The American Psychiatric Association (APA) has revised its diagnostic manual, known as the Diagnostic and Statistical Manual (DSM). The fifth edition (DSM-5) was published on 18 May.

The DSM is one of the two main international sets of diagnostic criteria for autism. It is influential, but it is not the main set used in the UK.

This article looks at the key changes in DSM-5 relating to autism and Asperger syndrome, and what The National Autistic Society thinks about the changes. At the bottom of the page you'll find links to further information including an opportunity to see the web chat we held with Professor Gillian Baird.

What does DSM-5 say?

Diagnostic criteria are revised periodically by a team of experts, taking into account the most up-to-date research.

DSM-5 includes some changes which could affect the way diagnoses will be given to people on the autism spectrum. However, it's important to add that although DSM is influential, the main set of criteria used in the UK is the World Health Organisation’s International Classification of Diseases (ICD). We don’t expect that there will be any immediate changes to the way that autism and Asperger syndrome are diagnosed in this country.

Here are some of the main changes in DSM-5.

  • In DSM-5, the terms  ‘autistic disorder’, ‘Asperger disorder’, ‘childhood disintegrative disorder’ and ‘PDD-NOS’ have been replaced by the collective term 'autism spectrum disorder'. For many people, the term ‘Asperger syndrome’ is part of their day-to-day vocabulary and their identity, so we understand concerns around the removal of Asperger syndrome as a distinct category from the manual. All individuals who currently have a diagnosis on the autism spectrum, including those with Asperger syndrome, will retain their diagnosis. No one will ‘lose’ their diagnosis because of the changes in DSM-5.
  • The previous use of three domains of impairments has been reduced to two domains: 
    - social communication and interaction.
    - restricted, repetitive patterns of behaviour, interests or activities.
  • Sensory behaviours are included in the criteria for the first time, under the 'restricted, repetitive patterns of behaviours' descriptors.
  • The emphasis during diagnosis will change from giving a name to a condition to identifying all the needs that someone has and how these affect their life.
  • DSM-5 has introduced 'dimensional elements' which give an indication of how much someone’s condition affects them. This will help to identify how much support an individual needs.
  • DSM-5 includes a new condition called 'social communication disorder' (more details below). 

What does the NAS think of DSM-5?

The new diagnostic criteria

The National Autistic Society welcomes the overall approach to streamline diagnostic criteria and make them simpler, to develop dimensional measures of severity and to recognise the full range of health problems someone is experiencing, as well as any other factors that impact on their diagnosis.

Severity levels

DSM-5 lists three levels of severity in each of the two domains ('social communication' and 'restrictive, repetitive behaviours').

The levels are:

Level 1 - requiring support
Level 2 - requiring substantial support
Level 3 - requiring very substantial support

We raised a number of concerns regarding the severity levels at the proposal stage, including:

1. the level of details 
2. consistency of the levels in relation to the diagnostic criteria 
3. the absence of key areas, including sensory aspects, within the severity levels.  
4. the criteria for the severity
5. the linking of a clinical diagnosis to recommendations of support may create expectations for people on the autism spectrum that services will be provided when this will not always be the case (at least in the UK), due to high eligibility thresholds or because decisions about such support may be taken by professionals who have no relation to the diagnostic process.
6. creating a direct link between a clinical decision over diagnosis and a recommendation for support could affect clinical impartiality. In the UK we are aware of situations where clinical professionals have felt under pressure from their employers to under-assess needs in order to ration limited resources.

While there have been improvements in the final version of DSM-5, we believe that the complexity of the presentation of autism spectrum disorders means that the severity of impairment can only be assessed on an individual basis. It must not be linked to simplistic recommendations for support.

DSM-5 does explain that severity levels may vary by context and also fluctuate over time, and it also states that the descriptive severity categories should not be used to determine eligibility for and provision of services stating that 'these can only be developed at an individual level and through discussion of personal priorities and targets'.

Asperger syndrome and other sub-groups

Our experts, Dr Lorna Wing and Dr Judith Gould, submitted a paper to the American Psychiatric Association, jointly written with Professor Christopher Gillberg.

The paper called for a stronger focus on social imagination, diagnosis in infancy and adulthood, and on the possible under-diagnosis of girls and women with autism. The authors recommended that sub-group names for particular autism spectrum disorders be kept in DSM-5, including a description of Asperger syndrome, to make it very clear that this continues to be a part of the autism spectrum. 

Early in 2010, The National Autistic Society asked for people’s views on the proposals so that we could also respond to the APA. We said that any changes could have an impact on people’s sense of identity, and that it would be important to consider the views of people with an autism spectrum disorder when revising the manual.

We also said that it would be important to recognise when people on the autism spectrum have to make a considerable effort to manage in a 'neurotypical' world - even if they appear to cope on a superficial level.

Social communication disorder

The APA has created a new diagnosis of social communication disorder. This is characterised by difficulties with verbal and non-verbal communication that cannot be explained by low cognitive ability. It includes difficulty in learning and using spoken and written language, as well as inappropriate responses in conversation. It does not include restricted, repetitive patterns of behaviour, interests or activities. The disorder limits effective communication, social relationships, academic achievement or occupational performance.

From our experience of assessing people with social and communication disorders, we believe that communication problems are rarely the basis of difficulties with social interaction, but communication problems are often rooted in difficulties with social communication and interaction. Moreover, our assessments suggest that in fact these people usually do have restricted or repetitive behaviours and interests, but have been able to mask them, particularly where someone is more able intellectually.   

Therefore, though social communication disorder is not part of the autism spectrum, we believe that it is likely that this group could be a sub-group of people on the autism spectrum. When the aim of DSM-5 is to avoid having autism sub-groups, we do not believe it is helpful to have created this additional diagnosis of social communication disorder.

Further information