The 0.1% smokescreen

Gender researchers try to rewrite well-established numbers, inadvertently explaining the birth of a new industry

After feedback from people whose opinions I trust, the final post has become much longer than my initial draft. I asked Google Gemini’s help to summarize the article, which appears first.

---

Google Gemini’s TL;DR version:

“The article argues against the idea that the number of adolescents receiving cross-sex hormones is "very, very small" (0.1% of all adolescents) and, therefore, not a significant issue. This is the narrative that has been pushed by a recent Research Letter published in JAMA Pediatrics (referred to as Hughes in the article) and amplified by outlets like NPR and the Associated Press. The author contends that this statistic is misleading and obscures a larger, more concerning narrative.

The main points of the article are as follows:

• Misleading Baseline: Comparing the number of adolescents on hormones to all adolescents is the wrong comparison. A more relevant comparison is the number of adolescents on hormones compared to those who self-identify as transgender. Even then, the author argues, the true comparison should be to the much smaller subset of those who meet clinical diagnostic criteria for gender dysphoria and then qualify for medical interventions.

• Inflated Numbers: The 0.1% figure cited in the JAMA Pediatrics study is a lowball estimate and likely doesn't account for recent increases in medicalization or the use of “endocrine disorder” as a catch-all diagnosis. [This is an issue that Ben Ryan discussed in his Substack.]

• Clinical vs. Non-Clinical: There’s a significant difference between those who self-identify as transgender and those who are clinically diagnosed with gender dysphoria and meet the criteria for medical interventions. Historically, the latter group is much smaller (100-300 times smaller), with children self-identifying at much higher rates compared to adults. This means many adolescents receiving hormones would not have met established clinical criteria.

• Reviews by WPATH experts: Reviews and studies by WPATH (World Professional Association for Transgender Health) experts suggest that the number of youth being medicalized far exceeds (by a factor of 20 or more) the number who should be based on clinical diagnoses.

• Business Incentives: Inflating the perceived size of the transgender population creates a strong financial incentive for opening and operating pediatric gender clinics. This can lead to over-medicalization and potentially harm vulnerable youth.

• Lack of Evidence: The author highlights the lack of robust evidence supporting the efficacy of hormonal interventions for adolescents, citing the Cass Review and other systematic reviews.

• Media Credulity: The author criticizes media outlets (like NPR) for uncritically accepting the “0.1% is small” narrative without proper fact-checking or consideration of the larger context.

In essence, the article claims that focusing on the 0.1% figure is a smokescreen to hide the fact that far too many adolescents are being medicalized without proper diagnosis or evidence of benefit. It also points to potential financial incentives driving this trend and challenges the media’s reporting on the issue.”

*****

At least my arguments seem clear to our AI overlords. And so I will keep it at the top for time-deprived humans. And for those humans who have the time to read the entire thing, the complete post appears below.

---

In a new Research Letter published in JAMA Pediatrics, researchers (Hughes hereafter) found that, by the time they reach 17, about 100 out of 100,000 minors under private insurance, or about 0.1%, were receiving cross-sex hormones. These numbers led them to conclude that “receipt of… hormones was rare.” NPR duly interviewed the study’s lead author and framed the issue similarly: “How many transgender teens in the U.S. are receiving medical care related to gender transitions? According to a peer-reviewed research letter published Monday in JAMA Pediatrics, the answer is very, very few.” If NPR hadn’t been so credulous of the narrative and had instead tried to dig a little bit more, they would have found that the numbers that the researchers found helps explain the birth of the pediatric gender industry: how, from zero clinics just a few years back, we now have more than 100, as Reuters found in their analysis of the industry in 2022.

The researchers’ claim is that since the numbers are “very, very small,” why are the critics so fixated on it? It's only 0.1% of the population! Stop trying to interfere with the “care” we're providing, you transphobic bigots (incidentally, this “care,” the Grande Dame of Gender Medicine now informs us, may not be “effective” in the medical sense).¹

Pardon me for injecting a sense of reality here. As a percentage of the entire adolescent population, the number of adolescents on hormones has to be small. As Hughes mentions, 1.4% of the entire adolescent population identifies as transgender or gender diverse (TGD), and “[s]ome adolescents who identify as TGD require medical interventions.”

