Pubertal Suppression for Transgender Youth and Risk of Suicidal Ideation

We thank Dr. Clarke for his interest in our article. We appreciate the opportunity to address his comments and correct the inaccurate information contained in them.

Dr. Clarke incorrectly states that this manuscript found an increase in recent serious suicide attempts among those who accessed pubertal suppression during adolescence. Though the raw values were higher for some of these outcomes, this was not a statistically significant finding, and thus the appropriate conclusion is that the study found no statistically significant association between access to pubertal suppression and greater odds of any measure of adverse mental health outcomes.

Dr. Clarke claims that the study participants responses were “biased,” based on the nature of the USTS sampling methodology. Though he does not clarify this comment, it appears he is referencing the fact that this is a non-probability sample. Non-probability samples are frequently used when studying minority populations, as they allow investigators to recruit large sample sizes and ask in-depth questions specific to the minority population being studied, which is generally not possible with probability samples. One major strength of this study is that it was able to recruit the largest sample of transgender people to date (over 27,000 participants) and ask in-depth questions, which allowed us to adjust for a wide range of potential confounders. This would not have been possible with a probability sample (e.g., the TransPOP sample, which uses random digit dialing for recruitment but was unable to recruit as many participants as USTS, or the CDC YRBSS, which is unable to ask as many in-depth questions about gender-related experiences because it must cover a wide range of other health topics).

Dr. Clarke expresses concern that “USTS respondents demonstrably did not know what puberty blockers were.” While it is not clear what this assertion was based upon, we would like to reiterate that our analyses excluded respondents who reported accessing pubertal suppression later than current guidelines advise. This intervention, when used at the recommended time, is done after counseling and education.

The most concerning claim in Dr. Clarke’s comment is his assertion that there is controversy regarding whether pubertal suppression should be made available for transgender youth who meet criteria under existing medical guidelines. Though there is controversy in the media and the political realm, there is clear consensus across the relevant major medical societies. The American Psychiatric Association, The American Academy of Child & Adolescent Psychiatry, The Endocrine Society, and The Pediatric Endocrine Society are some of the key organizations that oppose attempts to restrict access to pubertal suppression for transgender youth.

I am writing to express my concern that Pediatrics has named “Pubertal Suppression and Risk of Suicidal Ideation in Transgender Youth” by Turban et al. its Paper of the Year for 2020. The study’s conclusion that puberty blockers are beneficial is severely compromised by methodologic flaws. 1 To the best of my knowledge, Turban et al. failed to respond to the posted comments of their paper.

Turban et al.’s conclusions also contradict the recent extensive review of evidence by the UK High Court, which found puberty blockers to be an experimental treatment with significant risks.2 Contrary to Turban et al.’s recommendations to provide puberty blockers to any child or adolescent who wants it, the UK High Court ruled that young people can rarely consent to these treatments, requiring a case-by-case judicial review for any patients 16 and younger.

The following is a brief summary of the flaws in the Turban et al.’s study, which render their conclusions misleading:

1. The source study, the United States Transgender Survey 2015 (USTS), employed a non representative, biased convenience sample. The results from this survey are unreliable.3
2. Over 70% of the USTS respondents demonstrably did not know what puberty blockers were, claiming to have commenced treatment after age 18. Although Turban et al. attempted to control for this, a proper adjustment was not possible.
3. There was no control for underlying mental health. Since more stable individuals are more likely to be eligible for puberty suppression, one cannot discern mental health benefits or harms of puberty suppression without controlling for pre-treatment mental health.
4. Turban et al. ignored their own finding that a history of puberty suppression was associated with an increase in recent serious suicide attempts.

The question of whether and when it is appropriate to prescribe puberty blockers for Gender Dysphoria (GD) is far from settled, especially in light of the recent dramatic change in the presentation of GD. While in the past, most of the patients were male, currently, the majority of newly diagnosed cases are natal females with recent post-pubescent onset of GD and a high prevalence of comorbid autism spectrum conditions. These previously rare cases were considered ineligible for “Dutch Protocol,” which serves as the basis for the practice of prescribing puberty blockers for GD. One the principal authors of the protocol, Annelou de Vries, recently wrote a commentary in Pediatrics, alerting clinicians to the potentially inappropriate application of the protocol to a population for which it was not designed.4

Since the publication of Turban et al.’s work, Finland has issued much more cautious treatment guidelines that prioritize psychological support over puberty blockade and cross-sex hormone treatments.5 A similar change is underway in the UK following the High Court ruling.

