How WPATH Thumbed The Scale Of Youth Gender Medicine Research

This isn’t how science should work

Update, 5/29/2026: It turns out that there’s some real uncertainty over the role of the Endocrine Society here. See here for a more information on that.

---

In 2014, Johanna Olson-Kennedy, medical director of the Center for Transyouth Health and Development at Children’s Hospital Los Angeles, was taking the lead on an ambitious research proposal known as “The Impact of Early Medical Treatment in Transgender Youth.” Olson-Kennedy and three other established youth gender-medicine practitioners — her co-investigators Stephen Rosenthal of UCSF Benioff Children’s Hospital, Norman Spack of Boston Children’s Hospital, and Robert Garofalo of Lurie Children’s Hospital in Chicago — were seeking millions in funding from the National Institutes of Health to conduct a yearslong effort to study the outcomes of youth who came through their four clinics. They planned to publish separate studies on the effects of puberty blockers and cross-sex hormones.

Suffice it to say that this sort of effort requires a mountain of paperwork, and I obtained some of that paperwork via a Freedom of Information Act request. The packet reveals that during the application process, the team hit a potential snag when the NIH reviewers evaluating the proposal expressed qualms about its being an observational rather than an experimental study.

This distinction is quite important for understanding what happened next and why it matters.

An observational study simply tracks a group of patients over time as they receive treatment. The problem is, if the patients’ conditions improve it’s difficult to know whether the medication itself is responsible. People who are sick often get better over time, without intervention. Or it could be that, while the subjects were receiving the treatment, they were also receiving other forms of care or advice from their doctors. There can also be strange and not fully understood mind-body mechanisms at work, including the placebo effect — sometimes people appear to get better simply because they think they’re getting treated. Observational studies might, in some cases, also be susceptible to the Hawthorne effect, in which patients who know they are being observed modify their responses to certain queries as a result. These problems can be addressed by attempting to statistically control for unobserved factors that might be contributing to patient outcomes, but there are real limits to what observational studies can reveal, and they can’t generally provide strong causal evidence about the efficacy of medical treatments.

An experimental study, on the other hand, often involves randomly assigning study participants to either receive the treatment being evaluated or something else. Sometimes the “something else” is a placebo — that is, half of the headache sufferers get a headache pill, while half get a sugar pill — and sometimes it is some other treatment that the new treatment is being compared to.

In the strongest, most idealized version of this type of study, every effort is made to treat the patients identically with the exception of the treatment in question, and patients, doctors, and almost everyone else directly involved are “blinded” as to who is in which group until the experiment itself is complete and it is time to analyze the results. In theory, this prevents patients in one group or another from being treated differently by doctors and others who interact with them. Researchers go to significant lengths to complete what are known as “double-blind randomized controlled trials,” the gold standard for medical research. In some cases, they even perform “sham” surgeries for the sake of RCTs.

In a letter responding to the NIH’s critique, Olson-Kennedy and her colleagues acknowledged that “by our estimation [it] was the most significant issue raised by the initial review,” but then argued against the idea of withholding puberty blockers from any of their study participants:

However, the current standards of care from the Endocrine Society and the World Professional Association of Transgender Health (WPATH) were developed from a panel of scientific experts and a consensus-building process based largely upon clinical experience and existing minimal research, and each supports the positive impact of intervention and the negative impact of no intervention. In the context of these guidelines, we strongly believe that a non-intervention control group, or certainly any control group as part of a traditional randomized trial, would be both impractical and unethical (see the Endocrine Society and WPATH letters of support regarding our cohort design in Appendix A). [Italics mine.]

It’s true that you can’t blind a study participant (or their family members) as to whether or not they have really received puberty blockers — not for long, at least. But methodologists have come up with ways to run solid, nonblinded (or “open -label”) experimental studies, so the choice isn’t just between a blind experimental study and an observational one.

Olson-Kennedy’s more important argument here is that, because the available evidence and guidelines “support[ed] the positive impact of intervention and the negative impact of no intervention,” it would be unethical to deny study participants access to puberty blockers.

It’s true that by the time Olson-Kennedy and her team were seeking funding, the Endocrine Society (2009) and WPATH (2012) had both published guidelines supporting the use of puberty blockers to treat gender dysphoria in at least some instances. But those guidelines — written by teams dominated by early practitioners of youth gender medicine — were not carefully written, as an independent Cass Review assessment later found (see online supplemental table 3). They did not include many of the guardrails that seek to attenuate the role of human bias in guideline development, and they recommended puberty blockers for gender dysphoric youth in the absence of quality evidence supporting this approach. Jennifer Block’s reporting in The British Medical Journal has more information about all of this.

