You are here:
-A A +A

Redefinition: Autism, Asperger’s, and the DSM-5

Connie Anderson, Ph.D.
IAN Community Scientific Liaison
Date Last Revised: 
December 2, 2013
Date Published: 
April 20, 2012

The way autism spectrum disorders are defined has changed, and this has caused concern among families, advocates, and providers. Will someone who is placed firmly on the autism spectrum today still be on it tomorrow? What are the implications for access to services, for research, or for an affected person’s identity? And how has the participation of thousands of families in research projects such as the Simons Simplex Collection (SSC) and the Interactive Autism Network (IAN) helped to ensure that the new definition captures the entire range of the autism spectrum?

The psychiatric manual and how it changes

A woman's hands on a book as she readsThe Diagnostic and Statistical Manual (DSM) is a psychiatric handbook. It is used by doctors, therapists, insurance companies, and others in the U.S. to diagnose mental, behavioral, and developmental disorders.1 A similar diagnostic code, the World Health Organization’s International Classification of Diseases, also describes such conditions,2 but in the U.S. the DSM is much more frequently used.

Under the direction of the American Psychiatric Association (APA), the DSM is created by groups of experts who compile checklists that describe and define various conditions, such as major depressive disorder or post traumatic stress disorder. These checklists, or criteria, are based on the combined experience of many doctors and mental health professionals, and tested in the field.

The reason the agreement of experts plays such a major role is that many disorders of the brain, such as schizophrenia, anxiety, or autism, are still not well understood in a biological sense. As one observer put it, psychiatry doesn’t have what the rest of medicine has: “the biochemical markers that allow doctors to sort the staph from the strep, the malignant from the benign.” 3 Instead, professionals dealing with psychiatric conditions judge according to DSM criteria, measuring a person’s behavior and functioning against the descriptions provided.

Every so often, the DSM is updated, as knowledge about psychiatric conditions grows and our understanding deepens. The manic depression of one era later becomes bipolar disorder I, bipolar disorder II, and other more specific conditions with more refined criteria.4 Because the DSM is created by human beings, pooling their knowledge and making judgments, social beliefs and attitudes can change the DSM, too. For example, homosexuality was listed as a mental illness in the DSM-II, but disappeared from the pages of later editions.5

The last major revision of the DSM criteria was published in 1994 (DSM-IV), with some additions to descriptive text in 2000. (The “text revised” version that’s been in use since then is known as the DSM-IV-TR.1) The publication of the fifth edition (DSM-5) came in 2013, and the criteria for all the autism spectrum disorders have changed. Some feel the descriptions of autism and related conditions in the DSM-IV were not very good and change was long overdue. Others weree worried that the changes will mean some people will lose their ASD diagnosis, and that all the careful work to count and measure rates of autism in the population will be undone.

What are the changes being made? What are the reasons for the changes? How has research informed these changes, and what will the consequences of the changes be?

How DSM-5 solves DSM-IV problems

First of all, it is important to note that changes to the DSM are being made in a very careful way. Experts working on the DSM-5 Task Force were assigned to Work Groups according to their areas of knowledge.6 The Neurodevelopmental Disorders Work Group, which is drafting the new criteria for autism, reminds us that their recommendations “reflect the work of dozens of the nation’s top scientific and research minds and are supported by more than a decade of intensive study and analysis.” 7

Families participating in the Simons Simplex Collection (SSC) and the Interactive Autism Network (IAN) projects have been part of this. The information they have generously shared has been used in studies providing information vital to the Work Group.*

The changes will address a number of problems with the way autism and related conditions are defined in the DSM-IV.

The biggest change is that existing diagnoses on the spectrum – Autistic Disorder, Asperger’s Disorder, Pervasive Developmental Disorder–Not Otherwise Specified (PDD-NOS) and Childhood Disintegrative Disorder (CDD) – are merged into one: Autism Spectrum Disorder (ASD).8 There are some important reasons for this. It turns out that clinicians using the DSM-IV have been very good at identifying that a person belongs on the autism spectrum in general, but not very consistent when choosing a specific diagnosis such as Asperger’s syndrome or PDD-NOS. What this means out in the world is that a person might be given a diagnosis of autism by one person, a diagnosis of Asperger’s by the next, and a diagnosis of PDD-NOS by a third.

A study across twelve renowned university autism clinics provided convincing evidence of this problem.9 Each clinic was working on the Simons Simplex Collection autism genetics project, and using the same tests and standardized measures to gather information about children with ASD. In fact, the project’s success depended on all of the sites doing their work using the same approach, so all the data could be brought together in a meaningful, usable way at the end. However, sites were permitted to make their best diagnosis based on clinical experience and judgment. This was because researchers wanted to test what would happen. Would these expert sites all have the same way of telling autism, Asperger’s, and PDD-NOS apart?