When the alarm bells are clearly starting to ring internationally about the use of puberty blockers in GD, such an award appears to lack prudence and foresight. I believe that it would be helpful if the Journal could in future present a more balanced perspective on this issue.

References

1. Turban JL, King D, Carswell JM, Keuroghlian AS. Pubertal Suppression for Transgender Youth and Risk of Suicidal Ideation. Pediatrics. 2020; 145(2): e20191725. doi:10.1542/peds.2019-1725.
2. Judiciary.uk. https://www.judiciary.uk/wp-content/uploads/2020/12/Bell-v-Tavistock-Jud.... Published 2020. Accessed January 3, 2021.

3. D’Angelo R, Syrulnik E, Ayad S, Marchiano L, Kenny DT, Clarke P. One Size Does Not Fit All: In Support of Psychotherapy for Gender Dysphoria. Archives of Sexual Behaviour. 2020; https://doi.org/10.1007/s10508-020-01844-2

4. De Vries ALC. Challenges in Timing Puberty Suppression for Gender- Nonconforming Adolescents. Pediatrics. 2020.; http://pediatrics.aappublications.org/lookup/doi/10.1542/peds.2020-010111

5. Palko [Finnish Health Authority / Council for Choices in Health Care in Finland]. Medical Treatment Methods for Dysphoria Associated with Variations in Gender Identity in Minors: Recommendation. 2020.;
https://palveluvalikoima.fi/documents/1237350/22895008/Summary_minors_en.... Published 2020. Accessed January 3, 2021.

Dear editors and authors,
In response to the article “Pubertal Suppression for Transgender Youth and Risk of Suicidal Ideation”, I would like to thank the authors for exploring this topic. But I still have two questions about this article.

First, I have great concerns about the method of self reports which was used in this questionnaire. The answers might reflect the desire of the transgender social group rather than true psychological or physiological benefits. In this article, they asked “Have you ever wanted any of the health care listed below (including puberty blockers) for your gender identity or gender transition? Participants might be led to think “I would be happier and would not have the suicide ideation if I ever had the puberty blockers in my adolescence.” Therefore, the socially desirable option could be overreported (1). In addition, none of similar studies listed in this article has used the method of self reports, either.

We also noticed “Suicidality (past 12 mo): Attempt resulting in inpatient care” in Table 3 indicated 45.5% in the participants who had puberty blockers, while 22.8% in those who didn’t have such therapy. A higher odds ratio is noted; however, without significant difference (table 2). This might be attributed to the small sample size of the puberty blocker users (2, 3). Nevertheless, suicide attempts resulting in inpatient care would be an important indicator to know the true outcomes of puberty blockers. We would suggest a further investigation on this issue to clarify the outcome of puberty blockers rather than concluding based on “no significance”.

Sincerely,

I Cheng

1. NHS. Writing an effective questionnaire. England. 2018.
2. du Prel JB, Hommel G, Rohrig B, Blettner M. Confidence interval or p-value?: part 4 of a series on evaluation of scientific publications. Dtsch Arztebl Int 2009;106(19):335-9.
3. Szumilas M. Explaining odds ratios. J Can Acad Child Adolesc Psychiatry 2010;19(3):227-9.

Transgender Youth, Pubertal Suppression, and Mental Health Outcomes: The Next Steps

In response to the article “Pubertal Suppression for Transgender Youth and Risk of Suicidal Ideation”, I would like to thank the authors for exploring this topic. As the first study to examine associations between suicidality and access to pubertal suppression, this work adds significant findings to the continued discourse regarding transgender care. Of note, the two previous studies regarding transgender treatment and mental health had significant limitations including lack of diversity, few participants, and notably no control group. Modern studies must encompass a representative proportion of the population to provide accurate data. Recruiting the largest survey of transgender adults from all 50 states provides a more representative sample of the United States.

While inclusion is important, highlighting the differences in outcomes amongst the various races of participants would further the discourse. Measuring whether significant differences in household income, education level, family support, and employment status by race could add data to determine socioeconomic factors which influence mental health outcomes.

