American Academy of Pediatrics policy and trans- kids:Fact-checking

The American Academy of Pediatrics (AAP) recently published a policy statement entitled, Ensuring comprehensive care and support for transgender and gender-diverse children and adolescents (Rafferty, 2018).  It was quite a remarkable document: Although almost all clinics and professional associations in the world use what’s called the watchful waiting approach to helping GD children, the AAP statement rejected that consensus, endorsing only gender affirmation.  With AAP taking such a dramatic departure from other professional associations, I was immediately curious about what evidence led them to that conclusion.  (Extraordinary claims require extraordinary evidence, and all that.)  As I read the works on which they based their policy however, I was pretty surprised…rather alarmed, actually:  These documents simply did not say what AAP claimed they did.  In fact, the references that AAP cited as the basis of their policy instead outright contradicted that policy, repeatedly endorsing watchful waiting. 

The AAP statement was also remarkable in what it left out—namely, the outcomes research on GD children.  There have been eleven follow-up studies of GD children, of which AAP cited one [Wallien & Cohen-Kettenis (2008)], doing so without actually mentioning the outcome data it contained.  The literature on outcomes was neither reviewed, summarized, nor subjected to meta-analysis to be considered in the aggregate—It was merely disappeared.  (I have presented the complete list of the outcome studies on this blog before; they appear again at the bottom of this page together with their results, for reference.)  As they make clear, every follow-up study of GD children, without exception, found the same thing: By puberty, the majority of GD children ceased to want to transition.  AAP is, of course, free to establish whatever policy it likes on whatever basis it likes.  But any assertion that their policy is based on evidence is demonstrably false, as detailed below. 

AAP divided clinical approaches into three types—conversion therapy, watchful waiting, and gender affirmation.  It rejected the first two and endorsed gender affirmation as the only acceptable alternative.  Most readers will likely be familiar already with attempts to use conversion therapy to change sexual orientation.  With regard to gender identity, AAP wrote:

“[C]onversion” or “reparative” treatment models are used to prevent children and adolescents from identifying as transgender or to dissuade them from exhibiting gender-diverse expressions….Reparative approaches have been proven to be not only unsuccessful³⁸ but also deleterious and are considered outside the mainstream of traditional medical practice.^{29, 39–42}

AAP’s citations are:

38.  Haldeman DC. The practice and ethics of sexual orientation conversion therapy. J Consult Clin Psychol. 1994;62(2):221–227

29.  Adelson SL; American Academy of Child and Adolescent Psychiatry (AACAP) Committee on Quality Issues (CQI). Practice parameter on gay, lesbian, or bisexual sexual orientation, gender nonconformity, and gender discordance in children and adolescents. J Am Acad Child Adolesc Psychiatry. 2012;51(9):957–974

39.  Byne W. Regulations restrict practice of conversion therapy. LGBT Health. 2016;3(2):97–99

40.  Cohen-Kettenis PT, Delemarrevan de Waal HA, Gooren LJ. The treatment of adolescent transsexuals: changing insights. J Sex Med. 2008;5(8):1892–1897

41.  Bryant K. Making gender identity disorder of childhood: historical lessons for contemporary debates. Sex Res Soc Policy. 2006;3(3):23–39

42.  World Professional Association for Transgender Health. WPATH De-Psychopathologisation Statement. Minneapolis, MN: World Professional Association for Transgender Health; 2010. Available at: https://www.wpath.org/policies. Accessed April 16, 2017

AAP's claims struck me as odd because there are no studies of conversion therapy for gender identity.  Studies of conversion therapy have been limited to sexual orientation—specifically, the sexual orientation of adults—not to gender identity, and not of children in any case.  The article AAP cited to support their claim (reference number 38) is indeed a classic and well-known review, but it is a review of sexual orientation research only.  Neither gender identity, nor even children, received even a single mention in it.  Indeed, the narrower scope of that article should be clear to anyone reading even just its title: “The practice and ethics of sexual orientation conversion therapy” (Haldeman, 1994, p. 221, italics added).

