Reidentification With Birth-Registered Sex in a Western Australian Pediatric Gender Clinic Cohort

Key Points

Question

What proportion of young people seeking gender-affirming care subsequently reidentify with their birth-registered sex?

Findings

In this cohort study of 548 individuals with closed referrals to a pediatric gender clinic, 29 (5.3%) reportedly reidentified with their birth-registered sex before or during assessment. Two of these individuals reidentified during medical treatment, which corresponds to 1.0% of all patients who initiated medical treatment.

Meaning

Although these findings suggest that sex reidentification rates may be very low during medical treatment, longitudinal follow-up studies, including qualitative self-report, are needed to understand different pathways of the gender identity experience.

This cohort study explores the frequency and associated characteristics of reidentification with birth-registered sex among pediatric patients in a gender clinic setting.

Abstract

Importance

Some young people who identify as transgender and seek gender-affirming medical care subsequently reidentify with their sex registered at birth. Evidence regarding the frequency and characteristics of this experience is lacking.

Objective

To determine the frequency of reidentification and explore associated characteristics in a pediatric gender clinic setting.

Design, Setting, and Participants

This retrospective cohort study examined all referrals to the Child and Adolescent Health Service Gender Diversity Service at Perth Children’s Hospital between January 1, 2014, and December 31, 2020. The Gender Diversity Service is the sole statewide specialist service in Western Australia that provides children and adolescents up to age 18 years with multidisciplinary assessment, information, support, and gender-affirming medical care. All closed referrals for this study were audited between May 1, 2021, and August 8, 2022.

Exposure

Reidentification with birth-registered sex.

Main Outcomes and Measures

The number of referrals closed due to reported reidentification with birth-registered sex was determined, as well as descriptives and frequencies of patient demographics (age, birth-registered sex), informant source, International Statistical Classification of Diseases, Tenth Revision gender-related diagnoses, pubertal status, any gender-affirming medical treatment received, and whether subsequent re-referrals were received.

Results

Of 552 closed referrals during the study period, a reason for closure could be determined for 548 patients, including 211 birth-registered males (mean [SD] age, 13.88 [2.00] years) and 337 birth-registered females (mean [SD] age, 15.81 [2.22] years). Patients who reidentified with their birth-registered sex comprised 5.3% (29 of 548; 95% CI, 3.6%-7.5%) of all referral closures. Except for 2 patients, reidentification occurred before or during early stages of assessment (93.1%; 95% CI, 77.2%-99.2%). Two patients who reidentified with their birth-registered sex did so following initiation of puberty suppression or gender-affirming hormone treatment (1.0% of 196 patients who initiated any gender-affirming medical treatment; 95% CI, 0.1%-3.6%).

Conclusions and Relevance

These findings from a pediatric gender clinic audit indicate that a small proportion of patients, and a very small proportion of those who initiated medical gender-affirming treatment, reidentified with their birth-registered sex during the study period. Longitudinal follow-up studies, including qualitative self-report, are required to understand different pathways of gender identity experience.

Introduction

Some people who identify as transgender (hereafter referred to as trans) may later identify with their birth-registered sex and no longer consider themselves to be trans. Some of these individuals may take steps of social gender transition, commence gender-affirming medical treatment, undergo gender-affirming surgical procedures, or none of the above. After reidentifying with their birth-registered sex, some individuals have described their time identifying as trans and/or steps of social, medical, or surgical affirmation as a positive, meaningful, necessary, or neutral part of their personal history; others have described it as something they regret.¹ Terms such as desistence and detransition have been used historically in the literature regarding these experiences. These terms are inconsistently defined, and some people may find them pejorative or medicalized. We refer to the discontinuation of identifying as trans as reidentification with one’s birth-registered sex.

Defined here, reidentification includes people who have discontinued a trans identity, regardless of what (if any) social or legal changes or medical or surgical treatments have been accessed during transition. Reidentification does not include people who still self-identify as trans (binary or nonbinary), regardless of whether they stopped treatment or attending a clinic. This definition of reidentification contrasts with typologies of regret for medical or surgical transition steps.^2,3,4

There is particular interest in understanding the proportion of people who seek and/or access medical and surgical gender-affirming treatment who later reidentify with their birth-registered sex and their subjective experiences, including treatment satisfaction or regret. Ethical research in trans populations is founded on the understanding that any gender identity outcome (trans, binary or nonbinary, or cisgender [not transgender]) is equally acceptable, healthy, and valid.

