More Than a Checklist: How SLPs Can Improve Autism Assessment Accuracy

As speech-language pathologists (SLPs), we are often the first to receive a “suspected autism" referral, but the landscape of diagnosis is shifting beneath our feet. While the DSM-5 provides the criteria, our clinical reality is now caught between a traditional medical model and a social model that views neurodivergence as a difference rather than a disorder.

It’s tempting to open the DSM-5, check the boxes, and move toward a diagnosis. But in today’s changing society, an accurate diagnosis requires much more than a checklist. It requires us to look at the whole person, challenge our own biases, and sometimes, be the dissenting voice in the room.

The model tug-of-war: medical vs. social

We are currently operating in a world where two different lenses vie for our attention. Most of us are well-versed in the medical model. It’s logical and evidence-based; it allows us to categorize needs so we can determine which treatments are most appropriate. It is due to these categories that we can track prevalence, and we would not have the objective data to determine if a specific behavioral intervention, like Applied Behavior Analysis (ABA), actually works to achieve specific therapy outcomes.

However, we can’t ignore the social model, which has gained massive traction through social media. This perspective views neurodivergence not as a disorder, but simply as a difference that does not necessarily require intervention. While the social model promotes much-needed inclusion, it has its own caveats. It often encourages self-identification without a clinical diagnosis from a provider. 

This can be problematic, as the ability to differentially diagnose autism is difficult, even for experienced providers. More importantly, the social model can sometimes overlook the reality that autism is frequently correlated with cognitive impairments and/or language delays requiring intensive, specialized support for the individual to thrive.

Our job is to find the middle path. We use the medical model to secure the highly specialized care a child needs, while using the social model to respect a person’s preferences and autonomy.

Why diagnosis is harder today

The standard assessment process is under pressure from several modern challenges. Clinicians today face a landscape where the lines between clinical diagnosis and self-identification are increasingly blurred. While social media has increased awareness, it has also led to a rise in identification based on generalized measures.¹ Our challenge is to differentiate between common behavioral variations and the specific clinical indicators of neurodivergence. Our role is to provide the diagnostic precision that a digital checklist cannot offer. By doing so, we ensure that individuals with significant support needs receive the highly specialized, interdisciplinary care required for them to truly thrive.

At the same time, the stigma surrounding neurodivergence is shifting. In many cases, increased awareness has helped individuals feel seen, supported, and accepted for their differences. This positive cultural evolution requires clinicians to be even more precise in their reasoning. Cultural variables and varying communication habits across different backgrounds can easily skew our perception of a child's response to stimulus.² Without a robust, dynamic approach, these factors, combined with a general lack of interdisciplinary data, can lead to significant diagnostic errors.

Differential diagnosis: looking beyond the label

The core of our work isn't just identifying autism—it’s distinguishing it from other developmental disorders (DD). This is where clinical reasoning must be at its sharpest.

Many developmental disorders share overlapping features with autism spectrum disorder (ASD), including language delays, cognitive differences, and behavioral challenges. In early childhood especially, these presentations can look remarkably similar, making them difficult to clearly assess.³

That’s why differential diagnosis requires a rule-in and rule-out mindset. We’re not just asking, “Does this child meet the criteria?” We’re asking, “What else could explain this presentation?”

For example:

• A child with limited social interaction may have had minimal peer exposure rather than a social communication disorder.

• Language regression may be tied to environmental stressors or parental discord rather than neurodevelopmental differences.

• Behavioral dysregulation may reflect sensory needs, inconsistent routines, or even simple preferences rather than a divergence.

Without this level of clinical reasoning, it’s easy to misattribute symptoms to autism when another explanation is more accurate.

ASD vs. Developmental Disorder

Differentiating between ASD and general DD is one of our greatest challenges. Both are frequently correlated with:

• Cognitive impairments

• Learning and language delays

• Behavioral and physical difficulties

Because these characteristics overlap so heavily, differentiating between them in early presentation is notoriously difficult. The number and severity of these comorbidities further complicate accurate diagnosis.

The "rule-out" mindset

To avoid misidentification, we must shift from trying to "rule in" autism to utilizing a rigorous rule-out criteria. We must ask: Could these behaviors be explained by environmental factors, cultural communication norms, or a primary language disorder?

Why does this precision matter? Accurate diagnosis directly impacts access to services. When we fail to differentiate accurately, we risk placing children in inappropriate programs while those who truly need specialized autism services lose out on placement. We are doing our patients a disservice if we do not challenge each other to think deeply about the data we see.

The dream team protocol

If I’ve learned one thing in my career, it’s that a single-provider report lacks the robustness of a team effort. When we assess in a silo, we only see through one lens. 

In a perfect world, we’d have a dream team of six professionals, including psychiatry, psychology, speech therapy, occupational therapy (OT), and more. If that’s not possible, we need a triangulation of data at the bare minimum: speech, psychology, and OT.

A note on scope of practice

It’s a mistake I see often, but SLPs cannot independently diagnose autism within the medical model. Insurance requirements typically demand a diagnosis from a medical doctor, psychiatrist, or psychologist. We are a vital part of the team, but we are not the sole arbiters.

