(2023-02-23) Singal The New Highly Touted Study On Hormones For Transgender Teens Doesnt Really Tell Us Much Of Anything

Jesse Singal: The New, Highly Touted Study On Hormones For Transgender Teens Doesn’t Really Tell Us Much Of Anything. As I noted in that post, earlier this month The New England Journal of Medicine published a highly anticipated study called “Psychosocial Functioning in Transgender Youth after 2 Years of Hormones.” The research team has spent years following a cohort of kids who have been administered puberty blockers or hormones (hormone therapy) at four participating clinics

According to the authors, the kids showed key improvements two years later.

But that’s a questionable interpretation of the results for a number of reasons

In their study protocol, including a version that they submitted into a preregistration database, the researchers hypothesized that members of this cohort would experience improvement on eight measures. Then, in the published NEJM paper, the researchers changed their hypothesis and six of those variables were nowhere to be found. The two remaining — anxiety and depression — moved in a positive direction for trans boys (natal females) but not trans girls (natal males).

cherry-pick five that did show some improvement might not mean anything at all

For this post, though, let’s temporarily set aside this potentially crippling issue.

At the risk of repeating myself, I am opposed to the sorts of policies Donald Trump is proposing — both outright restrictions on youth gender medicine and his even more radical proposal to codify into federal policy a ban on even adults changing their legal sex. That latter part, in particular, is downright cruel and pointless

To be clear, this New England Journal of Medicine study is a significant improvement over what passes for research in the area of youth gender medicine (though that’s a low bar to clear).

This post will be organized around the following main points:

The Kids In This Study Had An Alarmingly High Teen Suicide Rate

of this study’s 315 kids

The NEJM researchers did administer a suicidal ideation scale to capture this in richer detail, but as we know from the last post, they simply didn’t report that data

Most Of The Improvements The Cohort Experienced Were Small

Life satisfaction: Increase of 4.64 out of 100 points

*Depression: Decrease of 2.54 out of 63 points

Anxiety: Decrease of 2.92 out of 100 points*

The first question you should ask in a situation like this, where you have statistically significant improvements that look small in magnitude, is whether they matter. You can have a statistically significant effect that isn’t clinically significant, meaning it wouldn’t represent noticeable improvement or worsening in the condition being measured

This is a genuinely complicated situation to interpret, partially because the baseline numbers are such a mixed bag: It’s not fair to harp on the lack of improvement in a kid who wasn’t doing that poorly to begin with

This is an issue in some youth gender medicine clinical research, which typically involves cohorts who have been screened for serious teen mental health problems beforehand: They don’t have much room to improve, so the deck is somewhat statistically stacked against researchers seeking to demonstrate mental health improvements

It would also be useful to know if the relatively large proportion of kids who didn’t provide data at 24 months, which is about a third of them, differed at other time points from the rest of the group, because if the kids with missing 24-month data were on average doing better or worse than their peers in the study, that could seriously skew the results.

At the end of the day, it seems hard to deny that a lot of kids stayed unwell despite two years of regular access to a gender clinic and to a medication designed to improve their mental health outcomes

It’s Impossible To Attribute The Improvements Observed In This Study To Hormones Rather Than Other Forms Of Treatment That Took Place At These Clinics

This is a longitudinal study without a comparison group.

Every kid in this study was seen at one of these multidisciplinary clinics. So it stands to reason that the ones with serious anxiety and depression issues were likely provided with access to psychotherapy, medication, or both.

This means that even when it comes to the subset of kids whose improvements were sizable, we have to ask: Did their symptoms abate because of the hormones, the medication, or the therapy? Or was it some combination of all three?

There’s no way to know.

There’s one other potential problem with pinning the improvements noticed in this study on the powers of gender-affirming hormones, per se. It was pointed out to me by a depression researcher with whom I exchanged some emails

He pointed out that the researchers “utterly ignore the obvious point that testosterone has large mood-elevating, anti-anxiety, and antidepressant effects!”

The One Bigger Improvement Was In A Variable That Might Not Mean All That Much

*The authors do highlight one more impressive-seeming finding, though:

Increasing appearance congruence is a primary goal of GAH, and we observed appearance congruence improve over 2 years of treatment*

*Specifically, over two years the kids in the study experienced about a one-point improvement on the five-point appearance congruence scale of the Transgender Congruence Scale.

The TCS is a 10- or 12-item instrument with the following items, as summarized in a table from a 2021 study in Sexuality Research and Social Policy*

it seems almost impossible to imagine how someone’s scores on an item like this wouldn’t “improve” as the physical changes of hormones took hold and brought their body in line with who they felt they were on the inside

I’m just not sure how impressed we should be by this.

That’s doubly true when you notice that the appearance congruence scale doesn’t appear to really correlate with other, more robust measures of well-being, anyway. Or at least that’s what the authors of the 2021 paper found

Ritchie wrote in an email. “The causality could easily have just gone the other way (people who felt better, mentally speaking, tended to worry less about their appearance).”

The Researchers Don’t Even Consider The Possibility These Treatments Are Less Effective Than They Thought — Their Only Answer Is “More Hormones

We’re now many years and many hundreds of thousands of dollars into this research effort. The researchers have published what was supposed to be one of their blockbuster studies, and it is missing absolutely crucial evidence we need in order to evaluate these treatments