Despite clinical, social, and media concern regarding reidentification,^5,6 there are few studies of its frequency and antecedents,^7,8 especially among children and adolescents.^8,9 Much of the existing literature on the longitudinal gender identities of children has major methodological, theoretical, and ethical limitations.^8,10,11 A person’s expressed reidentification with birth-registered sex may involve many factors, including sociocultural and familial pressures,⁷ clinical model of care (desistence-oriented vs gender-affirming), and definitions of reidentification.¹²

This study explores patients at the Child and Adolescent Health Service Gender Diversity Service (GDS) at Perth Children’s Hospital, Western Australia, who are discharged upon reidentification with their birth-registered sex. The GDS is a statewide pediatric service, providing multidisciplinary assessment, information, support, and medical gender-affirming treatment with puberty suppression and/or testosterone, estrogen, and antiandrogens, where appropriate, as per international and Australian best-practice guidelines.^13,14,15 The GDS is currently the sole provider of gender-affirming medical treatment for individuals younger than 18 years in Western Australia.

The GDS philosophy is to empower patients and support open family communication and participation throughout assessment and treatment. Trans identity is understood as valid and normal; there are also clear, repeated messages that it is acceptable to express shifts in gender identity, and it is acceptable to change one’s mind about treatment, consistent with recommended best practice.^11,12,13,14 Multidisciplinary assessment includes a comprehensive mental health assessment over several appointments by both a clinical psychologist and a consultant psychiatrist, medical assessment by an endocrinologist, fertility counseling, optional speech therapy, thorough education about requested treatments, family discussions to support decision-making, and multidisciplinary team consensus regarding treatment eligibility. For estrogen or testosterone treatment, a minimum 2-year history of expressing clear and stable gender identity and treatment wishes is required. For patients who commence gender-affirming treatment, there is regular endocrinologist follow-up and psychological support to monitor well-being and allow reflection on identity and wishes. Patients with more intensive mental health care needs attend elsewhere in liaison with GDS.

Methods

This cohort study was approved by the Child and Adolescent Mental Health Service (CAMHS) as a quality improvement project involving the secondary use of medical records; the CAMHS therefore waived the need for institutional ethics approval and participant informed consent. The CAMHS has granted publication approval. This study is reported in line with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational cohort studies.

Referral Closure Codes

All referrals to GDS for the 7-year period from January 1, 2014, to December 31, 2020, were examined. The audit took place from May 1, 2021, to August 8, 2022. We sought to determine a reason for the closure of each referral, consistent across a minimum of 3 evidence sources, including referral information, correspondence, the Gender Identity Longitudinal Experience Cohort database,¹⁶ the state Psychiatric Services Online Information System database, and the medical record. Where possible, a single primary reason for closure was determined. The audit protocol and codebook are available in eMethods 1 and 2 in Supplement 1.

All recorded reasons for referral closure are explored separately. Reasons included transfer to adult health services at age 18 years, inappropriate referral, prepubertal deferral (closure due to very young age, with invitation to return closer to puberty, as needed), relocation, aging out (unable to receive services due to age and wait-listing), declining referral, and reidentification with birth-registered sex. As a state-funded service, referrals are never closed due to prohibitive cost or insurance issues.

Defining Reidentification With Birth-Registered Sex

Closures were coded as reidentification if there were documented statements from the young person or caregivers that the patient no longer identified as trans or had affirmed their identity congruent with sex registered at birth (eg, “I do not feel that I am actually transgender”; “Family…let us know that [birth-registered female patient] has been comfortable living as female”). eMethods 2 in Supplement 1 provides full definitions and keywords used to identify possible reidentification. We did not classify as reidentification those whose language about identity evolved (eg, a shift from binary to nonbinary trans identity) unless they identified as their birth-registered sex. Finally, our reidentification coding was separate from any considerations of medical treatment. We did not include individuals who were referred for assessment but decided against having medical gender-affirming treatment or those who started treatment but then stopped, paused, or changed their treatment if they maintained a trans identity. Additionally, we recorded whether reidentification was communicated by the young person directly or whether it was recorded based on reports from caregivers or other adults, gender-related diagnoses, pubertal status at closure, gender-affirming medical treatment commenced by GDS, and any subsequent re-referral to GDS.