AAP continued, saying that conversion approaches for GD children have already been rejected by medical consensus, citing five sources.  This claim struck me just as odd, however—I recalled associations banning conversion therapy for sexual orientation, but not for gender identity, exactly because there is no evidence for generalizing from adult sexual orientation to childhood gender identity.  So, I started checking AAP’s citations for that, and these sources too pertained only to sexual orientation, not gender identity (specifics below).  What AAP’s sources did repeatedly emphasize was that:

(1)   Sexual orientation of adults is unaffected by conversion therapy and any other [known] intervention;

(2)   Gender dysphoria in childhood before puberty desists in the majority of cases, becoming (cis-gendered) homosexuality in adulthood, again regardless of any [known] intervention; and

(3)   Gender dysphoria in childhood persisting after puberty tends to persist entirely. 

That is, in the context of GD children, it simply makes no sense to refer to externally induced “conversion”: The majority of children “convert” to cisgender or “desist” from transgender regardless of any attempt to change them.  “Conversion” only makes sense with regard to adult sexual orientation because (unlike childhood gender identity), adult homosexuality never or nearly never spontaneously changes to heterosexuality.  Although gender identity and sexual orientation may often be analogous and discussed together with regard to social or political values and to civil rights, they are nonetheless distinct—with distinct origins, needs, and responses to medical and mental health care choices.  Although AAP emphasized to the reader that “gender identity is not synonymous with ‘sexual orientation’” (Rafferty, 2018, p. 3), they went ahead to treat them as such nonetheless.

To return to checking AAP’s fidelity to its sources: Reference 29 was a practice guideline from the Committee on Quality Issues of the American Academy of Child and Adolescent Psychiatry (AACAP).  Despite AAP applying this source to gender identity, AACAP was quite unambiguous regarding their intent to speak to sexual orientation and only to sexual orientation: “Principle 6. Clinicians should be aware that there is no evidence that sexual orientation can be altered through therapy, and that attempts to do so may be harmful.  There is no established evidence that change in a predominant, enduring homosexual pattern of development is possible.  Although sexual fantasies can, to some degree, be suppressed or repressed by those who are ashamed of or in conflict about them, sexual desire is not a choice.  However, behavior, social role, and—to a degree—identity and self-acceptance are. Although operant conditioning modifies sexual fetishes, it does not alter homosexuality.  Psychiatric efforts to alter sexual orientation through ‘reparative therapy’ in adults have found little or no change in sexual orientation, while causing significant risk of harm to self-esteem” (AACAP, 2012, p. 967, italics added). 

Whereas AAP cites AACAP to support gender affirmation as the only alternative for treating GD children, AACAP’s actual view was decidedly neutral, noting the lack of evidence: “Given the lack of empirical evidence from randomized, controlled trials of the efficacy of treatment aimed at eliminating gender discordance, the potential risks of treatment, and longitudinal evidence that gender discordance persists in only a small minority of untreated cases arising in childhood, further research is needed on predictors of persistence and desistence of childhood gender discordance as well as the long-term risks and benefits of intervention before any treatment to eliminate gender discordance can be endorsed” (AACAP, 2012, p. 969).  Moreover, whereas AAP rejected watchful waiting, what AACAP recommended was: “In general, it is desirable to help adolescents who may be experiencing gender distress and dysphoria to defer sex reassignment until adulthood” (AACAP, 2012, p. 969).  So, not only did AAP attribute to AACAP something AACAP never said, but also AAP withheld from readers AACAP’s actual view.

Next, in reference 39, Byne (2016) also addressed only sexual orientation, doing so very clearly: “Reparative therapy is a subset of conversion therapies based on the premise that same-sex attraction are reparations for childhood trauma. Thus, practitioners of reparative therapy believe that exploring, isolating, and repairing these childhood emotional wounds will often result in reducing same-sex attractions” (Byne, 2016, p. 97).  Byne does not say this of gender identity, as the AAP statement misrepresents.

In AAP reference 40, Cohen-Kettenis et al. (2008) did finally pertain to gender identity; however, this article never mentions conversion therapy. (!)  Rather, in this study, the authors presented that clinic’s lowering of their minimum age for cross-sex hormone treatment from age 18 to 16, which they did on the basis of a series of studies showing the high rates of success with this age group.  Although it did strike me as odd that AAP picked as support against conversion therapy an article that did not mention conversion therapy, I could imagine AAP cited the article as an example of what the “mainstream of traditional medical practice” consists of (the logic being that conversion therapy falls outside what an ‘ideal’ clinic like this one provides).  However, what this clinic provides is the very watchful waiting approach that AAP rejected.  The approach espoused by Cohen-Kettenis (and the other clinics mentioned in the source—Gent, Boston, Oslo, and now formerly, Toronto) is to make puberty-halting interventions available at age 12 because: “[P]ubertal suppression may give adolescents, together with the attending health professional, more time to explore their gender identity, without the distress of the developing secondary sex characteristics. The precision of the diagnosis may thus be improved” (Cohen-Kettenis et al., 2008, p. 1894).

