The 41% trans suicide attempt rate: A tale of flawed data and lazy journalists

If there is one constant in reports about transgender people, it’s the prevalence of suicidal intent.  Nearly all media accounts cite an average 41% suicide attempt rate. A Google keyword search for “transgender 41% suicide” results in over 43,000 hits.

Often, the attempted suicide rate is presented in the context of a story about a young person desperately needing to medically “transition” to the opposite sex. Caitlyn Jenner mentioned the 41% when accepting the ESPY courage award last month, and gender specialists like Johanna Olson routinely bring up suicide as a rationale for hormone and surgical treatments.

In every one of the stories I’ve read, the unspoken or explicit assumption is that transition cures suicidality.

A parent reading one of these stories will be terrified. The very notion that a child might attempt suicide is the worst possible nightmare a mother or father could imagine. The message being hammered  over and over again is that we can only save our young people by supporting their transition to the opposite sex, no questions asked, if that’s what they say they want.  Period. End of discussion. (I have personally been accused of contributing to the risk of youth suicide, simply by raising the questions I do in my blogs.)

So where does this 41% figure come from?

Many media accounts don’t cite the source of the 41% number, but this one does: an analysis released in January 2014 by The Williams Institute, in collaboration with the American Foundation for Suicide Prevention. The report, Suicide Attempts Among Transgender and Gender Non-Conforming Adults, drew its data from a 2008 U.S. National Transgender Discrimination Survey of 6456 self-identified transgender and gender non-conforming adults (ages 18+). Of these, 2566 (40%) were biological females at birth. (As always, natal females will be the main focus of my post.)

A suicide attempt rate of 41% is an emergency. Surely the Williams Institute analysis is conclusive enough to warrant the burgeoning number of gender clinics hurrying to diagnose and start “transitioning” young people who identify as transgender? Does the data convincingly show that gender dysphoria is alleviated by “passing” as the opposite gender, and that medical transition lowers suicide rates?

It does no such thing.

The authors of the study were well aware of its limitations, as I’ll show in this post. But you don’t have to take my word for it: read the AFSP/Williams Institute analysis yourself. It’s written in accessible language and is only 18 pages long, much of which is summarized in easy-to-understand tables. This material could be absorbed by even an average journalist, who presumably is paid to be at least marginally interested in the actual findings of the survey. Even a part-time, unpaid, obscure blogger like me can digest it in under an hour.

But it seems the actual “reporters”—even the ones who cite the source of the 41% figure–don’t analyze the report beyond such generalities as: The results are staggering..disturbing…alarming…

Yes, they are. A 41% lifetime suicide attempt rate is horrific, especially when compared to a 4.7% suicide rate for the US population as a whole, and a 10-20% rate for lesbian, gay, bisexual people (these numbers are according to the authors of the survey). What, exactly, does the survey tell us about attempted suicide in the gender nonconforming (GNC) and trans community?  What is causing this high rate of suicidality?  As with most things, the devil is in the details.

I will not attempt to cover all aspects of the Williams Institute analysis in this post, but will highlight a few of the more interesting nuggets of information I gleaned;  in particular, weaknesses and findings that have not been addressed in other accounts I’ve read.

  • The authors note that the survey was flawed because only one binary, Yes/No question was asked: “Have you ever attempted suicide?” More careful and rigorous studies always follow up with in-person interviews, and when self-harming behaviors (not intended to end life) are controlled for, the actual suicide attempt rate is typically halved—meaning the suicide attempt rate could be as low as 20%.
  • The highest suicide attempt rate of all–60+%–was GNC and trans people who self-report a mental disability. No big surprise there; it’s well known that having certain mental conditions is a risk factor for suicidality. But by the authors’ own admission, the survey made no effort to ask for further details about these mental health issues. The status of having a mental condition was self reported, with no corroboration from medical records or a provider. Nor was there any attempt to discover whether the actual rate of mental illness was objectively higher (via diagnosis by a mental health provider) than reported by the subjects.
  • People who had either sought or received transition-related services had a higher suicide attempt rate than people who have not. And the survey did not ask whether suicide attempts occurred before or after services were sought or received.
  • The data suggest that natal females seem not to be helped at all, in terms of self harm, by being either “stealth” trans or passing as male.  (This is the opposite finding from that of natal males.)

