Why Do So Many Autistic Girls Go Undiagnosed?
In this episode of our Question of the Week podcast, BU researchers Gael Orsmond, a Sargent College of Health & Rehabilitation Sciences professor of occupational therapy, and Sargent PhD student Sharada Krishnan offer a multifaceted perspective on how and why young women may have a different autistic experience compared to young men.
In recent years, an increasing amount of anecdotal evidence has surfaced about the differences in how autism presents between girls and boys. On social media apps such as TikTok, women are now sharing their stories of going undiagnosed until adulthood, leading many to question why late diagnosis seems to occur primarily in women. While some researchers believe this may be a result of biological differences, others point to bias in testing and ways that autistic women might present their disability differently. Are autistic girls being overlooked and underdiagnosed? BU researchers Gael Orsmond and Sharada Krishnan weigh in on this question.
- The criteria for autism in boys and girls is the same, but autism is diagnosed three to four times more often in boys than in girls. There’s debate in the field about whether these differences in rates between boys and girls are accurate.
- It’s possible a genetic, protective effect of being female makes autism occur less frequently in females. Additionally, societal factors, such as that the criteria were developed primarily based on research about boys, and social expectations that women are more social than boys, could explain why so many autistic girls go undiagnosed.
- Masking is when a person hides or camouflages their neurodiverse traits and replaces them with behaviors that might appear more neurotypical. Recent research points to females being more able to mask their autistic signs than males.
Dana Ferrante: This is Question of the Week from BU Today.
Gemma Acheampong: Hi, I’m Gemma Acheampong and this week on the podcast I’m asking the question: why do so many autistic girls go undiagnosed? But before we get into that, I want to share my own story and the rabbit hole that led me to propose this podcast.
Like many before the pandemic, I was not an avid TikTok user. But boredom from continuous quarantines and a brief yet stressful stint of unemployment found me scrolling through trying to use the app enough so that the magical algorithm would somehow ensure I’d be constantly entertained. And entertained I was.
But as time went on, I noticed that my “For You page”—a page of videos suggested by TikTok’s algorithm—started filling up with anecdotes from autistic women who shared their experiences growing up and the challenges they faced in adulthood. Some were funny.
TikTok Clip: “You’re autistic? Everyone’s a little autistic.”
“You’re six feet tall? Well, everyone’s a little six feet tall.”
“No, you’re five foot four!”
Acheampong: While others were sobering, as these women shared their difficulty adapting to society largely designed for the neurotypical mind.
Clip: “To other people, it seems like social interactions are really easy for me, but no, it’s terribly hard. I’m just good at making it seem comfortable for someone else, which isn’t necessarily a good thing, but it is a survival mechanism and a coping strategy I’ve used.”
Acheampong: These voices seem to fall into a few camps, those who had been diagnosed as children, those diagnosed as adults, and those struggling to figure out if they should seek out their own diagnoses.
And as time went on, I found myself identifying with that latter group and began to ask a question many on the app were asking: why did it seem like women were being diagnosed later in life or not getting diagnosed at all in comparison to men? After my deep dive of what I’ll call “TikTok research,” I turned to my next best source for information, Google.
I found that the criteria for diagnosing autism spectrum disorder (ASD) are based on data derived almost entirely from studies of boys. I also learned that the disorder is thought to be at least four times more common in boys than girls. And many girls are diagnosed later in childhood because their symptoms diverge from stereotypical autistic behavior.
However, there are limits to the everyday Googling and curiosity. For one, published research is often dense and written with other researchers and scientists in mind. I knew that in order to answer my question, I needed to talk to researchers directly for the full picture. So I reached out to BU researchers working directly in the autism field, Gael Orsmond, the associate dean of academic affairs for Sargent, BU’s College of Health & Rehabilitation Sciences, and PhD student Sharada Krishnan. Their current study, called the ROAD Ahead study, focuses on high school students on the autism spectrum who plan to graduate high school with a regular high school diploma.
The study isn’t focused on female high school students specifically, but having worked with this subset of autistic teens for years, I thought they might have some interesting perspectives on why this diagnostic difference seems to be ever-present. And as you’ll hear later on, TikTok and I weren’t too far off.
First of all, I’d like to say thanks for joining me on this podcast episode. To start, can you guys tell me a bit about yourselves and an overview of autism?
