With a side-order of lamentation over shoddy reporting.

This post was sparked by a piece in the NZ Herald this morning about the suicide of a young man in Auckland. This of course was a tragic event, and one that his mother is understandably having great trouble coming to terms with. However, the article’s sub head is appalling:

Eighteen months after Toran Henry killed himself, his mother finally gets to have her say on whether an anti-depressant had an influence on his death. Chris Barton looks at the case that has raised questions about the drug’s use

Excuse me? Exactly what bearing on the safety of a medication does the opinion of someone’s mother have? And “Chris Barton looks at the case that has raised questions about the drug’s use” … Ummmmm … no. This case is not “the case” that has raised questions; we’ve been talking about this issue for years now. This case is a tragic one of a young man taking his life, despite receiving treatment for depression.

Before going any further however, a disclaimer: I know nothing of the case other than what’s reported in the media. Though I imagine might well know the psychiatrist involved, having worked extensively in that area of Auckland, and have even worked at the service named when I was a trainee some 12 years ago, I do not know who the psychiatrist is, and no longer work in that particular health service – let alone that team. – Hell, I’m not even in the same country any more.

The points I want to make in this post are really two: first, around the issue of antidepressant medications and suicide; and second, the bad reporting evident in the linked piece from the NZ Herald.

Antidepressants and suicide

Depression is, potentially, a fatal illness. We would like it to not be, but sadly it is – at times despite our best efforts. Another wrinkle is that as I discussed in my very first Midweek Medicine post, suicide is not always due to mental illness. Thus attributing cause for many suicides is complex, and predicting or preventing any particular suicide very difficult. That said, what part do antidepressants play in this story?

If someone is suicidal because they are depressed, the best way to stop them being suicidal is – no surprise – get rid of the depression. That is one of the reasons we use antidepressant treatments – which include various medications. There is contention around antidepressant medications in younger people, on two fronts: firstly there’s a question as to whether they work as well for adolescents as they do for adults; and secondly there’s the issue of suicidality potentially increased by the medication itself.

Efficacy

We establish efficacy of a treatment by means of clinical trials, preferably randomised controlled trials, which eliminate as many confounding variables and sources of bias as possible, leaving us with just the effect of the active substance. Most trials of most treatments in medicine involve adults. Antidepressants are no exception. In addition, younger people presenting with psychiatric illness often don’t present as clearly as adults, and with everything that goes on psychologically during adolescence it can be difficult sometimes to be completely sure of what is or isn’t going on. If, as I mused recently, the recent suggestions that antidepressants are only effective in severe depression is something to do with biological illness versus psychological distress (with severity acting as a proxy for biological illness – though undoubtedly less than 100% accurate), then it would seem quite plausible that with the tumult of adolescence and the atypical presentations, a lot of adolescent depression might not represent true biological illness such as melancholia or bipolar disorder. Consequently we might expect (though this is still quite firmly in the realm of supposition, I hasten to add) that antidepressants wouldn’t be quite as effective – compared to placebo – as they might be in adults.

- And that said, there actually is evidence of efficacy; it’s just perhaps less clear-cut, and you might often try cognitive therapy instead, for example. However, I will note that I’ve sometimes wanted to go down that path, referred a young person to an adolescent psychiatry service, and the next thing I hear they have been started on antidepressants – and they’re the adolescent experts, not me – I deal more with adults. Anyway ….

Suicidality

Back in the late 90s, as a trainee, I went looking on the Internet for what was said about antidepressants and suicide or homicide. I also looked through the psychiatric literature. There was a marked disjunction: teh intertoobz sed antidepressants (well, most especially fluoxetine and paroxetine) caused suicides and homicides. David Healy’s name came up an awful lot. He was (and it appears, still is) the go-to guy whenever there’s a court case involving suicide or homicide and bright spark wheels out the fact that the person was taking an antidepressant. By contrast, the studies looking at this found no clear relationship. There was even basic biochemical stuff suggesting the opposite: low serotonin found in the cerebro-spinal fluid of people who had suicided.

