IAN Research Findings: Children with ASDs Have Different Weight, Height Profile Than Unaffected Siblings

Date Last Revised: February 26, 2008
Date Published: November 15, 2007

Preliminary analysis shows that younger children with ASDs tend to be taller, and older children with ASD tend to be heavier, than their unaffected siblings.

The IAN Weight and Height Questionnaire was launched on the afternoon of October 22, 2007. Families participating in IAN Research were notified, and got out their scales and tape measures. By the end of the next day, October 23, families had submitted height and weight measurements for more than 1,000 children. By November 21, IAN had received data for nearly 2,000 children with ASDs and more than 1,200 of their unaffected siblings.  

Why is Body Size an Important Topic to Investigate?

There are at least two major reasons body size, which is measured by head circumference, weight, height, and body mass index (BMI), is an important topic to investigate in autism. First of all, differences in any of these may help point the way to specific subtypes of autism -- different kinds of autism characterized by measurable physical, biochemical, or other factors. In addition, issues related to weight and BMI may be related to the future health of individuals with ASDs.

A recent paper, "Clinical, Morphological, and Biochemical Correlates of Head Circumference in Autism," is an excellent example of research tying a certain physical picture in ASD to other important characteristics. Studying 241 individuals with autism, ages 3 to 16, researchers found that head size, as well as overall body size, tended to be large and that those with the largest head size also tended to have more impaired adaptive behaviors and less impaired IQ, motor development, and verbal language, as well as a family history of allergic/immune disorders. 1 Another study, this one focusing on 28 children with autism and 25 with Pervasive Developmental Disorder - Not Otherwise Specified (PDD-NOS), discovered growth of body length was accelerated in very young children with ASDs.2

In a preliminary look at the data, what have we found?

Children With ASDS: Taller, Heavier Than Typical Peers...But Only at Certain Ages

To compare children with ASDs and their unaffected siblings  the IAN project used both height and Body Mass Index (BMI). (BMI is a measure that is used to calculate whether someone's weight is more or less than would be expected for his or her height.)


As a group, children with ASDs appear to be taller than their unaffected siblings. Children with Asperger's Syndrome and Autism, specifically, appear taller than those with PDD-NOS who, in turn, do not differ from unaffected siblings in height. (See Figure 1.)

Figure 1. Height Status: Children with ASDs and Unaffected Siblings -- Nov. 21, 2007

A bar chart showing Height Status: Children with ASDs and Unaffected Siblings

For the purposes of this analysis, "very short" is defined as being equal to or below the 5th percentile for height, while "very tall" is equated with being above the 95th percentile.  (These are the same types of percentiles your pediatrician would talk about when going over your child's growth chart with you.)

In late October, about a week into data collection, we conducted an analysis which showed a much bigger difference in height between children with various ASD diagnoses than the current one does. (See Figure 2.)

Figure 2. Height Status: Children with ASDs and Unaffected Siblings -- Oct. 28, 2007

Bar chart for Height Status: Children with ASDs and Unaffected Siblings

When we had only 328 children with Autism, 153 children with PDD-NOS, and 142 children with Asperger's Syndrome to go on, the differences between the children with Asperger's and those with PDD-NOS, for example, looked fairly large. As we collected information from more families about additional children, more than doubling our original numbers, these differences evened out to reflect a more accurate picture.

This illustrates perfectly why it is so important to have the maximum number of people participating in this or any study. According to the laws of statistics, the more participants in a study, the more likely researchers are to arrive at findings that are not due to chance, but to a real difference in whatever is being measured. The Asperger's group that seemed extraordinarily taller than the other groups was made up of only 142 individuals, only 40 of whom were in the very tall group.

Because this group was small, statistics reported about it were less likely to be accurate. Once the number of Asperger's individuals included in the analysis grew, it was clear they were not quite so much taller than the other children as it had first appeared. Note, however, that it is still clear that more children with ASDs are above the 95th percentile for height compared with their unaffected siblings.

(For those of you interested in statistics and probabilities, the p value for the difference in height categories -- very short, normal, very tall -- between children with ASDs and unaffected children was less than 0.001.)

Does this difference between children with ASDs hold across all age groups? The answer is no. When we divide children with ASDs and their siblings into groups by gender and age, we find that it is only the children in the 3-6 year old group that are taller when compared to their unaffected siblings. This holds true for both boys and girls. It does not hold true for older children of either gender. 20% of boys with ASD, ages 3 to 6, are above the 95th percentile for height. This compares to only 16% of male siblings in the same age range.  Furthermore, while 16% of unaffected male siblings are in the "very short" category, only 5% of boys with ASD are in that category. (See Tables 1 and 2.)

(Please note: the IAN Project did not collect weight and height data on infants or toddlers under age 3, so we do not know if the children with ASD in this category were taller/longer than their siblings.)