Statistical Analysis

Statistical analysis consisted primarily of frequencies and descriptive statistics (age, birth-registered sex) and was conducted using RStudio, version 4.2.1 (Posit) running Tidyverse, version 1.3.2. Race and ethnicity data were not collected as the personal information accessible through the quality improvement project is limited. Differences in reidentification proportion by birth-registered sex were tested with a χ² test (janitor package, version 2.1.0). Differences in age at reidentification closure by birth-registered sex were tested using a Mann-Whitney U test (stats package, version 4.2.1). Percentages and 95% CIs were calculated using the Hmisc package, version 5.1.0 binconf function (exact method). A 2-sided P < .05 was considered significant.

To assess frequency of gender-affirming treatment delivery, data were cross referenced with a separate dataset of all treatment initiated by any GDS patients within the audit dates. At the audit date, many patients receiving medical gender-affirming treatments were actively engaged in care, their referrals had not been closed, and they are therefore not described in referral closure data.

Results

Referral Closures

The Figure illustrates the flow of referrals audited. As assessment and treatment durations differed, closure status did not necessarily correspond to referral date. Of 995 referrals received (2014-2020), 443 remained open to GDS (active or wait-listed) on December 31, 2020, and were not audited or were closed to address administrative errors (eg, duplicate entries). The remaining 552 referrals were closed. A reason for closure could not be determined for 4 referrals due to insufficient documentation (0.7% of the 552 closed referrals; 2 birth-registered female, 2 birth-registered male), resulting in a final closed referral set of 548 individuals (337 birth-registered female, 211 birth-registered male).

Figure. Patient Flow Diagram.

Referral active indicates individuals undergoing assessment or treatment on the audit date; referral wait-listed, individuals on the waiting list for assessment on the audit date; administrative closure, referrals closed due to administrative errors (eg, duplicate entries); closure reason not determined, referrals for which there was insufficient evidence to determine closure reason; other closure reasons, non-reidentification referral closures (eg, transfer to adult health services at age 18 years, inappropriate referral, prepubertal deferral, relocation, aging out, declining referral).

Reidentification Frequency

Twenty-nine referrals (5.3%; 95% CI, 3.6%-7.5%) were closed because the patients reidentified with their birth-registered sex (20 birth-registered female, 9 birth-registered male). Reidentification closures made up a slightly greater proportion of referrals for those registered female at birth (5.9%; 95% CI, 3.7%-9.0%) compared with birth-registered males (4.3%; 95% CI, 2.0%-7.9%). However, we found no association between birth-registered sex and proportion of reidentification closures (χ²[1] = 0.43; P = .51). Birth-registered females were significantly older at reidentification closure (mean [SD] age, 15.81 [2.22] years) than birth-registered males (mean [SD] age, 13.88 [2.00] years; Mann-Whitney U = 137; P = .03) (Table 1).

Table 1. Age at Referral Closure for Reidentified Patients^a.

Measure	Age, y
	Birth-registered male (n = 9)	Birth-registered female (n = 20)	Total (N = 29)
Mean (SD)	13.88 (2.00)	15.81 (2.22)	15.21 (2.30)
Median (IQR)	14.29 (12.06-15.71)	16.24 (14.79-17.50)	15.45 (13.16-17.08)
Minimum	10.99	10.63	10.63
Maximum	16.40	18.73	18.73

^a After reidentification disclosure, patients may continue to be active with the Gender Diversity Service, receiving standard psychological support for several months before referral closure.

Reidentification Informant

Of 29 reidentification closures, 14 (48.3% of closures due to reidentification [95% CI, 29.4%-67.5%]; 2.6% of all closures [95% CI, 1.4%-4.2%]) documented the young person directly communicating their change in identity to GDS clinicians (eg, “[Patient said] she no longer felt any need to go through gender transition of any kind. She described feeling much more comfortable with being female.”). The remainder were based on documentation of GDS clinician interactions with caregivers (eg, “[Father reports patient told him] ‘I do not feel that I am actually transgender’ and [is] no longer wishing for gender-affirming treatment.”) or other service professionals (eg, community psychologists) reporting on their own clinical interactions (eg, “Case manager states [birth-registered female patient] is currently claiming her female gender identity.”).

