Whenever I give a lecture about autism, whether to doctors, teachers, or parents, I always get the same question: Why is there so much autism? Why is it becoming more common? Everyone cites the CDC (Centers for Disease Control and Prevention)-one child in sixty-eight has it. But since April 26, 2018, that number has changed to one in fifty-nine.
The estimated prevalence of autism as of 2018.
There is a lot of press coverage on this issue, too. There are more than enough hypotheses, and they depend on the author’s position. These include the notorious vaccinations, chemical additives in food, and the detrimental effects of the growing attachment of even the youngest to a phone or tablet screen. Scientists are more cautious.
I wanted to share some important and interesting information on the subject. This information usually escapes the attention of non-specialists.
Let’s first clarify two important medical statistical terms.
Prevalence is the total number of cases of a certain disease in a population in a certain period of time.
Incidence is the number of new cases of a disease in a population in a certain period of time.
Incidence provides information about the risk of a disease, and Prevalence shows how often the disease is prevalent.
In the case of a chronic condition like autism, incidence is very difficult to use. After all, the problem begins long before it is diagnosed, and this gap between onset and diagnosis is influenced by many factors other than biological risk per se.
This is why prevalence is used in autism statistics.
Let us now try to identify the main factors that have led to the increase in prevalence of autism.
Changes in classifications and criteria
Autism is a clinical diagnosis. We do not yet have biological markers that can be reliably used in clinical practice, although research on the subject is ongoing.
Autism is diagnosed based on the criteria outlined in the DSM (Diagnostic and Statistical Manual of Mental Disorders). The problem is that the DSM itself is not a static structure, but very much a mobile one, subject to constant changes and revisions. As the criteria in the DSM change, so does the diagnosis, and hence the prevalence of autism.
The prevalence of autism first began to be studied in the sixties, in England and the United States. At that time, the signs described in Dr. Kanner’s research were used for diagnosis. Accordingly, the most severe cases were identified and taken into account. In 1966, the prevalence of autism in the child population in England was defined as 1 in 2,500. In the classifications of the time, autism did not exist as an independent diagnosis, but was considered part of schizophrenia.
Autism was not included in DSM-1 (1952) or DSM-2 (1968), and only in DSM-3 in 1980 did it take its place in the classification as an independent diagnostic entity. It introduced only one form, infantile autism, which was described using six criteria. A patient had to meet all six to receive a diagnosis of autism.
One of these criteria was the onset of symptoms before 30 months (two and a half years of age). This immediately limited the diagnosis. Anyone with symptoms seen later in life was rejected.
In 1987, a revised edition came out called the DSM-3R. In it, the framework for diagnosing autism was greatly expanded. To establish a diagnosis, already out of 16 criteria, a patient had to meet only eight. And the 30-month barrier was removed. Pervasive Developmental Disorder, Nonspecific (PDD NOS) was also introduced. The prevalence of autism immediately jumped to 1 in 1,400 (i.e., almost double).
In 1994, the next edition of the classification, DSM-4, came out. It expanded the category of autism even further, with the addition of Asperger’s syndrome. The diagnosis became even more common.
The official statistics (1 in 68) still used the DSM-4TR criteria (revised edition of 2000), not the DSM-5, which came out in 2013.
And even the most recent CDC information on the subject, released April 26, 2018, used essentially the same criteria, although DSM-5 was also involved when possible.