The results were striking. There were few differences in results of standardized autism assessments and measures between sites. But when it came to deciding who had autism, Asperger’s, or PDD-NOS, the sites were not consistent at all. They were not random, because each site seemed to have its own formula for distinguishing the different ASDs, but that formula was not the same across sites. They weighed different factors, such as a child’s age or IQ, in different ways and came up with different answers.

Considering this, research based on these categories may not be that meaningful, especially epidemiological research that comes up with rates of autism, Asperger’s, or PDD-NOS. If the same person might be placed in a different “pot” by different professionals, it’s a problem.

This provides a good reason for collapsing these rather mushy categories into one. The idea is that all conditions on the autism spectrum are based on some definite and unusual behaviors and traits. People who share these core features should not be split into groups based on factors that aren’t essential to ASD, such as their IQ or whether they talked on time. Instead, such features will be noted as “clinical specifiers” that help describe a person’s condition and ability to function.8

The DSM-5 collapses the DSM-IV’s three essential “autism domains” into two. Restrictive and repetitive behaviors and interests will still be a domain, but social deficits and communication deficits, which used to be separate, are being combined. This makes sense, as people cannot communicate without being social, and cannot be social without communicating. It was always a challenge to tease these two factors apart. Several social/communication criteria are being merged and streamlined. Language delay, which often occurs in people who are not on the spectrum, and doesn’t always occur in people who are on the spectrum, is removed. Also, because people on the spectrum are so diverse in terms of their gifts and challenges, severity levels are assigned to each domain.8 Lastly, people with the social/communication deficits of autism but no issues with restricted and repetitive behaviors, who might have received a PDD-NOS diagnosis in the past, will now receive a new diagnosis called Social Communication Disorder instead of ASD.10

The DSM-5 acknowledges in the criteria that sensory issues are a key feature of autism, and makes them part of the restrictive and repetitive behaviors and interests domain. Sensory features are mentioned in the explanatory text of DSM-IV-TR, but not in the criteria list. They are called out specifically in DSM-5, which adds to the checklist: “Hyper- or hypo-reactivity to sensory input or unusual interest in sensory aspects of the environment (such as apparent indifference to pain/heat/cold, adverse response to specific sounds or textures, excessive smelling or touching of objects, fascination with lights or spinning objects.)” 8

The DSM-5 criteria for ASD and attention deficit hyperactivity disorder (ADHD) make it possible for an individual to be diagnosed with both.11 This acknowledges the fact that many children are diagnosed with both ASD and ADHD, despite the fact that the DSM-IV specifically states that a diagnosis of ADHD cannot be made in someone with autism.1

Debates and nervous anticipation

The idea that the definition of autism, Asperger’s, and the other conditions is changing has led to concern among many, from families to psychotherapists to educators. What will the changes mean when implemented in the real world?

A major fear voiced by families of children with ASD is that their child will lose a diagnosis, as well as the services and help that come with it. Unfortunately, the early release in the media of one unpublished research study’s results, shared at a conference in Iceland, fueled these concerns.12 The study claimed that a large number of high-functioning individuals would lose their ASD diagnosis. However, the study was based on fairly old data – from children diagnosed in 1993 – and experts rushed to reassure families that any conclusions were premature.13 At the same time, advocacy organizations issued position statements making it clear that they were paying attention, and were being especially watchful for any negative consequences caused by changes to the DSM.14,15,16

On a live computer chat held by the staff of Autism Speaks, families shared their worries. One mom said, “I am very scared about these new diagnostic criteria. My son is PDD-NOS. I have read as many as 85% of those children will be ruled out with the new criteria. My son has a 2-year developmental delay. He is ten, and reads at a first grade level, cannot write legibly, already receives basically no services outside of Supplemental Security Income (SSI) and Medicaid. What will happen to all of these kids?” The staff at Autism Speaks reassured her, saying that children with clear delays will certainly not lose access to services, and that no one will be re-diagnosed the second the DSM-5 comes out in any case. (Generally, that only happens at some key point when a person needs to qualify for services of some kind.)17

Another mom was afraid that her child would lose his diagnosis because intensive therapies had helped him so much. Would he be “kicked out” because he was doing so well, even though he still needed therapy to gain new ground? This was harder to answer, as there has been more debate about what might happen to high functioning children, especially those who have made great strides thanks to interventions.17

Those working on the DSM-5 were aware of these fears. Francesca Happé, a member of the Neurodevelopmental Disorders Work Group, wrote, “A major concern for us in the work group is that no individual currently diagnosed with Asperger disorder or PDD-NOS who needs support will lose that support because of changes to the DSM-5. It is our intention that all individuals with clinical levels of social-communicative impairment plus restricted/repetitive behavior will meet criteria for ASD, and that their individual levels of intellectual and language functioning will be noted and considered alongside this diagnosis.” 10

Research guides the process

Of course, the revisions to any edition of the DSM aren’t put into place without review. First of all, there is field testing. Before a new edition of the DSM is finalized, therapists and clinicians from all over the U.S. volunteer to use the draft criteria for a limited time to see how well they work in the real world. If problems are identified, changes can be made.