Diagnosis

The International Statistical Classification of Diseases, Tenth Revision¹⁷ is used according to Australian national health coding standards. As GDS does not record a gender diagnosis at the first appointment, patients who have attended more than 1 appointment are more likely to have a recorded diagnosis. Just more than one-half of all patients who reidentified (51.7% [15 of 29]; 95% CI, 32.5%-70.6%) were recorded as meeting diagnostic criteria for an International Statistical Classification of Diseases, Tenth Revision code of gender identity disorder (Table 2).

Table 2. Gender-Related Diagnoses^a Received by Patients Who Later Reidentify During Gender Diversity Service Care.

Diagnosis	No. (%)
	Birth-registered male (n = 9)	Birth-registered female (n = 20)	Total (N = 29)
No gender diagnosis	5 (55.6)	9 (45.0)	14 (48.3)
Any gender diagnosis	4 (44.4)	11 (55.0)	15 (51.7)
F64.0: transsexualism	0	1 (5.0)	1 (3.4)
F64.2: gender identity disorder of childhood	2 (22.2)	0	2 (6.9)
F64.8: other gender identity disorders	1 (11.1)	3 (15.0)	4 (13.8)
F64.9: gender identity disorder, unspecified	1 (11.1)	7 (35.0)	8 (27.6)

^a By International Statistical Classification of Diseases, Tenth Revision.

Pubertal Status

For some young people, experiences of gender incongruence and trans identification in earlier childhood may not continue through puberty and into adulthood, while other young people may remain stable long term in their trans identity experienced from childhood.^18,19,20 At GDS, individuals referred as young children (prepubertal) are deferred by the service, with advice to refer again closer to puberty, if appropriate. Of the 29 patients who reidentified, 3 were prepubertal (1 birth-registered female, 2 birth-registered males), and the remainder were peripubertal or postpubertal.

Care Pathways and Services

Table 3 details the care stage at the time that reidentification was communicated. All but 2 patients (93.1%; 95% CI, 77.2%-99.2%) who reidentified with their birth-registered sex did so either while waiting for initial assessment or during the assessment process before the delivery of any gender-affirming medical treatment.

Table 3. Treatment Pathways of Patients Who Later Reidentify.

Treatment	No. (%)
	Birth-registered male (n = 9)	Birth-registered female (n = 20)	Total (N = 29)
Before initial assessmenta	0	3 (15.0)	3 (10.3)
Triage assessmentb	5 (55.6)	4 (20.0)	9 (31.0)
Mental health assessmentc	4 (44.4)	11 (55.0)	15 (51.7)
Puberty suppressiond	0	1 (5.0)	1 (3.4)
Gender-affirming hormonese	0	1 (5.0)	1 (3.4)

^a Patients who reidentified after referral was made but before an initial assessment could be provided.

^b Patients who were seen for an initial assessment with a mental health nurse or psychologist but who were never formally active with the Gender Diversity Service (GDS) or seen for further comprehensive mental health assessment for gender-affirming medical treatment.

^c Patients who were formally active with the GDS and received comprehensive mental health assessment but who never received any medical treatment through GDS.

^d Patients who initiated puberty suppression but not gender-affirming hormones prior to reidentifying.

^e Patients who initiated gender-affirming hormones (with or without puberty suppression) prior to reidentifying.

One patient received puberty suppression for 9 months before reidentification disclosure to GDS clinicians, then attended GDS mental health appointments for a further 6 months before closure. Another patient received puberty suppression for 4 months, then testosterone treatment for 1 month, before reidentification disclosure to GDS clinicians and cessation of both treatments. This patient attended GDS mental health appointments for a further 10 months before closure.

Before the end of 2020, 196 patients commenced treatment at GDS; 163 initiated puberty suppression, and 132 initiated gender-affirming hormones (including 99 who were treated with both). The 2 patients from this study who ceased gender-affirming medical treatment because of reidentification therefore represent 1.2% (95% CI, 0.1%-4.4%) of all GDS patients who had ever initiated puberty suppression and 0.8% (95% CI, 0.0%-4.1%) of patients who had ever, at this time, initiated gender-affirming estrogen or testosterone hormone treatment, or 1.0% (95% CI, 0.1%-3.6%) of those who had initiated either.

