Tuesday, 1 March 2011

To epidemic or not to epidemic?

One of my pet hates is when headlines are used to lure me into a story which when I read it turns out to be a bit of a disappointment. I know, I know, I should expect it in these days of soundbites and spin doctors. In this case my attention was turned by a story posted by the Research Council of Norway titled 'No autism epidemic'.

It is a bit of an odd story this one because it describes the views and opinions of Maj-Britt Posserud, a researcher working on a council funded study on Autism Spectrum Disorders (ASD) as part of the “Barn i Bergen” (Children in Bergen) project. The study shows that the diagnosis of ASD may apply to as much as one per cent of the population (similar to other studies I have blogged about). Dr Posserud (I think she is a PhD or MD) is quite well-known in autism research circles; exemplified by some of her previous publications here and here.

In the Research Council press release Dr Posserud discusses the issue of the number of autism cases and how most of the increase is probably due to better ways of finding cases of autism rather than any true increase; to quote: "Our conclusion is that the rise in ASD can be explained mainly by the use of more thorough mapping methods and, consequently, that we are not seeing the emergence of an autism epidemic,” says Ms Posserud.

OK I hear you say, what's your problem? Well I have a few (grumble, grumble).

First of all the press release as I read it (please do correct me if I am wrong) contains no reference to any study or results. It talks about views and opinions but the reader cannot access where this information comes from. Again looking on PubMed, the only possible reference that this might be connected to is this study from 2010.

Second is the use of the sensational headline - 'no autism epidemic'. Epidemic occurs when new cases of a certain disease in a given human population, and during a given period, substantially exceed what is expected based on recent experience. I interpret this as meaning that the number of cases changes between point A and point B over and above what would be expected. Reading through her most recent study (albeit only the abstract), Dr Posserud and her colleagues seem only to have looked at one point during their study and not two, in a sort of 'snapshot' view of numbers of cases. Granted they do talk about figures from a 1998 study (which I assume is this one). It is perhaps overstepping the mark a little however to compare data between the 2 studies given differences in the age of participants (1998: 3-14 years vs. 2010: 7-9 years), differences in the screening and diagnostic instruments used to find and diagnoses cases of autism and differences in the way that cases are selected or referred. Chalk and cheese?

Now I am not giving any opinion as to what is going on with autism in terms of the number of cases being diagnosed. Everyday one or more study seems to emerge saying the increase is real or not real; discussing issues like greater awareness, diagnostic substitution, etc.

The issue of numbers of people with autism is a complicated one with many, many different factors potentially affecting the figures - the fact that we are better at screening for autism, the increased interest in autism and its symptoms, and yes, the fact that we live in a very different environment these days compared with the 'olden days'. What I would suggest is that this issue is not however helped by headlines which offer such a definitive, stark message without providing sound corroborating data. Perhaps I am just nitpicking (I really should take up a hobby), perhaps it's all just words and semantics; but time and time again it feels like science is being used to feed arguments where the answers aren't just black and white.