Fast-forward some years, and it became clear that in fact there could be something to it; the Herald article notes: “… analysis of adult placebo controlled drug trials by the United States Food and Drug Administration (FDA) in 2006 which show a 2.3 fold increased rate of suicidal behaviours in people aged 18-24 years.” Yes, there is a black box warning in the USA on paroxetine. Yes, it does appear that at least some antidepressants can increase suicidal thinking, and suicidal behaviour. The evidence does not show, however, that the rate of suicide is increased. Here’s some NZ data:

- from the NZ Ministry of Health’s suicide statistics

Overall, the total population three-year moving average suicide rate peaked in 1927–1929 (18.5 deaths per 100,000 population). After that point, the suicide rate dropped and then stabilised, with slight fluctuations, until 1971–1973 (10.2 deaths per 100,000 population). After 1971–1973, the suicide rate increased again, reaching another peak in 1996–1998 (16.7 deaths per 100,000 population). After this point, the suicide rate declined up until the most recent period, 2001–2003 (14.2 deaths per 100,000 population) by 15.0 percent.

Note: The first antidepressants were introduced in the 1950s. There was no increase in the suicide rate (there wasn’t a decrease either, but that’s grist for another mill). The suicide rate began to increase in the early 70s. Fluoxetine, the first SSRI wasn’t licensed by the FDA until 1987. If anything, the suicide rate plateaued then, but more likely that’s just a blip in an ongoing increase that continued until the late 1990s – since which time the rate has decreased (“despite” all the SSRIs we’re using). In other words, antidepressant medications do not appear to affect the overall suicide rate.

It’s worth putting this graph in as well:

Note that the female suicide rate essentially hasn’t changed, whereas the course of the male rate is for all intents and purposes identical to that of the overall rate. This is despite the fact that women are more likely to be diagnosed with, and treated for depression. This lends more weight to the contention that antidepressant medications do not increase the suicide rate – despite carrying a risk of increasing thoughts of suicide.

Is it new?

Ok, but they can increase suicidal thinking and behaviour, right? Probably yes. And that’s something new, right? Well actually … no. Old wisdom was to watch a patient really closely during the first couple of weeks of antidepressant treatment, because of a possible increase in suicidality. So this is not new. David Healy testifying at every court case he can, and writing 20 books and so on – that’s new. Threatening to file  private manslaughter prosecution against a psychiatrist who tried to help your son – that’s new. – And very sad.

In summary: suicide is complex. Depression can be complex – and moreso in adolescents. Antidepressants can be effective, but exactly who they’re most useful for is sometime a bit murky. They have their problems – and that does include, at least sometimes, and it seems particularly in the young, an increase in suicidal thinking (at least early on in treatment). They are associated with an increase in suicidal thinking and behaviour, yes, but not completed suicide.

So, now let’s turn to …

Reporting

First, I must get off my chest the horrible horrible writing style in this article. Yes, in headlines one uses the present tense even when describing past events, but for an entire article? Spanning some 18 months or more?? And flicking into past tense occasionally, then back to present tense even within the same sentence or two? Owwwwch! It is simply painful to read, and very hard to follow.

Right, that dealt with …

There is nothing advanced by this piece of writing. The young man’s mother is portrayed simply as fighting for “the truth” about her son (though at least they didn’t – overtly – attack or demonise the psychiatric service), and the writer has clearly not read about antidepressants and suicide – or has been incapable of grasping the timeline (which would go with the dreadful changing of past/present tenses in the article) or the nuance of the relationship between antidepressants, suicide, and suicidal thoughts or behaviour. All it does is to potentially raise in readers unwarranted fears about antidepressants, and point them towards Dr. David Healy, who is a different sort of “brave maverick doctor”: a voice in the wilderness crying out against psychiatry’s use of a terrible treatment. He’s mentioned four times in the article, and there really is no counterpoint mentioned, other than the health service (who must be bad anyway, right?).

Apparently, responsible and thoughtful journalism is to much to expect – but hope springs eternal.