The Cass Review Into Gender Identity Services for Children - Part 4

Regret, persistence, detransition, and further mistakes in the Cass review

May 10, 2024

This is part 4 of my series looking at the Cass review into gender identity services for children in the UK. You can find the other parts here (I will update as I add sections):

One of the biggest questions when it comes to transition is something that is rarely asked for in other areas of medicine - how many people regret their care? What proportion of people who end up taking medications or having surgery to assist them in transitioning genders will go on to regret having those medical interventions?

There are obviously some health risks that go along with medical transition. For someone who maintains their trans identity long-term, those risks are usually seen as a reasonable tradeoff, but if people - especially teens - decide later in life that they aren’t in fact trans, these concerns become a bigger issue. While this is a question rarely asked for medical care for cisgender people, it is something that has become central to the debate about transgender medicine.

The Cass review looked at this issues in many places. There is discussion in the review of how many children used to persist in their gender identity, how likely kids are to regret their transition, and how many people ultimately detransition after starting on a transition journey.

Unfortunately, the Cass review’s covering of these topics is extremely problematic. The authors cite conversion clinics as evidence that transgender children rarely experience dysphoria as adults, use incredibly low-quality research to support the idea that detransition is more common than we might imagine, and almost entirely avoid the data which shows that most transgender people - including children - persist in their identities and do not regret their transition.

Historical Transition

Throughout the Cass review, the authors present the idea that the population of children with gender identity issues has drastically changed over the last few decades. Part of this is the “exponential” increase that I discussed before, but the review also talks about a key belief that is very common in opponents of healthcare for transgender children: the idea that most trans kids used to change their minds as time went on. The story goes that in the 70s, 80s, and 90s, most children who were referred to gender clinics ended up as cisgender adults, many of whom are now homosexual.

Basically, the Cass review argues that most kids used to grow out of gender dysphoria. This is used as part of the argument against providing them with medical care - there’s no point treating children who are going to change their minds anyway. Cross-gender behaviour used to be a phase, that was a good thing, now everything has changed and that’s bad.

This idea is first presented right at the start of the review:

“The main focus of the early services was on therapeutic work with children and families, with only a small proportion with persisting gender incongruence being referred for hormone treatment from around age 16. Several studies from that period (Green et al., 1987; Zucker, 1985) suggested that in a minority (approximately 15%) of pre-pubertal children presenting with gender incongruence, this persisted into adulthood. The majority of these children became same-sex attracted, cisgender adults. These early studies were criticised on the basis that not all the children had a formal diagnosis of gender incongruence or gender dysphoria, but a review of the literature (Ristori & Steensma, 2016) noted that later studies (Drummond et al., 2008; Steensma & Cohen-Kettenis, 2015; Wallien et al., 2008) also found persistence rates of 10-33% in cohorts who had met formal diagnostic criteria at initial assessment, and had longer follow-up periods.” (Page 67)

These rates of desistence - children no longer feeling dysphoric after puberty - are extremely high. The narrative presented by the Cass review is that, prior to the current practice of gender-affirming care, most children changed their minds.

But is that narrative true? Let’s look at some of the actual research cited here. The paragraph starts off incredibly weakly. Green et al 1987 is a long-term review of boys who were referred to a gender identity service in the late 60s and early 70s, which found that many of them no longer engaged in “cross-gender” behaviour as adults. Instead, the study found, these boys had grown up to become male homosexuals.

This may not be surprising when you consider what these behaviours actually were. “Cross-gender” activities included having mostly female friends, playing with dolls, a lack of participation in “rough play” (specifically sports), and improvising feminine clothing. These diagnoses were based exclusively on parental report. What this study shows is that boys who were considered effeminate as children in the early 70s mostly grew up to be homosexual and bisexual men - it’s hard to relate this to a modern understanding of gender identity at all.