The first group of clinicians to prescribe puberty blockers as a treatment for gender dysphoria certainly thought this treatment worked, despite that lack of evidence, and sought to codify that view into the relevant professional guidelines. Olson-Kennedy’s co-investigator Norman Spack, for example, was one of the earliest American youth gender clinicians to embrace puberty blockers, and he also co-authored the Endocrine Society guidelines. Given the prominence of the Endocrine Society, those guidelines went a long way toward legitimizing a medical protocol that, in 2009, could only be described as experimental. “There was an attitudinal shift to be able to say that the Endocrine Society supports this,” Spack told Endocrine News in 2013 (I came across that quote via the sociologist and youth gender medicine critic Michael Biggs).

There are good, well-understood reasons not to automatically trust physicians who believe a new treatment works but who lack data. That’s why the “evidence pyramid” underpinning the practice of evidence-based medicine puts “background information/expert opinion” at the bottom.

Doctors, like everyone else, are susceptible to bias. And absent a carefully controlled experimental setting, they don’t generally collect rigorous data on their patients, nor attempt to follow up with those whom they are no longer in contact with. All of which can give physicians a skewed understanding of the effectiveness of the care they are providing.

So while Olson-Kennedy’s argument to the NIH is larded with signifiers of authority — phrases like “panel of scientific experts” and the names of leading professional organizations — if you actually examine the timeline and the names involved, it really just boils down to “The small group of early practitioners of youth gender medicine believed puberty blockers worked as a treatment for youth gender dysphoria and enshrined their belief in the early guidelines.”

Yes, a couple of early, influential articles were published by the Dutch clinicians who pioneered the puberty-blocking protocol in 2011 and 2014, but they were methodologically weak observational studies of a carefully screened group of Dutch youth, and they did not tell a conclusive story: In many cases, the youth entered the study with solid mental health and they still had solid mental health when they were last observed. Even setting aside other, potentially backbreaking issues with this Dutch research that wouldn’t be recognized until years later, the studies didn’t come close to providing the quality of evidence required to justify the stance that non-intervention would be unethical.

That’s especially true given that, as Olson-Kennedy and her team explained in their application, they were seeking, in some cases, to begin puberty blockers at an earlier age than the usual Dutch threshold of 12 years old. As they wrote, “no data have been collected in the U.S. on the physiologic and mental health impact, safety, or tolerability of pubertal-blocking medical interventions… for transgender youth, particularly in children younger than age 12, leaving a gap in the evidence for this practice,” emphasis theirs.

Olson-Kennedy and her colleagues, then, simultaneously argued that there existed “no data” on the “mental health impact, safety, or tolerability” of blockers, particularly in this cohort — again, these are all their exact words — but also that it would be unethical to withhold this medication from any patients. These two claims are very much in tension.

As the team understood, 2014 was a crucial time to develop genuine evidence about youth gender medicine. And there were options available for an experimental study, at least hypothetically. For example, the Dutch clinicians had, before blockers came on the scene, provided counseling and other mental-health support to gender dysphoric teenagers until they reached age 16, at which point they could begin hormones if they were deemed to be good candidates for medical intervention (the counseling did not include attempts to convince the youth they weren’t trans, or anything like that). Olson-Kennedy and her colleagues could have at least attempted to run an experiment comparing that approach (or a version of it) to the newer protocol involving earlier medical intervention via puberty blockers. This sort of comparative study would not have been perfect, but it could have provided more useful data than an observational study.

Instead, they argued that an experimental study that withheld puberty blockers would be unethical, and to buttress their argument they pointed the NIH to “letters of support” from the same two groups that had published the early guidelines. The letters, which were not included in the original packet the NIH sent me but which a follow-up FOIA request yielded many (many, many) months later, are telling. They show that by 2014, despite a lack of quality evidence to support puberty blockers as a treatment for youth gender dysphoria, both organizations had decided not just that puberty blockers worked, but that they worked so well that it would be unethical to deny them to study participants. Given how much uncertainty remains about these treatments more than a decade later, this is a striking example of how individuals and institutions have, in their embrace of youth gender medicine, contravened normal standards of medical evidence.

The Endocrine Society letter was authored by the endocrinologist Wylie Hembree, who died in 2022 and who at the time was the Chair of the Society’s Task Force on Clinical Practice Guidelines for Transgender Care. He was also lead author of the 2009 guidelines. Writing on Endocrine Society letterhead, In his letter, he argued to the NIH that “Non-intervention in adolescents with Gender Dysphoria has been demonstrated to result in harmful physical and psychological outcomes, especially at the onset and during pubertal development,” and that “Including a randomized non-treatment arm would be unethical based upon existing data documenting adverse effects of failure to treat.” But whether non-treated adolescents with GD sometimes suffer bad outcomes is a different question from whether puberty blockers (followed by cross-sex hormones) effectively treat GD in the long run and are worth the potential downsides, and it was those latter questions (among others) that Olson-Kennedy and her team were seeking to answer. (Correction, 5/26/2026: The letter wasn’t written on Endocrine Society letterhead, but Hembree did list himself as “Chair, Endocrine Society Task Force on Clinical Practice Guideline Transgender Care,” and Olson-Kennedy referred to the letter as having come from the Endocrine Society. I’m reaching out to the Endocrine Society again to attempt to clarify this.)