Researchers also take a hand, testing the new criteria in a variety of ways. This has certainly been the case with the new ASD criteria.

For example, a British study used statistics to analyze autism symptoms among 708 individuals with mild to severe autistic features. They wanted to find out if the DSM-5 criteria were a better “fit” than the DSM-IV criteria.18 They found the DSM-5 criteria, with its two domains instead of three, are superior because “among higher-functioning individuals, ASD is a dyad, not a triad, with distinct social communication and repetitive behavior dimensions.” They also found that the sensory aspects of ASD did indeed “fit” in the restrictive and repetitive behaviors and interests domain.

A study conducted in Finland evaluated more than 4,000 children according to both the DSM-IV criteria and the proposed DSM-5 criteria (which have already changed since then, due to earlier feedback received.)19 They found that the DSM-5 did not identify as many individuals with Asperger’s and high functioning autism as the DSM-IV did. For example, out of 11 children with autism who have an IQ of 70 or above, only 8 would keep their diagnosis under the DSM-5. Out of 11 children with an Asperger’s diagnosis under DSM-IV, none would receive an ASD diagnosis under DSM-5. The researchers recommended minor changes to the DSM-5 criteria which they showed would correct this.

Focusing on similar issues, a U.S. study drew on data from the thousands of families participating in the IAN project.20 Using information about symptoms and behaviors of 8,911 children on the autism spectrum and 5,863 of their unaffected siblings, these researchers set out to evaluate the DSM-5 criteria for ASD. They found that a model that includes a category (ASD vs. non-ASD) and two symptom domains (restricted/repetitive behavior and social/communication) fits the data best, providing support for the DSM-5’s view of ASD.

Like the Finnish researchers, however, they found that the DSM-5 criteria should be slightly less strict. According to the current criteria, about 12% of less severely affected individuals, including those with Asperger’s and girls, would be missed. (Girls with ASD often appear less impaired than boys with ASD, but quite impaired when compared to typical girls;21 under-diagnosis of girls has been a problem for a while. Read related article.) The study team asserted that requiring one less symptom from either the restricted/repetitive behavior or social communication domain would give the DSM-5 the best outcome of all. This is the ideal point where there are both few “false positives” (that is, people who don’t really have a condition but get a diagnosis) and few “false negatives” (that is, people who have a condition but aren’t given a diagnosis).

An issue of identity: Asperger’s

For some, the Asperger’s label especially has become a touchstone – shorthand for a certain type of high-functioning autism and a key to identity. There are Asperger’s support groups, Asperger’s discussion forums, Asperger’s camps, and Asperger’s school programs. Temple Grandin – a famous person on the autism spectrum and the subject of an HBO movie22 – said she’d “throw PDD-NOS in the garbage can,” but was much more reluctant to let go of Asperger’s.23

Fortunately, the loss of scientific meaning doesn’t equal the loss of meaning altogether. The term “Asperger’s” will still be available to those who find it valuable. It just won’t be a medical diagnosis.

Thanks to families

Families, advocates, and researchers are keeping a close eye on the evolution of “Autism Spectrum Disorder” in the DSM-5. With their input and their feedback, the final product should serve the autism community well.

We would like to offer a special thank you to families taking part in the Simons Simplex Collection project and IAN. Your participation in either study (or both!) has made it scientifically possible to test the DSM-5 criteria, and to improve them.

*See, for example, “A Multisite Study of the Clinical Diagnosis of Different Autism Spectrum Disorders,” 9Stability of Initial Autism Spectrum Disorder Diagnoses in Community Settings,” 24Trends in Autism Spectrum Disorder Diagnoses: 1994-2007,” 25 and “Validation of Proposed DSM-5 Criteria for Autism Spectrum Disorder.” 20