The second reference, Zucker 1985, is more shocking. This is a book chapter authored by Dr. Kenneth Zucker, a Canadian psychologist who headed Toronto’s gender clinic until it was shuttered amid accusations of conversion therapy. It’s hard to know how this relates to Dr. Zucker’s current thinking - the chapter is from nearly 40 years ago, after all. However, the chapter does seem to promote what can only be described as anti-transgender conversion therapy in fairly explicit terms:

“In general, it would seem that preventing either transsexualism or transvestism is a goal that will never gather systematic opposition, but to make the same claim for homosexuality is highly unlikely. Although many parents would probably prefer that their child not become homosexual, my own experience suggests that the range of acceptance for such an outcome is considerable.” (Zucker, 1985, p.116)

“In my view, offering treatment to a child (either on his or her own or through parental consent) can be justified for a relatively simple reason. Crossgender identification constitutes a potentially problematic developmental condition. Taken to its extreme, the outcome appears to be transsexualism. To make children feel more comfortable about their sex does not, in my view, constitute an unreasonable treatment goal. Although there is considerable disagreement about how one might achieve this aim, the goal itself seems relatively benign.” (Zucker, 1985, p.117)

The chapter goes on to note various methods of extinguishing cross-gender behaviour, including everything from recommending different toys to parents to ‘treatments’ involving “reinforcement and punishment” which is not specifically described. There are two issues with the Cass review citing a document that explicitly promotes conversion therapy. The first is that the review repudiates conversion therapy, arguing that “If an individual were to carry out such practices they would be acting outside of professional guidance, and this would be a matter for the relevant regulator” (Page 151). It is hard to square this stated aversion to conversion therapy with the review’s positive citation of a book chapter endorsing the practice.

Secondly, the review uses this book chapter to review the number of children who persist in their gender identity into adulthood. Whether or not the actual practices at Dr. Zucker’s clinic were conversion therapy as his detractors claim, the data on children who were treated at this time and in this manner is unlikely to have any bearing on modern gender clinics. It’s a bit like parole officers asking recently released convicts if they’re doing crimes again - you can’t rely on such a biased form of self-report.

After citing these extremely problematic papers, the Cass review then goes on to talk about this 2016 review that aggregated together 5 separate studies. These 5 studies are:

The same 1987 Green clinic paper that is discussed above.
Drummond et al 2008, a follow-up of girls treated at Dr. Zucker’s clinic (he is the last author) between the 70s and early 00s.
Follow-up data from Dr. Money’s clinic, which suffered some very similar criticisms to Dr. Zucker (some of the examples of problematic therapies in Dr. Zucker’s book chapter come from Dr. Money’s clinic).
Singh 2012, another follow-up from Dr. Zucker’s clinic.
Wallien et al 2008, which looks at patients referred to the Amsterdam clinic in the 80s, 90s, and early 00s.

Oddly enough, the primary analysis of the 2016 review does not discuss Steensma at length, but since this concerns the same patient population as Wallien et al (and has one of the same authors), there’s no great issue there.

So what we see here is that the Cass review is double-counting the same clinics, with the same patients, to make the dubious claim that in prior periods gender identity was vastly different to today. The highest estimate in the review - which is the 33% figure - does not actually appear in any of these papers that I can see. This likely comes from the Wallien paper, which notes that 27% of its sample were still dysphoric in adulthood, but they also had no response from 30% of their clinic patients. In fact, if you only include people who actually participated in this research study, about 40% of them were still experiencing clinically diagnosed gender dysphoria as adults.

This study is also somewhat challenging to interpret because the authors looked at whether children who mostly had clinically diagnosed gender identity disorder (as per the DSM III) still had rates of dysphoria consistent with a clinical diagnosis as adults based on a single survey. It’s hard to know how this might relate to whether adults who used to be dysphoric children self-identify (as trans or otherwise), which makes it very difficult to compare these figures to a more contemporary understanding of gender identity more than 20 years later.

