After reading the Viewpoint by Liao and Fombonne on autism overdiagnosis and its potential harms (JAMA Pediatrics, 2026) and the article laid out by Uta Frith (Psychological Medicine, 2026), I want to raise some concerns with their arguments.
I am a psychologist who conducts autism evaluations across the full range of presentation, including many patients who they would characterize as “milder.”
My primary concern is that the authors act as if there are a fix number of diagnoses available, and if someone with less visible needs get a diagnosis, it then prevents a child with more significant support needs from being diagnosed. This is an illogical argument, and the authors should not frame it as a zero-sum problem when it is not, in fact, zero-sum. I agree that there is a shortage of evaluators and services, but that reflects a lack of investment in healthcare infrastructure and funding, not an oversupply of diagnoses.
I cannot understand why the authors would choose to blame this on families and clinicians who are trying to understand and improve the lives of their clients rather than systemic underfunding, policy, and large-scale decisions about the limited available resources. It is clear that these authors have never seen the life-changing understanding that a late-diagnosed autism evaluation can bring to a person. Would they propose that we also limit the number of diagnoses given to learning disabilities, depression, and other diagnoses we give to preserve “the system”?
My second issue is that these authors assume that camouflaging is just that “a person understands acceptable behavior and then adjusts their more natural behavioral pattern accordingly.” They assume that the presence of camouflaging is evidence of diagnostic overreach rather than a genuine clinical phenomenon. This is another significant misunderstanding, and one that is clearly from mental health providers who have not stayed up to date on what autism actually looks like in autistic adults without intellectual disabilities.
The research on camouflaging describes it as a largely automatic, effortful process associated with significant measurable psychological costs, including burnout and disrupted interoceptive awareness (Khudiakova et al, 20205; Trevisan et al, 2021). Camouflaging is well documented among autistic girls, women, and adults identified later in life, a population which has been systematically under-studied for many years. The misunderstanding of camouflaging presented in this letter reflects a broader pattern in the field, where diagnostic frameworks built around externalizing, early-childhood presentations have not kept pace with what we now know about masking and internalized presentations, particularly in adults, people of color, and in people assigned female at birth.
I agree that diagnostic practice should be rigorous and should draw on multiple informants and longitudinal information where possible. But any conclusion that “milder” or camouflaged presentations should be diagnosed less often because of resource scarcity or a “self-fulfilling prophecy,” is not supported by evidence, and it does not address what is actually causing limited resources: insufficient investment in evaluation and support services.
Jessica Myszak, PhD