Table 1. Height: Affected vs. Unaffected Boys, Ages 3-6 Table 1


Table 2. Height: Affected vs. Unaffected Girls, Ages 3-6 Table 2

*Note: Very Short is defined as less than or equal to the 5th percentile for height (similar to a pediatrician's growth charts); Very Tall is defined as more than the 95th percentile for height.

These findings are particularly interesting in light of other recent research that finds early accelerated growth of either body length or head circumference in children with ASD.


Contrary to expectations that there might be a large number of children with ASDs who are very undernourished due to gastrointestinal (GI) problems, feeding problems, or pickiness regarding foods, preliminary IAN data show no major difference in the number of children with ASDs with a BMI below the 5th percentile as compared with their unaffected siblings.

In fact, the data show that children on the autism spectrum are more likely to be obese than are unaffected siblings. ("Obese" is defined as having a BMI in the 95th percentile or above.) About 15% of unaffected siblings fall into the obese category, while 23% of children with an ASD do so.

Some of this difference may not be due to differences in body type, but to other factors affecting children with an ASD. In fact, 40% of parents who are concerned that their child is underweight, and 37% of parents who worry that their child is overweight, report that their child has a health problem or is taking medication that contributes to the weight issue. It will therefore be important for researchers to investigate the extent to which the apparent higher rate of obesity in autism is accounted for by factors such as medication side-effects.

Children with Autism, PDD-NOS, and Asperger's Syndrome all appear more likely to be obese than unaffected siblings. (See Figure 3.)

Figure 3. Weight/BMI Status: Children with ASDs and Unaffected Siblings -- Nov. 21, 2007

Bar chart for Weight/BMI Status: Children with ASDs and Unaffected Siblings

This is an intriguing result, with both scientific and practical implications. These data may provide support for the findings of the studies mentioned above, for example, and certainly should be taken into account by those guiding individuals with ASDs in their daily choices regarding exercise and nutrition. According to the data, children with ASDs may be at even higher risk for obesity than other children their age.

(For those of you interested in statistics and probabilities, the p value for the difference in BMI categories -- very low BMI, normal, and very high BMI -- between children with ASDs and unaffected children was 0.008.)

Do BMI differences vary by age group, as height differences do? Yes. It is only children with ASD age 6 to 18 who tend to be heavier for their age, relative to height, than unaffected siblings. When we calculate and compare Body Mass Index (BMI) separately for boys and girls in different age groups, we find that there is little difference between affected and unaffected children in the 3 to 6 year old age group, regardless of gender. Older boys with ASD, age 6 to 18, have a definite tendency to have a very high BMI compared to unaffected male siblings of the same age. Older girls show a similar tendency to higher BMIs, although the difference is less striking than for boys, and only reached statistical significance for girls age 12 to 18. (See Tables 3 and 4.)

Table 3. Body Mass Index: Affected vs. Unaffected Boys, Ages 12-18 


Table 3

Table 4. Body Mass Index: Affected vs. Unaffected Girls, Ages 12-18 

Table 4

**Note: Underweight is defined as less than the 5th percentile for BMI; At Risk for Overweight is defined as between the 85th and 95th percentiles for BMI; and Overweight is defined as over the 95th percentile for BMI.

In the context of other recent research on body size or length in Autism, these findings are very relevant. We encourage autism researchers studying these topics to apply for access to the IAN database in order to further explore this issue.

Please Note: These Findings Are Preliminary

The analyses presented here by the Interactive Autism Network are preliminary. They are based on information submitted via the Internet by parents of children with autism spectrum disorders (ASDs) from the United States who choose to participate. They may not generalize to the larger population of parents of children with ASDs. The data have not been peer-reviewed -- that is, undergone evaluation by researchers expert in a particular field -- or been submitted for publication. IAN views participating families as research partners, and shares such preliminary information to thank them and demonstrate the importance of their ongoing involvement.

We encourage autism researchers investigating these topics to apply for access to the IAN database. Contact researchteam@ianproject.org.


  1. Sacco, R., Militerni, R., Frolli, A., Bravaccio, C., Gritti, A., Elia, M., Curatolo, P, Manzi, B., et al. (2007). Clinical, morphological, and biochemical correlates of head circumference in autism. Biological Psychiatry, 62, 1038-1047.  View Abstract
  2. Van Daalen, E., Swinkels, S.H., Dietz, C., van Engeland, H., & Buitelaar, J.K. (2007). Body length and head growth in the first year of life in autism. Pediatric Neurology, 37(5), 324-330. View Abstract
These archived articles were originally published as part of the Interactive Autism Network (IAN) research project. 
The project is closed and no longer accepting participants.

More about Autism Research

Center for Autism and Related Disorders

Subscribe to news and updates

<< Back to the Archives