Vaccines aren’t responsible for rise in autism cases; change in diagnostic criteria is the most plausible explanation

While Andrew Wakefield’s 1998 study purporting to link autism to the MMR vaccine has since been retracted for numerous flaws, including data fabrication and ethical violations, the claim that vaccines cause autism still persists to this day.

One example of this is a meme that circulated on social media in July 2023. It showed increasing trends in the number of vaccines in the Centers for Diseases Control and Prevention (CDC) schedule and in the “autism rate” from 1983 to 2018. Entrepreneur Grant Cardone, among others, shared it on Twitter and on Instagram, where it received more than 70,000 likes according to analytics tool Crowdtangle.

In the fashion of “just asking questions”, the meme’s caption alluded to “a connection” between vaccination and autism. The caption, combined with the vaccine and autism statistics and the statement that some groups would “deny” the connection, implied that getting vaccinated increases the risk of, or directly causes, autism in children. However, this claim is baseless, as we show below.

The meme exaggerated the increase in vaccination and autism from 1983 to 2018

The meme claimed the CDC childhood immunization schedules included 10, 32, and 74 vaccines in 1983, 2013, and 2018, respectively. However, this is inaccurate: there were four recommended vaccines in 1983 and 13 in 2018.

What the meme may have actually been referring to is the total number of doses: each vaccine requires multiple doses, just like certain COVID-19 vaccines required two initial doses and subsequent boosters.

Looking at the number of recommended doses in the CDC schedule, there were 11 doses for all vaccines in 1983. In 2013, there were about 33 doses for children from zero to 18 years excluding the annual flu shot, and about 51 doses including the flu shot. In 2018, there were around 32 doses excluding influenza and between 50 and 60 doses including influenza (this is because some children required an annual two-dose influenza vaccination up to their ninth year).

Thus, while the number of vaccines and doses did increase from the 1980s to 2010s, the figures cited in the meme inflated that increase.

The meme also cited a prevalence of one person in 10,000 diagnosed with autism in 1983, one in 88 in 2013, and one in 36 in 2018.

Numbers for the prevalence of autism in the 80s are difficult to obtain and it is unclear where the one in 10,000 figure is coming from. Scientific publications report a prevalence ranging between four to six cases per 10,000, equivalent to a prevalence of one in 2,500 to one in 166^[1,2,3]. Therefore, the scientific literature reported a much higher autism rate in the 1980s than what the meme claimed.

It is also unclear what the source of the 2013 prevalence is. The CDC reported a prevalence of 1 in 69 in 2012 and 1 in 59 in 2014. It is thus likely that the autism prevalence in 2013 was about the same. It is thus higher than the figure mentioned in the meme.

According to the CDC, the prevalence in 2018 was one in 44, lower than the figure reported in the meme.

To summarize, the meme markedly inflated the increase in autism cases over the past forty years. While the figures in the meme would suggest a 277-fold prevalence increase from 1983 to 2018, data from the CDC and from the scientific literature suggested that the increase was four to 57-fold.

That said, it is accurate that the autism prevalence reported by the CDC increased over time. The rise in the number of autism cases is well-documented and occurs globally^[4]. But as we will explain below, this is no indication that vaccines are responsible.

A correlation between two rising trends alone isn’t sufficient to show a causal relationship

Regardless of whether the figures presented in the meme are accurate, the figures by themselves don’t demonstrate a causal relationship.

The fact that two phenomena occur at the same time, or that variables increase at the same rate, doesn’t necessarily mean that one caused the other. In fact, many variables are correlated by pure chance.

As an illustration that correlation doesn’t equal causation, we can say that many things increased in the almost forty years from 1983 to 2018, like, for example, the U.S. population, the number of cell phone users, the average surface temperature of the U.S. Yet a reasonable person wouldn’t conclude that these increases are connected with an increase in autism.

Vaccines don’t cause autism

In spite of recurring claims to the opposite effect and claims that studies haven’t looked into this, there is now a scientific consensus—established from numerous studies showing no association between vaccines and autism—that vaccines don’t cause autism. The CDC stated that vaccines aren’t a known cause of autism. The U.S. National Academy of Medicine as well as the American Academy of Child and Adolescent Psychiatry also shared that conclusion.

In a previous article, we explained that the belief connecting vaccines and autism found its origin in a fraudulent, later retracted, study claiming an association between the MMR vaccine and autism. Later scientific studies challenged and overturned that finding.

A 2010 study showed that vaccination doesn’t negatively impact neuropsychological outcomes like speech and language, attention, and behavior in children. In fact, the study found that children with the greatest exposure to vaccines actually showed better neuropsychological outcomes^[5]. While the authors suggested that familial or socioeconomic factors could explain why highly vaccinated children have the best neuropsychological outcomes, this observation nevertheless runs counter to the claim that vaccinated children have poorer neuropsychological outcomes than unvaccinated children.

