Models of Gender-Affirming Care in Asia
- The study — what it maps, where it stands, and how the method guards against its author’s position (you are here)
- Call for language extractors and verifiers — paid, part-time roles for seven languages
This record grows as the study moves through registration, extraction and release.
COMPARATIVE POLICY STUDY · REGISTERED
Eden Openly is preparing a comparative study of what the health systems of South Korea, Japan, Taiwan, Hong Kong, Thailand, Malaysia and Singapore formally require of a person seeking gender-affirming hormone therapy or surgery, and what each system provides and costs. No data have been extracted. The protocol was registered publicly on the Open Science Framework on 30 September 2026, before any extraction began.
What we are doing
The study reads the rules themselves: statutes, court decisions, ministry and insurer rules, professional guidance and clinic protocols, in their original languages. For each system it records what is required, who imposes each requirement (the law, the profession, the insurer or the clinic), and what is not documented at all.
From those records it builds each system’s pathway to a first prescription and to each surgical procedure, and computes the documented minimum burden along it: time, clinical encounters, gatekeepers and direct cost.
We have found nothing comparable for Asia. The closest published work, a 2026 scoping review of clinical care in East Asia (Hong et al., LGBT Health), mapped English-language clinical studies; its authors note that it excluded grey literature and scholarship in Chinese, Japanese and Korean, and may have missed policy detail as a result. This study starts from exactly that material.
What it does not do
- It does not judge which system is better. There is no ranking, score or league table.
- It does not assess clinical outcomes or the quality of care.
- It does not survey patients or clinicians. It works from published documents only.
- It does not treat one clinic’s practice as a country’s rule. Every value carries the scope it applies to.
Scope
This is a plain-language summary of the registered scope. It is not the protocol, and where the two differ, the registered protocol governs.
The questions
- Primary. What does each system formally require of a person seeking gender-affirming hormone therapy or surgery, and what does it formally provide?
- On a fixed set of key requirements, where do comparable pathways in the seven systems differ, and what pattern does the difference take?
- Which requirements are imposed only by individual clinics, with no matching requirement from the law, the profession or the insurer?
- Where do the published rules and the reported experience of access diverge?
- What is the documented minimum burden of reaching a first prescription, and each procedure, in time, clinical appointments, gatekeepers and direct cost?
The care covered
| Group | Interventions | Depth |
|---|---|---|
| Hormones | Feminising and masculinising hormone therapy; puberty suppression for adolescents | Full pathway, requirements, cost |
| Chest | Mastectomy and chest reconstruction; breast augmentation | Full pathway, requirements, cost |
| Genital and gonadal | Vaginoplasty, orchiectomy, phalloplasty, metoidioplasty, hysterectomy, oophorectomy | Full pathway, requirements, cost |
| Face and voice | Facial feminisation surgery, tracheal shave, voice surgery | Requirements, coverage and cost |
| Hair removal | Laser, electrolysis | Coverage and cost |
Adults and adolescents are both covered. Care for differences of sex development is not: it is a distinct clinical and legal pathway.
What is recorded for each system
| Domain | Examples | |
|---|---|---|
| A | Legal and regulatory frame | Legal gender recognition and its conditions; the legal basis of care; restrictions for minors; religious or personal-law jurisdiction; conscription |
| B | Entry to care | Whether a psychiatric diagnosis is required; how many assessors; waiting or “real-life” periods; surgical prerequisites; provision for non-binary people |
| C | Adolescents | Minimum ages; who must authorise each intervention; whether an adolescent service exists |
| D | Delivery | Public or private provision; how many centres, and where; who prescribes; telemedicine |
| E | Cost and coverage | Insurance coverage and its conditions; published prices; fertility preservation; hair removal |
| F | Documented clinical practice | Which medicines are registered; monitoring; dispensing rules |
| G | Friction | Reported waiting times; how concentrated services are; documented barriers; self-sourced hormones |
| H | Cross-border continuity (extension) | Importing hormones; accepting foreign assessments and prescriptions; aftercare for surgery abroad |
| J | After transition (extension) | Screening invitations; visibility of diagnosis codes; claims for sex-coded treatment |
How each answer is recorded
- Who imposes it. Each requirement is recorded against the level that imposes it: the law or regulator, the profession, the insurer, or the clinic.
- Whom it applies to. Every answer carries its scope, such as adults only, the public insurer only, or one named clinic, so that one clinic’s rule is never read as a country’s.
- How well it is documented. Each answer is documented, ambiguous, in conflict between sources, or not found. “Not found” means the registered searches found nothing, and is kept separate from a source that says a requirement is not imposed.
- The source. Every answer links to the passage it came from.
How systems are compared
Only like with like: for example, an adult on the main public insurance route in one system against the same person in another, and separately a self-paying adult. Time and cost are reported as documented lower and upper bounds, not single figures, and cost also as a share of median earnings. Differences are described, never scored or ranked, and there are no cross-country averages.
