Intended for healthcare professionals

Rapid response to:


Increase in autism is due to changes in diagnosis, study claims

BMJ 2004; 328 doi: (Published 12 February 2004) Cite this as: BMJ 2004;328:364

Rapid Response:

Do the Math!

The author of this study wants us to believe that there was a 20% per
year decrease in other developmental disorders and a 20% per year increase
in Autism diagnosis. This would suggest that there is an exact one to one
correspondence in the shifting of the diagnosis.

However, it is mathamatically impossible for this to occur on a per
year basis. A close match is possible only for ONE of the years in the
eight (8) year time frame of the study. This leads me to believe that the
study is bogus an the author's motivation is highly suspect.

For example, if we start with 100 autistic children and 100 other
type children, using the 20% figure, the first year we might say there are
20 more children with Autism and 20 less in the Other category. The next
year we start with 120 with Autism and 80 Other. Twenty percent of 120 is
24, while 20% of 80 is 16. The number 24 is not equal to 16, not even
close. The next year after that, we get about 28 and 12 and the
difference keeps increasing year to year. The math does not add up so the
conclusion of the study wrong.

Competing interests:
None declared

Competing interests: No competing interests

26 February 2004
Dan Radulovic
Riverview, FL 33569
12026 Pennfield place