This all seems finicky, but it is very important. It is not just that the Cass review positively cites a book chapter that promotes conversion therapy - it is that these clinics define the context for the entire review. The Cass review refers several times to how drastically the clinical population has changed from the “early services” of gender therapy, but never discusses how these services were provided in a context where “transsexualism” was a formal mental health diagnosis, homosexuality was criminalized or recently decriminalized, and gender-related care mostly focused on convincing children that they were not, in fact, transgender. In fact, if you look at more modern clinics - especially in the late 90s and early 00s - the proportion of children desisting from their transgender identity had already dropped substantially from the 70s.

Regret and Detransition

As I’ve written before, the concepts of regret and detransition are complex and interrelated. People may regret their transition without detransitioning. They may detransition without regret. They may both regret and detransition and then go on to retransition later, especially if the reasons for their regret are external (i.e. a lack of parental support as a teen). It’s not a simple good/bad experience, because the experience of transition is itself very complex. Nevertheless, these concepts are vital in the discussion of transition-related care.

The Cass review refers to regret and detransition in many places, but focuses specifically on detransition about halfway through in section 15. One major concern throughout the review is that clinicians are not paying close enough attention to patients who might have regrets with their medical transition:

“Based on the MPRG review of the notes, this history/journey was rarely examined closely by GIDS for signs of difficulty, regret or wishes to alter any aspect of the child/young person’s gender journey trajectory.” (Page 159)

The review also argues in several places that the NHS’s own data is too weak to reliably estimate of the number of people who transition that later detransition, largely due to a lack of long-term follow-up:

“Estimates of the percentage of individuals who embark on a medical pathway and subsequently have regrets or detransition are hard to determine from GDC clinic data alone. There are several reasons for this: those who do detransition may not choose to return to the gender clinic and are hence lost to follow-up. The Review has heard from a number of clinicians working in adult gender services that the time to detransition ranges from 5-10 years, so follow-up intervals on studies on medical treatment are too short to capture this. The inflection point for the increase in presentations to gender services for children and young people was 2014, so even studies with longer follow-up intervals will not capture the outcomes of this more recent cohort.” (Page 188)

However, these arguments make little sense. As I’ve previously pointed out, the statement about the “inflection point for the increase in presentations” is incorrect. In addition, the argument about loss to follow-up and detransition taking 5-10 years is confusing when you look at the data that the Cass review includes from something called the MPRG.

The MPRG - Multi Professional Review Group - was a consequence of the court case that started the Cass review. It was an effort by the NHS to ensure that clinicians prescribing medications to trans kids had followed appropriately robust steps in their treatment prior to getting hormones or puberty blockers. The MPRG reviewed about 170 children between 2021 and 2023, of whom roughly 75% were treated entirely correctly, and about 25% had some measure of concerns.

One of the things that the MPRG noted was the average time that children had been treated by the gender identity services prior to their review, and the time that they had been waiting before even seeing the service. On average, children who were referred to endocrinology - remember, all this is happening before the clinic is even allowed to prescribe any medications - were originally referred for gender assessment at age 8. They then waited on a waiting list for 1.5 years, before seeing the gender identity service just shy of their 10th birthday. On average, these children then spent another four years in treatment, before first being referred to the endocrinology team to discuss whether they should potentially have medical treatment.

We also know that the rate of detransition within the NHS gender clinics was extremely low. Of the entire group of 3,306 patients who were analyzed as part of the Cass review, fewer than 20 (<0.5%) were reported to have detransitioned based on an analysis of their clinical records. There were around 140 (4.2%) children who declined treatment in the sample, some of whom may have detransitioned as well, but either way the overwhelming majority of people seen through this service did not detransition during the time period studied. Even if we include every person who was lost to follow-up, declined treatment, and detransitioned in this sample, the rate would still be below 25% of the total.

