撤回：ノルウェー研究、テストステロン投与後の性別移行中止18例および医学的治療なしでの22%の診療終了を報告

ノルウェー国立性別不一致センターに紹介された子供と青少年の治療軌跡

Authors: Cecilie Bjertness Nyquist, Leila Torgersen, Linda W. David, Trond Haaken Diseth, Kjersti Gulbrandsen, Anne Waehre
Journal: Acta Paediatrica
Year: 2025
Source: pubmed.ncbi.nlm.nih.gov/39648282/

Summary

Clinicians at Norway's national gender service used its own records to trace every child and teenager referred over 20 years. Of 1,258 referred, 62% went on to cross-sex hormones. Nearly all of the 11% given puberty blockers (97%) moved on to hormones. 22% of those who attended left without any medical treatment. Eighteen females stopped testosterone and detransitioned, and 11 people in the group died, though causes of death were not recorded. The authors admit the detransition count is probably too low, because people who stop often drop out of the clinic's sight. In July 2026 the journal withdrew (retracted) the paper after a research-integrity panel found it lacked ethics committee approval and patient consent. Five authors were found to have committed misconduct, which they are appealing. Its figures should not be quoted as settled evidence.

Key findings

この研究にはノルウェーの医学研究倫理委員会の承認と、患者からの適切な同意がなかったため、学術誌は2026年7月に論文を撤回しました。撤回はデータの誤りを指摘したものではありません。
2000年から2020年に紹介された子どもと10代の若者1,258人のうち、861人（68.4%）は女性として生まれ、紹介時の平均年齢は14.4歳でした。
135人（10.7%）が思春期ブロッカーを開始し、男性として生まれた人のほうが多く、そのうち97%が異性ホルモンに進みました。
紹介された全員のうち783人（62.2%）が異性ホルモンを開始しました。
少なくとも1回受診した1,198人のうち264人（22.0%）は、性別に関連する医学的治療を受けないままクリニックを離れました。
ホルモンをやめたと記録された18人（開始した783人の2.3%）は全員が女性として生まれ、2015年から2021年の間にテストステロンを始めていました。11人はもはや自分をトランスだとは考えておらず、7人は引き続きそう考えていました（5人は今の変化に満足、2人は妊孕性への不安）。
1,258人のうち11人（0.9%）が2023年末までに亡くなっていますが、死因は入手できませんでした。

Abstract

Aim: We aimed to describe treatment trajectories, detransition and mortality rate among children and adolescents referred to the Norwegian National Center for Gender Incongruence (NCGI). Methods: The cohort included all 1258 persons under 18 years at referral to the NCGI from 2000 to 2020. Trajectories were registered until end of 2023. Results: In total, 861/1258 (68.4%) were assigned female gender at birth (AFAB). Mean age at referral was 14.4 years. Puberty suppression with gonadotropin-releasing hormone agonists (GnRHa) was initiated among 135/1258 (10.7%), significantly more persons assigned male gender at birth (AMAB) than AFAB (p < 0.001). Gender-affirming hormonal treatment (GAHT) was initiated in 783/1258 (62.2%). The continuation rate from GnRHa to GAHT was 97%. Discharge rate from NCGI without gender-affirming medical treatment among those who attended at least one appointment, was 264/1198 (22.0%). Eighteen AFAB detransitioned after initiated GAHT, eleven due to a cessation of transgender identity. Mortality rate in the cohort until end of 2023 was 11/1258 (0.9%). Conclusion: Different trajectories including medical pathways and assessments without gender-affirming treatment were observed. GAHT was initiated in 783/1258 (62.2%), including eighteen AFAB detransitioning after testosterone treatment. There was a high continuation rate from GnRHa to GAHT. Various trajectories highlights the need for long-term follow-up in care.

Conclusion

This study provides valuable insights to gender-affirming care for children and adolescents, owing to its large sample size, including all children and adolescents referred to a national gender clinic over two decades. The study reports varying trajectories, encompassing both medical pathways and assessments ended without gender-affirming medical treatment. Our findings highlight the necessity for long-term follow-up and support of people referred to gender clinics. There is a need for ongoing research on gender-affirming treatment in young people to evaluate the short- and long-term benefits and risks of early treatment interventions, as well as patient-reported outcome measures.

Methodological limitations

Retracted for ethics breaches: The journal withdrew the article on 30 July 2026 after a research-integrity panel from Oslo University Hospital, Akershus University Hospital and the University of Oslo found it lacked approval from a regional ethics committee and proper consent. Five Oslo-based authors were found to have committed misconduct and are appealing (Retraction Watch, September 2026). The problem is ethics, not known data errors.
Records filled in afterwards: The data came from a clinic register built by going back through patient files, so some information may be missing, as the authors say.
Treatment elsewhere missed: Treatment from other providers was recorded only if clinic staff were told and wrote it down. People who treated themselves or got treatment privately or abroad were missed.
Detransition likely undercounted: People who stop treatment may stop coming to the clinic, and what happened after discharge was often unknown. So 18 detransitions (2.3%) is a minimum. The authors point to a Finnish study where 7.9% stopped, for comparison.
Uneven, often short follow-up: Referrals ran to 2020 and tracking to 2023, so follow-up ranged from about 3 to over 20 years. Recent referrals had little time to start or stop treatment, and the authors say recent percentages cannot be interpreted.
Rules changed over time: The criteria for treatment changed over two decades, so the combined rates mix very different eras of practice.
No information on when it began: The study could not separate gender distress that began in childhood from distress that began in the teens, which may matter for later desistance (no longer having gender distress).
Too few cases to draw conclusions: The authors say the detransition numbers are too small to draw any conclusions. All 18 were born female, and detransition was only counted when a stop in hormones was recorded.
Causes of death missing: Deaths came from the national register without causes, so the 0.9% death rate, including any suicides, cannot be interpreted.
Already filtered by referral: All young people were referred by local child mental health clinics after a first assessment. This may favour more severe cases and limits how far results apply to young people never referred.
No patients' own views: The study recorded treatment steps only, with no measures of gender distress, mental health, quality of life or satisfaction.
Clinic judging itself: The authors were clinicians and researchers at the national clinic whose practice is described. One author (Waehre) declared she had been an outside reviewer for Sweden's 2020–2022 national evidence review on gender dysphoria.

detrans.ai

Summary

Key findings

Abstract

Conclusion