Strategic layering: beyond the single test

One of the biggest pitfalls in autism assessment is relying on a single tool. Best practice isn't about picking one test. It's about layering them to find patterns, look for bias, and ensure results are consistent across observers.

Clinical reasoning: more than just scores

Standardized tests only tell half the story. I always use informal assessments and direct observation to support or refute the test scores. To get a truly robust picture, I take data on the things tests miss:

• Transitions: How many prompts and how much time does it take to move between activities?

• Joint attention: Are they maintaining and initiating attention with a partner?

• Behavioral frequency: A tally of noncompliant behaviors, protests, and the duration of tantrums.

The environment and the parent factor

I keep my assessment space open and uncluttered. I hide stimulus items under a blanket or a table so I am the one with access to them. This allows me to see if a child is under-stimulated or over-stimulated before I introduce new toys.

I also (and I know this is controversial) often ask parents to observe from a different room or a "buffer space." Why? Because parents influence too many variables. Children often feel they have to "perform" for their parents, and parents, with the best of intentions, often prompt answers or unintentionally escalate behaviors. I need to see what that child can do independently.

Holistic inquiry: sleep, nutrition, and screens

If we aren't asking about home life, we aren't doing a complete assessment. I ask about:

1. Sleep: If a child isn't sleeping, they can’t think or maintain attention. Attention is tied to memory, and memory is tied to learning.

2. Nutrition: Does the child only eat orange foods? Crunchy foods? Nutrition influences the nervous system’s ability to regulate.

3. Screen time: Research has proven that excessive screen time in children from ages birth to three years negatively impacts language development, sleep, and attention. We need to know if we're looking at a neurodivergent brain or an environment that isn't conducive to language growth.

Case study: differential diagnosis in practice

About a year ago, I was part of an interdisciplinary team that assessed a 3-year-old boy. On the surface, the referral data pointed directly toward an autism diagnosis.

The initial presentation (Intake)

• Reported regression: The mother noted the child had stopped talking as much.

• Social withdrawal: He was terrified to be away from his mother and reportedly did not know how to interact with other children.

• Communication deficits: He demonstrated limited eye contact and a significant delay in English.

The "rule-out" investigation 

Instead of taking these markers at face value, our team investigated environmental and cultural variables. We discovered significant parental discord regarding language: the father had pressured the mother to stop speaking her first language to the child. The child felt "scared" and chose silence to avoid getting in trouble.

Furthermore, we found the child had zero prior exposure to peers—no preschool, no daycare, and no relatives in the country. His "lack of social skills" wasn't an intrinsic deficit, but a total lack of opportunity.

Clinical discovery and outcome

We observed the child in a play setting for 30 minutes, using commenting rather than questioning. Once the pressure was removed, he warmed up and engaged in reciprocal play. His limited eye contact was also determined to be consistent with cultural norms.

The determination: not autism. By utilizing the rule-out mindset, we avoided a misdiagnosis. Instead, we provided caregiver training for a language-rich home and recommended preschool to provide the peer immersion he lacked.

The service of honesty

Being an expert means being open to the possibility that the "obvious" answer might be wrong. While there is often a stigma around the word "debate," in a clinical team setting, debate is simply critical thinking in action. It is about having an open mind to information that may differ from your initial impressions.

We owe it to our patients to speak the truth in love. If we sugarcoat our findings, skip the hard questions, or fail to challenge each other to think deeply about the data, we are doing our patients a disservice. Accurate diagnosis is an act of kindness, and it ensures we are painting a whole, honest picture of the individual to determine the specific interventions that will truly help them thrive.

If you are interested in deepening your expertise in differential diagnosis and neurodiversity-affirming care, you can find more in-depth information in my Medbridge course series:

• Assessing Autism in a Changing Society: History and Diagnostic Challenges – Explore the evolution of autism diagnoses and how shifting DSM criteria influence prevalence and service access.

• Assessing Autism in a Changing Society: Differential Diagnosis Today – Learn to distinguish ASD from other developmental disorders while navigating cultural factors and misinformation.

• Assessing Autism in a Changing Society: Best Practices in Assessment – Examine the integration of medical and social models and interdisciplinary team strategies to improve diagnostic accuracy.

References

1. Hus, Y., & Segal, O. (2021). Challenges surrounding the diagnosis of autism in children. Neuropsychiatric Disease and Treatment, 17, 3509–3529. https://pubmed.ncbi.nlm.nih.gov/34898983/

2.  Lombardo, M. V., & Mandelli, V. (2022). Rethinking our concepts and assumptions about autism. Frontiers in Psychiatry, 13, 903489. https://pubmed.ncbi.nlm.nih.gov/35722549/

3. Boucher, J. (2022). Autism spectrum disorders: Characteristics, causes and practical issues (3rd ed.). SAGE Publications Ltd.

4. Muppalla, S. K., Vuppalapati, S., Reddy Pulliahgaru, A., & Sreenivasulu, H. (2023). Effects of excessive screen time on child development: An updated review and strategies for management. Cureus, 15(6), e40608. https://pmc.ncbi.nlm.nih.gov/articles/PMC10353947/