The biannual announcement in 2018 of yet another increase in the prevalence of autism spectrum disorder (ASD) has practically become a new rite of spring.
At the start of the millennium, the rate of ASD reported by the US Centers for Disease Control (CDC) was 1 in 150 and this rate remained steady for 2002. Then, starting in 2004, there have been increases announced every two years: 1 in 125 in 2004, 1 in 110 in 2006, 1 in 88 in 2008, 1 in 68 in 2010 and 2012.
Last week, the data for 2014 were announced, with another increase to 1 in 59. It is noteworthy that the CDC prevalence data are based on 8-year-old children, so that the 2014 data (announced in 2018) are for children born in 2006.
What in the world is going on? Why do the reported rates of autism keep rising? And, will they keep rising despite new programs designed to detect and treat it at increasingly younger ages?
Although scientists have not yet discovered what causes autism, a number of proposed causes have been disproven. For example, the popularized view that MMR vaccines cause autism is no longer a tenable "cause" and the article originally presenting the “vaccines cause autism” discovery by Andrew Wakefield was withdrawn under scandalous circumstances.,
Thankfully, the rise in autism cannot be due to thimerosal either, a mercury based preservative thought to cause autism that was removed from vaccines starting in 2000. If thimerosal was causing autism, we would have expected the rate to decrease after this form of mercury was removed from vaccines.
Despite this, the view that vaccines cause autism persists and some point to the rise in autism as a reason to fear vaccinations. And, of course, the rate of autism keeps increasing.
But knowing what isn’t causing the increase doesn’t provide insight into why the numbers keep going up—and are likely to go up even further, perhaps even more rapidly in the near future. To be sure, there is scientific consensus that at least part of the increase can be attributed to more accurate diagnosis.
And, it is certainly likely that the real rate is increasing as advances in medical science enables more and more babies to survive severe illness than ever before. It is possible that some of these babies, including those born very prematurely, could subsequently develop autism. Of course, the overwhelming majority of these “at risk” babies do not develop autism.
But another possible contributor to the reported increase is, perhaps, an unintended consequence of pushing early identification to earlier and earlier ages, when it is much more difficult to distinguish between autism, other disabilities such as intellectual disability, and even normal differences in development such as late talking.
In 2010, the American Academy of Pediatrics recommended universal screening for autism in the US, even though the accuracy—and stability—for an autism diagnosis in 2-year-olds was unknown. Parents—and pediatricians—often refer 2-year-olds for autism testing when they talk late. Although late talking is indeed a symptom of autism, the overwhelming majority of children who talk late do not have autism!. Skilled clinicians can distinguish between autism, and other forms of late talking in 2-year-olds.
On the other hand, one cannot help wondering whether some of the increase is because more and more late-talking 2-year-olds are identified as having autism. After all, a 3- or 4-year-old who spins, ignores verbal commands, relies on routines, and throws tantrums—all symptoms of autism—stands out. But many, many 2-year-olds who do not have autism do these things. In fact, one could argue that these are rather typical behavior during the “terrible twos” that toddlers soon grow out of.
Of course, all parents of late-talking children should be sure to have their child screened for autism—and all other clinical conditions associated with late talking: speech disorder, language disorder, hearing loss and intellectual disability while also bearing in mind that many late-talking children do not have any disability whatsoever. But never forego assessment simply because the odds are favorable that a late-talking child does not have autism.
But, also be prepared for an even more dramatic increase in the next round of “autism data” and a HUGE increase in the reported incidence in 2020—and beyond—because the latest 2018 data do not include children who were screened in 2010 at age 2 when the new guidelines for screening 2-year-olds were initiated. That is, national data are for autism in 8-year-olds, but are not reported until four years after a birth cohort turns eight. This year’s increase is based on children who turned eight in 2014 and were thus born in 2006. The 2020 data will be for children born in 2008, and then screened as 2-year-olds in 2010. They will turn eight in 2016—and be reported in the national data in 2020.
Therefore, it is easy to predict that the rate will then double or perhaps even more because as many as 1 in 10 children talk late, as compared to the current autism data indicating a rate of 1 in 59. And, as mentioned earlier, because nearly all 2-year-olds display some “symptoms” of autism, it is likely that at least some of these children will be mistakenly included in the autism data—and raise rates astronomically.
Better data are needed so that more precise rates of autism can be tracked. And more importantly, the US can better determine whether the efforts to reduce autism through early identification and early intervention are actually working. Of course, it is vitally important that more accurate measures of autism in 2-year-olds be developed and that clinicians be better trained on how to distinguish autism from other disabilities in two years who talk late, but do not have autism.
 Autism Spectrum Disorder. Data and Statistics. (2018). Centers for Disease Control and Prevention. https://www.cdc.gov/ncbddd/autism/data.html accessed 4.29.2018.
 The Editors of The Lancet. Comment: RETRACTION:—Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. The Lancet. 2010; 375(9713): 445. Accessed 05/08/2018.
 Godlee, F., Smith, J., Marcovitch, H. Wakefield’s article linking MMR vaccine and autism was fraudulent. BMJ. 2011; 342: c7452. Accessed 01/25/2018.
 Dachel, A. (2014). The Big Autism Cover-Up: How and Why the Media Is Lying to the American Public. Skyhorse Publishing, Inc.
 Camarata, S. M. (2014). Late-talking Children: A Symptom Or a Stage? MIT Press.
 Camarata, S. (2017). The Intuitive Parent: Why the Best Thing for Your Child is You. Penguin.