Shocking news from Danish autism epidemiolgists!

18 Mar

Danish epidemiolgists have looked at criminal behavior in autistics, and the results are nothing short of startling:

In their paper, Pervasive developmental disorders and criminal behaviour: A case control study, authors S.E. Mouridsen, B. Rich, T. Isager, and N.J. Nedergaard, show:

The prevalence and pattern of criminal behaviour in a population of 313 former child psychiatric in-patients with pervasive developmental disorders were studied. The patients were divided into three subgroups and compared with 933 matched controls from the general population. Age at follow-up was between 25 years and 59 years. An account of convictions in the nationwide Danish Register of Criminality was used as a measure of criminal behaviour. Among 113 cases with childhood autism,.9% had been convicted. In atypical autism (n = 86) and Asperger’s syndrome (n = 114) the percentages were 8.1% and 18.4%, respectively. The corresponding rate of convictions in the comparison groups was 18.9%, 14.7%, and 19.6% respectively. Particular attention is given to arson in Asperger’s syndrome (p =.0009). © 2008 Sage Publications.

Yes, autsitics are as much as 20 times less likely to be convicted of crimes as their “typical” counterparts (0.9% compared to 18.9%).

I ran across this abstract and, given the current hype over Danish epidemiologists I couldn’t resist presenting it in this sensational mode. Besides, the title may draw some readers from the contingent bent on portraying child autistics as a looming threat to the well being of society.

23 Responses to “Shocking news from Danish autism epidemiolgists!”

  1. Rose March 19, 2010 at 01:28 #

    Aspergers is about equal with the norm. Actually, that’s kind of refreshing…our kids are no worse than anyone elses.

  2. Morgan March 19, 2010 at 01:40 #

    I assume this will be explained as further evidence of autistics’ “characteristic deficits” in . . . .

    (I’ll check back after you repair the link.)

  3. Sullivan March 19, 2010 at 01:50 #


    thanks for the head’s up on the link!

    There isn’t much more than the abstract I quoted, unless you have access to the journal, which I don’t. Otherwise I would try to figure out what is going on with that arson comment (which would be headline news on some blogs…)

  4. Science Mom March 19, 2010 at 03:47 #

    Sullivan, I don’t know about an arson comment but the study results about that may be relevant. In the Asperger’s case/control analyses, Case=114 Control=342, in the arson category, the convictions for the cases was 5 (4.4%) and controls was 0 (0%), p=0.0009.

  5. chaoticidealism March 19, 2010 at 04:16 #

    Stands to reason. Most crime is a social activity. You can’t mug somebody if you’re at home playing video games, memorizing train schedules, or studying biochemistry.

  6. Laurentius Rex March 19, 2010 at 10:33 #

    I am going to pick faults with the study because it equates crime with criminal conviction.

    Firstly there are the crimes which practically everybody commits at some time or another simply by ommission in an overregulated society where nobody knows the details of the law even those charged with enforcing it, putting ones feet on a seat in a train for instance has been considered a crime, rather silly really.

    Then there is the nature of serious crime versus very petty crime (such as putting ones feet on the seat in a train) this is the bugbear of all crime statistics and the ones that politicians massage most, there being in some statistical sets no difference between a crime of cussing someone out, and a crime of pulling a knife on them, both appear as crimes of violence (believe it or not)

    What is or is not a crime is socially defined, notwithstanding the laws, in that some laws are never enforced and some are rigidly enforced, that varying between state, but also between police district.

    The fallibility of the study then is the ill defined nature of what is a “crime”

    The second fallibility is the measure of autism, in that only former psyciatric inpatients are included in that measure, and to be accurate there would need to be consideration of underdiagnosis in the whole population.

    It’s not of much worth as a serious study though it might do as a piece of minor journalism or a party piece I suspect.

    Just because a study has a result I might like, does not mean I like the study. A sceptic can never allow that.

  7. farmwifetwo March 19, 2010 at 13:45 #

    If you have never been taught right from wrong. If you don’t have the “switch” to regulate right/wrong without training. If you don’t understand consequences occur with behaviour.

    If you are taught “behaviour is communication” and it can never be “negative”.

    Guess what…..

    Temple Grandin when I heard her speak mentioned the parent that wanted her to get a woman’s son out of a medical facility b/c he’d been researching/making bombs and the FBI found out. The FBI not knowing what to do with him, that’s where he ended up. Dr Grandin said “NO”, and blamed the parent. Dr Grandin also said time and time again, in her speach and in taking questions, autism isn’t a free pass past the social niceties or rules.

    I know many think the hacker from England shouldn’t be tried b/c he has Asperger’s.

    Autism may be the reason for behaviour…. but it’s never the excuse. The Child Psych calls it the “little devil”, those that can’t see that their own actions are wrong. Those that keep saying “It’s not my fault”. We deal with it day in and day out at our house and he told me it’ll be that way forever. High functioning ASD’s, NLD’s, ADHD’s, ODD, Separation Disorder’s etc… have a higher risk of “little devil’s”, “it isn’t my fault”, the “me, I know all”. I’ve also been told that by other professionals, that the highest functioning autistics they have met, have the highest behavioural issues.

