In the past two decades, there have been a lot of new studies and professional analyses on trans issues. This is very good. What is not good is that 1) people keep wildly misinterpreting some of them, and 2) occasionally some very bad studies get promoted. I’m going to help you cut through that. I’ve collected and analyzed four examples, some brought up in popular discourse and some not, some better and some worse. This will help you avoid common mistakes in understanding the science behind transgender issues.
Long-Term Follow-Up of Transsexual Persons Undergoing Sex Reassignment Surgery (Dhejne et al., 2011)
This study is possibly the single most commonly miscited. It’s so bad that the primary author, Cecilia Dhejne, has gone on record multiple times to explain how it’s misused.
First, let’s talk about the study itself. The group they looked at were “324 sex-reassigned persons (191 male-to-females, 133 female-to-males) in Sweden” and it was a long-term follow-up from 1973–2003. The goal was to see how the sex-reassigned persons are doing after a long time, specifically with regard to mortality (how often they die), morbidity (levels of sickness), and criminal rate. In a nutshell, the study was asking “how are these transsexuals doing after 30 years, compared to the general population?” This is the conclusion they reached:
Persons with transsexualism, after sex reassignment, have considerably higher risks for mortality, suicidal behaviour, and psychiatric morbidity than the general population. Our findings suggest that sex reassignment, although alleviating gender dysphoria, may not suffice as treatment for transsexualism, and should inspire improved psychiatric and somatic care after sex reassignment for this patient group.
There’s been a frequent misinterpretation that this study shows that people who get sex reassignment surgery have a higher risk for suicidal behavior post-operation than pre-operation. This is false. It was not intended to compare how they were doing before surgery and after surgery. It didn’t even measure the mortality, morbidity, and criminal rate of pre-surgery participants. This is comparing them, post-operation, with the general population. It does not measure if surgery improved their well-being; the statement that sex reassignment alleviates gender dysphoria is based on other studies. The conclusion is that they need more help, not to be denied sex reassignment surgery. The study even outright says at one point:
the results should not be interpreted such as sex reassignment per se increases morbidity and mortality.
This misinterpretation is so common that Dhejne has spoken up multiple times, as noted above. On Reddit, she noted:
Despite the paper clearly stating that the study was not designed to evaluate whether or not gender-affirming is beneficial, it has been interpreted as such.
In an interview with Cristan Williams at Transadvocate, there’s this illuminating exchange:
Williams: Before I contacted you for this interview, were you aware of the way your work was being misrepresented?
Dhejne: Yes! It’s very frustrating! I’ve even seen professors use my work to support ridiculous claims. I’ve often had to respond myself by commenting on articles, speaking with journalists, and talking about this problem at conferences. The Huffington Post wrote an article about the way my research is misrepresented. At the same time, I know of instances where ethical researchers and clinicians have used this study to expand and improve access to trans healthcare and impact systems of anti-trans oppression.
Of course trans medical and psychological care is efficacious. A 2010 meta-analysis confirmed by studies thereafter show that medical gender confirming interventions reduces gender dysphoria.
This study shows that transgender people need more help, not less.
Decision Memo for Gender Dysphoria and Gender Reassignment Surgery
This is a memo put out by the Obama administration’s Centers for Medicare & Medicaid Services (CMS) in 2016. They had received a request to set a standardized policy for how to handle people on Medicare who seek gender reassignment surgery. So, they did a review of the literature, seeking to answer the question:
Is there sufficient evidence to conclude that gender reassignment surgery improves health outcomes for Medicare beneficiaries with gender dysphoria?
They declined to issue a National Coverage Declaration (NCD) “because the clinical evidence is inconclusive for the Medicare population.” There are two ways this memo is misused. The first is to point out its characterization of the existing evidence, in particular, that “the four best designed and conducted studies that assessed quality of life before and after surgery using validated (albeit non-specific) psychometric studies did not demonstrate clinically significant changes or differences in psychometric test results after GRS.” Of course, this is a strict set of criteria, which it admits it’s looking for:
We believe at minimum study designs should have a pre-test/post-test longitudinal design accompanied by characterization of all patients lost to follow-up over the entire treatment series as well as those patients who did not complete questionnaires, and the use of psychometric quality-of-life tools which are well validated with linkage to “hard” (objective) patient outcomes in this particular patient population (Trentacosti 2007, PRO 2009) (Appendices C and D).
