Transgender Youth Referred to Clinics for Gender-Affirming Medical Care in Canada

Trans Youth CAN! is a 2-year prospective cohort of youth referred to Canadian medical clinics for hormonal suppression or gender-affirming hormone therapy. Participants enrolled between September 2017 and June 2019 at 10 clinics in Halifax, Montreal, Ottawa, Toronto, Hamilton, London, Winnipeg, Calgary, Edmonton, and Vancouver. All clinics have one or more pediatric specialists and links to mental health providers. The study protocol was approved by each institution’s Research Ethics Board (REB).

Inclusion criteria were (1) gender dysphoria,⁴³  (2) age <16 years, (3) pubertal or postpubertal, (4) not previously using GnRHa, estrogen or testosterone therapy other than contraceptives, and (5) enrolled after referral for hormonal suppression or hormones. Recruitment processes differed based on clinic and REB requirements, and did not always allow eligibility prescreening. Clinic staff invited potential participants to contact the research assistant before their appointment; 174 youth participated (163 in English, 11 in French). With youth agreement, 160 matched parents enrolled to provide family and parental data. Informed consent was obtained, including parental and/or guardian consent, when locally required.

Data were from the baseline interviewer-administered youth survey,⁴⁴  parent and/or caregiver survey,⁴⁵  and extracts from medical records.⁴⁶  Detail on each measure is provided as Supplemental Table 1.

Demographics were coded from data self-reported by youth in survey interviews.

Gender identity was collected as open text and a categorical item (male/boy, female/girl, or nonbinary). Youth self-reported ages they experienced discord between identified and assigned gender and began to live in their gender. Legal changes to name and gender marker were reported. Parental report captured types of professional and support persons with whom the youth and family had met to discuss gender. Transmasculine (assigned female at birth) and transfeminine (assigned male) were coded as gender groups for statistical comparisons.

Z scores for weight-, height- and BMI-for-age were calculated by using data from the clinical record and the WHO Growth Charts for Canada, 2014 revision.⁴⁷  Because hormone therapy can affect adult height, z scores were calculated by using both male and female growth curves, reflecting growth relative to birth-assigned sex and gender identity.

Self-rated physical and mental health questionnaire items came from the Canadian Community Health Survey.⁴⁸  The 5-item Modified Depression Scale (α = .7924 in our data) captures current depressive symptoms in adolescents, with a dichotomous score encoding probable depression.⁴⁹  The 5-item Overall Anxiety Severity and Impairment Score (OASIS) (α = .8692) captured past-week anxiety, with a cut-point of ≥8 for probable anxiety disorder.^50,51  Past-year self-harm was assessed with 5 items. Lifetime and past-year suicidal ideation and attempts were self-reported by using Canadian Community Health Survey items; substance use was self-reported. Previous or current mental or physical health diagnoses from the clinic or referring provider were extracted from the medical record.

Youth self-reported school and living situations, whether they had left home because of family rejection, support sources, and parental support for their gender. Types of group spaces avoided (eg, school, transit) were summed from a checklist. Reduced access to material resources was assessed by using 3 measures: food insecurity, household poverty (Statistics Canada low-income measure [LIM]⁵² ), and 5-item youth deprivation index. Trans-specific external stressors on families were assessed with the Stressors on Families of Trans Youth Checklist.⁵³ 

Prescriptions provided, or reasons for no prescription, were extracted from the medical record.

Data were analyzed by using SAS version 9.4 (SAS Institute, Inc, Cary, NC).⁵⁴  A SAS macro for World Health Organization Growth Charts for Canada (2014) was used to calculate Z scores.⁵⁵  Sample weights were derived (see eMethods in Supplemental Materials) to account for variable-length recruitment periods across clinics (mean: 12.9 months; range: 6–22). Individual weights were identical within each clinic (range: 0.5809–2.1301) and summed to the total sample size of 174. Statistical analyses were weighted for inference to the larger patient population.

For the full sample and transmasculine and transfeminine gender subgroups, frequencies were calculated for categorical variables and means and SDs for continuous variables. To compare gender subgroups at α = .05, Rao-Scott χ² tests were used for proportions and t tests with Satterthwaite estimators of variance for means. Fisher’s exact test was used with unweighted data when expected cell count was <5. Results for sex- or gender-specific measures are presented only for relevant groups.

Of 174 patient participants, 137 were assigned female at birth and 37 were assigned male. Results focus on weighted estimates for the patient population. An estimated 81.2% were transmasculine, and 18.8% transfeminine. A primarily nonbinary identity was reported by 7.8% of transmasculine and 10.5% of transfeminine patients (P = .68). Figure 1 displays age-related gender milestones; patients attended this referral visit an average of 3.9 years after they recognized their gender (SD = 2.6). Table 1 presents demographic and gender characteristics for the full population and gender groups. Of patients under age 16, 69.0% were age 14 or 15, 18.8% were Indigenous, 6.6% were racialized visible minorities, and 25.7% had 1 or more immigrant parents. Approximately half (55.4%) lived in urban environments. Although 84.0% lived in their gender full-time, few had legally changed their name (3.5%) or gender (5.4%). Youth had seen a wide range of providers to discuss gender, most commonly their family physician (68.5%). There were few statistically significant differences between gender groups, although transfeminine youth less frequently lived in their identified gender full-time (58.1% vs 90.1% of transmasculine youth; P < .001).