Where things stand
The study runs in four stages, in this order:
- Protocol registrationThe full protocol and codebook, frozen publicly on the Open Science Framework before any data are extracted. Registered 30 September 2026.
- South Korea pilotA registered feasibility phase that tests the method on real sources. The changes it may lead to are fixed in advance. We are recruiting the Korean extractor and verifier.We are here
- Extraction and verificationEvery non-English source read in the original by a paid native-language extractor, and checked by a second, independent reader.
- Correction round and releaseClinicians and community organisations check the draft for factual errors; each system’s data are then released openly, with a DOI.
The policy snapshot describes what was in force on 31 December 2026. Sources published by 31 March 2027 may be used.
The protocol
The protocol was registered on the Open Science Framework on 30 September 2026. The registration holds the protocol, the codebook, the search strings, the ethics statement and the review log of every independent check, with the disposition of each finding.
- Registration: doi.org/10.17605/OSF.IO/UVHDA
- OSF project (where amendments, the pilot report and each jurisdiction’s final search strings are posted): doi.org/10.17605/OSF.IO/UB82M
- Archived copy of the registration package on Zenodo: doi.org/10.5281/zenodo.23057857
Any change after registration is filed as a dated update to the registration, with its reason.
Clinic sampling seed. Where a system has more than 20 private clinics, the study reads a random sample of 20. The seed for that draw is drand round 6512726, published on 1 October 2026 at 00:00 UTC, after the registration went public, so it could not be known or chosen in advance. The value and the sampling script are posted in the OSF project, and the registration carries a dated update recording them. Anyone can rerun the draw.
If this page and the registered protocol ever differ, the protocol governs.
Our position, and why we don’t claim neutrality
We don’t claim to be neutral. Eden is a transgender woman receiving gender-affirming hormone therapy, and a dermatologist. She holds a declared prior in favour of access to gender-affirming care, and has a personal stake in the subject.
Her interests are listed in full, with the date each began, in our register of declared interests, and the protocol reproduces them as they stand at registration. Three bear on this study:
- She founded and funds a transgender health research institute that is being set up in South Korea, one of the seven systems studied.
- She founded and funds a non-profit transgender healthcare and education organisation.
- She co-owns a skincare business, and is planning a general aesthetic hospital, in a jurisdiction the study covers, that is not open and would not target transgender patients. The study records prices, coverage and waits for procedures, such as facial surgery and laser hair removal, that a hospital of that kind could also offer, so this interest bears directly on it. The study names no provider as a recommendation and ranks none, and if the hospital opens, Eden takes no part in coding it.
The study is self-funded. Eden accepts no third-party funding for it and pays every cost, including extractors and verifiers, herself.
A descriptive study can still be bent by what it chooses to look at. The protocol fixes the questions, the fields and the rules for comparison before any source is read.
How the method guards against it
- Rules before reading. The questions, every field, the search plan and the comparison rules are registered before extraction.
- Native-language extraction, independently checked. Every non-English source is read and recorded in the original by a paid native-language extractor, not by the author through machine translation. A second, independent reader first extracts a random 20% blind, then checks every key data point and pathway step. The author extracts English-language sources and keeps the coding consistent across the seven systems, but cannot change what a source is recorded as saying without the extractor’s agreement; every such change is logged. In Hong Kong and Singapore most sources are in English, so there the author extracts most of the data herself; every English data point is checked in full by an independent legal-English verifier, and the study reports this as a limitation.
- Missed sources tested, not assumed. Verifiers search independently for sources the author may have missed. If too many turn up, the whole jurisdiction is searched again.
- “Not found” is a result. Where nothing is documented, the study says so, and publishes the searches behind that statement.
- Every value linked to its source. The released data link each value to the document and passage it came from.
- AI assistance, disclosed. AI models helped draft the protocol and took part in reviewing it; every finding and how it was handled is logged. In the study itself, AI may help find sources and develop search terms, but its output is only ever a lead: every source is opened and read by a person, and AI never decides what a source says, how it is classified, or what the findings are. Extractors and verifiers are human, and machine translation is never used to extract. See How We Use AI.
- A single author, stated plainly. These safeguards reduce the risk of a one-person study. They do not remove it.
Taking part
We are recruiting paid, part-time language extractors and verifiers for Korean, Japanese, Chinese, Thai and Malay, a verifier for legal English and, on call, Tamil readers. The Korean roles come first. Read the call for extractors and verifiers.
Corrections and questions
If you think we’ve got something wrong, please get in touch through our editorial contact page. Corrections are logged on our corrections page.
Eden Openly is an independent clinical health-education publication. It receives no funding from, and has no commercial relationship with, any provider of gender-related healthcare or any pharmaceutical company. Where this page and the registered protocol differ, the protocol governs. Our full position on funding and interests is set out in our statement of independence, and every interest is listed, with its date, in our register of declared interests.