The long-awaited ICD-11 , a revision of the previous version ICD-10, has now been fully available for several months. The goal of classification systems is to improve correct diagnosis and thus ensure the indication and implementation of evidence-based therapies and care. The ICD (International Classification of Diseases) is the World Health Organization’s (WHO) classification system for mental disorders and somatic diseases, while the DSM (Diagnostic and Statistical Manual of Mental Disorders) is the American Psychiatric Association’s (APA) classification system. The latter has been available in its 5th edition since May 2013. For a long time, it was assumed that the ICD-11 would not differ significantly from those in the DSM-5 with regard to the diagnostic criteria for autistic disorders; accordingly, this was also communicated in scientific publications. The diagnostic criteria and their formulations for autism were widely discussed in advance, but were not made publicly available until they came into force on January 1, 2022.
- A central innovation concerns the classification of autistic disorders under the “disorders of neural and mental development” (DSM-5) or the “neural developmental disorders (ICD-11, “neurodevelopmental disorders”). These groups include, in addition to autism spectrum disorder, the 1) disorders of intelligence development, 2) disorders of speech and language development, 3) disorders of learning development, 4) developmental disorder of motor coordination, 5) attention deficit/hyperactivity disorder, and 6) stereotypic movement disorders. The grouping of autistic disorders into “neurodevelopmental disorders” was done because all of the above disorders often occur together and share many characteristics. Tic disorders, including Tourette syndrome, although previously listed with developmental disorders because of their frequent co-occurrence, were from now on classified with nervous system disorders.
- In accordance with the current state of research, the term autism spectrum disorder (ASD) is listed as an independent classification. The distinctions into eight differentiable subgroups (e.g., early childhood autism, Asperger syndrome, atypical autism) that were still included in ICD-10 and DSM-5 have been abandoned and combined into a single category. This is consistent with numerous studies showing that distinctions between subgroups have dubious diagnostic value and represent quantitative rather than qualitative differences. In addition, analogous to DSM-5, ICD-11 now adds various very detailed subclassifications (“specifiers”) to the diagnosis of ASD with respect to, for example, language and cognitive abilities, which takes into account the enormous heterogeneity of the disorders.
- The age at onset of the disorder is described as “early in development,” and the criterion “by age 3” is omitted.
- Comorbidity with ADHD is officially admitted.
A major criticism – and also major difference between DSM-5 and ICD-11 is that ICD-11 does not specify a required set of criteria for diagnosis. It only mentions features (“essential (required) features”) that characterize autism spectrum disorder: In autism spectrum disorder, “persistent deficits in the ability to initiate and maintain reciprocal social interactions and social communication are manifested by a range of restricted, repetitive, and inflexible behaviors, interests, or activities that are clearly atypical or excessive for the person’s age and sociocultural context. The onset of the disorder is developmental, typically in early childhood, but symptoms may fully manifest later when social demands exceed limited abilities (…)” . In the DSM-5 we formulate this similarly, but at the same time we emphasize that if there is evidence of good social and communication skills in childhood, the diagnosis may not be given. This is not required in the ICD- 11. Further, in both classification systems, the deficits must be persistent and severe enough to result in impairment in personal, family, social, educational, occupational, or other important areas of functioning and must be a consistent feature of the person’s functioning, observable in all domains, to warrant a diagnosis. There are then described in ICD-11 a number of “manifestations” that may be present (“manifestations may include limitations in the following…”). However, whether these sharpen the distinction from other mental and/or developmental disorders remains doubtful.
In the DSM-5, the criteria are operationalized more precisely in quantitative terms. A necessary number of abnormalities in the area of social communication and social interaction as well as in the area of restricted, repetitive behavior patterns, interests or activities is mentioned. Thus, in the area of social communication, there must necessarily be both 1. deficits in social-emotional reciprocity as well as 2. deficits in nonverbal communication behavior and 3. deficits in establishing, maintaining, and understanding relationships. In the area of restricted, repetitive behaviors, interests, or activities, it is specified that at least two of 4 characteristics must be present.
Although the criteria in both DSM-5 and ICD-11 are formulated rather generally and vaguely, DSM-5 at least specifies that a minimum set of features must be present. In ICD-11, the only requirement is that symptoms from each domain must be present, but this is not further specified. If only quite general symptom descriptions are used in the diagnostic assessment and no minimum level of abnormality is required, this reduces the specificity of the ASD diagnosis in differentiating it from other developmental and mental disorders. As a consequence, this means that in case of improper and non-guideline-compliant diagnostics (see below), the differentiation from other mental problems to the diagnosis of an autism spectrum disorder according to ICD-11 becomes significantly more difficult and the diagnosis could be assigned quite arbitrarily. However, this criticism also applies to the DSM-5. In the long run, there is a risk of increasing “meaninglessness” of the diagnosis, as it would hardly be linked to a specific symptomatology and thus implied evidence-based therapy option. It would also be feared that the already insufficiently available therapy and support services would no longer reach those with actual need.
The further explanations regarding the boundaries with normality (“boundary with normality (threshold)”) as well as the differential diagnostic boundaries (“boundaries with other disorders and conditions (differential diagnosis)”) are to be mentioned positively. Here, significantly more disorders are named than in the DSM-5 . However, it remains questionable whether these brief explanations help to compensate for the imprecise criteria and their non-specificity. Regarding the aforementioned subclassifications, which are of course necessary to further differentiate the formerly existing various autistic disorders that are now combined into one category, there are currently few, if any, studies on the clinical utility of these “specifiers”, or on whether they are actually the correct specifiers. Conceptualizing a single disorder with multiple subclassifications should therefore primarily encourage studies that examine the substantial heterogeneity of ASD both between different individuals with ASD and within the same individual with ASD over time.
To improve the correct diagnosis, indication and implementation of evidence-based therapies and care, the WGAS recommends to closely follow the diagnostic (https://www.awmf.org/leitlinien/detail/ll/028-018.html) and therapeutic (https://www.awmf.org/leitlinien/detail/ll/028-047.html) guidelines of the AWMF. Both with regard to diagnostics and therapy, a very structured, small-step, goal-oriented and differentiated approach is recommended, which certainly has the potential to improve the care of people with autism spectrum disorders and also to promote clarity in the diagnostic process. From the point of view of the WGAS, it should be demanded that when a diagnosis is made according to ICD-11, detailed reasons should be given why the symptomatology cannot be explained by one or more of the differential diagnoses mentioned.
 The following disorders with clear overlap to the symptomatology of autism spectrum disorder are mentioned: intellectual impairments, language disorders, motor disorders, attention disorders, stereotypic movement disorders, schizophrenia, schizotypal disorder, social anxiety disorders, selective mutism, obsessive-compulsive disorder, attachment disorders (reactive/disinhibitory), restrictive eating disorders, oppositional disorders, personality disorders, tic disorders, and Tourette syndrome, as well as overlap with other medical disorders or secondary syndromes.