Wednesday, February 27, 2008

Do Teens Commit Suicide More Often?

According to stereotype, yes. But not according to the data.

I’ve finally returned to my old suicide project, which had been moldering in a (virtual) drawer for quite a long time. With the help of a new co-author, I’ve been analyzing the results of a bunch of regressions. I was surprised to note that the variable for teenagers (percent of population from 10 to 19) consistently yielded negative coefficients, meaning higher teen population is associated with lower suicide rates. So I went back to the original data source (imagine that!) and found that the stereotype is dead wrong: suicide rates are notably lower for teenagers than adults.


Suicide rates do rise throughout the teen years, but they plateau at about age 20 and remain flat throughout the years 20 to 65. Then they jump again for the 65+ demographic.

So what explains the misperception? I glanced over the web resources on teen suicide, and by the numbers they technically do not seem to err. They note that suicides rates rise during the teen years, which is true, but it doesn’t indicate a particular problem among teens, because (as the graph shows) suicide rates remain high for adults. They also note that suicide constitutes a large percent of deaths among teens, but this is not terribly surprising, since teens are less likely to die of things like heart disease and cancer.

ADDENDUM: In case you're wondering, teen suicide rates have not been rising, either. They've been in decline since the late 1980s. (For this graph, I had to go beyond the years 1981-1998, which WISQARS maintains separately.)


[UPDATE: Graphic mysteriously deleted by Blogger now restored.]

12 comments:

KipEsquire said...

Any discussion about teen suicide needs to take sexual orientation into account. The academic literature is teeming with studies showing that gay teens, especially males, are far more likely to ideate about suicide, perhaps by a factor of ten, and also more likely to attempt suicide.

One would like to think that the decline in teen suicide rates that you mention is at least partially due to the gay rights movement and our society's increasing intolerance toward the intolerant.

Bob Murphy said...

Glen,

Do you have data on suicide attempts? I bet that would boost the relative frequency among teenagers, though it might not validate the stereotype.

Larry Sheldon said...

Regarding attempts....

It seems to me that there is great risk of being mislead in interpreting the number of attempts.

I have not thought or read about the subject in quite a while, but I have the sense that a lot (most?) of the "attempts" are more likely extreme pleas for help or attention and involve self-inflicted injuries that are not likely to be fatal.

But I should not be confused with somebody that actually knows what he talking about.

2008bill said...

You have only touched the surface of FALSE statistics that mindless people love to hear and believe. This is what Michael Males (youth today.org) has to say:



Every media story on youth is mindlessly, numbingly the same. There’s no sense of news editor or reporter initiative, responsibility, professionalism, fairness, interest in accuracy, or even avoiding the most primitive insanity and hostile exploitation. YouthFacts' fact-checking of media stories reveals a repeated series of indentical falsehoods recycled over and over.

This week’s People Magazine exploitative spread on “teen pregnancy” (1/14/08), ION Television’s pure-junk “Binge” feature (1/05/08), CBS News' interest-dictated "teen sex" report (1/6/08) that all hyped distortions while evading every important issue, and continued media fiction on “teens and drugs,” “underage drinking,” “youth violence,” “teen suicide,” “teen sex,” the formulaic litany of myths recycled over and over, all purport to be “news,” “shocking,” “the latest truth.” They are none of these.

These stories repeat the same reports done 15, 10, five, four, three, and two years ago, and one year and six months and three months and six weeks ago, in cloned redundancy. They rely for “facts” on self-interested sources that gain publicity and bucks from milking lurid, fabricated fears of youth. They ignore all social issues such as poverty and family abuses that crucially affect youth behavior. They butcher science, make up statistics, generalize from rare events, bully teens, and sensationalize cultural trivia. They allow adults, led by the blatantly non-objective reporter, to smugly moralize while evading discussion of any issue that might trouble their middle-aged viewers and sponsors.


Read much more on his website: http://www.youthfacts.org/media-final.html

EC said...

I cant deny to admit that this graphs seem oddly familiar, 309?

Glen Whitman said...

Hi, Enrique. Actually, I just made those graphs last week, so I couldn't have shown them in 309. But I did give you some suicide data to work with (from just one year, 1996 as I recall).

Anonymous said...

So what explains the misperception?

Perhaps it is because of two things:

1. A child's or teen's is generally perceived as more tragic than an adult's death. People feel that a child has unlimited potential, and they consider children "innocent." Whereas adults are rarely considered in this way.

2. People make assumptions based on their emotions (thanks to overly sensationalized media reporting) and fail to consider scientific data. That is if they are ever aware of any data in the first place.

I'm just supposin', really...

Anonymous said...

That little bump in suicide you see in the graph in 2004 is coicindentally occuring during the same year that the FDA put the black box suicide warning all the antidepressants in the U.S. The result of that action is that there was a huge drop in physicians prescribing antidepressants out of concern over lawsuits. Noone has established that the increase in suicide rate was caused by the change (decrease) in prescribing practices, but the correlation is strong.

Anonymous said...

As a physicist I have had to design numeous tests on various products. I won't go into the whole design statistics but maximum sensitivity to detect a failure normally happens at a 50% failure point or at a max rate of change of failure. The data shows a max slope and an aproximate 50% point for suicide rate within the 15-19 group. Might I suggest that with this degree of sensitivity any little variable could twang the data to prove anything. Now that you have demonstrated a high sensitivity test group, it requires that both side of any opinion very carefully define all the variables they can to test variations.

Glen Whitman said...

Thomas -- I'm not sure I understand your point. Were you responding to the anonymous commenter who made the statement about antidepressants? If so, then I agree; his hypothesis might be correct, but it would require more testing.

If you were commenting on my original post, I'll just observe that I was using total population statistics, not sample statistics. So the figures definitely show that teens have lower suicide rates than adults and seniors in the U.S. (unless you don't trust the data collection). I didn't posit any hypothesis as to why those rates are lower.

Anonymous said...

I've never heard that teens commit suicide more than older people. I thought it was well known that people become more suicidal as they age, especially among the elderly. Growing olds sucks. No surprise there.

The data I've been looking at isn't comparing the present teen demographic to the present adult graphic, but the present teen demographic to the past teen demographics and the present adult demographic to the past adult demographic. It's the generational cohort comparisons that are so striking.

The point the data shows is that teens are having increasing rates of suicide across the generations. Also, this is being carried over as these demographics age. For example, GenX had a sharp increase in teen suicides and now as they enter middle age that demographic is showing a sharp increase.

Glen Whitman said...

Benjamin, it seems like your point would be addressed by my second graph, the one in the addendum, which focuses on the suicide rate for ages 10-19. The graph is a few years out of date now, but what it shows is (a) that teen suicide rates did jump in the 1980s, and (b) that they've basically been in decline ever since. This seems to argue against the claim that "teens are having increasing rates of suicide across the generations." But maybe you have a longer time period in mind.