Right from the get-go on page 3, under Methods and Limitations, the authors acknowledge the fundamental flaws in the survey. They urge caution in interpreting their findings:

First, the…questionnaire included only a single item about suicidal behavior that asked, “Have you ever attempted suicide?” with dichotomized responses of Yes/No. Researchers have found that using this question alone in surveys can inflate the percentage of affirmative responses, since some respondents may use it to communicate self-harm behavior that is not a “suicide attempt,” such as seriously considering suicide, planning for suicide, or engaging in self-harm behavior without the intent to die …The National Comorbity Survey, a nationally representative survey, found that probing for intent to die through in-person interviews reduced the prevalence of lifetime suicide attempts from 4.6 percent to 2.7 percent of the adult sample … Without such probes, we were unable to determine the extent to which the 41 percent of NTDS participants who reported ever attempting suicide may overestimate the actual prevalence of attempts … In addition, the analysis was limited due to a lack of follow-up questions asked of respondents who reported having attempted suicide about such things as age and transgender/gender non-conforming status at the time of the attempt.

We could stop right here and say the survey’s main data point–the 41%–is worthless. If, in general population studies, it has been shown that, without followup questions, the rate of actual suicide attempts could be artificially inflated to nearly double, the real rate for GNC/trans people could be closer to 20%. In addition, the authors point out that, without some sense of when the self harm took place, there is no way to determine whether identifying as gender nonconforming or transgender was the key factor in the self-harming behavior.

But let’s not stop there. Even if the rate is closer to 20%, that is still unacceptably high. And self harm is a huge problem, whether the actual intent to end one’s life is present or not.

Second, the survey did not directly explore mental health status and history, which have been identified as important risk factors for both attempted and completed suicide in the general population. Further, research has shown that the impact of adverse life events, such as being attacked or raped, is most severe among people with co-existing mood, anxiety and other mental disordersThe lack of systematic mental health information in the NTDS data significantly limited our ability to identify the pathways to suicidal behavior among the respondents.

In other words, the survey is seriously flawed because there is no reliable information about the actual mental health status of the participants; and, since mental health problems are a known self-harm risk, there is no way to accurately figure out whether the suicide attempt rate is as high as it is due to co-occurring mental illness–not necessarily because of  “gender dysphoria.” Further, another high risk factor for suicidality is being physically or sexually assaulted, especially for people with mental health disorders.

So the authors tell us two things: the 41% figure should be interpreted with great caution; and the causes of the elevated self harm/suicide attempt rate (whatever that rate actually is)–the “pathways”–cannot be reliably determined.

Williams Table 12

What about people who have contemplated or received medical transition services? The survey tabulated everything from transition-related counseling to bottom surgery:

Respondents who said they had received transition related health care or wanted to have it someday were more likely to report having attempted suicide than those who said they did not want it. This pattern was observed across all transition-related services and procedures that were explored in the NTDS.

Williams Inst suicide table 5

People who said they “did not want” these services had a lower self-harm rate. Does medical transition, or seeking transition services, decrease or increase suicide attempt rates? We don’t know, because the survey didn’t ask respondents if the self harm occurred before or after such services were soughtas the authors note:

The survey did not provide information about the timing of reported suicide attempts in relation to receiving transition-related health care, which precluded investigation of transition-related explanations for these patterns.

This is very important: I have most often seen the 41% rate mentioned, with no caveats or analysis, to justify young people receiving medical transition services. It’s clear from the authors’ own words that this survey cannot be responsibly used as a basis for the presumption that medical transition reduces self harming behaviors over the lifespan.

And now to one of the more interesting findings in the Williams Institute report:  natal females (in contrast to natal males) who say other people generally don’t recognize them as trans or GNC have the same or higher suicide attempt rate as females who are more often recognized by others.

Williams Table 7

Trans men (FTM) were found to have the same prevalence of lifetime suicide attempts (46%) regardless of whether they thought others can tell they are transgender. … for respondents in the last two gender identity categories – female-assigned cross-dressers and gender non-conforming/genderqueer people assigned female at birth – the prevalence of lifetime suicide attempts was found to be higher among those who said other people “occasionally” or “never” can tell they are transgender or gender non-conforming, compared to those who said that other people “always,” “most of the time,” or “sometimes” can tell. 

