return to a madness in our method

Posted on Tuesday 24 June 2014

This is the sixth in a series:

  1. a madness to our method…
  2. are you listening?…
  3. another campaign?…
  4. read me him…
  5. a madness to our method – a new introduction…
First, my apologies for the length and number of posts about this BMJ article. It was so heavily pushed in the press and so confusing that I thought it was worth the monotony to chase down. In my original shot at this article [a madness to our method…], I just couldn’t follow what they were saying about how they came up with their method to identify suicide attempts in the database. I’ve highlighted the part I found confusing in red below in the abstract…
by Christine Y Lu, Fang Zhang , Matthew D Lakoma analyst, Jeanne M Madden, Donna Rusinak, Robert B Penfold, Gregory Simon, Brian K Ahmedani, Gregory Clarke, Enid M Hunkeler, Beth Waitzfelder, Ashli Owen-Smith, Marsha A Raebel, Rebecca Rossom, Karen J Coleman, Laurel A Copeland, Stephen B Soumerai
British Medical Journal. 2014 348:g3596.

Objective To investigate if the widely publicized warnings in 2003 from the US Food and Drug Administration about a possible increased risk of suicidality with antidepressant use in young people were associated with changes in antidepressant use, suicide attempts, and completed suicides among young people.
Design Quasi-experimental study assessing changes in outcomes after the warnings, controlling for pre-existing trends.
Setting Automated healthcare claims data [2000-2010] derived from the virtual data warehouse of 11 health plans in the US Mental Health Research Network.
Participants Study cohorts included adolescents [around 1.1 million], young adults [around 1.4 million], and adults [around 5 million].
Main outcome measures Rates of antidepressant dispensings, psychotropic drug poisonings [a validated proxy for suicide attempts], and completed suicides.
Results Trends in antidepressant use and poisonings changed abruptly after the warnings. In the second year after the warnings, relative changes in antidepressant use were −31.0% [95% confidence interval −33.0% to −29.0%] among adolescents, −24.3% [−25.4% to −23.2%] among young adults, and −14.5% [−16.0% to −12.9%] among adults. These reflected absolute reductions of 696, 1216, and 1621 dispensings per 100 000 people among adolescents, young adults, and adults, respectively. Simultaneously, there were significant, relative increases in psychotropic drug poisonings in adolescents [21.7%, 95% confidence interval 4.9% to 38.5%] and young adults [33.7%, 26.9% to 40.4%] but not among adults [5.2%, −6.5% to 16.9%]. These reflected absolute increases of 2 and 4 poisonings per 100 000 people among adolescents and young adults, respectively [approximately 77 additional poisonings in our cohort of 2.5 million young people]. Completed suicides did not change for any age group.
Conclusions Safety warnings about antidepressants and widespread media coverage decreased antidepressant use, and there were simultaneous increases in suicide attempts among young people. It is essential to monitor and reduce possible unintended consequences of FDA warnings and media reporting.
And in the text of the paper…
Study cohorts and outcome measures
Because previous studies showed that rates of depression diagnosis changed after the warnings and that outpatient claims are often incomplete for mental health conditions such as depression, to avoid introducing selection bias, we did not limit our cohorts to those with a coded diagnosis of depression.

…To examine changes in suicide attempts after the warnings, we used the same denominator population as defined previously. While encounters for suicide attempts can be identified in administrative databases using external cause of injury codes [E-codes], they are known to be incompletely captured in commercial plan databases. Our preliminary analysis found that E-code completeness varied across study sites, treatment settings, and years. Therefore, instead of deliberate self harm E-codes, we used poisoning by psychotropic agents [international classification of diseases, ninth revision, clinical modification [ICD-9] code 969], a more reliable proxy for population level suicide attempts. Poisoning by drugs or toxic substances is the most common method of suicide attempt leading to hospital admission and emergency room treatments. 35 36 Non-fatal poisoning by psychotropic drugs [predominantly tranquilizers] has a positive predictive value of 79.7% for suicide attempts [sensitivity was 38.3% and specificity was 99.3%], outperforming other types of injuries or poisonings..
So I was visiting a sick friend  out of town. And in the spaces when he was resting up, I started chasing the paper’s references backwards [as in the last post]. I thought their studying the inconsistency of the E-codes was solid. And I thought Patrick et al did a passable job of trying to locate a surrogate, though it felt forced. But when I got to the part where Lu et al rejected Patrick’s algorithm based on some of those old Pharma propaganda articles from back when they were trying to reverse the Black Box Warnings, I felt like I’d found the madness:

  1. Decline in treatment of pediatric depression after FDA advisory on risk of suicidality with SSRIs.
    by Libby AM, Brent DA, Morrato EH, Orton HD, Allen R, Valuck RJ.
    American Journal of Psychiatry. 2007 164[6]:884-891.
    [full text online]
  2. Persisting decline in depression treatment after FDA warnings.
    Libby AM, Orton HD, Valuck RJ.
    Archives of General Psychiatry. 2009 66[6]:633-639.
    [full text online]
They were assuming that the increasing rate of diagnosis of pediatric depression and prescribing antidepressants pre-Warning was correct and the Black Box Warning put a damper on things. Hallelujah for the damper is all I have to say about that. 

