April 2013: Proposed Changes to ADHD and LD in the DSM-5

Tannock, R. (2013). Rethinking ADHD and LD in DSM-5: Proposed changes in diagnostic criteria. Journal of Learning Disabilities, 46(1), 5–25.

Summary by Dr. Jack Fletcher


The author of this article is Rosemary Tannock, who is a Canada Research Chair at the Hospital for Sick Children and a member of the Neurodevelopmental Task Force. This task force is responsible for childhood disorders that include not only learning disabilities (LD) and attention-deficit/hyperactivity disorder (ADHD), but also intellectual disorders (mental retardation), autism spectrum disorders, and communication disorders.

In May 2013, the American Psychiatric Association will release the much-anticipated fifth edition of its Diagnostic and Statistical Manual (DSM-5). Replacing the fourth edition, released in 1994 (DSM-4), DSM-5 has been under development since 1999. The focus of Tannock’s article is changes to the diagnostic criteria for LD and ADHD and the extensive literature synthesized to develop the revisions for these two DSM-5 categories. As Tannock indicates, DSM-5 was revised based on the following four overarching principles:

  1. Because DSM-5 is an evidence-based tool to guide clinicians in a variety of fields in the assessment of behavioral and cognitive disorders, changes are based on both scientific information and clinical practice and are designed to balance these two domains.
  2. Because change is disruptive, only essential changes were made.
  3. Because there is insufficient evidence to base diagnoses on the causes of the disorder, diagnostic criteria instead focus on descriptions of specific behaviors.
  4. All changes were made after considering their potential adverse effects on a group of people who need services or other forms of assistance.

Tannocks’s article is an excellent review of the basis for changes to the two most common disorders that educators and school-based clinicians address. LD and ADHD both have high prevalence rates, with LD accounting for approximately half of all children identified for special education. ADHD is identified in the other health impaired category of special education, the fastest-growing eligibility category in special education, although not all children in this category are identified because of ADHD. Many children with ADHD also have academic problems and are identified in the specific learning disability or emotional/behavioral disturbance category if their behavioral problems are severe and there is educational need.

Why Does DSM-5 Matter to Educators?

U.S. schools’ responsibilities for students with special needs are specified in the Individual With Disabilities in Education Act (IDEA) and in Section 504 of the Rehabilitation Act of America, which provides guidelines associated with the development and implementation of 504 plans. On the surface, DSM-5 is irrelevant because IDEA and 504 supersede DSM-5, most obviously for the LD category. However, parents often obtain evaluations from outside professionals that may identify LD based on DSM-5 criteria. More importantly, parents of children identified with ADHD may obtain eligibility forms from a psychiatrist or other physician, based on DSM-5 criteria, indicating the diagnosis of ADHD, how it affects schooling, and recommendations for special education and other services. Although DSM-5 criteria indicate that a problem needs to be pervasive and interfere with adaptation, it does not specifically deal with the issue of educational need. Eligibility for special education under IDEA has two prongs: a disorder consistent with one of the 13 categories of special education and evidence of educational need that would necessitate a modified educational plan. Many parents and clinicians outside the school assume that the diagnosis is sufficient to determine eligibility for special education services. It is important to take into account the issue of educational need. This consideration ensures that the fundamental requirements of IDEA are met: a free and appropriate public education in the least restricted environment, based on individual educational needs.

Changes in DSM-5 Criteria for LD and ADHD

Neurodevelopmental disorders. As Tannock indicates, one major change to both categories is the recognition that they often co-occur in the same child, leading to their placement together as neurodevelopmental disorders. There is evidence that although the disorders are distinct, they share heritable features and often share treatment needs. In addition, by clustering them together as neurodevelopmental disorders with other disorders that have cognitive underpinnings (e.g., intellectual disabilities, communication disorders), these components can be separated from disruptive behavior identified as oppositional defiant disorder and conduct disorder, an important conceptual distinction that will hopefully facilitate identification of co-occurring LD and ADHD.

