Proposed Revisions to Gender Identity Disorder Diagnoses in the DSM-5

Archives of Sexual Behavior (2010) 39: 1253-1260

By Anne A. Lawrence, M.D., Ph.D.


In early 2010, the American Psychiatric Association (APA) announced proposed revisions to psychiatric diagnoses for the forthcoming fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). These revisions included proposed changes to Gender Identity Disorder (GID) diagnoses in adolescents and adults (APA, 2010c), as recommended by the APA's GID Subworkgroup. The Subworkgroup members laid the groundwork for their recommendations in articles published in Archives of Sexual Behavior (Cohen-Kettenis & Pfäfflin, 2010; Drescher, 2010a; Meyer-Bahlburg, 2010a; Zucker, 2010). The most important elements of the proposed revisions to GID diagnoses for adolescents and adults are:

  1. GIDs are conceptualized as reflecting a "marked incongruence between one's experienced/expressed gender and assigned gender" (APA, 2010c, Proposed Revision section), leading to the recommendation that the name Gender Incongruence (GI) replace GID.
  2. The clinical indicators for GID/GI have been revised to reflect empirical evidence, and the proposed number of indicators required for diagnosis has been specified.
  3. The presence of clinically significant distress or impairment is no longer required for the diagnosis of GID/GI but is proposed to be evaluated "separately and independently" (APA, 2010c, Rationale section, ¶ 9).
  4. Persons with a disorder of sex development (DSD) have become eligible for the diagnosis of GID/GI, and the presence or absence of a DSD is used to define subtypes.
  5. Subtypes based on sexual orientation have been eliminated.
  6. The status of the diagnosis of Gender Identity Disorder Not Otherwise Specified (GIDNOS) is uncertain, raising the possibility that the diagnosis may not be retained.

I will comment on these elements of the proposed revisions and will then offer alternative diagnostic criteria for the diagnoses of GID/GI and GIDNOS.

Revised Conceptualization of GIDs and Proposed Name Change

The proposed revisions conceptualize the defining feature of GIDs as "marked incongruence between one's experienced/expressed gender and assigned gender" (APA, 2010c, Proposed Revision section), leading to the recommendation that the disorder be renamed Gender Incongruence (GI). The DSM-IV (APA, 1994) and DSM-IV-TR (APA, 2000), in contrast, conceptualized the defining features of GID as "strong and persistent cross-gender identification" and "persistent discomfort with [one's] sex or sense of inappropriateness in the gender role of that sex" (APA, 2000, p. 581). The newly proposed conceptualization is, in effect, a reversion to that of the DSM-III-R (APA, 1987), which described GIDs as reflecting "an incongruence between assigned sex (i.e., the sex that is recorded on the birth certificate) and gender identity" (p. 71), but with the term gender identity now replaced by experienced/expressed gender and assigned sex replaced by assigned gender. I will discuss the following points:

  1. The term experienced/expressed gender is synonymous with gender identity, and the latter term is preferable, based on familiarity and clarity.
  2. Except in some persons with a DSD, the incongruence in question is between gender identity and biologic sex, and this formulation is preferable, based on familiarity and clarity.
  3. Two supposed benefits of the proposed new conceptualization are inclusion of persons with a DSD and provision of an "exit clause" for the diagnosis, but neither result would be desirable, and there would be a better way to accomplish the latter if it were desirable.
  4. The proposed name Gender Incongruence is merely a euphemism for GID (except, perhaps, in persons with a DSD), and the latter name is preferable, based on continuity, familiarity, and clarity.
  5. The phenomenon described by the proposed clinical indicators corresponds closely to the concept of gender dysphoria, suggesting that Gender Dysphoric Disorder could be an acceptable alternative name.

Experienced/expressed gender is synonymous with gender identity

In an explanatory statement, the Subworkgroup defined experienced/expressed gender as "what identity [emphasis added] one experiences and/or expresses" (APA, 2010c, Rationale section, ¶ 2). Moreover, in the Subworkgroup's document Dimensional Assessment for Gender Identity Disorder (Gender Incongruence) for Adolescents and Adults (APA, 2010c, Severity section), in which the proposed clinical indicators for GI (see Table 1) are operationalized as questionnaire items, the term gender identity replaces experienced/expressed gender in the items corresponding to indicators A1 and A2. The Subworkgroup members apparently concluded that gender identity is not only synonymous with experienced/expressed gender but is better understood by patients; I believe they were correct in their conclusion. Gender identity is a widely recognized and generally well understood term; there is little justification for replacing it with a newly invented synonym.