This makes sense given what the CMS was trying to do. To make a NCD is to set rules defining when gender reassignment surgery will and will not be covered by a government program. The evidence might suggest it can be useful, but the memo notes that there is too much noise in the studies to make a singular, standardized approach, and therefore let it be decided on a case-by-case basis. (It’s helpful to note that one of the most popular standards of care used for transgender people, the WPATH Standards of Care, is intended to be very flexible for similar reasons.)
A second way the memo is misused is to say that the evidence is inconclusive, period, as opposed to inconclusive for the Medicare population. This is not what the memo says, as is made clear by this note:
Comment: One group of commenters requested that CMS [Centers for Medicare & Medicaid Services] consider that, “The established medical consensus is that GRS [Gender Reassignment Surgery] is a safe, effective, and medically necessary treatment for many individuals with gender dysphoria, and for some individuals with severe dysphoria, it is the only effective treatment.”
Response: We acknowledge that GRS may be a reasonable and necessary service for certain beneficiaries with gender dysphoria. The current scientific information is not complete for CMS to make a NCD [National Coverage Determination] that identifies the precise patient population for whom the service would be reasonable and necessary.
They explain the reason that it’s so important that the Medicare population be specifically looked at instead of blindly applying studies:
The Medicare population is different from the general population in age (65 years and older) and/or disability as defined by the Social Security Administration. Due to the biology of aging, older adults may respond to health care treatments differently than younger adults. These differences can be due to, for example, multiple health conditions or co-morbidities, longer duration needed for healing, metabolic variances, and impact of reduced mobility. All of these factors can impact health outcomes. The disabled Medicare population, who are younger than age 65, is different from the general population and typical study populations due to the presence of the causes of disability such as psychiatric disorders, musculoskeletal health issues, and cardiovascular issues.
It’s clear that this memo was not intended to conclude if gender reassignment surgery can be a good idea. Using it for that purpose is wrong. You’d be better off looking at the numerous studies which do try to answer that question.
A follow-up study of boys with gender identity disorder (Singh, 2012)
This one is less misused and more a cautionary tale when promoting individual studies. There has been a new interest in the question of how many gender dysphoric children stay gender dysphoric into adolescence and adulthood (persist) and how many don’t (desist). The problem is, science is hard. Studies are hard to conduct and not-great studies can get released. Once they are, sometimes people hold up sub-ideal studies as presenting the truth. So let’s talk about this study.
The study looked at 139 natal boys (that is, people who when born the doctor went “it’s a boy!”) with gender identity disorder. At least, that’s what it claims in the opening section; later in the paper, it notes that only 63.3%, less than two-thirds, of the participants actually met the criteria for gender identity disorder. This is significant. You can expect a cohort centered around an actual diagnosis to have a more precise, less varied set of responses.
The problems are compounded when you look at how the study measured persistence and desistance. 64 of the participants, almost half, used the Gender Identity/Gender Dysphoria Questionnaire for Adolescents to determine if they were persisters or desisters at follow-up. They included children who didn’t have gender identity disorder on first assessment, but classified them as desisters if the questionnaire didn’t indicate that they had gender identity disorder at follow-up. This probably skewed the desistance rate higher.
There are further problems. The study examined natal boys at one point, and then followed up with them later. The follow-up period is very dispersed, with a range of almost 3 years later to 29 years later. Outcomes should be expected to be diverse if the follow-up period is so diverse. In addition, the youngest participant was only three years old at assessment. Children don’t tend to have a well-developed concept of gender until about six, as explained in Trans Kids and Teens by Elijah C. Neely, PhD.:
Most children are aware of their birth-assigned sex, gender, and normative gender expression expectations by ages two to three years (Grossman & D'Augelli, 2007). Most toddlers know whether they are girls or boys and begin using gendered pronouns around age three. However, children may not understand the notion of gender constancy until age six to seven years—meaning that before this age, children frequently believe that sex/gender is determined by the type of clothing someone wears or their haircut or hairstyle (Egan & Perry, 2001).
It’s not a godawful study. It’s a useful contribution to answering a hard and important question: What do gender variant and gender dysphoric kids look like after several years? But it’s far from perfect. Unless you’re a professional in the field, it’s much better to look at studies in bulk instead of individual ones in an attempt to understand what’s going on.
Parent reports of adolescents and young adults perceived to show signs of a rapid onset of gender dysphoria (Littman, 2018)
Now this study pisses me off. Not entirely because of the content of it. It’s an okay study! It’s an exploratory study; it’s intended to be not-that-great but prompt more research. It does so, with repeated calls to perform better studies on more specific topics. I know I railed against it previously, but that was a bit of a mistake. The study has flaws, and there are things I would’ve done differently if I ran it, like consult trans people in the process of conducting it. What truly pisses me off is that it has been repeatedly used to support the claim that gender dysphoria is being widely spread via “social contagion", the idea that being transgender can infect cisgender children and given them dysphoria, and that numerous trans children are actually just mentally disturbed cis children. That’s not what the study says at all!
