Abstract | Summary | Original Article

Analysis of prevalence trends of autism spectrum disorder in Minnesota.

Gurney JG, Fritz MS, Nes KK, Stevers P, Newschaff CJ, Shapiro EG.

Archives of Pediatric and Adolescent Medicine. July 2003; 157: 622-627.

Question: In the state of Minnesota, what changes in the rates of autism spectrum disorders (ASDs) have occurred during the period between 1981-1982 and 2001-2002 according to administrative data?

Background: Many education districts have reported enormous increases in the number of children identified as having an autism spectrum disorder. Some educators and policymakers have asked if, for instance, ASD has replaced the diagnoses of developmental delay or learning disability, making it appear that there are more children with ASDs. This study documents the growing rates of autism noted in the Minnesota school system and attempts to account for the increase.

Data Source: The database maintained by the Minnesota Department of Children, Families & Learning (CFL) of all children who receive special education.

Study Design: The annual data of year of age and disability category for the academic years 1981-82 through 2001-2002 were recorded. Individual children were not accounted for. Birth year- and age-year-specific rates were calculated using US Census Bureau data and the CFL data.

Method: The authors used the CFL data to determine the number of children designated as having an ASD and used census data to determine the number of children in the age groups during the years under study in order to calculate the rates of occurrence of ASDs.

Main Results: Between the school years 1981 and 1990 there was little change in the rate of children designated as having an ASD. However, the ASD rate rose from 3 in 10,000 children in 1991-1992 to 44 in 10,000 in 2001-2002.

Conclusions: This was a study conducted within a school system and its findings are useful to that type of setting. The chief conclusions were that:

1. there were great increases in the numbers of children identified as having an autism spectrum disorder in the Minnesota school system and the numbers showed no signs of leveling off or decreasing as time passed;
2. there was no evidence of diagnostic substitution; that is, children were not being identified as having an ASD instead of another condition. In such a case there would be increased numbers of children diagnosed with autism, but decreased numbers identified as having another condition;
3. there was evidence that autism spectrum disorders were under-diagnosed in this population; and
4. administrative requirements for reporting of cases of autism spectrum disorders based on newly enacted policies and laws may have influenced the increase in rates of ASDs in the population of children under study.

The latter three factors could not completely explain the increased rates of children with ASDs in the Minnesota school system.

This study adds to our knowledge of the prevalence of ASDs in small populations. Its methods, however, do not provide for good estimates in the general population.