Although no participants under the age of 18 were included in this research, the mental health outcomes of teens are an important area of study. In a 2016 Minnesota study, 80, 929 students were questioned about their gender identity and perceived health1. The study found that transgender/non-conforming teens reported significantly lower rates of health check-ups, poorer health status, and more nurse office visits than cisgender students1. Even further, students whose gender expression was incongruent with the accepted expectation from their sex assigned at birth were found to be at higher risk for poorer health outcomes1. These findings further support that transgender persons need suitable insurance coverage, culturally competent physicians, and familial financial support for their health and wellness. As stated in the research, insurance coverage and family support are important in order to access the GnRHa therapies which range in cost from a few thousand up to $25,000. Using data on mental health outcomes to shape policies and influence insurance coverage could help address the financial barriers to care.

Another area to further examine is the effect of experiencing assault as a transgender teen on adult mental health outcomes. In the United States, an estimated 47% of gender minorities report a lifetime history of sexual assault2. In a 2017 survey of 3,673 transgender and nonbinary US teens found a sexual assault prevalence of 36% in youth who had restroom and locker room restrictions2. Further work investigating assault and long-term mental health outcomes can be used to support advocacy efforts and shape current policies.

The field of Pediatrics will continue to see advancements in pubertal suppression therapies. As these therapies are implemented, we must still address the higher rates of mental health disparities, violence, and socioeconomic hardships for transgender persons. The research completed in this study provides an excellent basis for continued advocacy to address the mental health outcomes of transgender persons.

References

1. Health and Care Utilization of Transgender and Gender Nonconforming Youth: Population-Based Study: G. Nicole Rider, Barbara J. McMorris, Amy L. Gower, Eli Coleman and Marla E. Eisenberg; Pediatrics 2018;141; DOI: 10.1542/peds.2017-1683 originally published online February 5, 2018.

2. School Restroom and Locker Room Restrictions and Sexual Assault Risk Among Transgender Youth: Gabriel R. Murchison, Madina Agénor, Sari L. Reisner and Ryan J. Watson; Pediatrics 2019;143; DOI: 10.1542/peds.2018-2902 originally published online May 6, 2019.

Given the controversy surrounding the practice of puberty suppression for gender dysphoric adolescents, the article by Turban et al.1 creates more confusion than clarity. The authors imply causal evidence for a reduction in suicidal ideation with transgender adolescents who received puberty suppression (PS), yet they fail to acknowledge the exceedingly high rates in both groups of suicide ideation (75% and 90%) and suicide attempts (42% and 51%). The cross-sectional design using online survey data is insufficient to validate the efficacy of such a life-altering therapy. Because the data was collected by survey, there is no way of knowing how many would-be participants in either group actually succumbed to suicide. The so-called “lifetime suicidal ideation” is misleading, since at the time of the survey, the PS treatment group was significantly (p=001) younger (mean age 21.7 years) than the “ever wanted” PS control group (mean age 23.4 years), and the total age range of survey participants was 18 to 36 years. With mean age of hormone treatment initiation being 15.7 years in the PS group and 22.5 in the control group (p<0.001), it is obvious that the follow-up time for both groups was far too brief to assess “lifetime suicidal ideation.” The control group was not appropriately matched to the treatment group by age at time of survey or by age when hormone therapy was begun. Since there were over 30 controls for each PS case, they could have been selectively trimmed to be better matched.
What is more disturbing is that the PS treated group actually had double (45.5% versus 22.8%) the rates of the control group for serious (resulting in inpatient care) suicide attempts in the year preceding the data collection (Table 3).1 Adolescents who identify as transgender along with other “sexual minorities” (including lesbian, gay, and bisexual youth) are increasing at extraordinary rates (doubled between 2009 and 2017), and these adolescents are almost four times more likely than their heterosexual peers to commit suicide.2 The same study found that over 35% of adolescent suicide attempts in 2017 came from sexual minorities, who comprised only 14% of the adolescent population. Gender dysphoric youth and their families need to know the truth about therapies offered – that real long-term safety and efficacy studies do not exist. Research surrounding PS, cross-sex hormones, and “gender-confirming surgery” that potentially render recipients sterile, physically altered, and sexually dysfunctional must adhere to the same high standards applied to less controversial conditions, and these therapies should be considered experimental until such studies have been done. The prevailing narrative that these interventions are necessary to prevent suicide is without reasonable evidence.

