Diagnostic Agreement of a Structured Telehealth Adult Autism Assessment
This preregistered retrospective study demonstrates that a structured, neurodiversity-affirming telehealth assessment protocol for adults achieves substantial diagnostic agreement with blinded expert review, supporting the feasibility of combining affirming practices with criterion-based decision-making while highlighting the need for further validation, particularly among BIPOC populations.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
For many adults, the question of whether they are autistic remains an open one, often complicated by a lifetime of learning to hide their natural ways of thinking and interacting. Unlike children, whose behaviors are often more visible, adults have frequently developed sophisticated strategies to mask their traits, making it difficult for standard diagnostic tools to see the full picture. This challenge is compounded by the fact that many traditional assessment methods were designed for children or require in-person visits, creating barriers for those seeking answers later in life. As the understanding of autism has evolved to embrace the concept of neurodiversity—the idea that neurological differences are natural variations rather than defects—the need for assessment methods that respect an individual's lived experience while still providing clear, reliable diagnoses has become urgent. The central question for researchers and clinicians is whether a structured, remote evaluation can capture these complex adult experiences with the same accuracy as a traditional, face-to-face examination.
A team of researchers set out to test a specific approach to this problem: a structured, video-based assessment designed for adults who communicate primarily through spoken language. This method, developed by a telehealth organization, relies on a detailed interview process mapped to the official criteria used for diagnosing autism. The process involves two long sessions where clinicians ask specific questions about a person's developmental history, social interactions, and repetitive behaviors, while also paying close attention to how the person describes their own internal experiences and the effort they expend to navigate social situations. The goal was to see if this remote, neurodiversity-affirming framework could produce consistent results when reviewed by an independent expert who had not met the patient.
To find the answer, the researchers gathered a sample of one hundred adults who had recently completed this telehealth evaluation. The group was intentionally diverse, with more than half identifying as Black, Indigenous, or people of color, reflecting a conscious effort to include populations that have historically been underdiagnosed. Each participant had undergone a comprehensive evaluation involving interviews, questionnaires, and observations. After the original clinicians made their initial diagnoses, a blinded expert reviewer—a specialist with no connection to the original assessments—examined the de-identified records. This reviewer saw the same notes, transcripts, and reports but did not know what the original clinician had decided, nor did they interact with the participants directly. The study focused on whether the two independent judgments matched.
The results showed a high level of agreement between the original clinicians and the blinded expert. In eighty-six out of the one hundred cases, both raters reached the same conclusion: either both identified the person as autistic or both determined they were not. This represents an agreement rate of eighty-six percent. When the researchers looked at the consistency of these judgments using a statistical measure of reliability, the score indicated substantial agreement. The study found that the clinicians were particularly consistent when the expert also identified autism, with a very high rate of matching positive diagnoses. However, the agreement was slightly lower when both parties agreed that autism was not present, suggesting that distinguishing between autism and other conditions remains a complex task even with a structured approach.
The study also examined whether the results held up across different groups of people. While the overall agreement was strong, the researchers noted that the level of consistency was somewhat lower among participants who identified as people of color compared to those who did not. The data suggested that people of color were slightly more likely to be in the group where the two raters disagreed. The authors caution that this finding is preliminary and that the small number of people in specific subgroups makes it difficult to draw firm conclusions about any single racial or ethnic group. Instead, they view it as a signal that future assessments need to pay closer attention to how cultural context and different life experiences might influence how autism is recognized and documented.
Importantly, the researchers were careful to clarify what their study did and did not prove. Because the expert reviewer was looking at the same records created by the original clinicians, the study measured how well two people could agree when looking at the same evidence, not whether the diagnosis was absolutely correct against a separate, independent standard. The study did not test whether the telehealth method was better or worse than an in-person visit, nor did it prove that the specific interview questions were perfect. What it did demonstrate is that a structured, remote assessment can yield consistent and reproducible results when conducted by trained professionals using a clear, criterion-based framework.
The findings suggest that it is possible to conduct thorough, respectful autism assessments for adults over video without sacrificing the rigor needed for a reliable diagnosis. The protocol used in the study successfully integrated the client's own story and experiences with standard diagnostic rules, showing that a person-centered approach and a structured method can work together. While the study highlights the need for further research to ensure these methods work equally well for everyone, particularly for those from diverse backgrounds and those with different communication styles, it offers a promising path forward. It indicates that with the right training and structured tools, the growing number of adults seeking answers can receive high-quality evaluations that respect their unique perspectives while providing the clarity they need to understand themselves.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.