First, let’s talk about the purpose and design of the study. After explaining the circumstances which prompted this study, the author lays out the purpose in one sentence:
The purpose of this study was to collect data about parents’ observations, experiences, and perspectives about their adolescent and young adult (AYA) children showing signs of an apparent sudden or rapid onset of gender dysphoria that began during or after puberty, and develop hypotheses about factors that may contribute to the onset and/or expression of gender dysphoria among this demographic group.
The author wanted to see what parents were saying about this topic, systematize it, and try to figure out reasons why they might have this perception. To do so, the author reached out to a handful of websites known for hosting discussions about this topic and asked people to take and spread it. They received a lot of responses saying that their children’s dysphoria seemed to come out of the blue, that it developed after being around transgender people or learning about transgender people, and that it appeared related to preexisting mental health problems. That’s exactly what the author wanted. It’s not a sample of transgender children. The author wanted to see what parents were saying, and suggest reasons why.
Some of these reasons were “maybe the parents have a point.” But several of the reasons were the opposite. The study says it “is also possible that some parents might have had difficulty coping or could have been coping poorly or maladaptively with their child’s disclosure” in noting that parent-child relationships deteriorated. When it comes to parents claiming the transgender identification occurred out of the blue, the study says “it is possible that these indicators might have existed for some of the AYAs and that some parents either failed to notice or ignored these indicators when they occurred.” The author repeatedly calls for more studies, saying that it is possible that one thing or another can explain the results while rarely committing.
This is not a prevalence study and does not attempt to evaluate the prevalence of gender dysphoria in adolescents and young adults who had not exhibited childhood symptoms. Likewise, this study’s findings did not demonstrate the degree to which the onset of gender dysphoria symptoms may be socially mediated or associated with a maladaptive coping mechanism, although these hypotheses were discussed here. Gathering more data on the topics introduced is a key recommendation for further study.
It’s very open about the fact that it focuses only on the parents, and while it justifies this, it also acknowledges that to make a real conclusion we need input from the children.
Another limitation of this study is that it included only parental perspective. Ideally, data would be obtained from both the parent and the child and the absence of either perspective paints an incomplete account of events. Input from the youth would have yielded additional information. Further research that includes data collection from both parent and child is required to fully understand this condition… There are, however, obvious limitations to relying solely on parent report. It is possible that some of the participating parents may not have noticed symptoms of gender dysphoria before their AYA’s disclosure of a transgender identity; could have been experiencing shock, grief, or difficulty coping from the disclosure; or even could have chosen to deny or obscure knowledge of long term gender dysphoria. Readers should hold this possibility in mind… Although this research adds the necessary component of parent observation to our understanding of gender dysphoric adolescents and young adults, future study in this area should include both parent and child input.
At the end of the day, this line is most informative when it comes to the reception of this study (emphasis mine):
Emerging hypotheses include the possibility of a potential new subcategory of gender dysphoria (referred to as rapid-onset gender dysphoria) that has not yet been clinically validated
People who hype this study up, saying that it shows that rapid onset gender dysphoria is a thing are like the people who hype up every new nutrition study which finds that eggs are good, wait no they’re bad, actually they’re good now, ah a new study came out that says they’re bad—and so on. It’s an unhealthy approach to science. The difference is that nutrition studies like that might change how much people consume eggs, while this hurts transgender youth. The study itself is flawed but not too bad. The way people have used it? Genuinely awful.
What’s the take-away? Science sucks? Not at all! All the examples listed above contribute positively to our understanding of transgender issues. What I want people to take away from this is that science is very hard, very confusing, and bad actors can very easily distort the meaning of a technical work. In addition, an individual study or analysis should not be enough for the average person to change their beliefs or actions. I would not recommend basing your understanding of desistance on the A follow-up study of boys with gender identity disorder study I discussed above, but a meta-analysis trying to understand the desistance of children with dysphoria wouldn’t be amiss to include it.
The fact is that transgender issues are very contentious, and it’s best to trust experts like the American Psychological Association or the World Professional Association for Transgender Health. They aren’t always right, but they’re going to be right more often than someone basing their views off of an interesting study that showed up in their social media would be.