Researchers involved in the "Barn i Bergen" project got widely varying results when they used different methods to investigate the same group of children. The first sub-study concluded that 0.44 per cent of the children had ASD, whereas the result a few years later was 0.87 per cent.
"The clinical test revealed several additional cases of the disorder. This suggests that a diagnosis of ASD cannot be ruled out merely on the basis of interviews with the parents," Ms Posserud explains.
According to Ms Posserud, it is the children with normal intelligence who most often go unnoticed. These children were not included in the definition of autism a few decades ago when the diagnosis was only applied in the most serious cases. Today ASD covers difficulties with social interaction across a range of intellectual abilities. Since the definition has been expanded, many more people have been diagnosed with autism.Diagnostic criteria matter when measuring the autism rate in any population. When we use "liberal" (broad) criteria, we find far more individuals are "autistic" than when we apply strict, researcher developed criteria. In simple terms: we keep relaxing the criteria, we naturally end up with an "epidemic" of new autism cases.
This doesn't mean one set of criteria is better or worse than any other. What we must admit is that "autism" is a subjective diagnosis based only on the observed traits of an individual. Without a medical test for autism, the statistical measures of "incidence" will fluctuate. It is reasonable to assume the DSM-V will increase the number of diagnoses, and therefore a sudden incidence increase will have to be explained to the public.