Background: Until recently best estimate prevalence rates for autism spectrum disorders (ASD) were 0.5/1,000 for autism and 2.0/1,000 for the broader spectrum. Three recent studies have suggested a significantly higher prevalence rate for ASD of 6.0/1,000 (mean 95 % CI = 4.8-8.0).
Method: Possible determinants of the apparent increase in the prevalence of ASD are outlined. Methodological aspects of the three recent studies are examined.
Findings: Increased recognition, the broadening of the diagnostic concept over time and methodological differences across studies may account for most or all of the apparent increase in prevalence, although this cannot be quantified.
Conclusions: Findings from ongoing studies should help confirm or disconfirm the putative rate of 6.0/1,000 for all ASD. The possibility that autism has been over-diagnosed in recent studies needs to be ruled out. Notwithstanding these outstanding questions, it appears likely that the current true prevalence of ASD is considerably greater than previously recognised. This has significant implications for our scientific understanding of ASD and for families and services. Future directions for epidemiological research are outlined.