Dimensional disorders. A second change is the recognition that the defining attributes of LD (low achievement, inadequate instructional response) and ADHD (inattention, hyperactivity-impulsivity) are variations in typical development that represent the extremes of the distribution of these areas of child and adult development. This change clearly indicates that although it is easier to talk about LD and ADHD as though they are discrete categories, the boundaries that separate LD and ADHD (themselves separate disorders with different defining characteristics) from typical development are difficult to precisely define and vary, sometimes depending on context. Because of this imprecision, it is difficult to establish exact thresholds indicating LD and ADHD; there is presently no evidence of qualitative markers that would represent categorical markers of the disorders.

Role of cognitive processes. Cognitive skills (IQ as well as memory, language, and other skills) are clearly associated with LD and ADHD and vary systematically at the group level with the two disorders. ADHD is associated with poorer cognitive performance on measures of attention, self-regulation, and disinhibition. The correlates of LD depend on what aspects of reading, mathematics, and writing are impaired. But Tannock notes that the DSM-5 task force could find little evidence for definitional criteria based on specific cognitive skills because of overlap and lack of evidence that cognitive assessments could be used to rule in or rule out LD and ADHD. Thus, criteria for LD (and ADHD) based on assessments of cognitive processes were not provided. Moreover, the DSM-5 work group recommended abandoning reliance on a discrepancy between IQ and achievement because of the absence of validity evidence (Stuebing et al., 2002), especially in relation to treatment outcomes. IQ scores are not robust predictors of intervention response (Stuebing et al., 2009).

Changes specific to LD. An obvious change is that DSM-5 now refers to “learning disabilities” and not to the DSM-4 term “academic skills disorders,” aligning with traditional usage of the LD classification in IDEA and elsewhere. Whereas DSM-4 primarily recognized academic skills disorders in reading, mathematics, and written expression, DSM-5 recognizes LD and specific manifestations in reading (word reading accuracy, reading fluency, and reading comprehension), written language (spelling accuracy, grammar and punctuation accuracy, and organization of written expression), and mathematics (basic number sense, accuracy and fluency in recalling number facts, calculation accuracy and fluency, and mathematics reasoning). DSM-5 does not identify nonverbal LD because of limited evidence for the validity and utility of this classification (Spreen, 2011). DSM-5 also places difficulties with speaking and listening in the communication disorders category, which is appropriate and a desirable change to the IDEA definition of LD because such disorders are covered under the specific language impairment category. Within each academic category, DSM-5 applies the following four criteria: (1) persistence, despite the presence of adequate intervention, for at least 6 months; (2) low achievement, or scoring below the mean for age on a norm-referenced achievement test; (3) age of onset (i.e., the problem manifests during early years of schooling); and (4) no evidence that another condition (e.g., intellectual disability, sensory problem, other mental or neurological disorder, psychosocial adversity, lack of educational opportunity) provides a better explanation for the presence and persistence of low achievement.

Changes specific to ADHD. One major change to the DSM-4 ADHD criteria in DSM-5 is the removal of specific references to subtypes. Although inattention and hyperactivity-impulsivity are clearly separable attributes (and both reflect typical variation from a dimensional perspective), evidence for separate categories is not strong. DSM-5 maintains the idea of separable attributes but refers instead to “specifiers of current presentation.” When ADHD is identified by using DSM-5 criteria, it is with a hyperactive-impulsive presentation, an inattentive presentation (with either restricted or moderate levels of hyperactivity-impulsivity), and the combined presentation. To accommodate evidence of ADHD in adults, the diagnostic threshold was reduced for people older than 17. In addition, the age of onset criterion was revised to age 12 because the previous threshold of age 7 did not capture many initial presentations of ADHD. The wording of items was changed to reflect a broader range of impulsivity items and behavior appropriate for older adolescents and adults. Finally, autism spectrum disorders are no longer exclusionary because many with autism spectrum disorders also have ADHD behavior. To reduce errors in diagnosis by relying on a single informant, the pervasive criteria were modified to require at least two informants. ADHD is a pervasive disorder in which the problem behaviors occur in multiple situations.

Controversies Likely to Emerge

For ADHD, the subtype discussion will continue, although the evidence for categorical distinctions is weak. How to address the age of onset issue and the presentation of ADHD in adults will require more research, but DSM-5 seems consistent with the state of current scientific evidence. Issues about the prevalence of ADHD, the use of stimulants, and the best approaches to treatment of co-occurring disorders often seen with ADHD (e.g., LD, anxiety and mood disorders) will continue to be important topics, and it will be interesting to see whether any of this discussion shifts with the revised criteria.

