Autism in the DSM-5

, Marcy Willard¹ and Helena Huckabee¹

(1)

Emerge: Professionals in Autism, Behavior and Personal Growth, Glendale, CO, USA

Abstract

Clinical assessment is crucial for it is not only about diagnosis, but also a guide regarding treatment and prognosis. Classification is an important tool for communication between researchers, clinicians, and policy makers to ensure that they are referring to the same features, symptoms, and traits, when making a diagnosis (J Child Psychol Psychiatry 52(6):647–660, 2011a; J Autism Dev Disord 41(4):395–404, 2011b). Planning for the Diagnostic and Statistical Manual, Fifth Edition, (DSM-5) began in 2003 and included a series of conferences and work groups with experts from many disciplines and from locations all around the world. In his research review written in 2011, Rutter notes that a better attempt was made to review the science for DSM-5 than for DSM-IV or ICD 10. This review focuses only on one aspect of DSM-5, the diagnosis of ASDs. Rutter cites that subcategories of ASD “manifestly do not work” and argues for their elimination. Much research was reviewed in writing this text citing a substantial body of evidence that Asperger’s and Autism are not qualitatively distinct; rather different quantitative manifestations of the same disorder (J Autism Dev Disord 40(8): 921–929, 2010; J Child Psychol Psychiatry 40(2): 219–226, 1999; J Autism Dev Disord 34(4): 367–378, 2004; J Am Acad Child Adolesc Psychiatry 37(3): 271–277, 1998). This chapter examines the new severity classifications for diagnosing ASD which are different from the previous classification of Autism versus Asperger’s Disorder. The new criteria introduced in DSM-5 for identifying Social (Pragmatic) Communication Disorder is included in this chapter as well. Diagnostic criteria for ASD, including severity levels on a continuum of cognitive, language, and other skills, are discussed.

Keywords

ASD in the DSM-5Severity levels in ASDCognitive impairments in ASDSocial impairments in ASDSocial Communication Disorder vs. ASDWhat happened to Asperger’s?Can you keep your Asperger’s diagnosis?Diagnosing autism today

Studies on the diagnostic criteria for separate conditions along the Autism Spectrum date back to the late 1990s, suggesting that researchers have been considering whether or not to segment the Spectrum in this way for some time. This research is consistent with Rutter’s argument to eliminate subcategories, which the DSM-5 subcommittee did in 2013. Szatmari’s et al. (2009) for a lack of structural language impairment being the differentiating factor in Asperger’s and Tsai’s argument that PDD-NOS is distinctly different from Autism do argue against such a change, but seem to be in the minority. Some researchers in 2014 argue for an even broader categorization under neurodevelopmental disabilities because of the vast heterogeneity within Autism Spectrum Disorders. While this may work on some level, we need diagnosis to inform treatment and in order for third party payers to cover a portion of treatment cost. The steps DSM-5 takes to align diagnosis with the research seem to be appropriate. As clinicians become more familiar and learn to apply the new criteria appropriately, less controversy will remain. The International Meeting for Autism Research (IMFAR) is evidence of this phenomenon as the discussion of DSM-5 was minimal at IMFAR 2014 and much more prominent 2 years earlier in 2012. Volkmar, Klin, and McPartland (2014) argue that research and practice would benefit most if within this overarching category of “autisms” homogeneous subtypes of the disorder could be identified (p. 28 Eds. Volkmar et al., 2014). For now the authors remind us that DSM-5 states that those who met criteria under DSM-IV should continue to maintain the diagnosis and receive services (American Psychiatric Association, 2013).

Changes in Diagnostic Criteria: Pervasive Developmental Disorders to Autism Spectrum Disorder

First, discussion will focus on what seems to be the less controversial change in DSM-5 classification of ASD and is related to the factor structure and symptoms required to make a diagnosis. The DSM-5, released in 2013, includes a two-factor structure for identification of an Autism Spectrum Disorder instead of the three-factor structure included in DSM-IV-TR . The DSM-5 (2013) criterion A (first factor) states that social communication and social interaction deficits across multiple contexts must be present noting that the three criteria are “illustrative not exhaustive” and include deficits in social emotional reciprocity, nonverbal communicative behaviors, and in developing, maintaining, and understanding relationships. The second factor, criterion B, is restricted or repetitive patterns of behavior, interests, or activities noting that two of four criteria need be met. These include repetitive or stereotyped movements , insistence on sameness and routine, restricted and fixated interests, and hyper or hypo-reactivity to sensory input. Symptoms must be present in early development, though they may not be obvious until social demands increase. Finally, like all DSM diagnoses, symptoms must cause clinically significant impairment in functioning. In DSM-5, the social and communication domains become one factor, while these domains were two different factors in DSM-IV-TR. This change is supported by many research studies with factor analytic data showing that this improves the model (Mandy, Charman, & Skuse, 2012; Rutter, 2011a, 2011b). Additionally, severity scales and specifiers allow the clinician to provide more information on co-occurring conditions and level of symptom severity (and these will be discussed later in the review).