Table 1: GID Subworkgroup's Proposed Clinical Indicators for Gender Incongruencea

    A1. a marked incongruence between one's experienced/expressed gender and primary and/or secondary sex characteristics (or, in young adolescents, the anticipated secondary sex characteristics)

    A2. a strong desire to be rid of one's primary and/or secondary sex characteristics because of a marked incongruence with one's experienced/expressed gender (or, in young adolescents, a desire to prevent the development of the anticipated secondary sex characteristics)

    A3. a strong desire for the primary and/or secondary sex characteristics of the other gender

    A4. a strong desire to be of the other gender (or some alternative gender different from one's assigned gender)

    A5. a strong desire to be treated as the other gender (or some alternative gender different from one's assigned gender)

    A6. a strong conviction that one has the typical feelings and reactions of the other gender (or some alternative gender different from one's assigned gender)

 Note: Fulfillment of two or more indicators is required for diagnosis.
 afrom APA, 2010c, Proposed Revision section.

The incongruence in question is between gender identity and biologic sex

Except in persons with a DSD, in whom biologic sex may be ambiguous, the incongruence in question is between gender identity and biologic sex (reflected in assigned sex), not between gender identity and assigned gender, as the proposed language states. The Subworkgroup's own explanation concedes as much: "The term 'sex' has been replaced by assigned 'gender' in order to make the criteria applicable to individuals with a DSD" (APA, 2010c, Rationale section, ¶ 4). It is also notable that most of the proposed clinical indicators (see Table 1) reference sex, not gender. Indicators A1-A3 explicitly refer to dysphoria in relation to biologic sex characteristics, and a reference to biologic sex is also implied, if not explicitly stated, in indicators A4, A5, and perhaps A6. In any case, gender is not what is assigned at birth: Sex is assigned -- or, more accurately, recognized -- at birth, usually based on the morphology of the external genitalia. Birth certificates specify one's sex, not one's gender. If persons with a DSD were excluded from the diagnosis of GID/GI -- a desirable outcome, as I will argue later in this Letter -- then there could be no reasonable objection to framing the incongruence in relation to biologic sex, a formulation that is preferable on the basis of familiarity and clarity.

Framing the incongruence in relation to biologic sex would also prevent the diagnosis of GID/GI being given yet again to "transitioned individuals who have regrets" (APA, 2010c, Rationale section, ¶ 4), a possibility that the Subworkgroup members apparently consider desirable but that I regard as bizarre and confusing. In my opinion, it would make little sense to diagnose individuals with a mental disorder for wanting to live as members of their natal sex.

Two supposed benefits of the proposed conceptualization are not desirable

The Subworkgroup has offered two rationales for conceptualizing GID/GI as an incongruence between experienced/expressed gender and assigned gender: to "make the criteria applicable to individuals with a DSD" (APA, 2010c, Rationale section, ¶ 4) and to provide an "exit clause" from the diagnosis, allowing "individuals who have successfully transitioned to 'lose' the diagnosis after satisfactory treatment" (APA, 2010c, Rationale section, ¶ 4).

I will argue later in this Letter that it would be inadvisable to allow persons with a DSD to be eligible for the diagnosis of GID/GI. It would also be undesirable to provide an automatic "exit clause" from the diagnosis of GID/GI, simply on the basis of having undergone nominal gender reassignment. Individuals should "lose" the diagnosis -- to the extent that any psychiatric diagnosis can be lost -- only if they no longer experience clinically significant gender dysphoria. The proposed diagnostic criteria for GI, however, seemingly would automatically render the diagnosis inapplicable following a sought-after change in assigned gender, because the newly assigned gender would no longer be incongruent with the patient's experienced/expressed gender. Consider a gender dysphoric adult male who for medical or financial reasons was temporarily or permanently unable to undergo cross-sex hormone therapy or vaginoplasty. If this person nevertheless underwent a change in assigned gender (say, as a result of living in the desired gender role and achieving legal recognition as a member of the desired sex), her newly assigned gender would no longer be incongruent with her experienced/expressed gender, and she seemingly could no longer be diagnosed with GI, even if she were still severely gender dysphoric, due to a disparity between her gender identity and her primary and secondary sex characteristics. Surely this would not be a desirable outcome.