In other words, none of the dot points the Cass review lays out as reasons to completely discard this population make any sense. For a child to be seen at the gender identity services, they must have by definition spent more than a year on a waitlist. They then would on average spend at least 4 years in treatment. For children prescribed hormones or blockers, that timeline would be more like a decade in total from referral to discharge. The review has plenty of data showing persistence with more than 5 years of follow-up, and in this sample even if you were to assume that every person lost to follow-up desisted in their transition - something that we know is false - you’d still get a minority of children doing so.

This is actually supported by other data which the Cass review cites and then largely ignores. As part of its discussion about puberty blockers (which I will be getting to in a piece soon) the review cites the two most robust estimates of desistance in children from medical transition, which are these two studies from the Netherlands. The authors of these two studies had access to the linked medical record of every child or teen who had attended the Dutch clinics for gender dysphoria, and looked at how many of those people had continued with puberty blockers and then gone on to take cross-gender hormones as adults.

Both of these studies had extensive follow-up - the median length of observation for each paper was over 4 years. Given that the clinic had been running for 20 years when these studies were done, there were a reasonable number of people who have had gender-affirming medication for more than a decade in these datasets. Due to the linked nature of the data, and the Dutch centralized registry for medications, there is close to 100% follow-up of the individuals.

Of the young people treated at the Dutch clinic, only about 5% stopped taking medications. Even at the longer follow-ups, the rate of medical detransition was extremely low. This is an objective indicator of how rare regret is for children who start on a medical pathway.

Startlingly, the Cass review does not cite these papers in their discussion of detransition. Instead, more weight is given to the anonymous surveys of Redditors that I’ve discussed before, with the review even pulling out a graph from one of these atrociously bad studies. It is hard to square the review’s stated goal of relying on best evidence with their citation of some of the weakest academic literature that I’ve ever seen instead of the many better studies that discuss detransition and regret.

Modern Medical Care

Throughout the Cass review, there’s a strong implication that there are a group of as-yet-unidentified children who the authors are protecting. A group of kids who have been let down by their parents and clinicians but who will one day greatly regret transitioning genders. The review even goes as far as to note that some people who don’t regret their transition may simply not have an adequate frame of reference to understand the negative outcomes of their transition journey:

“In the absence of any experience as an adult ciswoman, they may have no frame of reference to cause them to regret or detransition, but at the same time they may have had a different outcome without medical intervention and would not have needed to take life-long hormones.” (Page 195)

But what the data shows - even the review’s own figures - is that this group is very small. Yes, there are children who will regret their transition. Some will go on to detransition, and may even suffer lifelong consequences of hormones and other medications. But these children represent the minority of those who turn up for treatment at gender clinics such as those run in the UK. Even the systematic review commissioned by the Cass review and conducted by the University of York into this subject found very low rates of regret and/or detransition. While the follow-up in most of the studies included in this review was genuinely quite short - this systematic review was done before much of the research I’ve discussed in this piece was published - the rates of desistance or regret were in most cases well below 10% of children who underwent medical transitions.

When you look at the studies here, this isn’t surprising at all. Once we moved on from the 70s and 80s, where behaving slightly out of the norm for your assigned gender was considered pathological, it’s clear that relatively few people substantially changed their mind about their transition journey. More modern evidence shows across the board that detransition and regret are far from the norm. It seems pretty likely that people regret transitioning less these days at least in part for social reasons - while the experience is still stigmatizing, it’s probably quite a bit better than in the 1970s. Also, at least in the UK, to get medications you had to really want it - remember, the average time to getting puberty blockers was over five years.

Of course, as with any subject, there is some complexity here. Some people may not regret their transition, but may regret their overall experience. Some people may not detransition, but may deeply regret their choices. It is also entirely possible that we will find at some point in the future that regret rates become much higher in people who transition - anything is possible.

But ultimately, we have to rely on the evidence that we have gathered so far. There is quite strong data showing that the speculation in the Cass review is broadly incorrect, including the review’s own evidence.

Gideon M-K: Health Nerd

Discussion about this post