For a long time, autism has been considered a condition that affects more boys than girls. Currently, there is only 1 girl diagnosed with autism for every 4 boys. For years, researchers have been trying to get to the bottom of why this divide exists.
Autism looks different in girls than boys
Different studies over the years have suggested differences in brain structures that made boys more likely to have autism, or that girls have a natural “protective effect” against autism. One study theorized that girls with more “male-typical” brain structures were more likely to have autism, helping reinforce the misconception that autism is mostly a male condition.
However, other research has shown that boys and girls present the symptoms of autism in different ways, and that girls are more likely to be able to “camouflage” those symptoms, except in the most severe patients. What this means is that those health care practitioners who are assessing these girls may be missing the symptoms of autism, simply for the reason that those symptoms do not look like what they are expecting to see in a child with autism.
Current assessments and data are boy-focused
It makes sense that if autism is thought of in the medical community as a mostly male disease, and boys present their symptoms more obviously than girls, that diagnostic assessments have historically skewed toward autism traits prevalent in boys. Therefore, when boys are assessed by a specialist, and their behaviors and delays look the same as those that were seen in previous boys with autism, a diagnosis can be relatively clear.
However, this pattern—using data based on boys to assess all children and primarily finding those same patterns in boys—may be creating a circular pattern of logic that shuts out girls who have autism, merely because they don’t present in a way that is easily seen through the lens of the most commonly used assessments.
Study suggests autism is under-detected in girls
One recent study, published recently in Science Alert, suggests that in the brains of mice, both male and female brains may be just as likely to have autism. What this means for the medical community is that the number of girls with autism is likely to be vastly underestimated, and that we need new data sets and new ways of diagnosing that include girls in the picture.
In the study, researchers studied two groups of mice, all from a previously-studied family line which had synaptic mutations and characteristics consistent with autism in humans, including certain neurologic deficits and lower signaling proteins than were found in mice with no characteristics of autism. The mice in the autism group also showed social deficits as well.
The key finding in this study is that there were no significant differences in the physical brain or neurological structures between males and females, nor were there significant differences in the social behavior of the male and female mice in the autism group.
For more information on this particular study, find the overview at: https://www.sciencealert.com/new-evidence-highlights-a-serious-flaw-in-our-perception-of-autism
Future studies need to include more girls
All of this suggests what many in the healthcare community, as well as parents of girls with possible developmental or behavioral delays have always suspected, which is that autism in girls continues to be vastly under-detected. We urgently need to start finding and generating data that shows the true prevalence of autism among girls, and developing diagnostic assessments that do a better job of recognizing the unique ways in which girls present with autism, so that girls have an equal chance to receive early interventions that can create major differences in their development throughout their lives.
The lead researchers in the study note that we need to include many more girls with autism in clinical studies, so we can record their symptoms and behaviors and eventually work those traits into existing diagnostic assessments.
We are committed to removing bias from autism diagnosis
At Cognoa, our data scientists have focused on balancing gender bias in diagnosis since day one. One of our main goals as a company is to help make sure that autism is not only diagnosed quickly and simply, but also equitably, so that no one is left out of the help they need because of their sex, race, ethnicity, or socio-economic status.
To this end, in developing Canvas Dx, the first FDA-authorized AI for diagnosis of autism, we actively sought thousands of diverse patients, including both boys and girls with varied presentations and comorbidities, as well as children from a range of racial, ethnic, and socioeconomic backgrounds. We have trained our AI to represent and account for differences in gender, race, ethnicity, and socioeconomics, recognizing, for example, that girls with autism show different traits than boys.
Every child deserves an equal chance at care
In addition, the AI diagnosis itself is not based on preconceived or subjective notions of who should have autism and who should not. Our AI is developed to even the playing field, assessing patients without bias or prejudice regarding who they are or how they present. It’s just clean, robust data going through a well-trained process to arrive at a likelihood of autism for all children.
In this way, we are cutting through old prejudices about autism, its presentation, and its prevalence, so that we can help provide equal help to all patients, no matter the structure of their brain, the state of their social development, or what body they happen to be born into.
Find out more about how Canvas Dx is trained at https://cognoaprodstg.wpengine.com/technology/.