Retransition and Subsequent Return

The GDS is the only known gender-affirming medical service for children and adolescents in Western Australia. Therefore, any patients who reidentified and later wish to resume assessment for treatment at younger than 18 years must return to GDS. One patient, whose referral closed due to reported reidentification, was referred again to GDS, expressing trans identity 14 months later. The second referral for this patient remained open at the audit date.

Of the 28 other individuals who reidentified, 13 were younger than 18 years at audit date and, as such, remained eligible to seek re-referral to GDS. It is unknown whether any of the 15 patients older than 18 years subsequently sought gender-affirming care from adult services.

Discussion

The aim of this cohort study was to explore the frequency of reidentification with birth-registered sex for pediatric gender service patients. Examining the reason for closure of 548 referrals over a 7-year period, we found that 5.3% of referrals (29 patients) were closed due to a patient or caregiver stating that the young person no longer identified as trans. However, only 2 patients whose referral closed due to reidentification had initiated any hormonal treatment. These 2 represented 1.0% of all patients who had initiated gender-affirming medical treatment with GDS between 2014 and 2020. The effects of treatments over the short time received (<1 year each) may be largely reversible.

Overall, the reidentification rate estimated in this clinical pediatric sample is somewhat lower than estimates from general practitioner data for UK adults²¹ and a large community-sourced adult sample.⁷ Broadly, similar rates have been observed in the clinical notes of UK adults seeking gender-affirming treatment, using a very similar operationalization of reidentification to our own.²²

Similarly, low reidentification estimates have been found in North American prepubertal children living in social transition^18,23 and in Dutch²⁴ as well as UK²⁵ children receiving puberty suppression treatment. In the study most similar to our own, Butler et al²⁶ used medical records to examine the pathways of individuals upon discharge from pediatric gender clinics in the UK. They calculated higher estimates of young people no longer identifying as trans (8.3%), including 5.3% who stopped gender-affirming treatment.

Caution should be applied when comparing these findings with those of other studies, however, due to inconsistent definitions (eg, regret, desistence, and detransition¹²), transition interventions (change in pronouns for children^18,23 vs genital or gonadal surgeries in adults^{2,27,28,29,30,31,32,33,34,35}), and whether patients identifying as nonbinary were excluded.¹⁸ Our definition of reidentification recognizes that cessation of treatment does not necessarily correspond to change in identity and that people who voice a shift from binary to nonbinary may still identify as trans. However, this approach is not universal.⁸ Variable follow-up periods also make comparison between studies difficult.

In the current study, 1 patient who reported reidentifying with their birth-registered sex was subsequently referred again with continuing gender incongruence. Temporary reidentification or retransitions have been previously noted in adults^7,22,31,36 (including in the 1970s³⁷). Interviews with adults who stopped or reversed transition identified that for some, initial transition was interrupted due to family disapproval, and these individuals later reengaged with medical transition care.³⁶ Additionally, there is evidence that reidentification can form part of a developing self-understanding and expression of a trans identity.^38,39

Similarly, studies have found that external pressures (including family and employment discrimination)^{2,7,28,31,36,37,40} or health concerns^{7,22,24,36,40,41} contribute to regretting, stopping, or reversing gender-affirming medical treatment in adults. On the basis of our study findings, we can make no statements about the reasons for reidentification, as we did not capture data on such pressures. However, the continued development of personal gender identity clearly accounts for reidentification for some people.^{7,24,36,41,42}

Additionally, there were no recorded personal explanations for former trans identification given by patients or families (eg, internalized homophobia, gender stereotypes, social influence, trauma, abuse, mental health conditions^38,41). For some, but not all, there was recorded denial of these factors.

It has been suggested that reidentification frequency in adults may be underestimated, as patients may quietly stop attending a clinic and cease medical treatment without reporting reidentification to care professionals.^36,41,43 In contrast, the GDS model of care involves active case management, with regularly scheduled appointments and clinician contact to follow up nonattendance. Any contact indicating that reidentification had occurred and the young person no longer wished for GDS care is recorded.