Moreover, to the extent that a formal diagnosis of GID/GI is necessary to justify access to clinical care, the language of the proposed criteria implies that this justification would automatically disappear once nominal gender (or sex) reassignment had occurred. Gender dysphoric patients may remain gender dysphoric and may require transition-related care for months or years after nominal gender (sex) reassignment has occurred; automatic loss of the diagnosis could jeopardize access to ongoing care.

Adding a distress/impairment criterion to the diagnosis -- which would be desirable for other reasons, as I will argue later in this Letter -- would be a more appropriate way to ensure that individuals who no longer experienced clinically significant gender dysphoria would not meet full criteria for the diagnosis and would no longer carry the diagnosis in the usual, unqualified sense. It is important to recognize, however, that individuals do not and should not necessarily lose all connection to a psychiatric diagnosis simply because they no longer meet full criteria for that diagnosis. The DSM-IV-TR (APA, 2000) explains that "a DSM-IV diagnosis . . . is not typically used to denote previous diagnoses from which the individual has recovered" (p. 1). It goes on to explain, however, that certain specifiers (i.e., Prior History, In Full Remission, and In Partial Remission) can be applied to such a diagnosis when "it may be useful to note a history of the criteria having been met for a disorder even when the individual is considered to be recovered from it" (APA, 2000, p. 1) or when the disorder is in full or partial remission.

It is hard to imagine a circumstance in which it would not be useful for a treating clinician to be aware that a patient had once met criteria for GID/GI. In my clinical experience, the fact of having once suffered from severe, persistent, clinically significant gender dysphoria, even decades earlier, remains psychologically relevant throughout life for almost all persons who have undergone sex reassignment. Therefore, retention of the diagnosis, accompanied by the specifier In Partial Remission, In Full Remission, or Prior History, would almost always be appropriate.

Gender Incongruence is merely a euphemism for GID

The members of the GID Subworkgroup apparently consider GI to be a genuine mental disorder: One Subworkgroup member has stated on an electronic mailing list that "no one [in the Subworkgroup] to my knowledge has recommended moving GI to a V-code" (Drescher, 2010b) -- that is, classifying GI as a nondisorder "Condition That May Be a Focus of Clinical Attention" (APA, 2000, p. 739). If GI is indeed considered to be a mental disorder, how is the disorder conceptualized? As I have argued above, the proposed language conceptualizes the disorder as an incongruence between gender identity and biologic sex. In the absence of a DSD, there is no persuasive evidence that biologic sex is disordered in persons who experience this type of incongruence. Consequently, the disorder reflects disordered gender identity, which is incongruent with nondisordered biologic sex. GI, therefore, is merely a euphemism for GID.

There are good reasons to employ the more explicit term. Using the term GID would maintain continuity with current clinical usage, with previous editions of the DSM (in which all severe manifestations of gender dysphoria have been considered GIDs, even if they have received other specific names, such as Transsexualism), and with the ICD-10 (World Health Organization, 1992), in which Disorders of Gender Identity is likewise a superordinate category. The term GID is familiar to and widely understood by both professionals and laypersons. It is also clear: It specifies precisely what is disordered. The term GI, in contrast, is vague -- perhaps intentionally so -- concerning what, if anything, is disordered.

The described phenomenon is gender dysphoria, which suggests another possible name

I have argued that the phenomenon described by the proposed clinical indicators is a disorder of gender identity and that there are good reasons to state this explicitly. Such a conceptualization is unacceptable to some transgender activists, however, and the Subworkgroup members understandably would prefer a conceptualization that these activists would find more acceptable. Because the phenomenon described by indicators A1-A6 corresponds closely to the concept of gender dysphoria, Gender Dysphoric Disorder might be another appropriate name for the disorder. Subworkgroup members Cohen-Kettenis and Pfäfflin (2010) also arrived at this conclusion.