Reference 41 presented a very interesting history spanning the 1960s–1990s about how feminine boys and tomboyish girls came to be recognized as mostly pre-homosexual, and how that status came to be entered into the DSM at the same time as homosexuality was being removed from the DSM.  Conversion therapy is never mentioned.  Indeed, to the extent that Bryant mentions treatment at all, it is to say that treatment is entirely irrelevant to his analysis: “An important omission from the DSM is a discussion of the kinds of treatment that GIDC children should receive. (This omission is a general orientation of the DSM and not unique to GIDC)” (Bryant, 2006, p. 35).  How this article supports AAP’s claim is a mystery.  Moreover, how AAP could cite a 2006 history discussing events of the 1990s and earlier to support a claim about the current consensus in this quickly evolving discussion remains all the more unfathomable.

Cited last in this section was a one-paragraph press release from the World Professional Association for Transgender Health.  Written during the early stages of the American Psychiatric Association’s (APA’s) update of the DSM, the statement asserted simply that “The WPATH Board of Directors strongly urges the de-psychopathologisation of gender variance worldwide.”  Very reasonable debate can (and should) be had regarding whether gender dysphoria should be removed from the DSM as homosexuality was, and WPATH was well within its purview to assert that it should.  Now that the DSM revision process is years completed however, history has seen that APA ultimately retained the diagnostic categories, rejecting WPATH’s urging.  This makes AAP’s logic entirely backwards: That WPATH’s request to depathologize gender dysphoria was rejected suggests that it is WPATH’s view—and therefore, AAP policy—which falls “outside the mainstream of traditional medical practice.” (!)

AAP based this entire line of reasoning on their belief that conversion therapy is being used “to prevent children and adolescents from identifying as transgender” (Rafferty, 2018, p. 4).  That claim is left without citation or support.  In contrast, what is said by AAP’s sources is “delaying affirmation should not be construed as conversion therapy or an attempt to change gender identity” in the first place (Byne, 2016, p. 2).  Nonetheless, AAP seems to be doing exactly that: Simply relabeling non-gender affirmation models as conversion clinics.

Although AAP (and anyone else) may reject (what they label to be) conversion therapy purely on the basis of political or personal values, there is no evidence to back the AAP’s stated claim about the existing science on gender identity at all, never mind gender identity of children.

AAP also rejected the watchful waiting approach, repeatedly calling it “outdated.”  The criticisms AAP provided, however, again defied the existing evidence, with even its own sources repeatedly calling that model the current standard.  According to AAP:

[G]ender affirmation is in contrast to the outdated approach in which a child’s gender-diverse assertions are held as “possibly true” until an arbitrary age (often after pubertal onset) when they can be considered valid, an approach that authors of the literature have termed “watchful waiting.” This outdated approach does not serve the child because critical support is withheld. Watchful waiting is based on binary notions of gender in which gender diversity and fluidity is pathologized; in watchful waiting, it is also assumed that notions of gender identity become fixed at a certain age. The approach is also influenced by a group of early studies with validity concerns, methodologic flaws, and limited follow-up on children who identified as TGD and, by adolescence, did not seek further treatment (“desisters”).^45,47

The citations from AAP’s reference list are:

45.  Ehrensaft D, Giammattei SV, Storck K, Tishelman AC, Keo-Meier C. Prepubertal social gender transitions: what we know; what we can learn—a view from a gender affirmative lens. Int J Transgend. 2018;19(2):251–268

47.  Olson KR. Prepubescent transgender children: what we do and do not know. J Am Acad Child Adolesc Psychiatry. 2016;55(3):155–156.e3