And later–buried  in the Executive Summary, we find this:

Importantly, our analyses suggest that the protective effect of non-recognition is especially significant for those on the trans feminine spectrum. For people on the trans masculine spectrum, however, our data suggest that this protective effect may not exist or, in some cases, may work in the opposite direction.

What does it mean to “not be recognized” as transgender or gender nonconforming? It could be one of two things: these natal females “pass” as male, or they are secretly gender nonconforming, perhaps cross dressing at home, in private. But in either case, being stealth or passing doesn’t seem to alleviate the urge to self harm.

I am going to guess that, for at least some of the natal females who answered this survey question, they did interpret  “people can’t tell”  to mean that they usually “pass” as male.  So for at least some of these females,  being perceived as male didn’t help them.  And FTMs, by all accounts, “pass” better than MTFs. Why wouldn’t passing relieve the distress for these female-born people? What is causing the misery in girls and women who are GNC or trans-identified?

If self harm risk remains elevated for many young women, whether they “pass” or not, wouldn’t a more compassionate and prudent approach be to help them–and their families–accept themselves as females who simply don’t fit societal gender norms? Many of these girls, prior to transition, live a lesbian lifestyle (even if they reject the label “lesbian”). How much kinder would it be to help them embrace the only bodies they will ever have, with the sexual preference they have, instead of endorsing extreme interventions that may never resolve their dysphoria?

Elsewhere in the survey, we learn that lack or loss of family support is a big factor in self-harm risk. This seems like a no-brainer. But support for what? Accepting a loved one’s gender expression or identity is not the same as getting on board with hormones and surgery. In fact, by encouraging the idea that they must medically transition (which entails lifelong and sometimes painful interventions) to be happy, might family and gender specialists even be increasing the risk of self harm?

It’s obvious that teens who are gender nonconforming are bullied and rejected because they don’t fit into the stereotypical boy or girl box–however they subjectively identify.  And like all kids, they just want to be accepted. Listen to Ash Haffner, the 16-year-old from Charlotte, North Carolina who died (as Joshua/Leelah Alcorn did) by bolting in front of a moving vehicle in February 2015, writing this days before her death:

if I die…I don’t want to be remembered as the faggot gay girl with all the scars on her arm. unfortunately thats who I am to alot of people. if those people would have just stayed silent and kept their ignorant thoughts in their heads then maybe i wouldn’t have those scars on my arm. maybe. it wasn’t always about what they had in their heads, it was what was inside of mine to. i just didn’t understand why i felt the way i did when i had a decent life. i may have come from a broken family but i always had a roof over my head and a loving mother who fully accepted me for who i was and never stopped trying. she was the only person who never gave up hope on me. but anyway, i don’t want to be remembered as the girl with problems, just remember me as someone who understood and stayed strong for as long as i could.”

Ash’s mother, who, according to media accounts, accepted her child as whatever gender Ash preferred, said:

“She was trying to figure out her identity,” Quick said. “She felt like a boy trapped in a girl’s body. She was caught somewhat in between. People weren’t really giving her the time to figure herself out. … All she wanted was for people to just accept her. Ash started enduring the most bullying when she cut her hair short.”

Ash’s “gender nonconformity”–her short hair, for crying out loud–is what caused the increase in bullying. Isn’t our challenge, as parents, as therapists, as a society, to  support our young people when they step outside stereotyped gender norms? To allow a girl to have a crew cut or wear boxer shorts? For a boy to wear a dress if he wants?


The Williams Institute analysis raises many more questions than it answers. It seems clear that being–or, more precisely, identifying as–some flavor of gender nonconforming or trans is correlated with a high rate of self harming behaviors. Mental health problems, coupled with a history of physical and/or sexual abuse or trauma, are associated with the highest risk of self harm. But judging by the evidence, gender specialists don’t seem to be taking those key risk factors into account when prescribing “transition” as an answer.

Whether the majority of these individuals who have self-harmed will ultimately benefit from medical transition is unknown (and for young people, this will not be known for years, if not decades), but there is absolutely nothing in the Williams survey analysis to indicate that medical transition will decrease suicidal intent or self harm.

In the words of the survey authors themselves:

Well-designed studies that specifically engage the transgender community will continue to be needed to identify and illuminate the health and mental health needs of transgender people, including access to appropriate health care services.

How about including the following things in “appropriate health services” for gender dysphoric young women?

