Gouvernement américain : preuves très incertaines sur la transition médicale des jeunes ; sociétés en désaccord

Traitement de la dysphorie de genre pédiatrique : examen des preuves et des meilleures pratiques

Revues systématiques & des données probantes Lignes directrices & politiques

Authors: U.S. Department of Health and Human Services (contributors: Evgenia Abbruzzese, Alex Byrne, Farr Curlin, Moti Gorin, Kristopher Kaliebe, Michael K. Laidlaw, Kathleen McDeavitt, Leor Sapir, Yuan Zhang)
Journal: U.S. Department of Health and Human Services (government report)
Year: 2025
Source: https://opa.hhs.gov/gender-dysphoria-report

Summary

The US Department of Health and Human Services ordered this review by executive order and published it in May 2025. It looks at the evidence for social transition, puberty blockers, hormones and surgery in under-18s. Its core gathers 17 earlier evidence reviews, which rated the evidence on mental health, quality of life and regret as very weak for every treatment. Other chapters cover policies abroad, WPATH, US practice, ethics and talk therapy. The main finding matches European reviews such as England's Cass Review. But the authors, named only in November 2025, include public critics of these treatments, and several were paid for legal or expert-witness work. US medical groups and academic critics say it misrepresents the evidence. Its evidence chapter fairly shows how uncertain the evidence is; its other chapters are argued opinion, not systematic research.

Key findings

Le rapport a rassemblé 17 revues de preuves antérieures : 2 sur la transition sociale, 9 sur les bloqueurs de puberté, 8 sur les hormones du sexe opposé, 3 sur la chirurgie et 5 sur la thérapie par la parole. Dix ont été jugées à faible risque de biais.
Pour chaque traitement, les preuves sur la détresse, la santé mentale, la qualité de vie et le regret ont été jugées très faibles. Seuls les effets corporels attendus, comme la mise en pause effective de la puberté, étaient étayés par des preuves solides.
Aucune des études des neuf revues sur les bloqueurs de puberté n'était un essai randomisé (où les personnes sont affectées au traitement par le hasard), et seules six avaient un groupe non traité de comparaison.
Le rapport énumère les préjudices possibles : infertilité, problèmes sexuels, formation osseuse plus faible, problèmes cardiaques et métaboliques, troubles de santé mentale, complications chirurgicales et regret. Une grande partie est déduite du fonctionnement du corps plutôt que mesurée chez des jeunes traités.
Le rapport dit que le véritable taux de regret est inconnu et que de meilleures données sont nécessaires. Il soutient aussi que le regret ou la satisfaction seuls ne peuvent pas montrer si un traitement est médicalement justifié.
La première version (1er mai 2025) ne nommait aucun auteur. La version du 19 novembre 2025 a nommé neuf contributeurs et publié neuf évaluations par des pairs avec réponses. Cinq contributeurs ont dit avoir été payés pour des travaux juridiques ou d'expertise judiciaire sur la médecine du genre.
Le gouvernement a invité l'American Academy of Pediatrics, l'American Psychiatric Association et l'Endocrine Society à l'examiner. Seuls les psychiatres ont accepté, et ils ont dit que les méthodes n'étaient pas assez ouvertes pour prendre les conclusions pour argent comptant.

Conclusion

The Review concludes that the overall quality of evidence on the effects of any intervention for pediatric gender dysphoria (social transition, puberty blockers, cross-sex hormones, surgery) on psychological outcomes, quality of life, regret or long-term health is very low, so reported benefits are likely to differ substantially from true effects. Evidence on harms in systematic reviews is also sparse, but the authors argue that known physiology points to risks including infertility, sexual dysfunction, impaired bone density accrual, cardiometabolic effects, surgical complications and regret. They judge the WPATH SOC-8 and Endocrine Society guidelines to lack developmental rigour and say U.S. medical associations overstated professional consensus. On ethics, they argue that respect for autonomy does not create a right to non-beneficial interventions. They describe psychotherapy as a noninvasive alternative that systematic reviews have not linked to adverse effects. The Review states that it is not a clinical practice guideline and makes no legislative or policy recommendations.

Methodological limitations

Ordered by a political decision: The report was ordered by an executive order that described these treatments in strongly negative terms and set a 90-day deadline. Critics argue this weakens its independence (Dowshen and others, Journal of Adolescent Health 2025).
Authors at first unnamed: The May 2025 version named no authors. The American Psychiatric Association's review and others said this stopped readers from judging the authors' expertise and conflicts of interest. Names came out only in November 2025.
Authors' paid work and past views: Five of the nine contributors said they had been paid for legal advice, court testimony or speaking fees about gender medicine. Several (including a co-founder of SEGM, a group critical of youth transition, and a fellow of the conservative Manhattan Institute) had already written or campaigned against medical transition for minors.
Peer review after release, run by the government: Review by outside experts happened after publication, and the government picked the reviewers and wrote the replies. The American Academy of Pediatrics and the Endocrine Society declined to take part. One invited reviewer (Smids) called it 'clearly not a neutral report' but said its reasoning was open. Another (Santen) said he used the Gemini AI tool to help frame part of his review.
A review of reviews: The evidence chapter summarises earlier reviews rather than re-checking the original studies. Many of those reviews cover the same small studies with no random assignment. Where reviews disagreed, the authors made their own ratings, and they treated phrases like 'no conclusions could be drawn' as meaning very weak evidence.
Some evidence from young adults: Reviews covering people up to age 26 were included. So some evidence comes from young adults rather than minors, although the authors report no difference in results.
Weak evidence is not proof of no effect: The authors admit that very weak evidence means uncertainty, not proof of harm or of no benefit. Critics argue the report is still framed as if it concludes against treatment.
Harms mostly reasoned, not measured: The authors admit evidence reviews say little about harms. Many listed risks come from how the body works rather than from measured results in treated young people.
Other chapters are opinion, not systematic: Chapters on WPATH, US clinical practice, whistleblowers, medical groups and ethics are argued essays, not systematic reviews. Critics say more than a fifth of references are news articles, blogs or social media (Dowshen and others 2025; Rider and others 2025). The authors say these were needed to document what clinics actually do.
Disputed wording and ideas: Critics object to the report's language (for example avoiding the term 'gender-affirming care') and its discussion of 'rapid-onset gender dysphoria' and exploratory talk therapy. The authors say their wording is neutral.
Corrections after release: Real errors were fixed after publication. Examples: 'irreversible' corrected to 'reversible'; a policy wrongly credited to the American Psychiatric Association instead of the American Psychological Association; UK and Finland recommendations swapped; and corrected depression scores for one study (Chen and others).
Critics also have a stake: Researchers who support gender-affirming care, Dowshen and others (2025; senior author Meredithe McNamara, Yale) and Rider and others (2025), say it misrepresents evidence and breaks scientific norms. The American Academy of Pediatrics said it misrepresents medical consensus. These critics also have professional and advocacy stakes in the care being reviewed.
Scope: The report covers minors only and says it is not a treatment guideline. Yet it has been cited in government policy, such as proposed rules on public health insurance (Medicaid and CHIP) coverage.

detrans.ai

Summary

Key findings

Conclusion