I was surprised first by the AAP’s claim that pubertal onset was somehow “arbitrary.”  The literature, including AAP’s sources, repeatedly indicated the pivotal importance of puberty, noting that outcomes strongly diverge at puberty.  According to AAP reference 29, in “prepubertal boys with gender discordance—including many without any mental health treatment—the cross gender wishes usually fade over time and do not persist into adulthood, with only 2.2% to 11.9%  continuing to experience gender discordance” (Adelson & AACAP, 2012, p. 963, italics added), whereas “when gender variance with the desire to be the other sex is present in adolescence, this desire usually does persist through adulthood” (Adelson & AACAP, 2012, p. 964, italics added).  Similarly, according to AAP reference 40, “Symptoms of GID at prepubertal ages decrease or even disappear in a considerable percentage of children (estimates range from 80–95%).  Therefore, any intervention in childhood would seem premature and inappropriate. However, GID persisting into early puberty appears to be highly persistent” (Cohen-Kettenis et al., 2008, p. 1895, italics added).  That follow-up studies of prepubertal transition differ from postpubertal transition is the very meaning of non-arbitrary.  AAP gave readers exactly the reverse of what was contained in its own sources.  If AAP were correct in saying that puberty is an arbitrarily selected age, then AAP will be able to offer another point with as much empirical backing as puberty has.

Next, it was not clear on what basis AAP could say that watchful waiting withholds support—AAP cited no support for its claim.  The people in such programs often receive substantial support during this period.  Also unclear is on what basis AAP could already know exactly which treatments are “critical” and which are not—Answering that question is the very purpose of this entire endeavor.  Indeed, the logic of AAP’s claim appears entirely circular:  If one were pre-convinced that the gender affirmation is the only acceptable alternative, then watchful waiting withholds critical support only in the sense that it delays gender affirmation, the method one has already decided to be critical.

Although AAP’s next claim did not have a citation appearing at the end of its sentence, binary notions of gender was mentioned both in references 45 and 47.  Specifically, both pointed out that existing outcome studies have been about people transitioning from one sex to the other, rather than from one sex to an in-between status or a combination of masculine/feminine features.  Neither reference presented this as a reason to reject the results from the existing studies of complete transition however (which is how AAP cast it).  Although it is indeed true that the outcome data have been about complete transition, some future study showing that partial transition shows a different outcome would not invalidate what is known about complete transition.  Indeed, data showing that partial transition gives better outcomes than complete transition would, once again, support the watchful waiting approach which AAP rejected.

Next was a vague reference alleging concerns and criticisms about early studies.  Had AAP indicated what those alleged concerns and flaws were (or which studies they were), then it would be possible to evaluate or address them.  Nonetheless, the argument is a red herring: Because all of the later studies showed the same result as did the early studies, any such allegation is necessarily moot.

Reference 47 was a one-and-a-half page commentary which off-handedly mentions criticisms previously made of three of the eleven outcome studies of GD children, but does not provide any analysis or discussion (Olsen, 2016).  The only specific claim was that studies (whether early or late) had limited follow-up periods—the logic being that had outcome researchers lengthened the follow-up period, then people who seemed to have desisted might have returned to the clinic as cases of “persistence-after-interruption.”  Although one could debate the merits of that prediction, AAP (and Olson) instead simply withheld from the reader the result from testing that prediction directly:  Steensma and Cohen-Kettenis (2015) conducted another analysis of their cohort, by then ages 19–28 (mean age 25.9 years), and found that 3.3% (5 people of the sample of 150) later returned.  That is, the childhood sample showing 70.0% desistence instead showed 66.7% desistance in long-term follow-up.  It is up to the reader to decide whether that difference challenges the aforementioned conclusion that that majority of GD children cease to want to transition by puberty or represents a grasping at straws.

Reference

Steensma, T. D., & Cohen-Kettenis, P. T. (2015). More than two developmental pathways in children with gender dysphoria?  Journal of the American Academy of Child & Adolescent Psychiatry, 52, 147–148.