The Glossary section of the DSM-IV-TR defines gender dysphoria as "a persistent aversion toward some or all of those physical characteristics or social roles that connote one's own biological sex" (APA, 2000, p. 823). Yet another definition can be derived from Criterion B of the DSM-IV-TR diagnosis of GID: "persistent discomfort with [one's] sex or sense of inappropriateness in the gender role of that sex" (APA, 2000, p. 581). Both of these definitions capture the spirit and most of the letter of indicators A1-A6. Thus, the opening text of Criterion A for Gender Dysphoric Disorder might read like this: "A strong feeling of discomfort with one's sex or the gender role associated with one's sex, of at least 6 months duration, as manifested by two or more of the following indicators" (see Table 2).

The name Gender Dysphoric Disorder probably would not satisfy all transgender activists: Many would resist having their condition described as a disorder, albeit this would be true whether the word disorder appeared in the condition's name or not. A particular complaint of some activists, however, is that the name GID implies that their identities themselves are disordered; employing an alternative name would avoid emphasizing this unpleasant reality and might circumvent some complaints. Gender Incongruence would be one such alternative name, but, as I have argued above, it is less than ideal; Gender Dysphoric Disorder would be a better alternative.

Another advantage of using the name Gender Dysphoric Disorder would be that, if a distress/impairment criterion were included as an additional diagnostic criterion -- as I will recommend later in this Letter -- then persons who met Criterion A but not the distress/impairment criterion could be ascertained to have gender dysphoria, whereas persons who met full criteria could be diagnosed with Gender Dysphoric Disorder. The DSM-5 Paraphilias Subworkgroup recommended making a similar distinction between paraphilias and paraphilic disorders (e.g., APA, 2010a, Rationale section, ¶ 1), based on nonfulfillment versus fulfillment of a distress/impairment criterion.


Table 2: Proposed Diagnostic Criteria for DSM-5 GID Diagnoses

Gender Identity Disorder [or Gender Dysphoric Disorder] in Adolescents or Adults

  1. A strong feeling of discomfort with one's sex or the gender role associated with one's sex, of at least 6 months duration, as manifested by two or more of the following indicators:

    1. a strong feeling of incongruence between one's gender identity and one's primary or secondary sex characteristics (or, in young adolescents, one's anticipated secondary sex characteristics)
    2. a strong desire to be rid of one's primary or secondary sex characteristics (or, in young adolescents, a strong desire to prevent the development of one's anticipated secondary sex characteristics)
    3. a strong desire for the primary or secondary sex characteristics of the other sex
    4. a strong desire to be the other sex
    5. a strong desire to be treated as the other sex
    6. a strong conviction that one has the typical feelings and reactions of the other sex.

  2. The condition is not concurrent with a physical intersex condition [or physical disorder of sex development].

  3. The condition causes clinically significant distress or impairment in social, occupational, or other important areas of functioning.

    Subtypes:

      Sexually attracted to males
      Sexually attracted to females
      Sexually attracted to both
      Sexually attracted to neither

Gender Identity Disorder Not Otherwise Specified

This category is included for coding disorders of gender identity that are not classifiable as a specific Gender Identity Disorder and that cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

Revised Clinical Indicators for GID/GI in Adolescents and Adults

The proposed changes to the clinical indicators for GID/GI represent significant improvements over the DSM-IV-TR in many respects. Based on the demonstration by Deogracias et al. (2007) that GID/gender dysphoria is best conceptualized as a unidimensional construct, the Subworkgroup has proposed that clinical indicators related to anatomic dysphoria (A1-A3; see Table 1) and cross-sex wishes or identification (A4-A6) be combined into a single, unified set and that the number of indicators required for diagnosis be explicitly specified.

The proposed indicators could be further improved in a few particulars, however. First, the proposed language for indicator A1 (see Table 1) does not make it clear whether the incongruence between experienced/expressed gender and primary or secondary sex characteristics is a sign (something the clinician observes) or a symptom (something the patient reports). Assuming the latter is intended, I would suggest inserting the words feeling of: "a feeling of marked incongruence between one's experienced/expressed gender [or gender identity] and primary and/or secondary sex characteristics."