Please rate the helpfulness of this article: 
Average: 4.7 (15 votes)
  1. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (DSM-IV-TR) (4th ed., text revision ed.). Washington, DC: American Psychiatric Association.
  2. American Psychological Association. (2009). ICD vs. DSM. Monitor on Psychology, 40(9), 63. Retrieved April 18, 2012. View Article
  3. Greenberg, G. (2012, January 29, 2012). Not diseases, but categories of suffering. New York Times. Retrieved April 18, 2012. View Article
  4. Goodwin, F. K., & Jamison, K. R. (2007). Manic-depressive illness: Bipolar disorders and recurrent depression (Second Edition ed.). New York, New York: Oxford University Press.
  5. Drescher, J. (2010). Queer diagnoses: Parallels and contrasts in the history of homosexuality, gender variance, and the diagnostic and statistical manual. Archives of Sexual Behavior, 39(2), 427-460. View Abstract
  6. American Psychiatric Association. (n.d.). DSM-5 task force members. Retrieved March 23, 2012. View Article
  7. American Psychiatric Association. (2012). News release: DSM-5 proposed criteria for autism spectrum disorder designed to provide more accurate diagnosis and treatment. Arlington, Virginia: American Psychiatric Association. Retrieved April 16, 2012. View News Release
  8. American Psychiatric Association. (n.d.). DSM-5 development: Autism spectrum disorder. Retrieved March 23, 2012. View Proposed Revision
  9. Lord, C., Petkova, E., Hus, V., Gan, W., Lu, F., Martin, D. M., et al. (2012). A multisite study of the clinical diagnosis of different autism spectrum disorders. Archives of General Psychiatry, 69(3), 306-313. View Abstract
  10. Happe, F. (2011). Criteria, categories, and continua: Autism and related disorders in DSM-5. Journal of the American Academy of Child and Adolescent Psychiatry, 50(6), 540-542. View Abstract
  11. American Psychiatric Association. (2010). DSM-5: Options being considered for ADHD. Retrieved 3/26, 2012. View Article
  12. Schwartz, C. (2012, January 20, 2012). A new definition of autism could exclude many now diagnosed, expert says. The Daily Beast. Retrieved April 18, 2012. View Article
  13. Carey, B. (2012, January 19, 2012). New definition of autism will exclude many, study suggests. New York Times. Retrieved April 18, 2012. View Article
  14. Autism Society. (2012, January 20, 2012). The autism society comments on the proposed DSM-5 revisions: Changing the definition of autism does not change the need for help. Retrieved April 18, 2012. View Statement
  15. Autism Speaks. (2012, February 2, 2012). Autism speaks statement on proposed revisions to the DSM definition of autism spectrum disorder. Retrieved April 18, 2012. View Statement
  16. National Autism Association. (2012, January 27, 2012). Press release: Changes in DSM-5 autism definition could negatively impact millions - autism organizations concerned that autism diagnostic changes will jeopardize services, impair tracking, and disrupt research around the globe. Retrieved April 18, 2012. View Statement
  17. Autism Speaks. (2012, January 20, 2012). Live chat with Geri Dawson, PhD & Lisa Goring analyzing DSM-5. Retrieved April 18, 2012. View Chat
  18. Mandy, W. P., Charman, T., & Skuse, D. H. (2012). Testing the construct validity of proposed criteria for DSM-5 autism spectrum disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 51(1), 41-50. View Abstract
  19. Mattila, M. L., Kielinen, M., Linna, S. L., Jussila, K., Ebeling, H., Bloigu, R., et al. (2011). Autism spectrum disorders according to DSM-IV-TR and comparison with DSM-5 draft criteria: An epidemiological study. Journal of the American Academy of Child and Adolescent Psychiatry, 50(6), 583-592.e11. View Abstract
  20. Frazier, T. W., Youngstrom, E. A., Speer, L., Embacher, R., Law, P., Constantino, J., et al. (2012). Validation of proposed DSM-5 criteria for autism spectrum disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 51(1), 28-40.e3. View Abstract
  21. Constantino, J. N., Zhang, Y., Frazier, T., Abbacchi, A. M., & Law, P. (2010). Sibling recurrence and the genetic epidemiology of autism. The American Journal of Psychiatry, 167(11), 1349-1356. View Article
  22. Ferguson, S., & Saines, E. G. (Producers). (2010, February 6, 2010). Temple Grandin. [Motion Picture] United States: HBO Films. View Trailer
  23. Wallis, C. (2009, November 3, 2009). A powerful identity, a vanishing diagnosis. New York Times. Retrieved April 18, 2012. View Article
  24. Daniels, A. M., Rosenberg, R. E., Law, J. K., Lord, C., Kaufmann, W. E., & Law, P. A. (2011). Stability of initial autism spectrum disorder diagnoses in community settings. Journal of Autism and Developmental Disorders, 41(1), 110-121. View Abstract
  25. Rosenberg, R., Daniels, A., Law, K., Law, P., & Kaufmann, W. (2009). Trends in autism spectrum disorder diagnoses: 1994-2007. Journal of Autism and Developmental Disorders, 39(8), 1099-1111. View Abstract