How can epidemiological studies contribute to understanding autism spectrum disorders?

More and more studies on the frequency of autism spectrum disorders (ASD) have been published recently, most of which show the increase in prevalence data. In this review, the author pointed out factors and parameters to be considered in analyzing frequency data, i.e., the enlargement of the concept of autism, prevalence and incidence, accuracy and precision in the initial screening, and the effect of the "vaccine debate". The proportion of high-functioning ASD has been growing higher and higher due to better recognition in the last few years, and the apparent increase might still be the tip of an iceberg. Future epidemiological studies should include themes on diversity of the longitudinal course and re-conceptualization of ASD by dimensional diagnosis.