Second, proposed indicators A4-A6 contain parenthetical explanations that they are applicable to persons who identify with "some alternative gender different from one's assigned gender" (see Table 1), rather than exclusively to persons who identify with "the other gender" (i.e., the other sex). There is no empirical basis, however, for this radical expansion of the types of gender identity to which these indicators might apply. The strongest argument for the validity of the clinical indicators, at least A2-A6, is that they were derived from items in an empirically validated instrument, the Gender Identity/Gender Dysphoria Questionnaire for Adolescents and Adults (GIDYQ; Deogracias et al., 2007). The relevant GIDYQ items, however, conceptualize gender identity in binary terms. The Subworkgroup's report did not state which specific questionnaire items from the GIDYQ were used to create indicators A4-A6, but one can make reasonable guesses, assuming that items with high factor loadings were selected (the quoted items and factor loadings are from Deogracias et al., pp. 374, 377-378): (1) A4 was probably derived from GIDYQ item 16 ("the wish or desire to be a man" [for females] or "a woman" [for males]), with factor loading .93; (2) A5 was probably derived from item 6 ("better for you to live as a man" [for females] or "as a woman" [for males]), with factor loading .93; (3) A6 was probably derived from item 10 ("felt more like a man than like a woman" [for females] or "more like a woman than like a man" [for males]), with factor loading .96.

These GIDYQ items conceptualize gender identity in binary terms. By transforming them into indicators that conceptualize gender identity as a "multi-category concept or spectrum" (APA, 2010c, Rationale section, ¶ 10), the Subworkgroup has unnecessarily discarded the presumption of validity. It is also worth noting that the few items in the GIDYQ that imply other than a binary gender identity are associated with considerably lower factor loadings, albeit still high in absolute terms (the quoted items and factor loadings are again from Deogracias et al., pp. 374, 377-378): (1) Item 25, "thought of yourself as a 'transgendered person'": factor loading .78; (2) Item 2, "feeling somewhere in between a woman and a man": factor loading .58; (3) Item 9, "at times feeling more like a man and at times feeling more like a woman": factor loading .58; (4) Item 11, "felt that you did not have anything in common with either men or women": factor loading .47.

The Subworkgroup conceded that, because the wording of the GIDYQ items "is not identical to the wording of the proposed indicators, further validation work will be required during field trials" (APA, 2010c, Rationale section, ¶ 1). The parenthetical additions to indicators A4-A6, however, do not involve mere changes in wording but changes in the very way gender identity is conceptualized. This undercuts the otherwise strong empirical basis of the proposed indicators. Consequently, I recommend that the parenthetical references to "some alternative gender" in indicators A4-A6 be deleted. Gender dysphoric persons who desire to be "some alternative gender" other than male or female undoubtedly do exist (e.g., Johnson & Wassersug, 2010) but should be diagnosed with GIDNOS, not GID/GI; I will discuss this recommendation in more detail later in this Letter.

Finally, indicators A4-A6 refer to the desire to be, the desire to be treated as, or the conviction that one has the feelings and reactions of the other gender. I believe, however, that the reference should be to the other sex. As noted earlier, the Subworkgroup stated that "the term 'sex' has been replaced by assigned 'gender' in order to make the criteria applicable to individuals with a DSD" (APA, 2010c, Rationale section, ¶ 4). If persons with a DSD and persons who desire to be some sex/gender other than male or female were excluded from the diagnosis, there would be no reason not to employ the more accurate term, sex. Most severely gender dysphoric persons want to assume, insofar as possible, the anatomic status of the other biologic sex (not just the associated gender role) and be treated by other people as members of the other biologic sex (not just as persons who have assumed that gender role). Admittedly, the situation with respect to indicator A6 is more equivocal -- gender is not clearly incorrect -- but sex works equally well or better, and its use would contribute to consistency of expression.

Absence of a Distress/Impairment Criterion

GID/GI is the only major diagnosis proposed by the DSM-5 Sexual and Gender Identity Disorders Workgroup that does not incorporate a distress/impairment criterion. The absence of such a criterion is surprising: Most theoretical examinations of the concept of mental disorders undertaken in connection with the DSM-5 revision process (e.g., First & Wakefield, 2010; Stein et al., 2010; Wakefield & First, 2003) have emphasized the need to include a distress/ impairment criterion to avoid problems of overdiagnosis. The absence of such a criterion is especially surprising, given that the Subworkgroup rejects the idea that a strong desire for sex reassignment necessarily indicates "inherent distress" (APA, 2010c, Rationale section, ¶ 9).