References:
1. Tuban JL, King D, Carswell JM, Keuroghlian AS. Pubertal suppression for trangender youth and risk of suicidal ideation. Pediatrics. 2020;145(2):e20191725.
2. Raiafman J, Charlton BM, Arrington-Sanders R, et al. Sexual orientation and suicide attempt disparities among U.S. adolescents: 2009-2017. Pediatrics. 2020;145(3):e20191658.

Turban et al. present an analysis of subjects from the database of the 2015 U.S. Trans Survey of more than 20,000 respondents. The authors extracted responses of 89 adult transgender subjects who experienced gender dysphoria (GD) in childhood and demonstrated that those who received puberty blockers (PB) fared better in measures of adult suicidality than those who did not.
This result is intuitive; adults more readily “pass” in society when development of secondary sex characteristics of the non-preferred biological birth sex was inhibited. The authors’ result supports this hypothesis.

Nevertheless, a word of caution is in order. The authors conclude that all who wish should be given PB. Presently, however, there is no method to predict persistence in GD. Estimates vary (1), but on average only about 15% persist; in contrast, when given PB, virtually 100% persist (2). Thus, an indiscriminate prescription of puberty blockers will significantly increase the number of adolescents who continue to full transition, which may worsen long-term outcomes in attempted suicides.

The differences in expected outcomes if PB is or is not prescribed can be estimated. Of 1000 children with GD, if all receive puberty suppression then we expect all 1000 to go on to full transition whereas without the pubertal inhibition only 150 (15% of 1000) will transition. As the authors correctly state in the paper, 40% of transpersons attempt suicide in a lifetime, which means that with PB administration to all, we expect 40% of 1000 = 400 persons to attempt suicide. The authors show, however, that because of the benefits of PB, this may perhaps be adjusted downward by a factor of 0.6; the expected outcome is then 240 attempted suicides. In contrast, if none of the 1000 subjects receive puberty suppression then only 60 persons (40% of 150) are expected to attempt suicide.

The estimate here excludes a number of details such as the baseline of suicide attempts in a control population, lifetime suicide attempts of desisters, and differences in suicide risk for different subgroups of transgender subjects (e.g. FtM, MtF), but these omissions are consistent with the assumptions in the analysis by Turban et al. and unlikely to affect the conclusion.

Thus, pending reliable prediction of who will desist or persist, prescription of PB is not warranted. Giving PB to all who wish it is expected to significantly increase the total number of suicide attempts, up to 240 per 1000, compared with the outcome when not giving puberty suppression to anyone, 60 per 1000.

1. Ristori, J., Steensma, T.D. (2016) “Gender dysphoria in childhood” International Review of Psychiatry, DOI: 10.3109/09540261.2015.1115754
2. Steensma, T. D., McGuire, J. K., Kreukels, B. P. C., Beekman, A. J., & Cohen-Kettenis, P. T. (2013). "Factors associated with desistence and persistence of childhood gender dysphoria: A quantitative follow-up study" Journal of the American Academy of Child and Adolescent Psychiatry, 52: 582–590

To the Editor and authors,

I read with great interest the article “Pubertal Suppression for Transgender Youth and Risk of Suicidal Ideation.” I am very grateful to Dr. Turban and colleagues for bringing attention to the relationship between mental health outcomes in the transgender community and receipt of available treatment with pubertal suppression during adolescence. Adolescents who identify as transgender have significant mental health disparities compared with their cisgender peers. It is of great importance that general pediatricians and Endocrinologists who are responsible for their care work together to increase access to all available treatments.

As pediatricians, we are inherently tasked with working to protect all children and being an advocate for the most vulnerable among them. Legislative advocacy is often used in this mission to increase these populations access to care and resources. It is disheartening when legislation is used instead to oppress the populations in the most need of our care like the transgender community. In the state of Georgia, legislation is now being drafted that would make it illegal for physicians to offer patients under the age of 18 surgical or medical gender-affirming interventions including the use of “puberty blockers.” As pubertal suppression is safe and completely reversible there is no reason that the decision to use this treatment or not should be made by legislators instead of medical professionals with patients and their families.

