Discrimination and Gay Suicide Rates

In this article I will argue that discrimination is at best a very partial explanation of why it is that gay people have higher rates of suicide than do straight people.

To begin with, it’s worth noting that most acts which constitute discrimination, ostracizing someone, insulting someone, emotionally or physically hurting someone, etc., are essentially forms of bullying. This fact matters because even though there is a substantial correlation between bullying and suicide rates the causal relationship between bullying and suicide is much weaker than this correlation suggests. To be specific, a one standard deviation increase in bullying predicts a 68% increase in the odds of someone attempting suicide. However, if you use a twin study design to control for genes and the family environment, and also control for psychopathology that pre-existed the bullying, we find that a one standard deviation increase in bullying only predicts a 25% increase in the odds of attempting suicide (Oreilly et al., 2021). Thus, only roughly a third (37%) of the commonly noted association between bullying and suicide is plausibly causal.

As we will see below, sexual minorities attempt suicide at a rate that is at least three times the rate at which heterosexuals attempt suicide. Given the weak causal impact that bullying has on suicide, for bullying to explain the elevated LGB suicide rate the average LGB person would need to be bullied at a level that fewer than one in one hundred thousand heterosexuals are. This is obviously completely implausible.

In fact, a meta-analysis of 15 studies found the vast majority (74%) of LGBT youth (age: 12 – 25) who have attempted suicide answer “no” when asked if they’ve personally been the victim of various sorts of anti-LGBT discrimination and bullying (Williams et al., 2021). And of course even for the 26% that did report having been victimized we can’t assume that this victimization caused their suicide attempt since, as already seen, the relationship between bullying and suicide is mostly non-causal.

That discrimination does not completely explain gay suicide rates is also evidenced by several studies finding that sexual orientation continues to be a risk factor for suicide after self reported victimization is controlled for.

For instance,  Garofalo et al. (1999) found that homosexuality continues to be a risk factor for attempting suicide after controlling for drug use and a scale measuring the degree to which one has been the victim of violence.

Similarly, Shields et al. (2012) analyzed data on 2,154 students attending high-school in San Francisco and found that LGB sexual orientation was associated with a 2.6 times elevated risk for suicide even after controlling for self reported victimization, race, gender, and drug use. Finally, Ballard et al. (2016) found that LGB status significantly predicted drug use and suicide risk even after controlling for self reported victimization in a large sample of American gay youth living in Appalachia.

It is also worth noting that the suicide rate of homosexuals is far greater than average even when only looking at those homosexuals who claim to have not significantly been victimized by discrimination. For instance, Ferlatte et al. (2015) analyzed data on 8,382 gay Canadian men and asked their participants about experiences of bullying, sexual violence, verbal abuse, physical violence, and discrimination at work. While experiences of marginalization correlated with suicidal behavior, the rate of suicidal ideation among those who reported having experienced no discrimination was roughly twice than of heterosexual Canadian men.

Similarly, Russell et al. (2010) found a very high rate of suicide attempt even among gay adolescents who self reported having experienced low levels of LGBT related school victimization.

Once again, I should add that the suicide rate of non-bullied LGB individuals should not be taken to be representative of what the LGB suicide rate would be if we got rid of discrimination because individuals who experience little bullying generally have better mental health than those who are bullied for reasons independent of bullying itself.

Another relevant observation is that the self reported happiness of homosexuals in the United States has slightly declined since the 1980s. Over the same time period we’ve seen a huge decrease in anti-gay discrimination and a large increase in gay legal rights, but all of this progress has seemingly done nothing for the psychological well being of homosexuals.

We reach a similar conclusion looking at suicide rates over time. Meyer et al. (2021) used data from a sample of 1,518 LBG American adults surveyed by Gallup. They divided their data into three age cohorts and found that the rate of life time suicide attempt had increased over time from 21% among those born in 1956-1963, to 24% among those born in 1974 – 1981, to 30% among those born in 1990-1997. The increase between the oldest and youngest cohort was statistically significant. This change is especially striking given that people born in the 50s have had far more time to accumulate a lifetime suicide attempt than have people born in the 90s.

In an analysis of data on 96,893 LGBT veterans, Lynch et al. (2020) found that the rate of suicide mortality had been fairly constant between 2006 and 2017 after going up and down in the early 2000’s.

Turning to data on adolescents, looking at data from Massachusetts, Liu et al. (2020) found that between 1995 and 2017 the trend in how the rate of suicide attempts was changing did not differ between LBG and heterosexual adolescents.

By contrast, using data from the states of Delaware, Illinois, Massachusetts, Maine, North Dakota, and Rhode Island, Raifman et al. (2020) found that the suicide attempt gap between LBG and hetero kids had decreased from a factor of 5.2 in 2009 to a factor of 3.8 in 2017. However, this change was not statistically significant. Moreover, the same paper finds that a different picture emerges if we define adolescents sexuality in terms of sexual contact rather than mere identity. Over this period, the gap in rates of past year suicide attempt between those who have and have not had sex with members of their own sex increased. In 2009, non-hetero behaving kids were 2.6 times as likely as hetero behaving kids to have attempted suicide. By 2017 this figure had increased to 3.3. However, this change was not statistically significant.

On the whole then, the evidence suggests that the gay-straight suicide gap has probably not narrowed over time while the absolute rate of suicide among gay people has either remained the same or increased.

The suicide rates of homosexuals in gay friendly areas also gives us reason to doubt that their mental problems are entirely a result of discrimination (Caputi et al., 2017; Shields et al., 2012; Bjokenstam et al., 2016; Ferlatte et al., 2015; Meyer et al., 2008; OMH., 2016; Bergen et al., 2013).

All this data suggests that, if anything, homosexuals in leftist nations and cities are at a higher relative risk for suicide than is the average homosexual in the United States. This may be surprising since bullying does have an effect, even if weak, on suicide. These findings can be reconciled by positing that the subcultures that gay people tend to establish in left wing communities are even more damaging for mental health than is being bullied, though this is just speculation.

Regardless, on the basis of the lines of evidence I’ve just reviewed it seems likely that discrimination either causally explains a modest fraction of the straight-LGB suicide gap or close to none of it.

As is well known, non-hetero sexualities has increased in popularity in recent years. The fact that suicide rates for non-hetero people are remaining so high even as we’ve created a world in which increasingly normal people identify as LGB and a culture which is generally accepting of homosexuality is cause for concern. It may be that abnormal sexualities, at least as they are currently being manifested in culture, cause defects in mental health. To the degree that this is plausible we should view this change in sexuality, at least in so far as it is real, as a problem. In the past we had a set of norms such that homosexuals were evidently no less happy than they are now but their numbers were not rapidly gaining. I’m naturally inclined towards a fairly liberal attitude about homosexuality, but the concerns I’ve just explained make me more sympathetic towards a more right wing approach.