The WPATH letter was written by Jamison Green, then the president of WPATH and the chair of its Ethics Committee (as well as a big name in trans activism more generally), and by Jamie Feldman, chair of the organization’s medicine and research committee. Both had been involved in the development of the 2012 WPATH guidelines. The letter read, in part, “There is no question that pubertal blockers block puberty,” the implication being that an experimental study would be unnecessary. The authors also argued that “We have generations of transgender people who have been denied puberty blocking treatment, and we know the trajectory of these people’s lives.” Citing the high suicide risk of transgender youth, they argued, “The prospect of entering a program for treatment with a chance that treatment will be withheld could endanger the mental and physical health of the subjects.” This is question-begging, given that Olson-Kennedy and her team were specifically investigating whether puberty blockers would ameliorate suicidality among study participants. (I reached out to both the Endocrine Society and WPATH for comment. I received a call from a public relations firm representing WPATH and, after a quick off-the-record chat about this story, didn’t hear anything further. I didn’t hear back from the Endocrine Society.)

There are, of course, situations in medical research where withholding treatment would be unethical. But that’s only the case when there’s a great deal of certainty about the treatment in question (the relevant technical term here is equipoise, or a situation of genuine uncertainty about which approach is best). There was no such certainty when it came to the use of puberty blockers: There were always obvious potential side effects worth grappling with concerning infertility, sexual dysfunction, bone health, and cognitive development, as critics of youth gender medicine have noted. Olson-Kennedy and her team, as we saw earlier, were well aware of these and other uncertainties, and even the aforementioned WPATH guidelines point out that “the long-term effects [of blockers used for gender dysphoria] can only be determined when the earliest treated patients reach the appropriate age.” It’s simply a misunderstanding of basic bioethical principles to claim, under conditions of such profound uncertainty, that it would be unethical not to treat.

These letters demonstrate that the Endocrine Society and WPATH both endorsed the idea that puberty blockers were already an established, effective treatment for gender dysphoric youth — that many of the very questions Olson-Kennedy and her colleagues were seeking to answer had already been answered. (Which, of course, would raise the question of why they needed federal research dollars in the first place.) They were not convincing responses to the issues raised by the NIH. Nonetheless, the NIH allowed the study to proceed with an observational design — no comparison group.

In August 2015, Olson-Kennedy and her co-investigators received an initial $5.7 million to begin their project. More than a decade and about $10 million later, they haven’t come close to answering the key questions they set out to answer. Their most important publication, a 2023 New England Journal of Medicine study on hormones, is rife with cherry-picking, outcome-switching, and other signs of shoddy and/or politicized science. And even if you accept the team’s highly suspect methodological choices, their results do not offer compelling evidence to support the administration of hormones to adolescents with gender dysphoria — not least because there is no comparison group, and the authors did not control for therapy and/or medications like antidepressants.

As for the blockers study, in 2024 Olson-Kennedy told Azeen Ghorayshi of The New York Times that “Puberty blockers did not lead to mental health improvements… most likely because the children were already doing well when the study began,” as paraphrased by Ghorayshi. Olson-Kennedy admitted that she and her team had withheld this data out of fear that it would be weaponized by conservatives seeking to ban youth gender medicine. The final version of that study still hasn’t been published, but a preprint of it (that is, a pre-peer-review version) published online last year appeared poised to repeat many of the mistakes of the 2023 NEJM study.

The failure of gatekeepers here can’t be overstated. WPATH and the Endocrine Society have attempted to present themselves as rigorous-minded groups that follow the science, but in this instance they acted much more like activist organizations (not for the last time, especially in WPATH’s case). Of course, it didn’t help matters that the project’s lead, Olson-Kennedy, had already publicly described these treatments as lifesaving years before she began studying them: As she told ABC News in 2011, “We often ask parents, ‘Would you rather have a dead son than a live daughter?’ ”

At the risk of repeating myself, I can’t emphasize enough how foolish it is to blindly “trust the experts” unless the experts in question have earned that trust. There are myriad ways the difficult process of scientific inquiry can go off the rails, and “The Impact of Early Medical Treatment in Transgender Youth” is a prime example.

---

Questions? Comments? I’m at singalminded@gmail.com. Image via the Endocrine Society.