Several reasons to doubt that rates of autism are going up

Do increases in diagnosed autism reflect significant increases in actual rates of autism, which here in the U.S. now stands at 1 out of 36?

Source: Wikipedia

There are several reasons to think that the answer is probably "no."

First, even the most recent peer-reviewed journal articles propose that expansions of the diagnostic criteria and of reporting practices are primarily responsible for apparent increases in autism rates.

A 2021 British study, for example, concludes:

Increases could be due to growth in prevalence or, more likely, increased reporting and application of diagnosis. Rising diagnosis among adults, females and higher functioning individuals suggest augmented recognition underpins these changes.

And a 2020 U.S. study of autism in the public schools reports:

Although the impact of environmental or genetic influences cannot be entirely ruled out, we identified significant shifts in eligibility trends that substantially contribute to the remarkable increase in autism prevalence.

And then there's this data from 2000-2010, which suggests that increasing numbers of cases once classified as intellectual disability are now classified as either autism or as autism plus intellectual disability:

Reclassification of those with intellectually disabilities, and greater identification of autism in higher functioning individuals, predicts increases in rates of severe autism and mild autism, but not increases in rates of moderate autism. 

That's because in moderate autism is hard to confuse with intellectual disability, and even harder to confuse with neurotypicality. In moderate autism there's often no intellectual disability as measured by non-verbal IQ tests, and even where there is a measured intellectual disability, it's overshadowed by the autism symptoms. Moderate autism, moreover, by definition involves more profound social and restrictive/repetitive behaviors than mild autism, and so is far less likely than mild autism to be completely overlooked.

So if those who are convinced that autism is becoming more prevalent want to convince the rest of us, they might start by demonstrating that rates of mid-spectrum autism--which is arguably the most prototypical subset of the spectrum--are rising in tandem with the rates of autism at the severe and mild tail ends of the spectrum.