Youth Hormone Study Reports Low Regret, but Discontinuation Outpaces It

Invitations went out between October 2023 and October 2025 to former patients who had joined a longitudinal research registry, with explicit encouragement for people who had quit hormones or left the clinic. Of those who completed the survey, 376 had started estrogen or testosterone as youth and 25 never had. About 70% of the hormone group were assigned female at birth and began testosterone, while about 30% were assigned male at birth and began estrogen. Nearly 92% started at 17 or younger, at an average age of 16; respondents were 14 to 28 and averaged almost 21 when surveyed. Follow-up since starting hormones averaged 4.9 years, ranging from slightly more than a month to almost 11 years. [1]

Most participants had gone through a structured evaluation with a mental-health professional together with parents or guardians, though a minority started hormones before or after being seen by the program. Minors needed the consent of every parent or guardian with legal medical decision-making authority as well as their own assent. The five-item Decision Regret Scale asks whether starting was the right call, whether the respondent regrets it and whether it caused harm; scores rise with regret. The researchers checked clinic records and surveys for current hormone use, binary versus nonbinary identity, the age a patient first identified as transgender or gender-diverse, and whether clinicians had documented prepubertal gender nonconformity or physical distress about gender. [1]

Three more participants had flagged regret or harm on an earlier annual survey but did not repeat it on the latest, bringing the lifetime count to 13. Seven of the 13 now identified only with their birth-assigned sex, and 12 had stopped hormones; the 13th marked both affirmative and negative items and may have misread the scale, the authors say. Nine wrote open-ended explanations. Seven attributed their regret or sense of harm to feeling unready to decide, citing age or maturity, mental-health problems or thin support from parents or therapists. Six named unwanted physical effects — body hair, a deeper voice or breast growth — and four cited side effects such as mood changes, vaginal atrophy or worry about fertility. [1]

By comparison, the 4.3 average regret score falls well under the 16.5 average recorded across studies of other medical choices that use the same scale, the authors note. Ten respondents endorsed regret and seven endorsed substantial harm on the latest survey. Among the 65 people no longer on hormones, nine (about 14%) said they regretted starting and six (about 9%) said it had harmed them a lot, yet 54 (83%) still said starting had been right. Those who stopped tended to be about a year older, to have started roughly a year earlier, and to be likelier to have used testosterone and to identify as nonbinary. Almost 59% of non-cisgender people who quit identified as nonbinary, against about 14% of those still on therapy. [1]

Regret scores averaged about 16 points higher among people who had stopped than among current users, and about six points higher among nonbinary than binary respondents. Testosterone and estrogen users did not differ significantly on regret. Nor did regret or continuing hormones show a statistically significant link to starting age, the age of first transgender or gender-diverse identification, or prepubertal signs of gender diversity. The authors say the data do not support the idea that teenagers without childhood-onset distress face greater regret risk over the period studied, while conceding that so few regret cases limits their ability to detect such links. [1]

Rogowska told PsyPost the project is the largest yet to concentrate on decisional regret in this group and that it largely confirms earlier clinic studies finding regret rates of 0% to 4%. She singled out one figure for more attention: 17.3% no longer on hormones. That is far above the roughly 2% discontinuation in a long-running Dutch clinic cohort and nearer U.S. insurance-claims estimates of about a quarter stopping within four years, she said, and among the people who did stop, roughly one in seven regretted it. She added that the overall average conceals this subgroup. [1]

The response pattern is a central caveat. Almost 300 of the 698 invited did not answer, so their hormone status and views are unknown. Initial clinic records showed respondents and nonrespondents were broadly alike in age, birth-assigned sex and ethnicity, but among enrollees, people assigned male at birth, and those whose records documented physical gender distress before puberty, were less likely to answer the follow-up survey. The authors say the political climate around transgender care could have discouraged some former patients or drawn people with strong views on either side, and that the roughly 57% response rate, counting the 25 who never started hormones, meets or exceeds typical long-term registry follow-up and beats most online patient surveys. A 2020 review they cite warns that satisfied patients may be overrepresented in such surveys. [1]

Because the research team included clinicians from the treating clinic, trust may have cut both ways, the authors acknowledge. Rogowska said people who quit or regret treatment may be less inclined to answer the clinic that treated them, which would bias the regret estimate downward. She also warned that the absence of associations with starting age or childhood-onset gender dysphoria should not be taken as proof those factors are irrelevant, since 13 regret cases leave little statistical power. Some regret in the wider literature surfaces years after treatment, she said, so a 4.9-year mean follow-up, with some respondents tracked only months, may be too short. She told PsyPost that rare decisional regret does not establish long-term benefit or safety, adding that the study gauges how people judge a past decision, not their physical or mental health. [1]

The study also has design limits beyond response. All participants came from one multidisciplinary clinic that required parental consent, and more than 93% were White, so findings may not carry to more diverse populations or to care settings with weaker family and mental-health support. The survey measured regret about the initial decision rather than satisfaction with particular physical effects, and it did not ask the wider group of people who stopped why they did so. On policy, Rogowska said the authors' conclusions reach further than one observational study at one site can support, in either direction. [1]

Whether the low regret rate holds across clinics is unresolved. Rogowska called for multi-site prospective cohorts that actively track nonrespondents and people who stop treatment, ideally led by researchers with no ties to the treating clinics. She also urged systematic accounting of why people discontinue, separating outside barriers from identity shifts or side effects, follow-up of 10 years or more, and closer study of the nonbinary subgroup. In this sample, she noted, nonbinary identity was tied to almost nine times higher odds of not currently receiving therapy. Whether longer tracking would raise measured regret, or whether independent multi-site research would reproduce these numbers, are open questions a future study could test. [1]

Sources

PsyPost: Relationships

Prepared with AI assistance from the cited sources.