At face value, the whole premise for the study is flawed. The SSRIs have only been shown to be effective in pediatric depression in Lilly’s earliest studies of Prozac, and in spite of their creative publications, the remainder have been ineffective. So, the idea that the SSRIs are even treatment for adolescent depression is in question, much less much a suicide attempt preventative. There is no direct linkage between the decrease in prescribing and their outcome parameter to validate the association implied. And the thing they actually measured is in itself a proxy for another proxy, based on the evidence from the most compromised of sources. My own takeaway from this article is that, once again, this was an attempt to answer a question using a huge dataset from a commercial administrative data, and it wasn’t up to the task. Having access to that much data is certainly tempting, but the absence of reliable E-coding doomed the study before it ever got off the ground. The errors intrinsic in any proxy, much less a second generation proxy, will probably never make them useful in answering subtle scientific questions.

There is a much larger question in this story, a question that has been present from the start – case studies versus population data. If you’ve seen Akathisia and suicidality in cases of adolescents put on SSRIs, and if you know of several completed suicides that you’re convinced were medication induced, how are you to look at a study like this even if you believe it?  This line of thinking presumes that the only thing one can do for a depressed teen is give them SSRIs, which has never been true. A blog is no place to launch into all the things one might do besides give a questionable symptomatic medication, how to actually approach a depressed teen. But even common sense tells us that there are a wide range of answers to that question that don’t only rely on medication. And if, after careful consideration, you decide to try an antidepressant, knowing that this kind of reaction can occur in some cases would certainly heavily inform how you would closely follow such a case.

There are many things one might say about this article, but this ending is certainly not one that would ever occur to me……
… it is disturbing that after the health advisories, warnings, and media reports about the relation between antidepressant use and suicidality in young people, we found substantial reductions in antidepressant treatment and simultaneous, small but meaningful increases in suicide attempts. It is essential to monitor and reduce possible unintended effects of FDA warnings and media reporting.
    June 24, 2014 | 11:52 PM

    I think it’s downright adorable that they end their article with a reference to “small but meaningful increases in suicide attempts.” Because of course, the scam of presenting “relative increases” and “relative decreases” led the press to report changes that were anything BUT small. Here’s what the Reuters wire story said:

    Antidepressant use fell 31 percent among adolescents and 24 percent among young adults after the FDA warnings, according to the study. Suicide attempts increased by almost 22 percent among adolescents and 33 percent among young adults in the same time period.
    I’m convinced this was by design. They knew the press would be unable to see through the “relative” statistical scam, and would report whopping decreases in meds, along with whopping increases in suicide attempts. And of course they could stand by and say it wasn’t their fault they were misinterpreted. Grinning like foxes all the way.

    June 25, 2014 | 9:01 AM


    At first, I thought thisarticle was something of a cut above the earlier attempts to dis the Black Box Warning. At least the explained their rationale and one could follow their calculations. But the deeper I go, the less I’m impressed. Their proxy makes no sense, nor does their reason for rejecting Patrick’s version. It just feels like a new incarnation of the earlier campaign. And the logic train is just as flawed…

    June 25, 2014 | 1:26 PM

    It’s observer bias all over again — they designed the study and interpreted the results to confirm their hypothesis.

    Coincidentally, here’s a blog about akathisia mostly in adults Please read the 69 comments and note how many times the person’s physician did not recognize akathisia. This is the state of the art regarding iatrogenic symptoms.

    June 27, 2014 | 11:31 PM

    Goddammit, does the BMJ have access to the Internet?

    Because I do, and I looked up the ICD-9 code that was used by these grifters as a proxy for suicide attempts: 968, Poisoning by Psychotropic Agents.

    That code does not cover most of the drugs Americans use to try & end it all. It doesn’t cover opiates, acetominophen, aspirin or NSAIDs (those are all Code 965, Analgesics). Doesn’t cover anticonvulsants (they’re Code 966). Doesn’t cover barbiturates, or sleep aids like Ambien (that’s Code 967, Sedative-Hypnotics).

    What DOES it cover? For starters … SSRI’s and other antidepressants (how perverse is that?) Okay, what else? Hallucinogens, including marijuana and LSD! Now I’m sure thousands of kids are dragged into the ER each year for problems related to marijuana … but I doubt even a handful were trying to toke themselves to death. Another huge distortion in the figures.

    It does cover stimulants like Adderall and Ritalin, which might be used to attempt suicide, but are overwhelmingly used for getting high or boosting school, work or athletic performance. It also covers “caffeine” — so yeah, all those kids who end up in the ER from chugging Red Bull also get counted towards the youth suicide epidemic.

    It does include benzos, which are a realistic means of suicide — but also madly popular for recreational use. It also includes antipsychotics, which I reckon could easily kill in overdose … and are also developing a track record for recreational use as well as medication.

    It’s little wonder that while teenage girls far outstrip boys in nonfatal suicide attempts, this “study” found its largest, most compelling effects among teen boys! What better evidence do you need that they are truly, madly, deeply off track?

Sorry, the comment form is closed at this time.