Friday, 3 March 2017

Lies, Damned Lies, and Suicide Statistics

     Ages ago, when I submitted my scientific studies for publication, they were in turn submitted to a couple of anonymous reviewers who, I presume, had undertaken studies in a similar field, and they made recommendations - not all appreciated - on how my data could be presented, and their possible implications. Likewise, when I was asked to review someone else's paper, I also made recommendations - which I hope were appreciated. That is how peer review works. So how could a prestigious journal publish a paper whose data make no sense whatsoever?

    Take, for instance, the recent paper by Raifman, Moscoe, Austin, and McConnell, entitled
Difference-in-Differences Analysis of the Association Between State Same-Sex Marriage Policies and Adolescent Suicide Attempts,
 which was widely touted as proving that the introduction of same-sex "marriage" (SSM) in America reduced adolescent suicide by 7% overall, and 14% among homosexuals and lesbians. Right away, an alarm bell rang. How could even a large reduction among a very small minority result in such a large overall reduction? It was time to examine the document in detail.
     The first thing to note is that the study was not about suicide as such. The authors eschewed those statistics because they were rare, often underreported, and also for the obvious reason that it is hard to discover a person's sexual orientation once he is dead. Instead, they focused on attempted suicides - which is a different matter entirely, as we shall see. But first, the introduction:
Among the 762 678 students (mean [SD] age, 16.0 [1.2] years; 366 063 males and 396 615 females) who participated in the YRBSS between 1999 and 2015, a weighted 8.6% of all high school students and 28.5% of 231 413 students who identified as sexual minorities reported suicide attempts before implementation of same-sex marriage policies.
     The wording is a bit misleading. The figure of 231,413 [30%] is not the number of sexual minorities, but the number who were asked about their sexuality. The true (?) figure was one in 8, or 12.7%, consisting of 2.3% homosexuals or lesbians and 6.4% bisexuals, with the final figure made up by adding the 4.0% who were unsure - though why they should be included, or why they should be bothered about SSM is far from clear.
     The first thing to note is that this is miles higher than the figures for adults, especially in the "bisexual" category. But who knows? The famous National Longitudinal Survey of Adolescent Health in the mid-1990s found that, of those who were wholly or predominantly homo at age 16, 72% of the males and 55% of girls were wholly or predominantly hetero at 22. It seems to me that this new study is a two-edged sword. If it can be cited as proving that SSM reduces the risk of suicide, it also confirms that same-sex attraction is more common in high school students, but for most it is a phase they outgrow.
     Be that as it may, in the current paper they examine the answers to the question, “During the past 12 months, how many times did you actually attempt suicide?” This was a question which was said to have a high test-retest reliability ie when they asked it again later on they got the same reply. Just the same, I had to read and reread the results to confirm they really said what they appeared to say:
Evidence from nationally representative 2015 Youth Risk Behavior Surveillance System (YRBSS) data indicates that more than 29% of gay, lesbian, and bisexual high school students reported attempting suicide within the past 12 months, relative to 6% of heterosexual students.
     Hit the pause button, and freeze on the lowest figure. 6% is nearly one in 16. Does that sound right? Is it conceivable that one in 16 ordinary, heterosexual teenagers not only want to, but actually try to, kill themselves every year? If that's the case, why aren't there bodies everywhere? Why aren't hospitals full of self-harming adolescents and counselors overwhelmed by disturbed high school pupils not yet at the suicidal stage? And just what is so terrible about growing up in America that at least one in 16 teenagers try to do away with themselves every year? If this has been a consistent finding over 17 years (and in earlier years it was said to be even higher), why haven't the authorities gone into panic mode and put into place prevention projects or, alternatively, seriously reconsidered their methodology?
     The overall suicide rate in the United States is 13.26 per 100,000 per year, but for adolescents it is much lower. To quote Raifman again,
[T]he mean annual suicide rate of 5.3 per 100 000 adolescents between the ages of 14 and 17 years between 1999 and 2014 is small relative to suicide attempts...
      That's putting it mildly. From this statistic it is easy to calculate the following:
  • 5.3 per 100,000 for 762,678 equates to just 40 suicides in 17 years. Rather than proving the SSM "saves lives", the figures are too small to be statistically significant.
  • The American Foundation for Suicide Prevention estimates that the ratio of attempted to actual suicide is about 4:1 for the elderly and 25:1 for the young. However, 6 per cent is 6 per 100, or 6,000 per 100,000. Applying the figure of 5.3 produces a minimum of 1,132 attempts for every successful suicide. These kids aren't trying very hard!
     Personally, I don't know what the 6%, or 8.6%, or 29% stand for, but I am quite sure they do not represent suicide attempts, or have any relationship to real suicides. They make no sense whatsoever.

    Where do the figures come from?
     They come from the Youth Risk Behavior Surveillance System or YRBSS, a series of surveys every two years of risky behaviour by teenagers. The surveys involved samples of variable size, sometimes of whole states, and sometimes of particular high density urban areas - useful for producing snapshots at the particular time, and establishing orders of magnitude, but not for comparing apples with apples. Only half the states enquired about sexual orientation, and most of them for only a few years prior to SSM introduction.
     What you won't find among the plethora of statistics is anything resembling the raw data. I would like to have seen a table of the actual figures according to state and year. They must have had something like that in order to do all their calculations. On the other hand, if they seriously reveal one in 16 attempting suicide, perhaps they aren't worth much anyway.

     Is it plausible?
    My guess is that the paper passed the peer review uncritically because it followed The Narrative, that most of the problems homosexuals and lesbians face are due to social stigma. It is superficially attractive. Being a member of a despised minority, having "the love that dare not speak its name" would be expected to be detrimental to your mental health, perhaps tempting you to end it all. But looking closely, however, we note that most homosexuals and lesbians do not get "married" when the law allows it and, in any case, these children are not in the age group considering marriage. Also, SSM is usually introduced in the wake of various social changes. Are we to assume that its introduction suddenly improves the overall atmosphere so much that these troubled teens suddenly find a reason to keep living after all?
     Perhaps we should examine the situation in those countries where it has been established the longest. Take, for example, this report from the Netherlands in 2011:
Research published by the government’s social policy unit SCP last year showed that 12% of young gay teenagers had attempted to commit suicide. That research also showed 9% of the population still have serious objections to homosexuality, one in five people don’t think gay people should be allowed to adopt children and one in 10 thinks same sex marriage should be abolished.
     Although it focused on violence against homosexuals, and tried to put a negative spin on the situation, it still remains that 91% did not have "serious objections" to homosexuality, and 90% were OK with SSM. The Netherlands must be the gay-friendliest country in the world, yet still had a 12% rate of attempted youth suicide among them. It is similar in other jurisdictions:
Along this line, a 2016 study published in the European Journal of Epidemiology examined how the hard-fought social affirmation that legalizing gay marriage was promised to provide impacted the well-being of Swedish same-sex married couples. They have suicide rates nearly three times that of their opposite-sex married peers. The authors caution their numbers are likely an underestimation. A similar Danish study found that same-sex married women suffered a six-fold increase in suicide risk over their opposite-sex married peers, and same-sex partnered men an eight-fold increase risk.
     Finally, there is one group of people who are persecuted far worse than homosexuals, so much so that they dare reveal their true selves to only a few confederates. These people are called "criminals". I don't know what the suicide or attempted suicide rate is among crooks, but I have never heard it suggested that they are killing themselves because other people hate them.