The proportion of Swedish children diagnosed with autism spectrum disorder rose from 0.5% in the early 2000s to more than 2% in 2014.
Researchers at Sahlgrenska Academy, University of Gothenburg have taken a closer look at the increase to determine whether it actually reflects greater prevalence of the phenotype.
Two different databases
They used two databases to address the question: a patient register of all 1,078,975 Swedes born in 1993-2002, and an ongoing study of 19,993 twins born during the same period.
After calculating the frequency of autism spectrum disorder in the patient register, the researchers interviewed the parents of the twins. The questions, which were taken from a validated instrument for assessing autism spectrum disorder, concerned the prevalence and severity of various symptoms.
Not more common
Published in the British Medical Journal, the study found that the symptoms reported by the parents of twins remained constant in 1993-2002 even though the patient register showed a yearly increase.
“In other words, the growing number of diagnoses does not reflect greater prevalence in the population,” says Christopher Gillberg, professor at Sahlgrenska Academy. “The actual phenotype is no more common than it was ten years ago.”
Three major explanations
There appear to be three explanations for the increase in diagnosis: the criteria are broader, people have a greater propensity to see doctors and clinicians are more knowledgeable.
“Our findings suggest that administrative rather than etiological changes account for the reporting trend,” Dr. Gillberg says.
“Autism phenotype versus registered diagnosis in Swedish children: prevalence trends over 10 years in general population samples” appeared in the April 28 issue of the British Medical Journal.
Christopher Gillberg can be reached via Ingrid Vinsa, Gillberg Neuropsychiatry Centre, University of Gothenburg
+46 31 342 59 70