Reference 45 did not support the claim that watchful-waiting is “outdated.”  Indeed, that source said the very opposite, referring to watchful waiting as the current approach:  “Put another way, if clinicians are straying from SOC 7 guidelines for social transitions, not abiding by the watchful waiting model favored by the standards, we will have adolescents who have been consistently living in their affirmed gender since age 3, 4, or 5” (Ehrensaft et al., 2018, p. 255).  Moreover, Ehrensaft et al. said there are cases in which they too would still use watchful waiting: “When a child’s gender identity is unclear, the watchful waiting approach can give the child and their family time to develop a clearer understanding and is not necessarily in contrast to the needs of the child” (p. 259).  Ehrensaft et al. are indeed critical of the watchful waiting model (which they feel is applied too conservatively), but they do not come close to the position the AAP policy espouses.  Where Ehrensaft summarizes the potential benefits and potential risks both to transitioning and not transitioning, the AAP presents an ironically binary narrative.

In its policy statement, AAP told neither the truth nor the whole truth, committing sins both of commission and of omission, asserting claims easily falsified by anyone caring to do any fact-checking at all.  AAP claimed, “This policy statement is focused specifically on children and youth that identify as TGD rather than the larger LGBTQ population” (p. 1); however, much of that evidence was about sexual orientation, not gender identity.  AAP claimed, “Current available research and expert opinion from clinical and research leaders…will serve as the basis for recommendations” (p. 1-2); however, they provided recommendations entirely unsupported and even in direct opposition to that research and opinion.

AAP is advocating for something far in excess of mainstream practice and medical consensus.  In the presence of compelling evidence, that would be exactly called for.  The problems in Rafferty (2018), however, do not constitute merely a misquote, a misinterpretation of an ambiguous statement, or a missing reference or two.  Rather, AAP’s statement is a systematic exclusion and misrepresentation of entire literatures.  Not only did AAP fail to provide extraordinary evidence, it failed to provide the evidence at all.  Indeed, AAP’s recommendations are despite the existing evidence. 

------------------------------------------------------------------------------------------------------------------------

Outcome Studies of GD Children and Their Results

Count	Group	Study
2/16 4/16 10/16	gay trans-/crossdress straight/uncertain	Lebovitz, P. S. (1972). Feminine behavior in boys: Aspects of its outcome. American Journal of Psychiatry, 128, 1283–1289.
2/16 2/16 12/16	trans- uncertain gay	Zuger, B. (1978). Effeminate behavior present in boys from childhood: Ten additional years of follow-up. Comprehensive Psychiatry, 19, 363–369.
0/5 5/5	trans- gay	Money, J., & Russo, A. J. (1979). Homosexual outcome of discordant gender identity/role: Longitudinal follow-up. Journal of Pediatric Psychology, 4, 29–41.
2/45 10/45 33/45	trans-/crossdress uncertain gay	Zuger, B. (1984). Early effeminate behavior in boys: Outcome and significance for homosexuality. Journal of Nervous and Mental Disease, 172, 90–97.
1/10 2/10 3/10 4/10	trans- gay uncertain straight	Davenport, C. W. (1986). A follow-up study of 10 feminine boys.  Archives of Sexual Behavior, 15, 511–517.
1/44 43/44	trans- cis-	Green, R. (1987). The "sissy boy syndrome" and the development of homosexuality. New Haven, CT: Yale University Press.
0/8 8/8	trans- cis-	Kosky, R. J. (1987). Gender-disordered children: Does inpatient treatment help? Medical Journal of Australia, 146, 565–569.
21/54 33/54	trans- cis-	Wallien, M. S. C., & Cohen-Kettenis, P. T. (2008). Psychosexual outcome of gender-dysphoric children. Journal of the American Academy of Child and Adolescent Psychiatry, 47, 1413–1423.
3/25 6/25 16/25	trans- lesbian/bi- straight	Drummond, K. D., Bradley, S. J., Badali-Peterson, M., & Zucker, K. J. (2008). A follow-up study of girls with gender identity disorder. Developmental   Psychology, 44, 34–45.
17/139 122/139	trans- cis-	Singh, D. (2012). A follow-up study of boys with gender identity disorder. Unpublished doctoral dissertation, University of Toronto.
47/127 80/127	trans- cis-	Steensma, T. D., McGuire, J. K., Kreukels, B. P. C., Beekman, A. J., & Cohen-Kettenis, P. T. (2013). Factors associated with desistence and persistence of childhood gender dysphoria: A quantitative follow-up study. Journal of the American Academy of Child and Adolescent Psychiatry, 52, 582–590.

*For brevity, the list uses “gay” for “gay and cis-”, “straight” for “straight and cis-”, etc.