A 2013 study also found that the amount of immunogen (the immunity-stimulating molecules contained in vaccines) didn’t affect the likelihood of developing autism. This directly contradicted the claim: receiving more vaccine doses didn’t increase the risk of children being diagnosed with autism at a later stage^[6].

A 2014 meta-analysis found no association between either the measles, mumps, rubella (MMR) vaccine or the amount of mercury in vaccines and the risk of developing autism^[7].

Rise in autism awareness and changes in diagnostic guidelines likely explain the increase in autism prevalence

Having established that vaccines don’t cause autism, we now turn to examining potential explanations for why autism diagnoses are rising.

Scientists are still trying to understand what exactly causes autism. The scientific evidence so far indicates that autism is likely the result of a combination of genetic and environmental factors. The diversity of symptoms and severity associated with ASD also suggest that there are multiple contributing factors involved rather than just one. Overall, the complexity in the way ASD arises and how it presents in different people makes it challenging to diagnose.

The most plausible reason for a rise in ASD diagnosis may have to do with a change in diagnostic criteria over time. These changes accounted for children who showed ASD symptoms later, as well as subtler features of ASD that weren’t considered before. In the U.S., the Diagnostic and Statistical Manual of Mental Disorders (DSM) serves as a guide to the diagnosis of mental disorders by professionals. Autism was first introduced in the third version (DSM-III) in 1980. A revised version of DSM-III in 1987 expanded the definition of autism. It also removed the requirement for the diagnosis to occur before the child’s thirtieth month. As a result, older autistic children who hadn’t been diagnosed could now be identified.

In 1994, the fourth version, DSM-IV, further broadened the definition of autism and introduced other variants, like Asperger’s syndrome.

DSM-V, released in 2013, changed the definition of autism once again, pooling together categories that were previously distinct under the broader notion of autism spectrum disorder (ASD).

Therefore, the definition of autism became broader over the years, encompassing many different situations. Because the definition is now broader, more people may be diagnosed as autistic than 40 years ago. Consistently, some scientific studies indicated that the changes in diagnostic guidelines are responsible for a part of the increase in autism prevalence^[8,9].

In fact, data show that the increase in autism diagnoses was accompanied by a decrease in diagnoses of “intellectual disability”^[10]. This suggests that people who would have been categorized as intellectually disabled in the past are now more accurately diagnosed as autistic thanks to our improved understanding of neurodevelopmental disorders.

Conclusion

Medical progress since the 1980s means that more diseases are now preventable by vaccination than ever. In keeping with this, the number of recommended vaccines for children has also increased over the past decades.

While autism diagnoses have also increased during the same period, this doesn’t mean that vaccines cause autism. On the contrary, the weight of scientific evidence overwhelmingly shows that vaccination isn’t associated with autism. The most plausible explanation for the rise in autism diagnoses so far is a broader diagnostic criteria thanks to improved understanding of the condition, rather than an actual rise in the incidence of autism in the population. An increase in public awareness of ASD is also likely to drive this trend.

REFERENCES

• 1 – Ritvo et al. (1989) The UCLA-University of Utah epidemiologic survey of autism: prevalence. American Journal of Psychiatry.
• 2 – Newschaffer et al. (2007) The Epidemiology of Autism Spectrum Disorders. Annual Review of Public Health.
• 3 – Blaxill (2004) What’s going on? The question of time trends in autism. Public Health Reports.
• 4 – Li (2022) Global, regional and national burden of autism spectrum disorder from 1990 to 2019: results from the Global Burden of Disease Study 2019. Epidemiology and Psychiatric Sciences.
• 5 – Smith & Woods (2010) On-time vaccine receipt in the first year does not adversely affect neuropsychological outcomes. Pediatrics.
• 6 – DeStefano et al. (2013) Increasing Exposure to Antibody-Stimulating Proteins and Polysaccharides in Vaccines Is Not Associated with Risk of Autism. The Journal of Pediatrics.
• 7 – Taylor et al. (2014) Vaccines are not associated with autism: an evidence-based meta-analysis of case-control and cohort studies. Vaccine.
• 8 – King & Bearman (2009) Diagnostic change and the increased prevalence of autism. International Journal of Epidemiology.
• 9 – Hansen et al. (2015) Explaining the Increase in the Prevalence of Autism Spectrum Disorders.The Proportion Attributable to Changes in Reporting Practices. JAMA Pediatrics.
• 10 – Thurm et al. (2019) State of the Field: Differentiating Intellectual Disability From Autism Spectrum Disorder. Frontiers in Psychiatry.