Are children and young people with gender dysphoria at higher risk of suicide?

Suicide risk is frequently raised as a reason to medically intervene using puberty blockers and hormones to ‘affirm’ the self-identity of young individuals with gender dysphoria. Given the tragedy of completed suicide, a medical intervention shown to reliably reduce chances of this outcome would be welcomed. It is vital that premises and claims around suicide are checked. Certain antidepressants used to be thought of as protective in adolescents, but were later shown to increase adolescent suicidality (Le Noury et al., 2015). There is a general lack of high quality research that examines whether children and young people with gender dysphoria are at higher risk of suicide, quite separately from the question of whether endocrine interventions ‘work’ to change it. Some of the available evidence is gathered and evaluated here.

Surveys:

There are several surveys of varying quality.  

RARE study (also known as ‘PACE’) (Nodin et al., 2015)

• This was a UK based convenience sample (participants were self-selected and recruited largely through email, web-based advertising, social networks and targeted promotional campaigns). The online survey included the responses of 27 young people under the age of 26 who self-identified as transgender.  It reported that amongst those 27, 13 (48%) had attempted suicide at least once.

Stonewall School Report (Bradlow et al., 2017)

• Participants who identified as lesbian, gay, bisexual or trans (LGBT), or who thought they might be, aged between 11 and 19 years old completed an online questionnaire regarding their experiences at school, online and home. 
• 3713 young people from England, Scotland and Wales were included in total, 594 of them identified as transgender.
• Rates of self-harm and suicidal ideation / attempts were high in all groups. 
• Rates of suicidal ideation (thoughts) were highest in young people identifying as transgender (92%) compared to lesbian, gay and bisexual (70%) young people who did not identify as trangender or non-binary. Reported suicide attempts (at any point) was higher in young people who identified as transgender (45%) compared with lesbian, gay and bisexual (22%).
• Unclear role of other risk factors for suicide, as these were not measured.

US large scale Survey (Toomey, Syvertsen and Shramko, 2018)

• This was the only peer-reviewed survey available.
• The study was based on secondary data analysis of three years worth of answers to the Profiles of Student Life: Attitudes and Behaviors survey, which began to include a question on gender identity in 2012.
• Large sample of 120,617 adolescents, which also recorded other demographic data. Less than 1% of respondents identified as transgender. 202 respondents (0.2%) identified as male to female transgender; 175 respondents (0.1%) as female to male transgender; and 344 respondents (0.3%) as nonbinary. Additionally, 1052 (0.9%) reported that they were not sure of their gender identity.
• Self-reported lifetime suicide behaviour was assessed by one question: “have you ever tried to kill yourself?” The study dichotomised the answers as either never (0 self-reported attempts) or ever (1 or more). 
• Showed higher rates of self-reported lifetime suicide attempts in female adolescents (17.6%) compared to males (9.8%).
• Females identifying as transgender were at the highest risk (50.9%), followed by non-binary individuals (sex not specified, 41.8%), males identifying as transgender (30%), and adolescents questioning their gender identity (sex not specified, 27.9%).
• The study also noted increased risk of suicidal behaviour for lesbian, bisexual and gay adolescents.

In summary, surveys suggest high prevalence of suicidal thoughts or attempts in LGBT groups. However, these are limited in their analysis (for example, in not always knowing how the questions are worded or understood by respondents, not gathering data on mental health comorbidity or other risk factors for the suicidal ideation). The methodology does not take into account variations in how suicidal ideation might be experienced or give information on psychiatric hospitalisation or completed suicides.

Is there a higher risk of completed suicide in LGBT youth? 

The UK keeps high quality data on suicide in the National Confidential Inquiry into Suicides and Homicides, which gives further information on this question. 

The data for all suicides of 10-19yr olds was examined for the period 2014-16 and published in 2020 (Rodway et al., 2020).

• In total, there were 595 suicides of 10-19yr olds in the period 2014-2016.
• 32 were known to be LGBT or questioning young people (5% of the total), 5 of whom were transgender (1% of the total).
• 44% of the LGBT group were female, compared to 28% in the whole group.
• Suicide in LGBT young people was significantly associated with other risk factors for suicide such as bullying (28%), abuse (16%) and self-harm (75%).
• LGBT young people were also significantly more likely to have suicide related Internet use.

The Office for National Statistics reported in 2015 that 3.3% of 16-24 yr olds identified as LGBT (the time-period correlating reasonably with the study). It would therefore appear there may be an increased risk of completed suicide in LGBT individuals (at least 5% of the total vs 3% expected), although this is modest. It might be explained by higher rates of other co-occurring risk factors. The proportion of girls completing suicide in the LGBT group is higher (44% vs 28% in the overall population) which may add complexity to the sex, sexual orientation and gender issues. It may suggest that young females who identify as sexual or gender minority (ie lesbian, bisexual, transmasculine, nonbinary or other gender) are more likely to have significant psychological problems, or (vice versa since cause and effect are not known) that girls/ young females with significant psychological problems are more likely to identify as sexual or gender minority.

What about long term risk of suicide?

