What are the strengths and weaknesses of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR) and International Classification of Diseases, Eleventh Revision (ICD-11) in diagnosing Autism Spectrum Disorder (ASD)?

Comparative Analysis of DSM-5-TR and ICD-11 for Autism Spectrum Disorder Diagnosis

Both DSM-5-TR and ICD-11 represent significant advances in autism diagnosis, but ICD-11's broader symptom combinations risk reducing diagnostic precision and increasing false positives, while DSM-5-TR maintains better specificity through more restrictive criteria, making it preferable for research contexts despite ICD-11's superior clinical utility features. 1

Major Structural Changes from Previous Versions

DSM-5-TR Evolution

• Eliminated the subcategories of autism (Asperger's disorder, PDD-NOS, autistic disorder) that existed in DSM-IV-TR, consolidating them into a single "Autism Spectrum Disorder" diagnosis 2
• Collapsed the previous three-domain model (social relatedness, communication/play, restricted interests) into a two-domain structure: (1) social communication/interaction deficits and (2) restricted, repetitive behaviors 2
• Added sensory sensitivities as a core diagnostic criterion within the restricted/repetitive behavior domain, recognizing that up to 90% of autistic individuals experience atypical sensory perception 2
• Introduced three severity levels to capture functional impact, though these levels show significant limitations in capturing within-category heterogeneity 3

ICD-11 Innovations

• Expanded from 11 to 21 disorder groupings in the mental, behavioral, and neurodevelopmental disorders chapter, providing more comprehensive classification 2, 4
• Eliminated the separate "childhood and adolescence onset" disorder grouping, moving autism to other categories to highlight developmental continuity across the lifespan 2
• Introduced dimensional symptom specifiers that allow rating severity across multiple domains at each assessment, moving beyond purely categorical diagnosis 2, 4
• Added course qualifiers with two components: episodicity and current clinical status, which inform treatment intensity decisions 5

Strengths of Each System

DSM-5-TR Strengths

• Maintains diagnostic precision through more restrictive criteria, reducing the risk of false positives that could limit access to ASD-specific services 1
• Better suited for research contexts where homogeneous samples are needed for studying etiology and biological pathways 1
• Provides detailed exclusion criteria to differentiate ASD from other conditions, including insight specifiers to guide treatment planning 5
• Cross-cultural validity has been empirically demonstrated in Finnish and UK samples, with the DSM-5 model fitting well and outperforming DSM-IV models 6
• Emphasizes observable, behavioral, and neurodevelopmental features that can be measured more objectively 1

ICD-11 Strengths

• Superior clinical utility with 82.5-83.9% of clinicians rating it as quite or extremely easy to use, accurate, clear, and understandable 4
• Dimensional qualifiers provide actionable information for treatment planning that categorical diagnoses alone cannot capture 5
• Allows systematic tracking of treatment response through severity ratings applied at each assessment 5
• Better suited for routine clinical practice and international epidemiological data collection 5
• Course pattern documentation guides long-term management strategies more effectively 5

Critical Weaknesses and Limitations

DSM-5-TR Weaknesses

• Large within-category heterogeneity despite attempts at severity levels—two children at the same severity level can have vastly different neuropsychological profiles 3
• Severity levels focus on intensity but not quality of symptoms, limiting ability to formulate reliable prognoses or plan individualized treatment 3
• More restrictive focus on research settings may reduce accessibility in real-world clinical contexts 5
• Remains fundamentally categorical despite dimensional additions, with arbitrary boundaries limiting reliability 4
• No biological validation—classifies based on observable symptoms rather than underlying pathophysiology 4

ICD-11 Weaknesses

• Allows extreme variety in symptom combinations, potentially reducing specificity and increasing heterogeneity to problematic levels 1
• Moves toward subjective inner experiences rather than observable behaviors, making objective measurement difficult 1
• Contains vague concepts leading to non-falsifiable diagnoses with high risk of false positives 1
• May be difficult to differentiate from other mental disorders and autism-like traits in clinical practice 1
• Among individuals with broader autism phenotype, the model fitted poorly in Finnish samples, suggesting cross-cultural variability for milder characteristics 6
• Could hamper research replication due to increased non-specificity and heterogeneity of diagnosed samples 1