As Jack Turban pointed out, many of the adolescents who identify as TGD are merely rejecting gender role stereotypes: they feel no gender-related distress and have no intention to change anything in their lives beyond their pronouns. Thus, only a small fraction of this 1.4% feel any gender-related distress, and a fraction of this fraction feel distressed enough to go to a medical provider and seek interventions. And presumably, after the “comprehensive mental health diagnosis” that Turban refers to, only a fraction of this fraction of a fraction goes on to receive cross-sex hormones. It is a mathematical reality, therefore, that, expressed as a fraction of the entire adolescent population, the percentage of adolescents receiving cross-sex hormones HAS to be small.

If it helps, I will try to illustrate this graphically (and not very elegantly!):

The 0.1% compares the adolescents who are being medicalized to all adolescents (shown by the red line). What we should really be comparing, however, are those adolescents who are prescribed cross-sex hormones to those self-identifying as transgender (the orange line). The analysis by Hughes suggests that this fraction is 0.1% ÷ 1.4% or 1-out-of-14. Is that “very, very small?” The short answer is an emphatic “No.”

As per several dozens of studies reviewed by many experts—ten of whom are authors of WPATH’s SOC8²—that fraction (i.e., number of people clinically evaluated and then prescribed hormones ÷ the total number of people self-identifying as transgender) used to be truly very small—something that would actually qualify for the “very, very small” adjective. Compared to that, the number of youth that are being medicalized today by the gender medicine industry, as per Hughes’s estimate, far exceeds—by over a factor of 20 or more—the number of youth who should have been medicalized after a proper clinical diagnosis.

(It has been argued that the 0.1% estimate is almost certainly a lowball estimate that does not account for recent increases in adolescents who are getting medicalized.³ However, let’s give Hughes the benefit of the doubt and assume that the percentage of adolescents who are getting medicalized remained steady at 0.1%.)

Clinical versus nonclinical estimates

Until the last decade or so, there was a clear distinction between individuals self-identifying as transgender and those who were, after comprehensive clinical evaluations, diagnosed with gender dysphoria and/or offered medical interventions. It was also common knowledge among experts in the area that the estimates for the former (individuals self-identifying as transgender) were several orders of magnitude greater than the latter (individuals who were prescribed medicalization after clinical diagnosis). As recently as 2021, in their systematic review (that was commissioned by WPATH) on the effect of hormones on mental health, Baker et al. distinguished between studies that “rely on clinical records” and those that “focus… on self-report among nonclinical populations” [emphasis added] (p. 2). (The Baker in question, incidentally, is Kellan Baker, one of the coauthors of the Yale Integrity Project’s “Evidence-Based Critique of the Cass Review.” This is not some random transphobe making up some arbitrary distinctions between “clinical” and “nonclinical” populations. So, when I use these terms elsewhere, please be aware that I am following the experts in this area.)

How large is the difference between the clinical and nonclinical populations? Studies that have tried to estimate the size of the transgender population have not only highlighted this distinction but also pointed to the huge difference between the estimates of the clinical and non-clinical populations. Put simply, the number of adults who identify as transgender or gender-diverse (TGD) in surveys has been noted to be around 100 times as large as the clinical population. The number of children self-identifying as TGD is about 300 times as large as the clinical population.

For example, one systematic review of 27 studies found that the estimate for “self-reported transgender identity” among adults was 128 times larger than the estimate for “transgender-related diagnoses” (0.871% vs. 0.0068%: i.e., while 0.871% of all adults, on average across these studies, self-identified as transgender, only 0.0068% of all adults received any “transgender-related diagnosis”⁴ after a comprehensive evaluation). Three of its authors—Reisner, Tangpricha, and Goodman—later became coauthors of WPATH’s Standards of Care 8 (SOC8), which Hughes cites approvingly as “the current standards of care.” Reisner is also a frequent research collaborator with Jack Turban. These are established WPATH experts who conducted this review. It is important to understand that these estimates of prevalence—whether in this review or the two other reviews discussed below—were reviewed and vetted by some of the biggest luminaries of WPATH, people who then went on to craft the organization’s standards of care. (We will set aside the many scandals associated with those standards of care for the time being.)