The more controversial change in DSM-5, as mentioned in the introduction, has to do not with factor structure, but with combining the PDDs into one ASD. DSM-IV-TR provided five diagnoses falling under Pervasive Developmental Disorders including Autistic Disorder, Asperger’s Disorder , Pervasive Developmental Disorder- Not Otherwise Specified, Childhood Disintegrative Disorder, and Rett’s Syndrome. Ongoing research on these different PDDs allowed the team of experts on the DSM-5 subcommittee to determine that one diagnosis would better conceptualize Autism Spectrum Disorder and in fact the differences in Autism/Asperger’s/PDD-NOS came down more to clinical preference/familiarity than actual differences in disorder. As Sally Ozonoff, author and expert Autism researcher and clinician, noted in her editorial on DSM-5 (Ozonoff, 2012) the changes have empirical basis from extensive literature reviews and data analysis and are not “capricious or arbitrary.” Dr. Ozonoff shared that the concept “Autism Spectrum Disorders” came about in a 1991 article by Happe and Frith. From the early 90s, studies on the difference between Autism and Asperger’s failed to find empirical differences between the diagnoses (Frith 2004) and in fact differences were quantitative (Prior et al, 1998).

The degree of impairment, severity of symptoms, and level of cognitive functioning were the most common differences between Asperger’s and autism leading to a dimensional classification removing subcategories (Eaves & Ho, 2004; Kamp-Becker et al., 2010; Prior et al., 1998; Stone et al., 1999; Tanguay, Robertson, & Derrick, 1998; Volkmar et al. 1994). Lord and Jones (2012) wrote that individuals receive diagnoses of autism, Asperger’s , and PDDNOS based on knowledge and biases of the clinician. If these are indeed on a severity continuum, it is logical to use one label. Rett’s Syndrome now has a known genetic cause, thus it is recognizable on a genetic level and no longer falls under ASD. As early as 1998, Volkmar discusses Rett’s inclusion in the neurological section. Researchers argue that little is known about Childhood Disintegrative Disorder to determine its relation to the rest of the Spectrum and it has a different outcome than Autism (Volkmar, 1998). As for the other disorders on the Spectrum evidence supports the “broader phenotype” of Autism (Bailey et al. 1998; Rutter, 2011a, 2011b; Yirmiya & Chairman 2010), the DSM-5 purports to include all those with clinically sound diagnoses under the new diagnosis ASD.

While individuals on the Spectrum have reported to the authors of this text that “Asperger’s” the word and the label is important as a piece of their identity, the research does support DSM-5’s attempt to combine dimensional and categorical models (Rutter, 2011a, 2011b). Encouraging clients to continue to use Asperger’s as a part of their identity, even when ASD is the diagnostic label, can help with discontent regarding the change. Let us recall that individuals with ASDs do not always adapt well to change. Helping clients understand that they are still “Aspies” may be helpful in some cases. Diagnostically, Asperger’s is subsumed on the Spectrum, but this does not mean that individuals should feel the need to give up the term “Aspie.” Rutter suggests that some individuals with Asperger’s do not view themselves as having a “disorder,” and when coping well, intervention may not be needed. He offers the title “Autism Spectrum Patterns” to include these individuals. Further investigation of research on DSM-5 and the continued “controversy” follows.

Controversies Surrounding DSM-5

As the Autism community waited for the release of DSM-5 in 2013, many articles were published in mainstream newspapers and magazines discussing the fear that changes in DSM criteria for Autism would lead to different diagnoses and in some cases a loss of services for those who previously met criteria for one of the Autism Spectrum Disorders (Autistic Disorder, Asperger‘s Disorder, Pervasive Developmental Disorder- Not Otherwise Specified, Childhood Disintegrative Disorder, Rett’s Syndrome). Papers were published noting the proposed changes confusing and poorly justified (Ghaziuddin, 2011). A literature review by Kulage, Smaldone, and Cohn (2014)) reported that out of 14 studies reviewed, more than 7 demonstrated reduction rates in diagnosis from 25 to 68 % when using DSM-5 criteria. Four studies included those with Asperger’s diagnoses and while a reduction in diagnosis was noted, it was not significant (Kulage et al., 2014). Kulage et al. conclude that the DSM-5 should establish a “gold standard” for diagnosis and criteria thresholds for receipt of services may have most implications for those formerly captured by the PDDNOS diagnostic label . The authors of this book believe that the “gold standard” is achieved when expert clinicians use appropriate diagnostic tools and evaluate comprehensively. Stone et al. (1999) agrees that “extensive experience” improves diagnostic clarification. Structured interviews as well as observations are very important in research (Rutter, 2011a, 2011b) and it is a loss to exclude one or the other.