Shared Fundamental Limitations

• Neither system has biological validation, resulting in biologically heterogeneous groups within the same diagnostic category 4
• Both classify mental phenomena based on self-reported or clinically observable symptoms rather than underlying pathophysiology 4
• Arbitrary boundaries between diagnostic categories persist despite efforts toward dimensionality 4
• Changes from previous versions remain relatively modest at their core 4
• Neither adequately addresses overlap with other neurodevelopmental conditions—approximately 50% of children with ASD also meet criteria for ADHD, yet diagnostic guidance for comorbidity remains limited 7

Specific Clinical Challenges

Comorbidity Recognition

• The mutual exclusion between ASD and ADHD has been eliminated, but both systems provide insufficient guidance for managing this common comorbidity 7
• More than 90% of individuals with autism have at least one comorbid medical or mental health condition, requiring screening for anxiety, depression, learning disabilities, language disorders, sleep disorders, oppositional defiant disorder, conduct disorders, and tic disorders 7
• Neither system adequately addresses how comorbid conditions alter diagnostic presentation or treatment approach 7

Assessment Reliability Issues

• IQ testing results are less reliable for individuals with severe ID or language impairment, as fewer such individuals were included in establishing score ranges 2
• In children under 5 years, IQ measures are not reliable, requiring use of "Global Developmental Delay" terminology instead 2
• Performance on standardized testing can be underestimated in cultural and linguistic minorities, as instruments may lack sensitivity for these populations 2

Recommended Improvements for Future Iterations

Enhance Diagnostic Precision

• Develop more finely grained, objective clinical symptom characterization that is relatable to neurobehavioral concepts rather than subjective inner experiences 1
• Identify core ASD subtypes/endophenotypes through quantitative description and objective measurement of symptoms 1
• Create explicit differential guidance for distinguishing ASD from conditions with overlapping features, particularly in geriatric populations where catatonic features and age-related psychomotor changes overlap 4

Improve Dimensional Approaches

• Expand dimensional severity scores beyond current limited domains to capture quality of symptoms, not just intensity 3
• Include dimensional ratings in statistical analyses to facilitate international comparison and meta-analysis 5
• Develop standardized dimensional measures that maintain reliability across cultural contexts 6

Address Comorbidity Systematically

• Provide explicit algorithms for diagnosing ASD when ADHD symptoms are present, given the 49% methylphenidate response rate in children with ASD and hyperactivity 7
• Include mandatory screening protocols for common comorbidities (anxiety, depression, sleep disorders) as part of diagnostic assessment 7
• Specify how comorbid conditions modify diagnostic thresholds and treatment approaches 7

Enhance Cross-Cultural Validity

• Conduct systematic cross-cultural validation studies beyond North American and UK samples, particularly for milder autistic characteristics where variability is greatest 6
• Develop culture-specific normative data for standardized assessment tools like ADOS, ADI-R, and M-CHAT 7
• Address linguistic and cultural factors that affect testing reliability in minority populations 2

Integrate Biological Markers

• Move toward Research Domain Criteria (RDoC) approach that links symptom domains to underlying biological and neurological mechanisms 8
• Incorporate precision medicine approaches based on precise diagnostic markers rather than purely behavioral observation 1
• Develop biomarkers that can validate diagnostic categories and reduce heterogeneity within diagnostic groups 1

Improve Functional Outcome Measurement

• Measure functional outcomes as primary endpoints rather than symptom reduction alone, as recommended by the National Institute of Mental Health 5
• Include adaptive functioning measures that capture real-world impact across developmental stages 2
• Develop longitudinal tracking systems that document how support needs change over time with intervention 2

Common Pitfalls to Avoid

• Do not rely solely on IQ composite scores—the profile of IQ subtests reveals cognitive strengths and weaknesses more effectively 2
• Avoid diagnosing based on single-source information; obtain data from multiple sources (parents, teachers, mental health professionals) documenting symptoms in more than one environment 7
• Do not assume diagnostic stability—reassess at least every 3 years in school-aged children as federal law requires, since support needs may change with intervention 2
• Recognize that improved adaptive skills may result from environmental modifications rather than changes in underlying condition 2
• Avoid using "Unspecified Intellectual Disability" as a permanent diagnosis when assessment is difficult due to sensory, physical, or behavioral factors—continue reassessment when possible 2