1-800-649-8481 Toll Free
DSM-5 is Here!

The Diagnostic and Statistical Manual of Mental Disorders (DSM-5) appeared in bookstores last week and pre-ordered copies arrived in the mail. For those who followed the DSM draft versions over the past year, the published Autism Spectrum Disorder diagnostic criteria came with a few surprises, some well-considered and helpful, and some . . . less so. Here’s a summary of a few of the ways the published version differs from the last draft:

Differences in Criteria

The text of criterion A says: “Persistent deficits in social communication and social interaction across multiple contexts, as manifested by the following . . .” and goes on to list the sub-criteria much as we saw them in the draft. Significantly, it does not say “manifested by all 3 of the following” as we expected from the draft. We might guess that the authors’ intent remains the same; but then why does criterion B specify “. . . as manifested by at least two of the following . . .”? Allen Frances, chairman of the DSM-IV PDD task force, opined last week:

The really fatal flaw here is that no instructions are given as to whether one item, two items, or all three items must be present to make the diagnosis of Autism Spectrum Disorder.”

Whether this really is a fatal flaw or not remains to be seen; Dr. Frances has established himself as something of an elder curmudgeon with respect to DSM-5 and some will dismiss his criticism as mean-spirited carping. But the fact remains that there is ambiguity here and the Task Force would do well to clarify its intention.

​A number of reports have suggested that the diagnosis would be well-served by relaxing the demand for “all 3 of the following” (e.g., Barton et al. 2013). However, it is not at all clear that the change from the draft was intended to allow for such a relaxation and, in any case, there is little to recommend a solution that leaves the decision (about how many sub-criteria must be met) up to individual clinicians’ interpretations.

“By history” and what it means

Criterion A further indicates that the deficits in social communication and social interaction may be manifested either “currently or by history.” (Criterion B, concerning “Restricted, repetitive patterns of behavior, interests, or activities” contains the same clause.) This formalizes a practice that some clinicians have, no doubt, always followed with respect to characteristics of autism; if individuals displayed the key features of autism in the past, even if some of those features are no longer present, the criterion is considered met.

Explicitly including the “by history” clause highlights a quandary faced by those seeking to make a diagnostic determination in older individuals (adolescents and adults) for whom no one is available to report on their early developmental history. When evaluating an adolescent for a first-time diagnosis of an autism spectrum disorder, if there is no one who can state whether formerly he had “difficulties in sharing imaginative play or in making friends” (A.3), for example, the clinician may lack important data that could help with the determination. Thus, in some cases, a diagnosis of ASD will be dependent on whether there is a caretaker available who knew the individual in early childhood. This quandary is not new, but those diagnosticians unaccustomed to considering the presence of features “by history” may find themselves facing an unfamiliar and uncomfortable dilemma.

When evaluating adolescents or adults (without a caregiver for early history), some clinicians have sought evidence of early indicators of ASD by means of self-report. This approach may be fruitful for some individuals with respect to “Restricted, repetitive patterns of behavior, interests, or activities;” however, the limited capacity for selfreflection seen in many individuals with ASD makes it unlikely that the approach will yield much useful data regarding “deficits in social communication and social interaction.”

Some may worry that the “by history” clause will inappropriately retain an ASD diagnosis in those “optimal outcome” individuals whose positive response to intervention has moved them off the autism spectrum. This concern, however, can be alleviated by careful attention to Criterion D which requires that “Symptoms cause clinically significant impairment in social, occupational, or other important areas of current functioning.” If there is no impairment in current functioning related to features that define Autism Spectrum Disorder, then Criterion D is not met and the diagnosis is not applied.

On balance, specifying that the criteria may be met “by history” is probably a positive development in that it aligns with current thinking about the developmental trajectory of ASD and may, one hopes, lead to improved consistency across diagnosticians.

“Illustrative, not exhaustive”

Criterion A concludes with the qualifier that the examples in the sub-criteria are “illustrative, not exhaustive.” (Criterion B includes the same qualifier.) So, considering sub-criterion A.1:

Deficits in social-emotional reciprocity, ranging, for example, from abnormal social approach and failure of normal back and forth conversation; to reduced sharing of interests, emotions, or affect; to failure to initiate or respond to social interactions.

This qualifier allows the diagnostician to take into account other ways in which one might display a “deficit in social-emotional reciprocity,” ways that are not explicitly listed in the examples in the text. While this introduces additional room for inconsistency across clinicians, it is an appropriate acknowledgement that a diagnostic manual cannot list all of the possible ways in which an important developmental difference like a “deficit in social-emotional reciprocity” might be manifested. When dealing with behaviorally defined conditions, no diagnostic manual, however detailed, will dispense with the need for clinical judgment, informed by the scientific literature and clinical experience.

Notes on previous diagnoses

The diagnostic criteria for ASD are immediately followed by a “Note:”

Individuals with a well established DSM-IV diagnosis of autistic disorder, Asperger’s disorder, or pervasive developmental disorder not otherwise specified should be given the diagnosis of autism spectrum disorder. Individuals who have marked deficits in social communication, but whose symptoms do not otherwise meet criteria for autism spectrum disorder, should be evaluated for social communication disorder.

While this concession may satisfy those who worried about previously identified individuals losing their diagnosis under the new criteria (“de-diagnosis”), others might feel that it tends to undermine confidence in the entire schema. The fact that one’s diagnosis of an autism spectrum disorder (or lack thereof) may depend on whether the evaluation was sought before – or after – DSM-5 was released, will be perceived by some as unfair or arbitrary. More importantly, this fact suggests that the scientific underpinning of diagnostic labeling is far less secure than we might have hoped.

The fears about “de-diagnosis” cannot be easily dismissed and the removal of a diagnosis could have a profound impact on an individual’s life. Admittedly, there is no easily-identified alternative solution to the dilemma faced by the DSM Task Force regarding the possibility of “de-diagnosis;” but the inclusion of the “Note” injects further confusion about the meaning of an autism diagnosis into a discourse that is already considerably fraught.

Perhaps DSM-5 has done us a favor by reminding us that the manual, and the entire enterprise of psychiatric diagnosis, are social constructions. Diagnostic criteria are not, and never have been, objective indicators that “carve nature at its joints” (Plato, a very long time ago). These criteria are tools that help us describe the challenges that individuals face and point us in the direction of things that we might do to better support them. And if they serve that purpose, perhaps we should be content.


Barton, M. L., Robins, D. L., Jashar, D., Brennan, L., & Fein, D. (2013). Sensitivity and specificity of proposed DSM-5 criteria for autism spectrum disorder in toddlers. Journal of Autism and Developmental Disorders, 43(5), 1184-1195. doi:10.1007/s10803-013- 1817-8

Stay connected to CA

CA works with autistic people and their families to help them thrive. Every day, we’re building a future where the most vulnerable Virginians can actively participate in our community and realize their full potential.

© 2023 Commonwealth Autism. All Rights Reserved.