Persistence and Desistance
On April 20, the Florida Department of Health released guidance saying that children should not be allowed to socially transition, let alone receive puberty blockers or hormones. It ought to go without saying that the guidance is factually shaky. It selectively cites relevant studies and misrepresents several of the studies it does cite. This much is easy to demonstrate: the first bit of evidence is from a single author with no relevant experience published in a Catholic bioethics journal, the sixth link directly contradicts the claim the guidance tries to make, and links seven and eight simply do not contribute to the argument it implies they do. It avoids several studies published within the past three years which point to the opposite conclusions. It is simply bad guidance.
In the face of this, a reasonable person operating in good faith would be inclined to ignore the guidance. Nevertheless, much discussion has been dedicated to one particular claim:
A paper published in the International Review of Psychiatry states that 80% of those seeking clinical care will lose their desire to identify with the nonbirth sex.
I have discussed desistance before, but I want to talk about it more.
First, I want to discuss the terms persistence and desistance. I will get around to the hard science, I promise. It is better to first situate the science in reality rather than act as though the categories of persistence and desistance were hand-delivered to us from God above.
If you start searching around for materials on “desistance” with no other qualifiers, you will get a lot of stuff about efforts to get people to stop committing crimes. In preventing crimes, it is important to keep offenders from becoming serial offenders, so people study what efforts make them desist from committing crimes. Most people know the term desist from cease and desist letters, which threaten prosecution and conviction if the recipient does not stop certain activities.
It might feel like a stretch to say that this implication carries over to discussions of desistance of gender dysphoria, but the simple fact is that it does. As Jules Gill-Peterson, editor of Transgender Studies Quarterly and author of Histories of the Transgender Child, has explained, a standard prerequisite for medically transitioning has always been to first try everything to get the person to desist. The logic is that being transgender is undesirable, so engaging with it positively should be a last resort. This mindset remains prevalent both in general society and among medical professionals. The legal approach to gender variance makes the comparison to criminal desistance even clearer, since in the United States expressing gender variance was illegal until relatively recently.
This context is important when we discuss desistance. The idea that most children desist comes as a relief to many, a justification for not changing anything at all. (After all, that is exactly how it is used in the Florida Department of Health guidance.) For some reason or another, being transgender is bad, a sin, a nuisance to others, an expense, a life path which leads only to worse outcomes. It is never good, an opportunity, inspiring, or even just a value-neutral (if uncommon) life path.
The other part of desistance that’s worth understanding is that it is a very whacky category. In the DSM-III, the diagnosis for gender identity disorder (GID) for “males” and “females” is nearly identical, except for “males” one can qualify in part by being interested in women’s clothing or stereotypically feminine activities, while the reverse does not apply to “females.” DSM-III-R introduced similar requirements for “females,” but still didn’t quite make it symmetrical. DSM-IV finally reached symmetrical requirements, but a person could still qualify for GID without any stated desire to be or insistence that they are a gender other than what they were assigned at birth. (Part of the reported rise of AFAB people presenting to gender identity clinics can be attributed to the end of this asymmetry.) I go over these three definitions because they are the most commonly used in studies which aim to measure desistance rates.
Because these definitions are so unusual, it makes for some questionable results. Desistance studies persistently include youth who are sub-threshold for GID. Because GID is already an uneven and loose category, including sub-threshold youth means the studies poorly target transgender individuals. That was always the point, though. The point was to find gender-atypical youth and attempt to get them to desist from nonnormative behavior. When we talk about desistance and persistence in the context of these studies, we are most properly talking about if already medicalized youth stay gender-variant into adolescence and adulthood in the face of hostile clinicians.
Now I’ll finally talk about the results of studies themselves. I know it might be weird to discuss that after essentially saying they are awful, but the fact is that they are awful in one direction. The general persistence rate as reported in these studies likely sits below the floor of a plausible real persistence rate. There are quite a few now, so if we just get a general idea for what they think the persistence rates are, we can say it’s somewhere above it, even higher if we assume an environment that isn’t actively trying to eliminate gender diversity. These two charts sum much of it up:
A few quick notes on the studies: Bakwin et al. (1968) is actually a composite, as done in Ristori & Steensma (2016). DeVries et al. (2011) wasn’t intended to measure desistance, but did follow 70 youth on puberty blockers and found all of them continued on to cross-sex hormones. Zucker (2018) is a composite, but was worth including because it explicitly counted only youth who actually met diagnostic criteria for GID. The studies have a large diversity of methods and definitions of persistence. This, plus the fact that they were conducted across a time-frame of 54 years, explains the diversity in results. You can see, however, that they tend to clump around 20-40%, which is confirmed by the stats: the average is 39% and the median is 23%.