Gael Orsmond: I’m Gael Orsmond, I’m a professor of occupational therapy here at Sargent College. I’m a clinical and developmental psychologist by training. I’ve been doing research focused on autism for the past 20 plus years.
Sharada Krishnan: Hi, I’m Sharada Krishnan, I also go by Shar. I am a PhD student in Dr. Orsmond’s lab. My background is in occupational therapy, and I worked on a neuroimaging study with autistic children and adolescents and am now moving more into the adolescent space to do work on Gael’s study.
Orsmond: Autism is a neurodevelopmental disorder. There are probably multiple causes. There are hypotheses about environmental factors, and probably most likely it’s a combination of environment and kind of genetics interacting that tends to cause autism. That question has been out there since I started autism research, which was 25 years ago.
And it’s amazing that we don’t know more yet. It can affect all aspects of one’s life, but since it’s a spectrum, individuals experience different effects depending upon their strengths and their specific challenges. The autism spectrum disorder is diagnosed according to criteria by the DSM-V (that’s the Diagnostic and Statistical Manual published by the American Psychiatric Association).
And there are two primary areas where we see impairments for a person to be diagnosed with autism. One is in social communication and interaction. So that can include challenges with social-emotional reciprocity, things like having a back-and-forth conversation. There can also be impairments in nonverbal communicative behaviors used for social interaction.
So, things like eye contact or not using as many gestures or as much facial expression to facilitate that communication and social interaction. And then difficulties in developing, maintaining, and understanding relationships. And then the second primary area is about repetitive and focused behaviors and interests. And this, again, can range from repetitive motor movement, can include things like an insistence on sameness or distress at small changes in one’s environment or routine.
It can include interests that are atypical in their intensity or their focus, and it can also include sensory challenges. The behaviors have to be present from a young age and impact the person’s everyday life.
Krishnan: In addition to all of that, we know from research that autistic people often have differences in their executive functioning abilities, so the ability to plan and organize and these sorts of higher order cognitive skills. And some of them might also have challenges with daily living skills.
Acheampong: I think it’s interesting you said that it has to have an impact on their lives. Would you say if it’s not having a huge impact, it probably wouldn’t register on the autism spectrum?
Orsmond: All the behaviors that I mentioned that are part of that criteria occur in the general population. So that’s the challenge, right? There’s a spectrum, and for a diagnosis you have to have a line somewhere. For example, there’s variability in eye contact among people in the general population, and some of that’s even cultural. Some of us have more or less facial expression. Some of us want more or less social interaction.
And then many of us have hobbies that are similar in content to some of the interests in people with autism. But for the diagnosis, it has to be something that’s really kind of impacting their life or there’s more intensity to their interests than in the general population. So for a diagnosis you really need a combination of these symptoms or behaviors to kind of meet the threshold.
In general, the gold standard for diagnosis is an assessment called the autism diagnostic observation schedule. And it’s an interactive assessment where the clinician rates the behaviors of the person according to specific criteria. That’s generally the route that someone would go through to get a formal diagnosis.
Acheampong: I’ve heard of the term masking, or camouflaging, in reference to autism. Can you speak more about what that might be, and what that entails?
Krishnan: So, masking is essentially when a neurodiverse person hides or camouflages those traits and replaces them with behaviors that might appear more neurotypical. So it’s covering up their autistic traits in a sense. An example of this might be that an autistic person is having a challenge understanding the intention of the person talking to them, maybe they don’t understand the tone of that person’s voice. But they are smiling through it because they don’t want to necessarily appear confused. We also know from research that it might be that females are more able to mask their autistic signs than males.
Acheampong: Is there any indication in the research why that might be? Do you think it’s a nature thing?
Orsmond: The criteria for autism are the same for boys and girls, but historically we’ve seen that autism is diagnosed three to four times more often in boys than girls. But in children with autism who have intellectual disability, the ratio is more like two boys to every one girl.
And we also know that girls with autism who do not have intellectual disability tend to be diagnosed later. And there’s debate in the field about whether these differences in rates between boys and girls are accurate. More from a genetic or biological perspective, it’s possible that there’s a protective effect of being female and that would suggest that these rates are accurate.
And that females are just less likely to develop autism symptoms due to genetic or biological factors. And then, as Shar mentioned, the other possibility is that autism is underdiagnosed in girls because they express autism in a way that doesn’t align with the diagnostic criteria. These criteria were developed primarily based on boys, and so girls with autism may be better able to mimic or copy the expected social behaviors.