There is limited information about the long-term risk of suicide as many recent studies have short follow up periods.  There is one robust, long term Swedish cohort study that followed up adult transsexual individuals (who had undergone legal gender reassignment and genital surgery) over a period of 30 years between 1973 to 2003 (Dhejne et al., 2011). Because of the time period covered, it is highly likely that those included would have experienced more barriers to transition, both social and medical (for example, an expectation of extensive psychotherapy before cross-sex hormones and surgery) and have been living in less tolerant times than now with more positive public imagery.

• This study included 324 transsexual individuals. The authors compared their outcomes to 3240 individuals who were matched for birth year and sex (the control group). The measure used to compare the risks of various outcomes between the groups is the adjusted hazard ratio. 
• There were 10 deaths by suicide in the transsexual group between 1973-2003. In the same time there were 5 deaths by suicide in the control group (5/3240). The risk of death by suicide was 19 times higher in the transsexual group compared to the non-transsexual controls.
• The number of recorded suicide attempts in the transsexual group was 29 (29/324). During the same time there were 44 suicide attempts in the control group (44/3240). This risk of suicide attempts was 4.9 times greater for transsexuals than for the controls. 
• The number of psychiatric hospitalisations in the transsexual group was 64 (64/324). During the same time there were 173 psychiatric hospitalisations in the control group (173/3240). Transsexuals had a 2.8 times higher risk of being hospitalised for mental ill health than controls.

The research suggests that transgender people have higher long-term rates of mental health problems and suicide regardless of having fully undergone legal gender reassignment and medical interventions including genital surgery. Other studies appear to support the finding that the risk of suicide for transgender people remains higher than the general population and deaths from suicide occur “during every stage of transition” (Wiepjes et al., 2020). 

It is disputed whether these findings show that transition ‘works’ to reduce suicidality and poor mental health. Proponents of medical transition may argue that suicidal behaviour risks would be even greater without endocrine or surgical interventions. Others may argue that medical transition does not ‘work’ to reduce suicidality, because other associated factors such as trauma or depression are at play, and the risks remain elevated despite hormones and surgery. It is not disputed that improvements in health and life expectancy are needed.

Do young people with gender dysphoria have a higher rate of mental health difficulties?

Several studies have shown higher rates of mental health problems in young people identifying as transgender.

• Kaltiala-Heino in Finland found that 75% of young people referred to a gender identity clinic service had severe mental health problems including depression, anxiety, psychosis, eating disorders and autistic spectrum disorders (Kaltiala-Heino et al., 2018).
• Holt, studying referrals to the Gender Identity Development Service (GIDS) in the UK found a similar range of comorbid conditions, including abuse (17%) and bullying (47%) (Holt, Skagerberg and Dunsford, 2016). 
• Comorbidity rates in different studies vary from a low of 33% (the Netherlands) to 75% (Scandinavia), averaging about 40-45%. These rates are similar to those of young people referred to Child and Adolescent Mental Health Services for other reasons (Kaltiala-Heino et al., 2018).
• Autistic spectrum disorders are particularly and consistently over-represented (van der Miesen et al., 2018).

Does treatment reduce the risk of suicide?

Again, information is limited, particularly by the lack of controlled studies. Studies are generally short in terms of follow up so they do not provide adequate data about longer term outcomes.  ‘Before-and-after’ studies are notoriously misleading as they virtually always show benefit in all branches of medicine (and alternative remedies), due to the phenomena of regression to the mean, placebo, expectation, Hawthorne effect and mood-altering drugs. 

An early study on puberty blockers, sometimes referred to as ‘The Dutch Study’ (de Vries et al., 2014), showed some promising results.  Behavioural, emotional and depressive symptoms reduced. It is worth noting that patient selection for this study excluded many young people who would be considered for puberty blockers today. Only those with an onset of gender dysphoria in early childhood were included in the original study, and psychological assessment and treatment were standard.  There was no control group, so it cannot be concluded that puberty blockers definitely improved function.

A study at GIDS in the UK (Costa et al., 2015) showed that all gender dysphoric young people improved psychologically over the time frame of the study (which followed participants for 18 months). The authors noted a slightly greater improvement in global functioning for the young people prescribed puberty blockers with psychological support, compared to those receiving psychological support alone. However, at baseline the young people who were given puberty blockers and psychological support had higher global functioning scores compared to the young people given only psychological support, a finding that failed to reach statistical significance possibly due to sample size. At the end of the study, there were also slightly higher functioning scores for the group given puberty blockers, however that difference also did not reach statistical significance. The study did not have a control group. Therefore its findings are unclear.

The GIDS Early Intervention Study of puberty blockers (Carmichael et al., 2021) found less positive results. During the study, 44 children prescribed puberty blockers were followed for up to 3 years.  No significant improvements on measures of psychological functioning were found.

A US study (Turban et al., 2020) claimed to show that puberty suppression was associated with reduced mental health morbidity but has been heavily criticised in its methodology and conclusions (Biggs, 2020).  The study suggested that adults who stated that they wished they had received puberty blockers during their adolescence were more likely to experience suicidal ideation compared to a group of individuals who reported having been prescribed puberty blockers.  

