The prevalence of autism spectrum disorder is the same among Somali and white children living in Minneapolis, but Somali children tend to develop a more severe form of the developmental disorder, according to a new report released Monday by University of Minnesota researchers.
The study’s data revealed that 1 in 32 Somali and 1 in 36 white children aged 7 to 9 were identified with autism in 2010 — numbers that are statistically indistinguishable, according to the researchers.
Both Somali and white children in Minneapolis were, however, more likely to have been identified with autism than their non-Somali black or Hispanic peers. The data showed that the prevalence of autism was 1 in 62 among the city’s black children and 1 in 80 among its Hispanic children in 2010.
Overall, 1 in 48 Minneapolis children were identified with autism in 2010. That number is fairly close to the national parent-reported prevalence of 1 in 50 that was reported in March 2013 by the Centers for Disease Control and Prevention (CDC). But it is much higher than the CDC’s more official 1 in 88 estimate, which is based on 2008 data from 14 communities across the United States. (That estimate is expected to be updated in 2014.)
The data for the U of M study was collected from the school and medical records of more than 5,000 Minneapolis children. Clinicians reviewed each child’s records for behavioral descriptions or other information consistent with autism to determine if the child had the disorder. Although it has obvious limitations, this method of identifying and tracking children with autism or other development disabilities has long been used by the CDC.
The ‘why’ is unknown
The authors of the study stress that although the data they collected revealed racial and ethnic differences in the prevalence of autism among Minneapolis’ children, it did not provide an explanation for why those differences exist.
“We just don’t know,” said Amy Hewitt, the study’s lead author and a senior research associate at the U of M’s Institute on Community Integration, in a phone interview with MinnPost. “That’s the answer. And we need to know why. That’s for future research.”
Another unexplained — and surprising — finding was that every Somali child with autism in the study had an intellectual disability (defined as an IQ score less than 70), compared to one-third of all the children with autism in the study.
“It’s definitely a very interesting finding, a very important finding,” said Hewitt. “It definitely tells us that autism spectrum disorder looks differently in kids from the Somali community, and we have to find out why and how that is.”
Somali parents in Minneapolis have long expressed concerns that their children were experiencing a high rate of autism. In 2009, the Minnesota Department of Health found that Somali three- and four-year-olds were two to seven times more likely to be receiving services for autism within the Minneapolis school system than were other children. That finding was the impetus behind the current study and report.
“This is taking a huge toll on our community,” said Dr. Edward Ehlinger, state health commissioner, in a phone interview shortly after the U of M study was released on Monday. “We will be advocating with the CDC and [the National Institutes of Health] to do more studies.”
Ehlinger expressed particular concern with the report’s finding that the average age of diagnosis for Minneapolis’ children with autism was five. Research has shown that children can be reliably diagnosed at age two. Nationally, the average age of diagnosis is three, according to the CDC.
“We need to do a better job at early identification,” Ehlinger said. “The earlier you get [a child] into intervention, the more impact the intervention will have.”
‘Reassuring’ but ‘misleading’
“This report is reassuring in the finding that autism rates in the Somali community are not greater than expected,” wrote Dr. Steve Miles, a U of M professor of medicine and bioethics, in an e-mail response to a request from MinnPost for his comments on the report’s findings. He was not involved in the study, although he has advocated for research on autism in the Somali community.
The study also shows, Miles added, “that the Somali community, which is disadvantaged with regard to health care, does merit continued public health and clinical attention.”
Miles believes, however, that the report’s data on the prevalence of autism among Minneapolis’ different ethnic groups is misleading.
“The correct number to use is not the prevalence but the confidence intervals, which reflects the precision of the numbers,” he pointed out. “All the confidence intervals overlap — there is no difference between the populations.”
Miles also found the intellectual disability part of the report “deficient.” “The failure to give confidence intervals or statistical tests on the intellectual disability on page 12 [of the report] is a terrible error,” he wrote, “and is probably gravely misleading. Furthermore, none of this data is properly corrected for socioeconomic status.”
Other research has found a strong correlation between poverty and intellectual disability, he pointed out.