Their interests might be more age- and gender-appropriate, like focused interests on dolls or pop stars, and so it doesn’t seem quite so unusual to the people around them. The debate is really still out. We don’t know what the truth is and there’s quite a bit of research going on about the gender differences.
Krishnan: And it could also be partly a societal expectation of women, like we expect that women have more pressure to fit in. And we have these gendered expectations for social behavior. So that could also be playing into it.
Acheampong: Your current research is focused on teens with autism and the difficulties they may face in their transition into adulthood. Can you speak more about the research you two are specifically doing and how that’s progressing?
Orsmond: For the past five years, we’ve been conducting the ROAD Ahead Study, which is funded by the US Department of Education’s Institute of Educational Sciences. ROAD Ahead stands for “Responsibilities of Adulthood.” We’re interested in how autistic youth learn to manage the responsibilities of adulthood as they transition and how schools provide services to these youth.
We’ve been studying high school students who are graduating with a regular high school diploma. Diploma-track autistic students are those who are more likely to go on to college, and we were interested in finding out what types of services they received in high school and how that prepared them for adulthood.
We’ve recruited a sample of about 40 youth who participated prior to graduation and then a year after graduation. One of the things Shar and I have been doing recently is working on a new way to think about how to measure the adult outcomes. Overall, we found that these young adults are doing quite well when we look at the outcomes in a way that both captures objective and subjective indicators. For example, they’re going to college, working full time, or they’re in a transition program.
They’re largely satisfied with their social life, they have fairly high levels of autonomy in their living situation, even if they’re still living at home.
Krishnan: It’s important to not only look at these objective measures of researcher-defined values, but also what autistic youth want for themselves and whether they are happy and satisfied in whatever they are doing, especially in terms of living situation.
Previous research has categorized youth as doing well if they’re living away from home in the year after they graduate from high school, but even if you think about a neurotypical student, a lot of them live at home after graduation, or they may not be employed because they are in a post-secondary education program, so we need to rethink what is developmentally appropriate for this population.
We’ve been doing some work looking at gender differences in how youth are achieving both of these objective and subjective outcomes. And really, what we found is there are not a whole lot of differences in gender. Granted we do have a very small sample size of females, and I think that in itself is indicative of this gender difference in autism diagnosis as a whole.
We do have a smaller number of females in our sample, but they really aren’t doing all that differently. One thing that is noticeable, however, is that females do appear to have less autonomy in their daily life than males do. And that, again, could go back to societal expectations of how we raise females versus how we raise males.
Acheampong: Do you think you’ll be doing more later on to see how they do after college?
Orsmond: I think that’s a great question. Currently, we don’t have funding to follow them beyond this time point. The grant was from the Department of Education, so they’re really interested in what schools are doing to prepare youth.
In a prior study that I worked on with some mentors from the University of Wisconsin and Brandeis, we followed families for 12 years. So there is quite a bit of data out there. But some of that data is old now, or it looks at outcomes across a range of ages, so 22 to 60.
And so what we’re really trying to do in this current project is focus specifically on young adulthood and contextualize those outcomes.
Acheampong: Does your study take into account not only what the school is doing, but also the home living situation and what the parents might be doing?
Orsmond: In that prior study that I mentioned, the focus was on the family. In this current study, it’s been on the school, given the funding source. But we do have other research going on in our lab, about how adult siblings support their family member on the autism spectrum and work with their families to plan for the future.
That family focus is definitely something that we have done in other projects.
Acheampong: I’ve seen the terms “high-functioning” and “low-functioning” used sometimes when people reference or talk about people with autism. How do you feel as a researcher about those terms?
Orsmond: So that’s always a challenge. As we mentioned, we started the ROAD Ahead study more than five years ago, and at that time, people were using the terms. Over the past five years, we’ve really come to realize that some people don’t like those terms, and they don’t like being referred to as “low-functioning.” Also that wording is really not very specific, so low-functioning meaning what?
I try to be more specific in the language that I use, such as a person who has nonverbal autism or they don’t speak or they have intellectual disability. There was also a study in 2016 by Kenny and colleagues, and they looked at which stakeholders preferred which terminology. And since then, I think a lot of people are trying to be more respectful and thoughtful about the terms they use.
We are tending to use more identity-first language, so “autistic person” or “on the spectrum.” We try to avoid using words like disorder or deficit, and we try to take a strength-based approach.