    We’re not talking about head banging etc… we’re talking about that bully on the playground.

    They also end up in trouble with the law the most often. The Child Psych has that info on the wall of his office. It’s a gov’t poster.

    So.. Yes.. their research is probably very much correct.

  8. Joseph March 19, 2010 at 16:23 #

    This data doesn’t jive with the AoA narrative, does it? They can always counter with lots of anecdotes, though.

    Larry is correct that autistics might not be convicted of crime as often, for one reason or another. Additionally, autistics might not have as much opportunity for crime.

    Then again, it could very well be a positive characteristics of autistics. There’s no need to assume autistic characteristics are always negative or neutral.

  9. Sullivan March 19, 2010 at 18:26 #

    Joseph, Laurentius Rex,

    a few points-

    I agree about the difference between conviction and actual commission of a crime. This is not presented here as a definitive study. I found it interesting though, and a worthwhile talking point.

    Certainly this is worth more weight than, say, a wealthy woman who approaches the authorities who find her potentially a danger to herself and her autistic child…but the woman gets free and two years later kills the child. I see that story isn’t being pushed as hard since it became public that she was suspected of being a danger two years ago.

  10. Joseph March 19, 2010 at 19:47 #

    That’s correct. That a study has limitations/confounds doesn’t make it a bad study per se. (The abstract doesn’t contain any unreasonable interpretations of the data.) What it does mean is that anyone looking to do a follow-up should improve upon what’s already been done, rather than simply repeating it.

    Do you know if they matched controls for IQ, for example? That’s an obvious next step.

    • Sullivan March 19, 2010 at 20:01 #


      I haven’t read the full paper. All they say is “matched” in the abstract.Another interesting side point here–

      the “in-patient” group has a large number of Asperger syndrome individuals. I’ve heard that “in-patient” in Denmark means something very different than what people in the US assume.

  11. David N. Brown March 19, 2010 at 20:20 #

    “I am going to pick faults with the study because it equates crime with criminal conviction.”
    An important point, but one which can work both ways: When the notion that those with the “double Y” chromosomal disorder was being considered, one possibility investigated was that those who did commit crime were caught more often.
    Something I can say from my own experience in elementary school is that when I broke the rules (usually by fighting back when other kids abused me),I don’t remember ever going to the trouble of trying to avoid being caught. I expected to be caught, which I suppose could be considered a theory of mind issue.

  12. David N. Brown March 19, 2010 at 20:26 #

    “I am going to pick faults with the study because it equates crime with criminal conviction.”
    A significant issue, but one which works both ways: If those with a disability are more likely to be CAUGHT, the group’s crime rate could appear higher than it is. I can say from my own experience just with school authorities that when I broke the rules (mostly by fighting back against bullies), I was perfectly willing to accept the consequences.

  13. David N. Brown March 19, 2010 at 22:51 #

    Sorry for near-duplicate comments… It looked like the spam filter got the first one.

  14. dedicated lurker March 19, 2010 at 23:23 #

    farmwifetwo: This would only be applicable if the ASD kids had the higher crime conviction rate. The study states the opposite.

  15. AutismNewsBeat March 20, 2010 at 03:10 #

    Is it more likely for judge or a prosecutor to seem extra-legal remedy for an offense if it is determined the suspect has autism? For instance, dismissing the case in exchange for enrollment in a treatment program, or being remanded to the custody of a guardian?

  16. Ian MacGregor March 20, 2010 at 03:38 #

    I would not get overly excited about this study. The matching was by gender, birthdate within a day. birthplace, and social group.

    The control group was committed had significantly more traffic violations, the study group had significantly more violations for arson. The respective p values were .0009 and .006. There were several situations were the study group was more likely to commit crimes than the control group and vive versa, but not significantly so.

  17. ebohlman March 20, 2010 at 18:10 #

    Now, what’s needed is a study comparing crime victimization rates between autistics and non-autistics. It often turns out to be the case that when a group is popularly perceived as having a high crime rate, they actually have a high victimization rate instead.

  18. Anne March 20, 2010 at 18:12 #

    I’m not sure how it would be possible to do a study on the rate of commission of crimes without conviction. Where would you find any record of that? Or would you just ask your subjects whether they’ve ever committed a crime, regardless of whether they were caught? I guess you could compare autistic and control groups on the rate of admission of wrongdoing.

  19. Ian MacGregor March 20, 2010 at 18:24 #

    Anne, the study could have been based on arrest records. The study used convictions because it is a higher standard of guilt.

  20. Ian MacGregor March 21, 2010 at 15:29 #

    I switched the p values. For the crime of arson the study group was more likely to be convicted, p value .0009 and for traffic violations , the control group was more likely to be convicted, the pvalue was .006.


  1. Kiln Guns | R1 Project Art Pottery - March 19, 2010

    […] Autism Blog – Shocking news from Danish autism epidemiolgists … […]

  2. GlassTV – Leave autism out of mass shootings - December 21, 2012

    […] and ethical; we just have trouble reading social signals. Research has also shown that we are less likely to engage in criminal behavior. If anything, those with autism are more likely to be bullied or treated with scorn because we are […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: