I do not have access to the update but I do have their original paper from 1999 published in Autism Vol 3(1) 7–16. I note that

The study included all children with variants of pervasive developmental disorders seen as inpatients at the University Clinics of Child Psychiatry in Copenhagen and Aarhus during the 25 year period 1960–84.

This suggests that their data may not be applicable across the entire spectrum. Firstly, not all autistic children become psychiatric inpatients. Secondly, the diagnostic criteria in use during that period were more restrictive than those we use today.

There was one confirmed death from an epileptic seizure and one suspected. Two subjects with epilepsy in residential care had swallowed dangerous objects and choked on them while unsupervised. One person with epilepsy died of meningitis and one drowned in a swimming pool, probably as a result of a seizure.

Only one of the subjects died in childhood, a boy aged 9 with subacute sclerosing panencephalitis. He died of pneumonia. The remaining five deaths comprised two suicides, one accidental overdose, one other case of pneumonia and one urethral bleeding from a person with renal cancer.

While all these deaths are a cause for concern the issue seems a little more complex to me than declaring that autism is a killer. In their conclusion the authors suggest that improving the quality of care could have prevented some of these deaths.

Altogether, mortality in our sample of patients was about double (SMR = 1.93) the expected rate, taking age, gender and temporal trends into consideration. This is a higher rate than found in earlier reports on deaths in autism, and may be due to the fact that these patients were followed for a longer period than has been done previously, that is, up to a mean of 31 years (range 14–48 years) of age.

On the other hand, we had expected to find an even higher mortality in this relatively severely handicapped group. The rather limited excess mortality is, however, in agreement with the general trend seen in Western countries for survival among the mentally retarded, for whom estimates of life expectancy have been increasing for the last decades (McGrother and Marshall, 1990; Carr, 1994). Nevertheless, some of the deaths reported here might have been avoided. The five deaths related to and probably caused by epilepsy (during seizures and accidents) demand special consideration. Some of these deaths are potentially preventable by careful monitoring of anti-epileptic treatment.

I find your conclusion, ... the underlying biomedical problems that cause symptoms of autism that simultaneously cause epilepsy and premature deaths, to be unsupported by the data. Of the twelve deaths in the original study as well as five with epilepsy there was one with hydrocephalus/epilepsy, one with Hodgkin’s disease, the boy with SSPE, the cancer victim also had tuberous sclerosis. The two suicides had a history of depressive illness and the accidental overdose, the only one without an additional illness or diagnosis was a drug addict of many years.

It is my experience that reading scientific papers reveals rather more complexity than the abstracts suggest.

the RATE OF EPILEPSY ITSELF is astronomically higher in autistic population. Therefore the rates of epilepsy-related deaths are astronimically higher in autistic population… Autism not only devastates, it actually KILLS.

“Astronomically” is an exaggeration. It might be one order of magnitude higher. In the general population, the prevalence of epilepsy is probably 0.5% to 1.0%. As of 2005, 6% of autistics registered with CalDDS were reported to have epilepsy (and this is gradually dropping in tandem with admin. prevalence rises.)

Excess mortality from epilepsy seems to be about 5 in 1,000 per year (Shackleton et al. 1999.) For an average autistic, it might be about 3 in 10,000 per year. So in a given year, you might have an additional 0.03% chance of dying (without counting all the other factors that might result in excess risk.)

Jake has been described over at AoA as a “Contributing Editor” since his Feb 19 2009 post, up to and including his Dec 22 2009 post.

Contributing Editors may have editorial powers, but may also be mere contributers. If Jake did call himself an ‘editor’ without the ‘contributing’ prefix, then he had no right to unless he has editorial powers.

Otherwise he’s just a contributer and should regard himself as such regardless of his title.

I was under the impression that was Mikes point.

I niether stated nor implied such a thing, nor is it the actual arguement you intially opposed. Simply because I disagree that it is autism itself (and not the reasons I stated which you appear to have convieniently ignored) that kills, does not mean I am okay with premature deaths in people with autism. Your habit of making stuff up as you go along is not amusing.

I have worked in services which have helped people with autism deal with potentially fatal co-morbid conditions, such as epilepsy, diabetes, asthma, anaemia, COPD and so on.

You were asked to supply your reference to your counter-claim that it is autism itself that kills. Do so.

Autism not only devastates, it actually KILLS.

(btw no, of course it is not “autism” that kills, but the underlying biomedical problems that cause symptoms of autism that simultaneously cause epilepsy and premature deaths…. Grow up and read some science.

“Sullivan,”

Have you forgotten this blog entry you wrote four months ago?

http://leftbrainrightbrain.co.uk/?p=2710

You should know better than everybody else here that I have no control over what gets put on AoA beyond my own writing.

I believe you recently referred to yourself on this blog as “editor”. What you describe above is “writer” or “contributor”.

http://dictionary.reference.com/browse/editor

a person having managerial and sometimes policy-making responsibility for the editorial part of a publishing firm or of a newspaper, magazine, or other publication.

Of course you did, you’re “Science”/Discover Blogs’ handy troll.

If you read that, you will see that he cries Conspiracy! instead actually provide evidence.