In another review of studies to estimate prevalence, a 2019 narrative review (all of whose six authors are coauthors of WPATH’s SOC8, including the SOC8’s lead author, Eli Coleman) found five different ways⁵ of classifying TGNB population estimates across 43 studies conducted in various locations at different times.⁶ The median estimate from the studies that estimated the percentage of those who received or requested medical interventions (in estimates arrived from countries as varied as the United States, Italy, Serbia, Netherlands, Belgium, Sweden, Spain, and Singapore) between 1968-2011 is 0.00526%. The median self-identification percentage among adults (0.515%) was nearly 100 times (97.9 times, to be more accurate) as large as the median estimate based on clinical records. Among children, the self-identification percentages were even higher: the median self-identification percentage (1.6%) was over 300 times larger (304.2 times, to be more accurate) than those based on clinical records.⁷

Since estimates from individual studies vary, a 2015 systematic review of 21 studies calculated the weighted average estimate of the clinical transgender population at 0.0046%. The review’s two lead authors—Arcelus and Bouman—are coauthors of WPATH’s SOC8.

To summarize, therefore, there are several reviews within the last 10 years—authored by the doyens of WPATH, no less—that have found the “prevalence of transgender,” when based on clinical criteria, is in the range of 5 per 100,000, with a meta-estimate of 4.6 per 100,000 (or 0.0046%). Like other studies in this area, I will use this estimate for the clinical population hereafter, just as other studies have done (for example, a 2019 study in AAP’s flagship journal, Pediatrics, cited this study, saying that “a recent meta-analysis suggests a global prevalence of 4.6 per 100,000 individuals” (p. 2)).

(Since it is difficult to wrap our heads around very small numbers, here’s one way to think about this meta-estimate of 0.0046%. Suppose you were an exceptionally good conversationalist, so much so that you met and had a conversation with three new people every single day of your life. Even for a social person like you, it would take 18 years on average to meet one person who would be part of the clinical population and qualify for medical interventions.)

So, even if we accept the findings of Hughes uncritically, the number of adolescents who are being prescribed cross-sex hormones in recent times is nearly 22 times (0.1% ÷ 0.0046% = 21.74) the number of adolescents who should have been prescribed these hormones based on clinical criteria.⁸

Calculating the actual numbers for the US population

Illustrating these percentages with concrete numbers might help. There are an estimated 25.4 million adolescents between 12-17 in the United States. The UCLA survey (that Hughes cites) estimates that about 1.4% of them, or about 356,000, identify as transgender and gender diverse (TGD). Hughes estimates that by the age of 17, 0.1% of these adolescents, or about 25,400, are being prescribed cross-sex hormones. However, if we go by the prevalence estimates established by (literally) dozens of prior studies and at least three reviews of the literature by ten different authors of WPATH SOC8, less than 1,200 adolescents (0.0046% of 25.4 million, or 1,168 to be exact)—in the entire United States—should have been offered these hormones. The remaining 24,232—i.e., more than 95% of the adolescents who were being medicalized as per Hughes’s analysis—were being prescribed the hormones without any medical justification—medical criteria that some of the top experts of WPATH agreed upon.

I have no problem with the idea that many more adolescents today are identifying as TGD. To the extent that it points to a more open society that does not adhere to stereotypical gender roles, it is to be applauded.⁹ And every gender-nonconforming child (or adult) has the right not to be discriminated against in any fashion. Their nonconformity should never be bullied, as is common for many of these children. I also have tremendous sympathy for those who go through significant gender-related distress.