I believe the risk of overdiagnosis is significant if a distress/impairment criterion is not included in the diagnostic criteria for GID/GI. In recommending that "the GI diagnosis be given on the basis of the A criterion alone and that distress and/or impairment . . . be evaluated separately and independently" (APA, 2010c, Rationale section, ¶ 9), the Subworkgroup members have declared that it is appropriate to diagnose a mental disorder in gender variant people who are not significantly distressed or impaired by their gender variance. To the extent that psychiatric diagnoses are inherently stigmatizing, such a declaration is troubling, even if one believes, as I do, that some inherent distress, albeit perhaps not always clinically significant distress, can be assumed for most of the proposed clinical indicators.

What kinds of persons could be diagnosed with GID/GI on the basis of indicators A1-A6 alone, despite experiencing little or no distress or impairment? Probably they would be persons who fulfilled only indicators A1 and A6, the only indicators that do not involve strongly felt, unmet cross-sex desires. Hypothetically, such persons might strongly identify with the other sex (and thus feel their identity to be incongruent with their anatomy) and might also believe they had the typical feelings and reactions of the other sex but might not feel any strong desire to undergo sex reassignment. Consequently, they might be neither distressed nor impaired by living as members of their natal sex. Natal females, in whom substantial sex-atypical expression is often socially tolerated, might be especially likely to experience little or no distress or impairment, despite identifying with and believing themselves similar to the other sex. I can see no justification for diagnosing GID/GI in such persons; adding a distress/impairment criterion would prevent this.

As noted earlier, if it were considered desirable to provide an "exit clause" from the diagnosis of GID/GI, a distress/impairment criterion would provide a straightforward way of accomplishing this. It would make better sense intuitively to lose a diagnosis because one no longer suffered any distress or impairment than because nominal gender reassignment had theoretically eliminated the incongruence between gender identity and assigned gender.

Eligibility of Persons with a DSD for the Diagnosis of GID/GI

The presence of a DSD, formerly called an intersex condition, was an exclusion criterion for the diagnosis of Transsexualism in the DSM-III (APA, 1980) but not in the DSM-III-R (APA, 1987). A DSD was reintroduced as an exclusion criterion for the diagnosis of GID in the DSM-IV (APA, 1994) and continued in the DSM-IV-TR (APA, 2000). The reintroduction of a DSD as an exclusion criterion was the result of an extensive review by the GID Subcommittee of the DSM-IV Task Force; one of the principal conclusions of this review was summarized by Meyer-Bahlburg (1994):

Intersex patients with significant gender identity problems or gender change do differ from nonintersex patients with GID in prevalence, in age at onset and presentation, and in the sex ratio, and the evidence available -- in spite of its methodological shortcomings -- makes it very likely that the development of gender problems in intersex patients is in most cases not directly comparable to GID as it develops in nonintersex patients. The two forms of gender identity problems are unlikely to be the same disorder. (p. 33)
Meyer-Bahlburg recommended that it would be best to "exempt all intersex patients from the GID diagnosis" (p. 35) -- that is, from being diagnosed with either GID or GIDNOS -- but this recommendation was not adopted: In the DSM-IV and DSM-IV-TR, intersex patients were excluded from the diagnosis of GID proper but could be diagnosed with GIDNOS.

Fifteen years later, Meyer-Bahlburg (2009) expressed conclusions almost identical to his earlier ones:

The differences between GIV [gender identity variants] in DSD individuals and GIV in non-DSD individuals with regard to phenomenon, context of presentation, epidemiology, etiology, and (pediatric) treatment settings are so large that, at this stage of our knowledge, identical diagnostic categories and treatment approaches are not justified, as has recently also been argued by Mazur, Colsman, and Sandberg (2007). (p. 231)
The review articles authored by the Subworkgroup members offered no compelling new evidence that these conclusions should be modified or disregarded: Meyer-Bahlburg (2010a) expressed no doubts about his previously stated conclusions, and Zucker (2010) recommended that a physical intersex condition continue to be an exclusion criterion for the diagnosis of GID in children. As noted by Meyer-Bahlburg (2009), Mazur et al. (2007) also concluded that, because of the many differences between GID in persons with and without intersex conditions, "it would be prudent to consider them as separate entities when initiating an evaluation" (p. 236). Accordingly, the proposed recommendation to make persons with a DSD eligible for the diagnosis of GID/GI seems ill-advised.