Earlier, I suggested that persistence rates would go up in less hostile environments. Given that in the countries these studies were conducted have gotten more accepting over time, these studies seem to reinforce that notion.
I’ve argued that desistance research is conceptually and methodologically questionable, and that from this we can reasonably infer that persistence rates are higher in general and will likely increase as acceptance increases. However, I’d like to end by suggesting that desistance rates aren’t able to make the argument that groups like the Florida Department of Health want them to. As Florence Ashley (2021) has argued, in addition to some of the points I’ve brought up in this article, the fact is that “corrective” and “wait-and-see” models of transgender care are harmful. They impose upon the child the idea that they are aberrant and that things would be better off by persistently denying themselves. They develop the mindset that there is something wrong with expressing themselves, even if it does not harm others. This mindset can have negative repercussions for the rest of their lives. The affirmative care model does not have symmetrical risks; both endogenous and exogenous puberties cause permanent changes, but the affirmative care model teachers kids to know and love themselves, while other models encourage denial at best, self-hatred at worst.
I think kids shouldn’t hate themselves, so the affirmative model is probably the best.
REFERENCES
Ashley, F. (2021). The clinical irrelevance of “desistance” research for transgender and gender creative youth. Psychology of Sexual Orientation and Gender Diversity. Advance online publication. https://doi.org/10.1037/sgd0000504. (https://www.florenceashley.com/uploads/1/2/4/4/124439164/ashley_the_clinical_irrelevance_of_%E2%80%9Cdesistance%E2%80%9D_research_for_transgender_and_gender_creative_youth.pdf)
Castro, C. D., Solerdelcoll, M., Plana, M. T., Halperin, I., Mora, M., Ribera, L., Castelo-Branco, C., Gómez-Gil, E., Vidal, A. “High persistence in Spanish transgender minors: 18 years of experience of the Gender Identity Unit of Catalonia,” Revista de Psiquiatría y Salud Mental, 2022, ISSN 1888-9891, https://doi.org/10.1016/j.rpsm.2022.02.001. (https://www.sciencedirect.com/science/article/pii/S1888989122000283)
Cohen-Kettenis, P. T. Gender identity disorder in DSM? J Am Acad Child Adolesc Psychiatry. 2001 Apr;40(4):391. doi: 10.1097/00004583-200104000-00006. PMID: 11314563.
De Vries, A. L., Steensma, T. D., Doreleijers, T. A., & Cohen‐Kettenis, P. T. (2010). Puberty suppression in adolescents with gender identity disorder: A prospective follow‐up study. The Journal of Sexual Medicine, 8(8), 2276-2283.
Drummond, K. D., Bradley, S. J., Peterson-Badali, M., & Zucker, K. J. (2008). A follow-up study of girls with gender identity disorder. Developmental psychology, 44(1), 34.
Ristori, J., & Steensma, T. D. (2016). Gender dysphoria in childhood. International Review of Psychiatry, 28(1), 13-20.
Singh, D. (2012). A follow-up study of boys with gender identity disorder (Doctoral dissertation, University of Toronto).
Wallien, M. S., & Cohen-Kettenis, P. T. (2008). Psychosexual outcome of gender-dysphoric children. Journal of the American Academy of Child & Adolescent Psychiatry, 47(12), 1413-1423.
Zucker, K. J. (2018): The myth of persistence: Response to “A critical commentary on follow-up studies and ‘desistance’ theories about transgender and gender nonconforming children” by Temple Newhook et al. (2018), International Journal of Transgenderism, DOI: 10.1080/15532739.2018.1468293
Zucker, K. J., & Bradley, S. J. (1995). Gender identity disorder and psychosexual problems in children and adolescents. Guilford Press.