However, the question here is not of societal acceptance of changing gender norms. Or whether some people suffer tremendously because of gender-related distress. They obviously do. But that does not absolve the medical community from answering questions about why they are being offered to at least 20 times the target population, and quite possibly many more adolescents, without any medical justification beyond that these people are transgender because we said so. This is even before we debate the efficacy of these medical interventions, the evidence of which is still based on, in the words of Dr. Hilary Cass, “shaky foundations.” As the systematic reviews that were published along with the Cass Review or were conducted by other countries have shown, there is no evidence that these hormones do anything at all, and that includes studies in the United States.

An evidence-based critique of "An Evidence-Based Critique of the Cass Review"

The issue is not whether the phenomenon is “rare” within the entire population

In absolute terms, 0.1% is rare—obviously! Establishing that does not require interviewing the researchers—a logical mind and a calculator would have sufficed. The question that NPR should have asked is why almost every adolescent who lands up at a Planned Parenthood center (the country’s leading provider of gender transition hormones for young adults) is allowed access to the hormones on the same day of the visit without any meaningful medical evaluation.

Unless gender clinicians in this country have suddenly discovered new clinical criteria in the last few years that have led to a more-than-20-fold increase in the prescription of cross-sex hormones among adolescents (Hint: they haven’t), what Hughes should have done after their analysis of the data is to explain (or at least question) this dramatic increase that they found in their analysis—as opposed to inanely emphasizing that 0.1% is a small percentage.

A more evidence-based reporting of the issue would be as follows: “In recent years, many adolescents—perhaps even over 5% as per a recent Pew Research survey—have identified as TGD. However, an overwhelming majority of them are part of what clinicians have long considered to be the “nonclinical population.” Dozens of prior studies have established that (based on DSM or ICD criteria) less than 0.005% of the total population of adolescents would be considered part of the clinical population, i.e., those who might have been offered access to medicalization (the lack of evidence for their benefits notwithstanding). However, the analysis of Hughes proves that more than 20 times that number are nowadays being prescribed hormones.”

It is easy to craft a narrative that seems to be “backed by math.” That is what Hughes has done. They want you to focus on just one aspect of the overall issue: “0.1% is rare! 0.1% is rare!!” At the risk of repeating myself, rare compared to what? Hughes would rather not have the media think about the larger narrative—a narrative that includes the lack of evidence for the interventions (which the systematic reviews from the UK and Sweden, for example, have shown conclusively) or how “very, very small” being used as a smokescreen to justify the absence of any medical evaluation at all the gender clinics or Planned Parenthood centers before they prescribe these hormones indiscriminately among a vulnerable segment of minors.

Activist-clinicians have tried emphasizing “very, very, small” before

Incidentally, this is not the first time that the modern crop of gender activist-clinicians has tried to insist that the percentage of adolescents receiving hormones is “small.” Take the (non-peer-reviewed) Yale Integrity Project “white paper” (McNamara hereafter) that criticized the Cass Review. One of McNamara’s criticisms is that contrary to the Review’s claim that the erstwhile NHS GIDS clinic was overwhelmed in recent years with adolescents presenting as transgender, the clinic was seeing less than 10% of UK adolescents who would have qualified for medicalization. To justify its claim, the paper estimated the proportion of adolescents who are eligible for “specialized, supportive interventions” (such a beautiful euphemism!) at a “conservative” 0.6%.

Why a “conservative” 0.6%? Anticipating that critics would argue that there is a difference between self-identification and clinical diagnosis, they claimed that “[y]outh disclosing self-identification as transgender… is distinct from our population of interest… we seek to describe youth who are transgender and may wish to consider the opportunity to discuss specialized, supportive interventions” (p. 18).

And how did they arrive at this magical 0.6%? They are from the UK 2021 census, where 0.54% of the respondents identified as transgender, and they then rounded up this percentage to arrive at 0.6% (in their own reporting of the data, the UK census rounded the number down to 0.5%, as is customary).¹⁰

Journalists shouldn’t be this credulous

We expect clinician-activists to inflate the numbers to maintain their livelihood. However, shouldn’t we expect journalists of major media organizations to do better? Before headlining the researchers’ claim that only a “very, very small” number of transgender minors are medicalized, Selena Simmons-Duffin might have paused to ask: If the numbers were really that “very, very small,” how come Reuters found that there are more than 100 pediatric gender clinics back in 2022, where there was none just a few years earlier?

