From the abstract of the cited study: "Changes in reporting practices can account for most (60%) of the increase in the observed prevalence of ASDs in children

From the abstract of the cited study: "Changes in reporting practices can account for most (60%) of the increase in the observed prevalence of ASDs in children born from 1980 through 1991 in Denmark." In other words, 40% of the increase could not be accounted for changes in reporting practices. That's quite a bit and not consistent with the tone of the essay. One has to be careful with these types of arguments, at least in the US, because clinicians tend to use different diagnostic standards, especially with things like autism. If they don't fit one definition, they use a "not otherwise specified" or "not elsewhere classified" diagnosis. That is, even if the DSM or ICD or whoever changes its criteria, it doesn't necessarily impact clinicians, who might have been using a different standard all along.