concerns.A third type of psychosexual problem involves adolescents who have been referred because of homosexual behavior or orientation. Many of these adolescents have a history of GID or a variation of it, perhaps akin to the unmasculinity described by Friedman (1988; see also Friedman and Downey 2002; Friedman and Stern 1980). Although the reason for referral varies, it is important to rule out continuing concerns about gender identity. For adolescents distressed about their sexual orientation, the diagnosis of sexual disorder not otherwise specified can be given.The last type of psychosexual problem is, as far as we know, the exclusive domain of adolescent males: cross-dressing associated with sexual arousal. As noted earlier, the extent of the cross-dressing varies and there is no problem, in principle, in employing the diagnosi