If she did, the Occam’s Razor explanation would have stared her in the face. The difference between 0.0046% and 0.1% is whether there is a business case for opening pediatric gender clinics—or none. A clinical population of 1,168 in the entire country isn’t much of a customer base.

Let me illustrate with a specific example—that of the erstwhile Transgender Center within the St. Louis Children’s Hospital (SLCH), where Jamie Reed became a whistleblower. SLCH serves the area of Greater St. Louis, whose population is around 2.8 million. Assuming that the percentage of children aged 12-17 in Greater St. Louis is approximately the same as in the entire country (i.e., around 7.5%),¹¹ the estimated number of children aged 12-17 in the Greater St. Louis area would be around 210,000. 0.0046% of 210,000 kids in the age range of 12-17 is… 9.6 kids (I checked, the decimal is at the correct place).

From these 9.6 children, subtract the 0.5 children who are uninsured (5.3% of the children are uninsured in Missouri), and all you are left with are 9 children aged 12-17 (or, more accurately, their parents) who can reliably pay the bills. The “non-profit” SLCH does not report profits, but its 2023 annual report boasted that their “total net patient revenue exceeded $1 billion for the first time, underscoring [their] financial health.” For a “financially healthy” organization, it makes no business sense whatsoever to open a specialized pediatric Transgender Center that serves just 9 kids (SLCH loves large patient volumes, like “[their] 600th pediatric heart transplant”).¹² The Transgender Center makes even less sense—financially and operationally—if the majority of these 9 kids will get cross-sex hormones and relatively few will receive the more lucrative puberty blockers¹³ or surgeries.

However, what if the market size were 22 times as large? That’s nearly 200 customers—and that’s a lowball estimate from 2022. The plucked-from-thin-air estimates of McNamara promise a much larger market to reach out to in the coming years, one that is “conservatively” six times as large. That’s nearly 1,200 patients in Greater St. Louis with insane profit margins, all for immediately affirming the child and getting them started on the pipeline of lifelong medicalization, all from a niche clinic that needs no special investments at all. Now, that’s a sustainable and growing business! That’s a business any self-respecting executive would enter.

Far from establishing that the numbers are “very, very small,” what the Hughes article manages to confirm is that among adolescents under private insurance, as of 2022, the gender medicine industry was pushing cross-sex hormones to (very conservatively) more than 20 times the number who would have received them if they had undergone a comprehensive medical evaluation. Hughes has conclusively given us evidence of an industry running amok, where adolescents can go to the largest provider of such hormones and come out with a prescription the same day without any comprehensive mental health diagnosis whatsoever—a mental health diagnosis that even Jack Turban says should have taken place.

Ms. Simons-Duffin, I imagine it is not quite the story that Hughes wanted you to cover, but isn’t that your job as a journalist—to go beyond the narrative that your interviewee wants to foist on you?

As for the Research Letter, it is not for me to speculate whether this team of researchers from Harvard¹⁴—or the article’s peer reviewers in a storied journal like JAMA Pediatrics—are operating from a position of ideology, improbity, or sheer ineptitude. My guess would be a combination of all of the above. Because absent such an explanation, it is simply inexplicable that researchers with PhDs would fail to carry out a simple back-of-the-envelope calculation that would have told them that what they are proposing as their research hypothesis fails to pass the basic smell test.

1. Because otherwise, as Annelou de Vries, one of the founders of the “Dutch Protocol”, mentioned in her latest article, “the (implicit) normative expectation that GAMT [gender-affirming medical treatment] should result in improvements across multiple physical, psychological, and psychosocial outcomes risks undermining the provision of this care practice.” I agree, Annelou! How unrealistic of the critics of gender medicine to demand objective measures of improvement in medical outcomes!