Other research reports that DSM-5 can lead to as sensitive and specific of a diagnosis as the DSM-IV-TR (Kent et al., 2013). The majority of studies reviewed the DSM-5 criteria; one by Kent et al. (a paper including researchers Ann LeCouteur and Lorna Wing) reported either good sensitivity or good specificity (only one study reported both). One purported reason for inconsistencies was that data in all studies was collected according to DSM-IV-TR. Kent et al. used the Diagnostic Interview for Social and Communication Disorders (DISCO ) developed by Wing and Gould, a semi-structured interview measure with solid reliability and validity. The study suggests that DSM-5 can achieve both sensitivity and specificity. The sensitivity was lower than specificity, particularly for PDDNOS, but authors found that by relaxing thresholds increased sensitivity could be achieved without hurting the improved specificity of DSM-5 over DSM-IV. Kent et al. additionally suggests careful consideration of all symptom information to increase sensitivity of the criteria. This research group continues to collect data on use of the DISCO to assess Autism Symptoms using DSM-5 and the data as of 2014 continues to be very promising with regard to DSM-5 diagnosis (IMFAR presentation “In search of Essential Behaviours for Diagnosis” Leekam et al.). This research is consistent with the viewpoint of the authors of this text and points to a need for expert assessment and diagnosis.

Wing et al. (2011) explored the degree to which the newest version of the DSM V covers all of the important issues in autism, finding it lacking regarding diagnosing infants and adults, and the unique presentation in girls (Wing, Gould, & Gillberg, 2011). The authors on this text agree that diagnosis across the lifespan requires specific expertise beyond what is readily available in terms of diagnostic criteria and updated research. Thus, the book has been developed to guide clinicians through a practical and clear approach from the time of the referral to an accurate diagnosis; as well as, to rule out or dually diagnose other conditions during an autism evaluation.

Another group of researchers, Huerta et al., used the ADI-R and ADOS to compare to DSM-5 criteria. The clinical diagnoses were determined by experienced psychologists and psychiatrists using DSM-5 criteria. Using over 4000 children who received DSM-IV diagnoses of PDDs, they found that 91 % continued to have an ASD diagnosis on DSM-5. The adequate sensitivity of diagnosis came across all groups including young children, girls, and cognitively “higher functioning” individuals. Researchers note that most children previously diagnosed would continue to be diagnosed and the specificity of diagnosis is improved (over DSM-IV-TR) particularly taking both clinical and parent report data into account (Huerta, Bishop, Duncan, Hus, & Lord, 2012). This study went on to examine the individuals who did not continue to meet criteria for a DSM-5 diagnosis. Those who did not meet did not demonstrate the required impairments in social and communication functioning and most did meet for restricted and repetitive behaviors (Huerta et al., 2012). The study showed that criteria A1 and A3 were most challenging to interpret and assign items effectively. This data was provided to the DSM-5 subcommittee as they refined the wording of the diagnostic criteria before DSM-5 was published.

In an editorial response to this article by Tsai and Ghaziuddin, researchers conclude from literature review that PDDNOS is qualitatively and quantitatively different from autism and thus cannot lie on a continuum (Tsai & Ghaziuddin, 2014). This is not consistent with the research cited above noting that severity of symptoms differentiates the diagnoses of Asperger’s, PDD, and Autism—indicating that a dimensional model removing subcategories works best. Further review of the literature supports the idea that Asperger’s is not distinct from Autism beyond severity, IQ, and language differences. Research on PDDNOS emphasizes that it is different only in that it is less severe than Autism and over time individuals may switch back and forth from classification of Autism to PDDNOS. Eaves and Ho (2004) go on to state that PDDNOS is less stable and “not well defined.” The debate continued on; however, the research was leaning heavily toward support of a severity model, rather than diagnosing separate and distinct disorders along the Spectrum.

A study by Young and Rodi (2014) included psychologists, speech pathologists, and teams including both and utilized DSM-5 criteria to reassess and confirm diagnoses. This study found high specificity and low sensitivity. It is noted that one difference between Huerta et al. and Young & Rodi is the use of “experienced psychologists and psychiatrists” versus “psychologists and speech pathologists and teams of both trained in the ADI-R.” As noted earlier in the text, a speech pathologist is not trained to independently diagnose an autism Spectrum disorder. However, of course, speech pathologists are often critical members of the diagnostic team, providing necessary insights about language discrepancies and delays that are part of diagnostic decision making. The article concludes that perhaps the diagnosticians were “too stringent” in applying criteria and appropriate training is important to ensure consistency in DSM-5 application (Young & Rodi, 2014).

Only gold members can continue reading. Log In or Register to continue