Turban’s study had many limitations, not least that it was a convenience sample, based on self-report and subject to recall bias. Participants either did not understand the terminology (70% claimed to have been prescribed puberty blockers at age 18 or over), or their doctors were prescribing them quite late, or for another reason. No correction for mental health comorbidity was attempted, which is particularly important as individuals with serious mental health comorbidities would have been less likely to receive a prescription for puberty blockers. Concerningly, the study also showed a trend towards higher rates of suicidal acts, especially more severe ones, in those people who recalled being prescribed puberty blockers. In other words, suicidal thoughts were more common in people who did not recall being prescribed puberty blockers, but suicidal acts, particularly serious ones, were more common in individuals who did recall receiving treatment with puberty blockers. This finding was dismissed by the authors as it did not reach statistical significance given the small numbers, but it is of high clinical concern and warrants further study.

In summary, the evidence that puberty blockers have any kind of beneficial effect on mental health is equivocal.  Positive results are frequently held up as evidence that young people must have access to puberty blockers, whilst more troubling results (as in Turban’s case) are dismissed as insignificant.  Without prospective, controlled studies working to pre-specified protocols that follow up over a longer period of time, it is hard to see how more reliable evidence of safety and effectiveness, with more precision around the sizes of risks and harms, especially for any subgroups, can be gathered.

When life is difficult, Samaritans are here – day or night, 365 days a year. You can call them for free on 116 123, email them at jo@samaritans.org, or visit samaritans.org to find your nearest branch.

References:

Biggs, M. (2020) ‘Puberty Blockers and Suicidality in Adolescents Suffering from Gender Dysphoria’, Archives of Sexual Behavior, 49(7), pp. 2227–2229. doi: 10.1007/s10508-020-01743-6.

Bradlow, J. et al. (2017) School Report:The experiences of lesbian, gay, bi and trans young people in Britain’s schools in 2017, Stonewall. Available at: https://www.stonewall.org.uk/system/files/the_school_report_2017.pdf.

Carmichael, P. et al. (2021) ‘Short-term outcomes of pubertal suppression in a selected cohort of 12 to 15 year old young people with persistent gender dysphoria in the UK’, PLOS ONE. Edited by G. L. Santana, 16(2), p. e0243894. doi: 10.1371/journal.pone.0243894.

Costa, R. et al. (2015) ‘Psychological Support, Puberty Suppression, and Psychosocial Functioning in Adolescents with Gender Dysphoria’, The Journal of Sexual Medicine, 12(11), pp. 2206–2214. doi: 10.1111/jsm.13034.

Dhejne, C. et al. (2011) ‘Long-Term Follow-Up of Transsexual Persons Undergoing Sex Reassignment Surgery: Cohort Study in Sweden’, PLoS ONE. Edited by J. Scott, 6(2), p. e16885. doi: 10.1371/journal.pone.0016885.

Holt, V., Skagerberg, E. and Dunsford, M. (2016) ‘Young people with features of gender dysphoria: Demographics and associated difficulties’, Clinical Child Psychology and Psychiatry, 21(1), pp. 108–118. doi: 10.1177/1359104514558431.

Kaltiala-Heino, R. et al. (2018) ‘Gender dysphoria in adolescence: current perspectives’, Adolescent Health, Medicine and Therapeutics, 9, pp. 31–41. doi: 10.2147/AHMT.S135432.

Le Noury J, et al. (2015) ‘Restoring Study 329: efficacy and harms of paroxetine and imipramine in treatment of major depression in adolescence’, BMJ, 351:h4320. doi:10.1136/bmj.h4320

van der Miesen, A. I. R. et al. (2018) ‘Autistic Symptoms in Children and Adolescents with Gender Dysphoria’, Journal of Autism and Developmental Disorders, 48(5), pp. 1537–1548. doi: 10.1007/s10803-017-3417-5.

Nodin, N. et al. (2015) The RaRE Research Report. LGB&T mental health risk and resilience explored. doi: 10.13140/RG.2.1.2810.0961.

Rodway, C. et al. (2020) ‘Children and young people who die by suicide: childhood-related antecedents, gender differences and service contact’, BJPsych Open, 6(3), p. e49. doi: 10.1192/bjo.2020.33.

Toomey, R. B., Syvertsen, A. K. and Shramko, M. (2018) ‘Transgender Adolescent Suicide Behavior’, Pediatrics, 142(4). doi: 10.1542/peds.2017-4218.

Turban, J. L. et al. (2020) ‘Pubertal Suppression for Transgender Youth and Risk of Suicidal Ideation’, Pediatrics, 145(2). doi: 10.1542/peds.2019-1725.

de Vries, A. L. C. et al. (2014) ‘Young Adult Psychological Outcome After Puberty Suppression and Gender Reassignment’, PEDIATRICS, 134(4), pp. 696–704. doi: 10.1542/peds.2013-2958.

Wiepjes, C. M. et al. (2020) ‘Trends in suicide death risk in transgender people: results from the Amsterdam Cohort of Gender Dysphoria study (1972–2017)’, Acta Psychiatrica Scandinavica, 141(6), pp. 486–491. doi: 10.1111/acps.13164.