One could argue that the DSM-IV and DSM-IV-TR already allowed males with two very different conditions and probably different etiologies -- homosexual and nonhomosexual gender dysphoria -- to receive the same GID diagnosis, distinguishing the two conditions using subtypes (Lawrence, 2010). The proposal to allow gender dysphoric persons with and without a DSD to receive the same GID/GI diagnosis and to distinguish their conditions using subtypes could be seen as merely the extension of an existing principle. But extending the diagnosis of GID/GI to persons with a DSD would only unnecessarily increase variability within a diagnostic category that already encompasses substantial variability.

The Subworkgroup would have been better advised to follow Meyer-Bahlburg's (1994) original recommendation and make persons with a DSD ineligible for any GID/GI diagnosis. Gender dysphoric patients with a DSD already have a medical diagnosis that provides access to treatment, unlike gender dysphoric persons without a DSD. The stigma associated with psychiatric diagnoses is yet another reason for excluding persons with a DSD from any GID/GI diagnosis, or at least from the diagnosis of GID/GI proper.

Elimination of Subtypes Based on Sexual Orientation

Beginning in 1980, subtypes based on sexual orientation have been used in connection with GID diagnoses in every edition of the DSM. Subtypes based on sexual orientation offer substantial descriptive, prognostic, and heuristic value; the Subworkgroup's proposal to eliminate them is ill-advised. In a review article (Lawrence, 2010), written at the request of the GID Subworkgroup (Meyer-Bahlburg, 2010b), I observed:

The most widely used and influential typologies for transsexualism and gender identity disorder (GID) in adolescents and adults employ either sexual orientation or age of onset of GID-related symptoms as bases for categorization. . . . Typologies based on sexual orientation and age of onset of GID-related symptoms are roughly comparable in ease and reliability of subtype assignment. Typologies based on sexual orientation, however, employ subtypes that are less ambiguous and better suited to objective confirmation and that offer more concise, comprehensive clinical description. Typologies based on sexual orientation are also superior in their ability to predict treatment-related outcomes and comorbid psychopathology and to facilitate research. Commonly expressed objections to typologies based on sexual orientation are unpersuasive when examined closely. The DSM should continue to employ subtypes based on sexual orientation for the diagnosis of GID in Adolescents or Adults or its successor diagnosis. (p. 514)

Uncertain Status of the Diagnosis of GIDNOS

The status of the diagnosis of GIDNOS is uncertain: According to the DSM-5 website, "Changes, or lack thereof, are still under review by the work group" (APA, 2010b). This suggests that the diagnosis of GIDNOS might not be retained in the DSM-5. I believe that the diagnosis should be retained.

There should be a diagnostic category other than GID/GI for gender dysphoric persons who do not identify as members of the other sex, desire to have the primary or secondary sex characteristics of the other sex, or desire to live as members of the other sex. Too little is known about such persons to confidently place them in the same diagnostic category as persons who unequivocally identify with and want to belong to the other sex (i.e., transsexuals), a population about which a great deal is known. Although there have been Internet-based surveys of persons who claim to identify as neither men nor women, I am not aware of any systematic studies involving actual clinical patients who so identify. Absent such studies, I believe it would be premature to diagnose such patients with GID/GI.

This kind of cautious approach to nosology explains why, for the last 30 years, the DSM has always provided one or more alternative GID categories for individuals who cannot confidently be classified with transsexuals for diagnostic purposes: Atypical Gender Identity Disorder in the DSM-III; Gender Identity Disorder of Adolescence or Adulthood, Nontranssexual Type and GIDNOS in the DSM-III-R; and GIDNOS in the DSM-IV and DSM-IV-TR. In my opinion, such an alternative diagnostic category (i.e., GIDNOS) should also exist in the DSM-5, and persons with gender dysphoria or gender identity problems who identify as neither men nor women and wish to live as neither men nor women should receive a diagnosis of GIDNOS, not GID/GI.

Proposed Revised Language for GID Diagnoses in the DSM-5

Table 2 displays the alternative diagnostic criteria I propose for GID diagnoses in the DSM-5, based on the foregoing analysis.

References

American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: Author.

American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., revised). Washington, DC: Author.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text revision). Washington, DC: Author.