Krishnan: I think there’s a key distinction to be made between disorder and disability. Because disorder implies that there’s something inherently wrong with the person.
If you look at the disability rights movement and that community, they [argue] a person has a disability because the environment is not built or made to match their needs and support needs. And so really to alleviate disability, what we need to be thinking about as a society is how to build society in a way that works for all people.
And so among autistic people, I think by and large there is a consensus that autism is a disability, because society is set up in a way that doesn’t meet their needs.
Acheampong: You’ve talked about this before, about the differences you noticed in your research between girls and boys, and that there’s probably not that much difference aside from girls maybe having a little bit less autonomy. Are there any other differences?
Orsmond: When we just looked at the differences between young women and young men in our sample, they’re going to college and working at kind of the same rate. They are satisfied with their social lives at about the same rate, that’s really about it.
I know from our prior research, in which we interviewed educators and family members, we heard a little bit of what Shar was mentioning in terms of the differences in how parents parent their child with autism. Parents of young women will talk about their concerns over their daughter’s vulnerability, so there’s some differences there.
Parents, when they talk about their autistic children, are talking about their children. Parents of young women are concerned about their daughters’ sexual behavior, for example. And parents of young men have often talked to us about their concerns about their son having encounters with police, because of their behaviors.
So, some of those are gendered expectations and concerns, and we definitely see that in our research as well.
Acheampong: I think it says more about what’s happening in society, and how we view men versus women in general. I wouldn’t say that’s an autistic thing—I’d say in general women and young girls are treated and raised a bit differently than young boys.
It’s interesting that you noticed that in your research, because hearing that I’d say, “Wow, that just says more about society than about autism.” I think you mentioned this much earlier, just how much more research needs to be done in regard to these differences to understand more about the difference between diagnosing genders.
Krishnan: There was a study and it showed that when you compared clinician ratings of autism symptoms to self-reported autism symptoms, clinicians tended to underrate symptoms as compared to the actual autistic female individuals themselves. And I think it does show that we really need to understand more about how females present, and the nuances of masking and how that potentially contributes to diagnostic differences.
Orsmond: In the literature there are also quite a few articles written by autistic women, who talk about camouflaging and the emotional stress of having to always present yourself in a way that doesn’t feel natural to how you typically operate. We need to do more research, but that research is really challenging to do.
Krishnan: Building off of that, we know from these anecdotal qualitative studies that autistic females who are diagnosed later in life often report feeling profoundly different from their peers, and so getting that diagnosis can be extremely validating and give them a sense of self-identity. There are definitely benefits to getting a diagnosis, even if it is later in life.
Orsmond: I was also going to add that we’re talking about males and females, and there’s quite a bit of research in the autism field looking at nonbinary identification, and the rates seem to be higher than in the general population. Even when we look at gender outcomes in our study, we need to be mindful of what we’re really looking at.
Acheampong: There’s talk about a revision to the DSM-V [the Diagnostic and Statistical Manual of Mental Disorders] that will be released sometime in March, which is the month we’re recording. The talk that I’ve seen about it online and on social media has been about concerns with the new conservative criteria. Do you think you’ll see those effects or changes in your research down the line?
Orsmond: When we’re doing research, we have to have a fairly narrow definition of autism and who we include in our studies, and that’s in part to compare studies and in part because those are the expectations for grant funders and publishers of research.
I don’t think it’ll affect research. It could affect the prevalence rates in the general population as you mentioned, but I think there’s always going to be questions about that; we’ve been changing our criteria for the past 20 years. We keep trying to get it right, but given it’s a spectrum, you have to draw a line somewhere about where it becomes a condition or a disability. And where you draw that line is the hard part.
Acheampong: I’d like to thank you both for talking to me about this. It’s something I’ve been very interested in, because of social media and because of my own personal life. It’s good to talk to researchers who are working with autism, and getting a broader picture of what is happening in the field.
Orsmond: Thanks, Gemma, it’s been nice talking with you as well.
Acheampong: Thanks to Gael Orsmond and Sharada Krishnan for joining us on this episode of Question of the Week.
This episode was edited by Doug Most, engineered by Andy Hallock, produced by Dana Ferrante, and hosted by me, Gemma Acheampong.
Thanks for listening, and see you in two weeks.
May 2, 2022
March 14, 2022
February 28, 2022