More articles like this are needed to continue to provide evidence demonstrating access to this safe and effective treatment results in better long-term outcomes. This article shows that patients who were aware of available treatments, sought out care, and received pubertal suppression had decreased odds of suicidal ideation and severe psychological distress. However, only 2.5% of those who wanted pubertal suppression received the treatment. Child advocates and legislators should be working together to make sure that 100% of adolescents wanting this treatment are able to receive it, which would reduce mental health disparities and improve long-term productivity.

The National Transgender Discrimination Survey found that 41% of transgender respondents attempted suicide, compared to 1.6% of the general population. This is a staggering statistic and pediatricians should work to protect transgender youth from being a part of this 41%. Dysphoria resulting from physical characteristics that does not align with one’s gender identity contributes to depression and anxiety. If there is a way to improve access to care and as a result improve mental health outcomes then it is neglectful for us to not do so.

The Endocrine Society and the World Professional Association for Transgender Health Standards of Care recommend that puberty suppression be offered to transgender patients and the AAP recommends that all transgender youth have access to comprehensive, gender-affirming health care. As these organizations are the leaders and experts surrounding this issue it is important that we take their advisement into consideration when shaping policies affecting these populations and that we advocate against policies, like the one proposed in Georgia, that disregard medical guidelines.

This article finds that adults who had been prescribed puberty blockers have lower suicidal ideation than those who wanted them but were not prescribed them. The association holds (under multivariate analysis) for one out of five measures of suicidality tested by the authors. It does not, as the authors claim, warrant the recommendation “that this treatment be available for transgender adolescents who want it.”

We know that psychological stability is a prerequisite for puberty blockers. WPATH, for example, stipulates that they can be prescribed only if “Any coexisting psychological, medical, or social problems that could interfere with treatment ... have been addressed, such that the adolescent’s situation and functioning are stable enough to start treatment” (WPATH 2012: 19; cf. de Vries et al. 2011: 277).

Because adolescents with greater suicidal ideation were less eligible for puberty blockers, this automatically created an initial negative association between the two—before the treatment took effect. Therefore authors’ finding, from adults surveyed many years after treatment, is compatible with three scenarios: puberty blockers reduced suicidal ideation; puberty blockers had no effect on suicidal ideation; puberty blockers increased suicidal ideation, albeit not enough to counteract the initial association between suicidal ideation and eligibility.

The authors acknowledge that “the study’s cross-sectional design… does not allow for determination of causation.” But this caution was not conveyed in the many news reports generated by the study. “Puberty blockers reduce suicidal thoughts in trans people” ran a typical headline (LGBTQ Nation 2020).

Aside from the spurious leap from association to causation, the analysis is inevitably limited by the poor quality of the data.

Firstly, the survey’s respondents are not sampled from any defined population. The convenience sample excludes those who underwent medical intervention but subsequently stopped identifying as transgender. It also excludes those who did commit suicide.

Secondly, the key questions on puberty blockers confused some of the respondents. The survey report cautions that “a large majority (73%) of respondents who reported having taken puberty blockers [in question 12.9] ... reported doing so after age 18 [in question 12.11] ... This indicates that the question may have been misinterpreted by some respondents who confused puberty blockers with the hormone therapy given to adults and older adolescents” (James et al. 2016: 126). To mitigate this problem, Turban et al. follow the survey report in ignoring those respondents who reported taking puberty blockers after the age of 18. No such adjustment is possible, however, for the question asking whether the respondent had ever wanted puberty blockers (question 12.8), which Turban et al. use to define the subset of respondents in their analysis. Therefore the comparison group will include an unknown number of respondents—possibly the majority—who actually wanted cross-sex hormones rather than puberty blockers.

Thirdly, many questions have a large number of missing values. Of the 89 respondents who took puberty blockers, only 11 answered the question on whether they had been hospitalized as a result of attempting suicide in the last 12 months (question 16.5).

In sum, this article contributes little to our knowledge on the effect of suppressing puberty on adolescents. The single study to show positive psychological effects has a sample size of 70 and lacks any control group (de Vries et al. 2011). The authors cite a second study (Costa et al. 2015), but that actually shows no statistical difference in improvement in psychological functioning between the group prescribed puberty blockers and the group given therapy (Biggs 2019).