American Psychiatric Association. (2010a). Proposed revision: 302.3, Transvestic Fetishism. Retrieved from http://www.dsm5.org/ProposedRevisions/Pages/ proposedrevision.aspx?rid=189

American Psychiatric Association. (2010b). Proposed revision: 302.6, Gender Identity Disorder Not Otherwise Specified. Retrieved from http://www.dsm5.org/ProposedRevisions/ Pages/proposedrevision.aspx?rid=194

American Psychiatric Association. (2010c). Proposed revision: 302.85, Gender Identity Disorder in Adolescents or Adults. Retrieved from http://www.dsm5.org/ProposedRevisions/ Pages/proposedrevision.aspx?rid=482

Cohen-Kettenis, P. T., & Pfäfflin, F. (2010). The DSM diagnostic criteria for gender identity disorder in adolescents and adults. Archives of Sexual Behavior, 39, 499-513.

Deogracias, J. J., Johnson, L. L., Meyer-Bahlburg, H. F. L., Kessler, S. J., Schober, J. M., & Zucker, K. J. (2007). The Gender Identity/Gender Dysphoria Questionnaire for Adolescents and Adults. Journal of Sex Research, 44, 370-379.

Drescher, J. (2010a). Queer diagnoses: Parallels and contrasts in the history of homosexuality, gender variance, and the diagnostic and statistical manual. Archives of Sexual Behavior, 39, 427-460.

Drescher, J. (2010b, February 20). Re: Hyperbole [Electronic mailing list message]. Sexnet listserv.

First, M. B., & Wakefield, J. C. (2010). Defining 'mental disorder' in DSM-V. Psychological Medicine. doi:10.1017/S0033291709992339

Johnson, T. W., & Wassersug, R. J. (2010). Gender identity disorder outside the binary: When gender identity disorder-not otherwise specified is not good enough [Letter to the editor]. Archives of Sexual Behavior, 39, 597-598.

Lawrence, A. A. (2010). Sexual orientation versus age of onset as bases for typologies (subtypes) for gender identity disorder in adolescents and adults. Archives of Sexual Behavior, 39, 514-545.

Mazur, T., Colsman, M., & Sandberg, D. E. (2007). Intersex: Definitions, examples, gender stability, and the case against merging with transsexualism. It R. Ettner, S. Monstrey, & A. E. Eyler (Eds.), Principles of transgender medicine and surgery (pp. 235-259). Binghamton, NY: Haworth Press.

Meyer-Bahlburg, H. F. (1994). Intersexuality and the diagnosis of gender identity disorder. Archives of Sexual Behavior, 23, 21-40.

Meyer-Bahlburg, H. F. (2009). Variants of gender differentiation in somatic disorders of sex development: Recommendations for version 7 of the World Professional Association for Transgender Health's Standards of Care. International Journal of Transgenderism, 11, 226-237.

Meyer-Bahlburg, H. F. (2010a). From mental disorder to iatrogenic hypogonadism: Dilemmas in conceptualizing gender identity variants as psychiatric conditions. Archives of Sexual Behavior, 39, 461-476.

Meyer-Bahlburg, H. F. (2010b, February 21). Re: DSM-5 GID issues - some clarifications [Electronic mailing list message]. Sexnet listserv.

Stein, D. J., Phillips, K. A., Bolton, D., Fulford, K. W. M., Sadler, J. Z., & Kendler, K. S. (2010). What is a mental/psychiatric disorder? From DSM-IV to DSM-V. Psychological Medicine. doi:10.1017/S0033291709992261

Wakefield, J. C., & First, M. B. (2003). Clarifying the distinction between disorder and nondisorder: Confronting the overdiagnosis (false positives) problem in DSM-V. In K. A. Phillips, M. B. First, & H. A. Pincus (Eds.), Advancing DSM: Dilemmas in psychiatric diagnosis (pp. 23-55). Arlington, VA: American Psychiatric Association.

World Health Organization. (1992).International statistical classification of diseases and related health problems (10th rev., Vol. 1). Geneva, Switzerland: Author.

Zucker, K. J. (2010). The DSM diagnostic criteria for gender identity disorder in children. Archives of Sexual Behavior, 39, 477-498.

© 2011 by Anne A. Lawrence, M.D., Ph.D. All rights reserved.