Health findings are presented in Table 2. Growth parameters reflected sex assigned at birth, with puberty effects apparent in comparisons with identified-gender standards; when analyzed on the female standard, mean height-for-age z score for transfeminine youth was 1.40 (SD = 0.85). Of youth, 74.9% reported good to excellent physical health, in contrast with 45.5% for mental health. Anxiety was the most common previous diagnosis (40.2%), and 65.2% had OASIS scores consistent with a probable anxiety disorder. Past-year self-harm (67.6%) and suicidal ideation (34.5%) were common, whereas physical health diagnoses were not. Most youth did not use substances. Although transmasculine youth were more likely to report symptoms consistent with depression (21.2% vs 10.8%; P = .03) and anxiety (66.1% vs 33.3%; P < .001), lifetime and past-year suicidal ideation or attempts were common regardless of gender. Transmasculine youth were more likely to report past-year “other” drug use, but gender groups did not differ statistically significantly for other substances. Past diagnoses differed significantly only for attention-deficit/hyperactivity disorder (ADHD) and autism.

Social context results are presented in Table 3. All youth were in school, with 3.2% homeschooled; 12.9% attended religious school. Household income was below Statistics Canada’s LIM for 27.2%, as reflected in proportions without access to essential resources or food security. Almost two-thirds (62.0%) reported strong support for their gender identity and expression from all parents, and 96.0% lived with parents. Of families, 64.3% experienced at least one external stressor related to the youth’s gender, most commonly school involvement (41.0%), unwanted parenting advice from friends and family (31.2%), and friends’ parents barring them spending time with the youth (22.7%). Most youth reported avoiding group spaces. Common sources of support included parents (91.9%), real-life (off-line) friends (85.8%), and nontrans friends (80.5%). Social context differed for gender groups. Transfeminine youth were more likely to attend a religious school, less likely to experience food insecurity or avoid group spaces, and had families less likely to experience trans-specific external stressors.

Table 4 presents medications prescribed. An estimated 62.4% received a prescription, with no difference by gender group (P = .45). Depot leuprolide acetate was the most common, given alone to 53.0% of transfeminine and 45.2% of transmasculine patients, or prescribed with testosterone (1.7% of transmasculine) or estrogen (6.6% of transfeminine). Of 46 participants not prescribed hormones at this visit, physician’s reason was included in 36 records. The most common reason was the clinic protocol of not prescribing at first visit. Additional reasons included being younger than the clinic’s age limit for estrogen or testosterone, unconfirmed gender dysphoria, waiting for information (bloodwork, pregnancy test, psychiatric assessment), waiting for parental consent, no desire for medical transition, and current medication working well (eg, continuous oral contraceptives suppressing menstruation). Youth’s reasons included fear of needles, not interested or uncertain on timing, waiting for additional medical care (eg, psychiatric, fertility preservation), and wanting time to consider options or discuss with family doctor. Parent’s reasons included disagreement with therapy, existing medications working, belief that gender dysphoria is a phase, and needing to confirm another parent’s consent. No physician, youth, or parent indicated cost as a reason for not starting medication.

An estimated 18.8% of clinic youth under 16 were transfeminine, similar to 18% observed for 14- to 18-year-olds in a national community survey⁵⁶  and 23% in a single-clinic records review,²⁴  but different from the higher transfeminine proportions among older youth⁵⁶  and approximately equal proportions among older youth and adults.⁵⁷  This may be in part explained by earlier puberty for transmasculine youth. Prepubertal nonnormative gender expression in assigned-female children is often supported,^58,59  and pressure to rapidly conform to feminine norms may contribute to urgent need for clinical services at puberty. In contrast, transfeminine nonconforming youth are often disapproved of, victimized, or unsupported^60,61 ; transmisogyny (which penalizes perceived femininity in those assigned male) may underlie such findings,⁶²  delaying coming out and acceptance.

Indigenous identities were indicated by 18.8%; as with trans community surveys in Canada,^56,63  this is higher than the 4.9% in the general population.⁶⁴  In contrast, 6.6% were classified as visible minorities, versus 27.2% among Canada’s school-aged population,⁶⁵  raising questions for further exploration regarding study participation, timing of identity development, parental disclosure, pathways of care seeking, or access to care. Our findings also raise questions regarding access to care at younger ages, as youth had higher levels of parental gender support than community samples.³⁷  Youth from low-income families were well-represented, with 27.1% in households below the LIM, versus 17.0% of Canadians under age 18.⁶⁶  Low income in our study was not associated with ethnoracial group (P = .35) but was associated with urban residence (P = .01).