2. In the order of the appearance of their names in SOC8, they are Eli Coleman (who is the first author of SOC8), WP Bouman, J Motmans, M Goodman, SL Reisner, V Tangpricha, N Adams, T Corneil, BPC Kreukels, and J Arcelus. Among them, Coleman is a past president of WPATH, and the editor-in-chief of the International Journal of Sexual Health, the official journal of WPATH. Bouman is a past president of WPATH.

3. As Ben Ryan pointed out, the study does not break down the prescription rates by year, and “[c]onsequently, it is likely that gender-transition-drug prescription rates for minors were higher, perhaps much higher, in 2022 than in the across-the-board figures the new study reported regarding the five-year span that began in 2018.” Furthermore, “[t]he investigators also did not apparently factor into their analysis prescriptions written under the diagnosis code “endocrine disorder not otherwise specified.” Healthcare providers have historically used this diagnosis at least some of the time when prescribing blockers or hormones to youth who identify as transgender. Thus, the omission of this code…may also have artificially reduced the prescribing rate reported in the research letter.”

4. The clinical diagnosis criteria in these studies ranged from GID (or GD) diagnosis under DSM /ICD criteria or by otherwise unspecified “GID diagnosis” or “transsexual diagnosis.” The studies that estimated the prevalence of self-reported transgender identity all relied on self-identification surveys. In the self-identification studies, the estimates became progressively higher as the survey questions for classification purposes went from “Self-identity as transgender” to “Incongruent gender identity,” “Desire to be treated as a person of different sex,” “Feeling as a person of different sex,” and “Not sure of gender identity.”

5. The five different ways of estimating TGNB populations were receipt of medicalization (hormones and/or surgery), receipt of transgender-specific diagnosis, self-identification among adults, self-identification among children, and legally changing name and/or gender.

6. Kellan Baker cited this narrative review in their 2021 systematic review.

7. These 100- or 300-fold (for adults and children respectively) orders of magnitude of difference between clinical and nonclinical populations continue to hold in current surveys. Hughes cited the Williams Insitute at UCLA School of Law survey that estimated the percentage of adults who identify as transgender at 0.5% (i.e., around 100 times larger than the median estimates based on clinical records) and the percentage of youth between 14-17 who identify as transgender at 1.4% (which is a little less than 300 times as large as the median estimates based on clinical records).

8. I will, for the time being, leave aside the question about the quality of evidence for medicalizing even these 4.6 out of 100,000 people who are clinically diagnosed with gender dysphoria. I will also not try to invoke the uncomfortable memories of the recent opioid crisis. Initially, as USAFacts.org points out, they were “prescribed sparingly,” and then the prescriptions became a flood, thanks to the “pill mills”—clinics, pharmacies, and clinicians inappropriately dispensing prescription medications.

9. As the Cass Review states, “For adolescents, exploration is a normal process, and rigid binary gender stereotypes can be unhelpful.” (p. 31)

10. Note that this 0.5% is exactly the same as the percentage of adults in the US who identify as transgender, as per the UCLA survey. Also, this percentage might have been overestimated, as the BBC recently reported.

11. The population of the US is around 340 million. The estimated population of children aged 12-17 is 25.4 million, which is around 7.47% of the total population.

12. The entire paragraph is quote-worthy: “Our operations continue to grow as well, with our operating room minutes exceeding 1.5 million, and we proudly completed our 600th pediatric heart transplant, marking significant milestones in our journey of care and excellence. Our total patient days rose to 115,578, reflecting a 2.09% increase from 2022. Additionally, our imaging volume grew by 5.7%, and therapy volume saw a notable increase of 15.1%. Impressively, our total net patient revenue exceeded $1 billion for the first time, underscoring our financial health and the trust placed in us by the communities we serve.”

13. The link is from an NPR news article from 2020 on the ridiculous prices of puberty blockers prescribed for central precocious puberty or CPP—which is the approved use for the drug.

14. A researcher from FOLX Health is excused. We shouldn’t expect any better from a company founded in 2019 to take advantage of the easy money sloshing around to deliver “affirmative,” “supportive,” or “inclusive” care for the “LGBTQIA+ community.”