Our data rely on the assumption that young people feel able to tell their caregivers or clinicians how they are feeling. As approximately one-half of the stated reidentification closures were communicated by caregivers, these data also rely on accurate caregiver report, as routinely sought in child psychiatry. As parents have shown some support to enable GDS attendance, many may not push their children to reidentify or report reidentification knowingly falsely. However, this avenue for bias remains in the current estimate of reidentification. Considering only directly communicated reidentification closures, the reidentification estimate is 2.5% of all referrals. Methodological improvements include supplementing clinically derived information through the universal administration of exit surveys, with young people and caregivers being surveyed separately.

Clinical Context

The GDS is currently the sole provider of pediatric gender-affirming medical treatment in the state of Western Australia. Hence, discharged patients who later request gender-affirming treatment when younger than 18 years may be referred again to GDS. As the GDS is a state-funded public service, external pressures of financial cost and insurance difficulties are not plausible explanations for expressions of reidentification.

The GDS care model is a comprehensive, multidisciplinary assessment process. Models of gender-affirming care delivery differ across regions. International collaboration is required to directly compare treatment models and investigate the role of setting, assessment process, and treatment eligibility criteria in reidentification frequency.

Limitations

This study had several limitations. The retrospective medical record review has methodological limitations, as it relied on accurate recording by clinicians. However, while cross referencing multiple sources for each patient, we found an unequivocal reason for closure for almost all (99.3%). Complete medical record audit of a clinic cohort has advantages over other forms of data collection, such as surveys, which may be liable to attrition and self-selection.

For this sample of young people referred to a pediatric gender service, trans identity at the outset was assumed for all participants. However, as some did not attend initial assessment, they may not have identified as trans or met diagnostic criteria for gender incongruence⁴⁴ or gender dysphoria⁴⁵ at referral. The sample cannot be assumed to represent trans youth more broadly. As clinic referral involves caregiver participation, this sample may have received more family support than young trans people in the general population.

Although the stated reason for closure was captured at the time of closure, eliminating recall bias, this methodology is not able to capture identity development that occurs after discharge or age older than 18 years. Some patients whose referrals were closed for other reasons may later have reidentified with their birth-registered sex (eg, patients may reidentify with their birth-registered sex as adults, or young prepubertal patients may reidentify with their birth-registered sex after prepubertal deferral). This could lead to an underestimation of the true rate of reidentification occurrence. Others whose recorded closure was for reidentification with birth-registered sex may have later identified again as trans and possibly sought gender-affirming treatment as adults. Similarly, we cannot account for patients discharged due to reidentification who subsequently identified as trans again as adolescents but did not return to GDS because they did not want medical treatment or because they accessed services interstate, overseas, or via the internet. The current results may therefore represent an overestimation of the true rate of reidentification.

Secondary use of existing data meant that we could not capture personal reflections regarding regret (not synonymous with reidentification^{22,36,38,42,43}) or quality-of-life outcomes. Adult follow-up of all former pediatric patients would be valuable to study the long-term course of identity and elicit factors associated with reidentification, disentangling personal shifts in identity from external pressures.

Conclusions

The study’s findings suggest that the specter of reidentification need not overshadow clinical care frameworks.¹¹ Care efforts are better directed toward considering how children and adolescents may be best supported in their well-being and quality of life, validating all future gender identity possibilities. This support includes necessary gender-affirming treatment for some adolescents.^{3,8,9,38,46,47} Young people and families should be supported to embrace attitudes of acceptance, whether gender identity remains stable or continues to evolve. Clinicians can reassure patients and families that they will be supported and receive quality health care if their identity or treatment wishes change. It is important that clinicians sensitively, and without stigma, meet the ongoing medical needs of trans and formerly trans-identifying individuals.^1,36 Clinicians should also refrain from making assumptions about the meaning of a person’s reidentification or treatment cessation and instead ask open questions that do not shame, blame, or stigmatize in order to understand each individual’s experience and care needs, creating a sense of safety in accessing medical care.³⁶ Finally, clinical practice can be enhanced by regular check-ins with all patients regarding satisfaction, current needs, and wishes regarding affirmation.

Supplement 1.

eMethods 1. Protocol

eMethods 2. Code Book

Supplement 2.

Data Sharing Statement