Consistent with existing reports,⁶⁷  mental health diagnoses were common. Highest were depression and anxiety, with prevalence similar to clinical data published on trans youth ages 10 to 17.⁶⁸  Personality disorders were not found to be relevant comorbidities, consistent with existing clinical samples.⁶⁹  Prevalences of self-harm, suicidal ideation, and suicide attempts observed were elevated compared with published estimates for cisgender youth.^6,68  Past-year suicide attempt (16.8%) was similar to findings for Ontario trans youth age 16-24 (19%; 95% confidence interval: 10–30)⁷⁰  but lower than the 36.1% observed in a national trans youth survey.⁶  These findings highlight the need for ongoing mental health supports to address psychological distress in younger patient populations; such supports differ from the psychological assessment process often required to access care. Community samples show lower suicide risk in older trans youth and adults,^6,70  with reductions associated with social, legal, and medical gender affirmation.^5,71 

Validated self-report symptom indicators of anxiety (71.4% vs 35.0%; P < .001) and depression (22.0% vs 7.6%; P = .03) were statistically higher among transmasculine youth, and diagnoses approached significance. Conversely, ADHD and autism spectrum disorder diagnoses were more likely in transfeminine youth (38.7% vs 19.8%; P = .04 and 23.4% vs 1.9%; P < .001). Findings mirror observations of a higher prevalence of neurodevelopmental disorders in clinical samples of trans youth, especially transfeminine youth.^72–75  and align with existing gender norms favoring ADHD and autism being diagnosed earlier and more often in those assigned male. The proportions of transmasculine and transfeminine youth who indicated a nonbinary identity did not differ (P = .68). Our findings of transmasculine disadvantage on some social context and mental health factors may be partially explained by the higher proportion of transmasculine youth living in their gender (90.1% vs 58.1%), thus experiencing others’ responses and negotiating access to gendered spaces and activities.

Hormonal suppression is most effective if initiated during puberty. Although youth identified gender issues at young ages, 69.0% were not seen for medications until age 14 or 15, and in clinics many youth are older at referral. This represents a missed opportunity to achieve the physiologic and psychological advantages hormonal suppression was designed to provide.²³  Our finding of high height-for-age z scores for transfeminine youth relative to female reference ranges suggests missed intervention opportunities for youth who may desire shorter stature. Estimates that 0.6% of the adult population is transgender,⁷⁶  76% will need gender-affirming medical care,⁷⁷  and ∼80% recognized their gender before age 14,^57,78  suggest that at age 14 ∼1 in every 277 youth is aware they are trans and will need gender-affirming care, although few are accessing care.

Finally, the high proportion of youth from low-income households, with food insecurity, or lacking access to essential resources indicates that these youth do access care in Canada. Although no participant gave cost as a reason for not starting medication, Canada’s “universal” health care system has cross-provincial variability in pharmacare (universal versus age-limited) with gaps in coverage. If needed at older ages, gender-affirming chest and genital surgeries would be publicly funded, although some may need additional supports for travel or aftercare costs. These findings support the need for administrative and financial supports to help patients and their families navigate care.

The study population is likely representative of youth <16 years of age referred for hormonal suppression or hormone therapy to specialized medical clinics in Canada. We included all major clinics providing such care during the study period. Although recruitment period varied by clinic, we weighted results to remove this potential bias. Some community clinics, family doctors, and general pediatricians have begun providing hormonal suppression for younger youth, including those in rural areas, often in consultation with subspecialists; such youth are not represented in this study.

Findings are from medical clinic-referred pubertal and postpubertal trans youth under age 16, and interpretation should be limited to this group. The majority of youth receiving gender-affirming medical care are older^21,22,24,25  and may attend without parents or be treated under adult protocols. Our study population differs from two more frequently studied groups: broader trans and gender-diverse communities and psychology clinic samples of gender-diverse children. The former includes those not needing gender-affirming medical care, or who identify as trans at later ages,^56,57  whereas the latter includes a more heterogeneous group of generally prepubertal children, including many who do not meet diagnostic criteria or will not later seek or be referred for gender-affirming medical care.^79–81 

Because some REBs did not allow eligibility prescreening, we were unable to calculate participation rate or examine differential participation across subgroups. Sample frequencies for assigned sex were comparable to those from a single-clinic records review,²⁴  but we were unable to similarly compare ethnoracial background. Although strategies to reduce participation barriers were implemented (parking, honoraria), participants may underrepresent rural patients. Responses to survey items regarding providers seen “prior to being seen at this clinic” may exclude those within the same clinic. One clinic may have less complete data on previous diagnoses, because an extended mental health waitlist resulted in high-risk patients being seen for hormonal suppression first. Finally, small subgroups limited statistical power (eg, (n = 37 transfeminine; n = 10 visible minority), although based on previous research and treatment differences we did make gender group comparisons.

We presented baseline youth characteristics from Trans Youth CAN!, the first prospective cohort study of trans and gender-diverse youth referred to medical clinics for hormonal suppression or gender-affirming hormone care in Canada. Findings are from patients under age 16, a group for whom hormonal suppression is a common therapy and outcomes are rarely studied. Future analyses will present follow-up data over a two-year period.

The Trans Youth CAN! Study Team would like to thank the trans youth and their families who have generously shared their time and experience with us. The authors acknowledge the contributions of the local site teams to participant recruitment, including site leads Ashley Vandermorris, Robert Stein, Natasha Johnson, and Jennifer Ducharme.