Pages

Sunday, June 07, 2015

Use of Anti-Inflammatories Associated with Threefold Increase in Homicides

Scene from Elephant, a fictional film by Gus Van Sant


Regular use of over-the-counter pain relievers like aspirin, ibuprofen, naproxen, and acetaminophen was associated with three times the risk of committing a homicide in a new Finnish study (Tiihonen et al., 2015). The association between NSAID use and murderous acts was far greater than the risk posed by antidepressants.

Clearly, drug companies are pushing dangerous, toxic chemicals and we should ban the substances that are causing school massacres Advil and Alleve and Tylenol are evil!!

Wait..... what?


Tiihonen and colleagues wanted to test the hypothesis that antidepressant treatment is associated with an increased risk of committing a homicide. Because, you know, the Scientology-backed Citizens Commission on Human Rights of Colorado thinks so (and their blog is cited in the paper!!):
After a high-profile homicide case, there is often discussion in the media on whether or not the killing was caused or facilitated by a psychotropic medication. Antidepressants have especially been blamed by non-scientific organizations for a large number of senseless acts of violence, e.g., 13 school shootings in the last decade in the U.S. and Finland [1].

The authors reviewed a database of all homicides investigated by the police in Finland between 2003 and 2011. A total of 959 offenders were included in the analysis. Each offender was matched to 10 controls selected from the Population Information System. Then the authors checked purchases in the Finnish Prescription Register. A participant was considered a "user" if they had a current purchase in the system.1

The main drug classes examined were antidepressants, benzodiazepines, and antipsychotics. The primary outcome measure was risk of offending for current use vs. no use of those drugs (with significance set to p<0.016 to correct for multiple comparisons). Seven other drug classes were examined as secondary outcome measures (with α adjusted to .005): opioid analgesics, non-opioid analgesics (e.g., NSAIDs), antiepileptics, lithium, stimulants, meds for addictive disorders, and non-benzo anxiolytics.

Lo and behold, current use of antidepressants in the adult offender population was associated with a 31% greater risk of committing a homicide, but this did not reach significance (p=0.022). On the other hand, benzodiazepine use was associated with a 45% greater risk (p<.001), while antipsychotics were not associated with greater risk of offending (p=0.54).

Most dangerous of all were pain relievers. Current use of opioid analgesics (like Oxycontin and Vicodin) was associated with 92% greater risk. Non-opioid analgesics were even worse: individuals taking these meds were at 206% greater risk of offending that's a threefold increase. 2  Taken in the context of this surprising result, the anti-psych-med faction doth complain too much about antidepressants.

Furthermore, analysis of young offenders (25 yrs or less) revealed that none of the medications were associated with greater risk of committing a homicide (benzos and opioids were p=.07 and .04 respectively). To repeat: In Finland at least, there was no association between antidepressant use and the risk of becoming a school shooter.

What are we to make of the provocative NSAIDs? More study is needed:
The surprisingly high risk associated with opioid and non-opioid analgesics deserves further attention in the treatment of pain among individuals with criminal history.

Drug-related murders in oxycodone abusers don't come as a great surprise, but aspirin-related violence is hard to explain...3


Footnotes

1 Having a purchase doesn't mean the individual was actually taking the drug before/during the time of the offense, however.

2 RR = 3.06; 95% CI: 1.78-5.24, p<0.001 for Advil, Tylenol, and the like. And the population-adjusted odds ratios (OR) weren't substantially different, although this wasn't reported for NSAIDs:
The analysis based on case-control design showed an adjusted OR of 1.30 (95% CI: 0.97-1.75) as the risk of homicide for the current use of an antidepressant, 2.52 (95% CI: 1.90-3.35) for benzodiazepines, 0.62 (95% CI: 0.41-0.93) for antipsychotics, and 2.16 (95% CI: 1.41-3.30) for opioid analgesics.

3 P.S. Just to be clear here, correlation ≠ causation. Disregarding the anomalous nature of the finding in the first place, it could be that murderers have more headaches and muscle pain, so they take more anti-inflammatories (rather than ibuprofen "causing" violence). But if the anti-med faction uses these results to argue that "antidepressants cause school shootings" then explain how ibuprofen raises the risk threefold...


Reference

Tiihonen, J., Lehti, M., Aaltonen, M., Kivivuori, J., Kautiainen, H., J. Virta, L., Hoti, F., Tanskanen, A., & Korhonen, P. (2015). Psychotropic drugs and homicide: A prospective cohort study from Finland. World Psychiatry, 14 (2), 245-247. DOI: 10.1002/wps.20220

4 comments:

  1. Interesting! But it seems to me that the data came from the prescription register, so the reference to over-the-counter medication seems a bit out of place. In fact, they don't appear to have had any data on over-the-counter analgesic use *at all* among either the offenders or the controls.

    ReplyDelete
  2. That part was a little puzzling to me as well. Maybe the non-opioid analgesics are also tracked in that database, whether the medications are by prescription only or available over-the-counter. The paper wasn't very clear about which drugs were included in this class, either. The press release said:

    "The study found, rather surprisingly, that the highest increase in the risk of committing a homicide was associated with opiate painkillers (+92%) and anti-inflammatory painkillers (+206%)."

    ReplyDelete
  3. Valtteri SuonmutkaJune 08, 2015 4:56 AM

    What I believe to be the case is that for example anti-inflammatory painkillers that have over 400 mg of ibuprofen per tablet are not sold over-the-counter but require a prescription. There is obviously no way they could have included over-the-counter drugs in their analyses (or I am not paranoid enough about the government's spying technologies).

    There might be several reasons for the result that painkillers are correlated with higher homicide rates. Just for the fun of it, I'll present two of my guesses that are not backed up by any data, these are based solely on my own anthropological observations.

    Firstly, I'd guess that mixing alcohol with drugs is more commong among people who are more likely to commit crimes (including homicide). These people tend to "milk" prescription drugs from doctors by over-exaggerating their pains (or even by making them up). Usually the goal is to get opioid based drugs, but in many cases the kind of a "first step" is that the doctor prescribes only strong non-opioid painkillers.

    Secondly, maybe people who tend to seek quick and easy solutions to problems might be more likely to resort to violence, and painkillers instead of trying to be "Zen" about pain. Most of homicides are, after all, done at the heat of the moment. People have been killed because they took a sip from the wrong whisky bottle.

    ReplyDelete
  4. Valtteri - Thanks for your observations, very interesting indeed. It's true that in 79.4% of the homicides, the perpetrator was intoxicated. Alcohol is a well-known risk factor for committing a homicide.

    I can't claim to know anything about the prescription database in Finland, so I looked up four of the authors' cited articles. And that didn't help! They mostly studied antipsychotic use in individuals with schizophrenia.

    One thing that's seems odd to me is that anyone can just double the dose of an over-the-counter anti-inflammatory and get the same effect as the stronger prescription version. And, among the major indications of NSAIDs are arthritis (mostly elderly people) and menstrual pain (women only, obviously). Yet it's rare that people in these demographics commit murder. In the paper, 88.5% of the offenders were male. Age information was not very precise:

    "The median age of offenders and controls was 36.3 years (range 13.3-88.0 years)."

    Overall, this finding seems pretty flukish to me...

    ReplyDelete