• This is the first time Minnesota has been a part of the ADDM network, and we are working on expanding our geographic area. The findings in this report reflect a small number of children concentrated in a large metropolitan area. The higher prevalence estimate compared to national estimates is not unanticipated for a large metropolitan area. 
  • In Minnesota, there are differences between the percentage of boys and girls identified, with more boys than girls. This is consistent with previous estimates. 
  • In Minnesota, we identify autism much later than when first concerns are reported. Although 73% of children had developmental concerns documented in their records before age 3, the average age of the first diagnosis was 4 years and 9 months. This lag is concerning due to what we know about the importance of early intervention. 
  • We found varying prevalence rates across racial and ethnic groups in Minnesota. The small number of children in some of these groups makes it difficult to determine whether the rates of children with autism truly are different across groups. Additionally, the confidence intervals around the prevalence estimates are large, and this tells us we should use extreme caution before drawing conclusions about any differences between groups. As the geographic area for MN-ADDM grows and includes more children, we will be better able to judge whether there are true differences in prevalence estimates. If differences are found, it will be important to focus on general health disparities that may influence these differences.  
  • We found that of the children who had IQ information in their records, 28% of children overall had an intellectual disability. When we looked at ethnic groups, we found that 43% of Somali children with autism and 18% of Hmong children with autism also had an intellectual disability. The sample sizes were too small to be able to tell whether this difference was real or whether it occurred by random chance. 
  • While the percentages appear different, there was no significant difference (using a p-value < .01) in presence of intellectual disability in Somali children when compared with non-Somali children with autism.

Key Terms

Confidence interval

A confidence interval gives an estimated range of values which is likely to include the information (e.g., prevalence) that we want to know about.  The interval is a range of values that you can have confidence contains the true value in the population. There are different levels of confidence intervals; the most common is 95%. With a 95% confidence interval, we say that we are 95% confident that the value in our population will be included in our interval.  5% of the time, it will not.  So, we could say, “We are 95% confident that the prevalence of ASD in 8-year-olds in our surveillance area was between 21 and 27 per 1,000 children in 2014. Larger samples typically have smaller confidence intervals.  Smaller samples usually have bigger confidence intervals.

Intellectual disability

Intellectual disability means that a person has difficulty learning at an expected level and functioning in daily life. In this report, intellectual disability is measured by intellectual quotient (IQ) test scores of less than or equal to 70.


Prevalence is a scientific term that describes the number of people with a disease or condition among a defined group at a specific period in time. Prevalence is usually expressed as a percentage or proportion of the defined group. For this project, we counted the number of 8-year-olds in 2014 who were identified with ASD and then divided that number by the total number of 8-year-olds in our surveillance area during 2014.