  • family therapy aimed at helping parents and young women come to terms with being “gender nonconforming” women
  • evaluation and therapy for underlying mental health issues apart from gender dysphoria
  • strong female role models for girls and young women that don’t entail conforming to porn star chic
  • support for and acceptance of lesbian identity– especially for girls who don’t look like the gender-conforming, makeup-wearing “lesbians” on the “L” Word

Given the flawed data available to us, the leap in logic to assume the only viable choice is to medically transition or die ought to shame any provider, researcher, or journalist worth their salt. The Williams Institute data, if looked at honestly, should instead spur providers to offer effective psychological health evaluation and treatment for both young people and their families, and the least invasive intervention possible.

 

 

102 thoughts on “The 41% trans suicide attempt rate: A tale of flawed data and lazy journalists

  1. Pingback: When is a girl not a girl? -Lily Maynard

  2. Add two more parents to “the concerned.” But HOW do we tell this to our child without risking the worse or being cancelled?

  3. Pingback: TMI: Genderqueer 11-year-olds can’t handle too much info about sterilizing treatments–but do get on with those treatments | 4thWaveNow

  4. Pingback: My first article- "A Mum's Voyage Through Transtopia" |Lily Maynard

  5. Pingback: Fox News Drinks Gay Pride Kool-Aid with Propaganda-Filled Trans Family Story – Uncaged Lion

  6. Pingback: Fox News Promotes Gender Ideology with Propaganda-Filled 'Trans' Family Story – Daily Citizen – Anzhel

  7. Pingback: Ni marginalisés, ni maltraités, ni vulnérables – TERF EST UNE INSULTE

  8. Pingback: Kjønnsskifte på barn, den medisinske skandalen media tier om – Samlivsrevolusjonen

  9. Pingback: Kevin’s Assortment of Gender Critical Links | Kevin Bryan Hash

  10. I’m more of your inclination than not, but I would ask you to be a little critical of your own biases here. I went and read the excellent Williams Institute summary you linked, and this is the very first paragraph in its “Discussion” (conclusion), using capitalization in place of bold text to add emphasis:
    “The most STRIKING finding of our analysis was the
    EXCEPTIONALLY HIGH PREVALENCE OF LIFETIME SUICIDE ATTEMPTS reported by ntds respondents. In looking
    at the percentages reporting a lifetime attempt within
    various subgroups of the overall sample, we repeatedly
    found “lows” in the range of 30 to 40 percent, while the
    “highs” exceeded 50 or even 60 percent. EVEN TAKING INTO CONSIDERATION that SOME DEGREE OF OVER-REPORTING likely occurred in the survey, the results suggest these
    transgender and gender non-conforming respondents
    have experienced EXCEPTIONALLY HIGH LEVELS of SUICIDALITY.”

    Meanwhile, from your own writing, using that source as your citation:
    “But it seems the actual “reporters”—even the ones who cite the source of the 41% figure–don’t analyze the report beyond such generalities as: The results are staggering..disturbing…alarming…”

    That’s not fair OR accurate. Wouldn’t you agree “striking”, especially paired with “exceptionally high levels of suicidality”, is a legitimate synonym of “staggering”, “disturbing”, AND “alarming”?

    Again, and genuinely, I am fairly aligned with you, but I’ve found that I need to constantly battle my own tendency to react strongly against the dominant misinformation in the narrative today. That is, when I’m not careful, I find it natural to be a reactionary who loses sight of the truths present on the other side of the line. And I’m trying to share that perspective with you, someone who has been more personally impacted by all of this, only out of respect. I hope you have a nice day.

    • Did you read the whole article? Suicidality is not downplayed. Rather, there is a discussion of what may becontributing to it. To wit: “Even if the rate is closer to 20%, that is still unacceptably high. And self harm is a huge problem, whether the actual intent to end one’s life is present or not.”

      The article discusses the fact that concomitant factors are not addressed (mental health status, in particular), something even the original Williams Institute analysis acknowledges:

      “Second, the survey did not directly explore mental health status and history, which have been identified as important risk factors for both attempted and completed suicide in the general population. Further, research has shown that the impact of adverse life events, such as being attacked or raped, is most severe among people with co-existing mood, anxiety and other mental disorders …The lack of systematic mental health information in the NTDS data significantly limited our ability to identify the pathways to suicidal behavior among the respondents.”

Leave a Reply to kaypasserCancel reply