LD is likely to be more controversial. Much of the controversy in the United States may reflect the alignment of DSM-5 with IDEA and current scientific thinking. There will be ongoing discussion about the role of cognitive assessments for both LD and ADHD. Cognitive assessment issues will be especially important for adults referred for accommodations for LD and ADHD because treatment opportunities are either difficult to document or are simply insufficient (Mapou, 2008). The review of evidence by Tannock and her colleagues is thorough, and it is incumbent on advocates of testing cognitive processes to demonstrate that such assessments facilitate diagnosis and/or treatment.

The abandonment of IQ-achievement discrepancy by DSM-5 is likely to cause discussion, especially from critics of response to intervention approaches who worry that children with high IQ and low achievement will not receive services. Many schools are hard pressed to provide services to clearly low achievers, so the provision of resources to children with average achievement is of concern. Unfortunately, the research base indicating how children who are considered “twice exceptional” or “gifted LD” should be identified is still sparse. Educators should be aware that the most typical approach is an IQ-achievement discrepancy, often in children with average achievement levels. Complex statistical issues affect the interpretation of any cognitive discrepancy because the aptitude and achievement scores are correlated and associated with measurement error, so that a straight discrepancy between scores (e.g., one standard deviation) is not the same at upper and lower levels of IQ.

Concerns have already been expressed about the use of the terms “dyslexia” and “dyscalculia” in DSM-5. The commentary for the revision will identify these terms, but they are not in the diagnostic criteria. DSM-5 clearly refers to children with deficits in reading word accuracy and fluency, which along with spelling are the cardinal identifying attributes of dyslexia. By referencing word-level deficits in reading and spelling, DSM-5 references dyslexia in all but the name.

The decision of DSM-5 to include persistence in the face of adequate instruction is well considered and begins to deal with the frustrating issue of how to deal with children and adults who have not had adequate instructional opportunities, which is also addressed in IDEA. This criterion does not equate with the IDEA provisions for response to intervention, a largely U.S. phenomenon and a much more specific service delivery framework. Including a persistence criterion makes LD in DSM-5 more than simple low achievement.

DSM-5 discusses a number of potential cut points for low achievement that range from 1.5 to 1.0 standard deviations below the population average (bottom 7% to 16%). Educators should be aware that any cut point at the lower end of the achievement distribution is arbitrary. If we add to states’ proposed cut points for LD the fact that many in the lowest 2% likely have intellectual deficiencies, the number of students with persistent low achievement does not represent the bottom 7% of low achievers on a norm-referenced test. The criteria for absolute low achievement should be higher if the goal is to actually identify the lowest 7%. If all sources of intervention are considered (e.g. Title 1, 504, bilingual), more than this number are currently served. These are thorny issues that make any effort to subject the identification of LD to a formula or threshold difficult and often lead to confusion about prevalence and its relation to resources.


Tannock’s summary of DSM-5 is important reading for people who serve children in schools. Not only is it important to understand changes in diagnostic criteria for ADHD and LD, but also her synthesis of the literature and thinking of the Neurodevelopmental Task Force is incisive and well aligned with current scientific research. However, her article summarizes a draft of DSM-5; those who are interested should read the final version of DSM-5 when it is released in May 2013.


Lovett, B., & Lewandowski, L. J. (2006). Gifted students with learning disabilities: Who are they? Journal of Learning Disabilities, 39(6), 515–527.

Mapou, R. L. (2008). Adult learning disabilities and ADHD: Research-informed assessment. New York, NY: Oxford Press.

Spreen, O. (2011). Nonverbal learning disabilities: A critical review. Child Neuropsychology, 17(5), 418–443.

Stuebing, K. K., Barth, A. E., Molfese, P. J., Weiss, B., & Fletcher, J. M. (2009). IQ is not strongly related to response to reading instruction: A meta-analytic interpretation. Exceptional Children, 76, 31–51.

Stuebing, K. K., Fletcher, J. M., LeDoux, J. M., Lyon, G. R., Shaywitz, S. E., & Shaywitz, B. A. (2002). Validity of IQ-discrepancy classifications of reading disabilities: A meta-analysis. American Educational Research Journal, 39, 469–518.