“Longitudinal clinical trials are needed to better understand the efficacy of pubertal suppression,” as the authors observe. It is remarkable that such a call is still necessary after more than two decades of this experimental treatment on adolescents suffering from gender dysphoria.

References not included in the target article
Biggs, M. (2019). A letter to the editor regarding the original article by Costa et al: Psychological support, puberty suppression, and psychosocial functioning in adolescents with gender dysphoria. Journal of Sexual Medicine, 16, 2043.
LGBTQ Nation (2020). Puberty blockers reduce suicidal thoughts in trans people: Republicans want to ban them. https://www.lgbtqnation.com/2020/01/puberty-blockers-reduce-suicidal-tho...
World Professional Association for Transgender Health (2012). Standards of care for the health of transsexual, transgender, and gender-nonconforming people, 7th version.

The article by Turban et al.1 is being widely interpreted as providing evidence that GnRH agonists are beneficial in preventing suicide in transgender youth. This includes an accompanying AAP news article2. This conclusion is not warranted from the article. While the authors acknowledge that, as a cross-sectional study, their data cannot establish causal relationship between pubertal blockade and suicidality, they fail to emphasize that those who received puberty blockade, AND those that did not, both had alarmingly high rates of suicidal ideation (50% or higher) within the last year, rates strikingly similar to those previously reported for transgender adults3. There was no difference between the study groups when comparing a more robust measure of suicidal risk: ideation with a plan. Furthermore, those receiving GnRH agonists had higher rates of hospitalization for suicide attempts when compared to those not receiving this medication. The argument that there was a lack of statistical power is an assumed explanation for these effects, and is often used by scientists when their hypothesis is not supported by the data. An equally plausible explanation is that suicidal risk is almost independent of taking GnRH agonists because pubertal blockade fails to address important co-occurring psychological issues. The study has several methodological weaknesses. Community-based samples of patients identifying as transgender are rife with ascertainment biases. Gender dysphoric youth who do not identify as transgender later in life are not in the sample. The influence of pubertal blockade on their identity and suicidality are not considered. It is not surprising that patients who identify as transgender later in life value an intervention that supported their early identification and that patients that desired but did not receive that intervention over-estimate its value. The most meaningful comparison should be between patients that receive the intervention and those that do not regardless of whether the treatment was offered and regardless of whether they identify as transgender later in life. Secondly, since gender dysphoria experienced by these patients is influenced by a myriad of factors, it is misleading to base conclusions about the benefit of receiving GnRH agonist on any univariate analyses. To achieve reliable conclusions on suicide risk, more complex modeling assessing multiple inter-related contributors is required. Notably, all but one of the univariate results is non-significant when controlling for but a few background factors. Although the odds ratio for lifetime suicidality remains significant, the Pearson correlation is surprisingly small, < .08, yielding a very small effect size (Cohen's d < .20) and does not assess current or future suicide risk. Thus, from this survey it is an unjustifiable stretch to conclude, or even suggest, that for gender dysphoric youth, pubertal blockade is a lifesaving intervention. Young adults who experience a gender identity that is discordant with biological sex leads to serious concerns about the long-term efficacy of puberty blockade. Further research into this complex and poorly understood population is needed. Until this research is published caution is required in recommending GnRH agonists to alter the natural course of puberty.

1. Turban et al. Pubertal Suppression for Transgender Youth and Risk of Suicidal Ideation. Pediatrics. Jan. 23, 2020, https://doi.org/10.1542/peds.2019-1725)
2. Jenco, Blocking puberty in transgender teens linked to lower likelihood of suicidal thoughts. AAP News January 23, 2020. https://www.aappublications.org/news/2020/01/23/pubertysuppression012320
3. Adams, Hitomi & Moody, Varied Reports of Adult Suicidality: Synthesizing and Describing the Peer-Reviewed and Gray Literature. Transgender Health 2.1:60-75, 2017

Fantastic attention and information for care of Transgender Youth patients. This will save lives. In this case, a strong suspicion should be had that the suicide risk for ALL youth may be significantly lowered with Pubertal Suppression, which I would not support. Proving this with an ethical study would be most challenging. Of course, that does not negate the findings here in any way. Thank you for this very strong work.