Diverging Views on Language Disorders Researchers debate whether the label “developmental language disorder” should replace “specific language impairment.” Features
Free
Features  |   December 01, 2018
Diverging Views on Language Disorders
Author Notes
  • Nancy Volkers is a freelance medical writer based in Vermont. nvolkers@nasw.org
    Nancy Volkers is a freelance medical writer based in Vermont. nvolkers@nasw.org×
Article Information
Language Disorders / Specific Language Impairment / Features
Features   |   December 01, 2018
Diverging Views on Language Disorders
The ASHA Leader, December 2018, Vol. 23, 44-53. doi:10.1044/leader.FTR1.23122018.44
The ASHA Leader, December 2018, Vol. 23, 44-53. doi:10.1044/leader.FTR1.23122018.44
Imagine this scenario:
In a nomadic hunter-gatherer society, where running a 10-minute mile is considered a crucial life skill, some elders notice there are children unable to do this, for no obvious reason. These children are not obese, they do not have motor impairments, intellectual or developmental disabilities, or congenital malformations. The elders are puzzled: Why can’t these children run fast enough? They categorize these children as having fast running impairment, or FRI, and start investigating ways to help them.
Over time, others start to question why other children—such as those with weight problems or motor disabilities—who can’t run a 10-minute mile aren’t included in the FRI group. They state their position: We need to come up with a better term to describe this impairment so that every child can get help, not just the ones that fit this narrow FRI definition.
The elders respond that they defined FRI in a narrow way specifically to discover why these children have an impairment with no apparent cause. They state their position: We can’t figure out how to help kids who don’t have other contributing problems, such as obesity or motor impairments, unless we exclude those factors. And we do that by testing what works with FRI.

SLI was never intended to be inclusive; it was created by excluding children whose language issues were associated with (but not necessarily caused by) other conditions, such as Down syndrome or sensorineural hearing loss.

Though not precise, this scenario is analogous to the debate over specific language impairment (SLI) versus developmental language disorder (DLD) in the speech-language pathology world. Communication sciences and disorders (CSD) researchers first began using the term specific language impairment, or SLI, in the 1980s to define a group of children who have language difficulties for no apparent reason: Their language impairment is not explained by brain injury, hearing loss, intellectual disability or another medical condition. SLI was never intended to be inclusive; it was created by excluding children whose language issues were associated with (but not necessarily caused by) other conditions, such as Down syndrome or sensorineural hearing loss.
“These children have been interesting to study for exactly that reason: [there is no obvious cause for their language problems],” says language researcher Mabel L. Rice, the Fred & Virginia Merrill Distinguished Professor of Advanced Studies at the University of Kansas. “Language impairment is their disability.”
However, a countering group of clinicians and researchers—looking at SLI through more of a clinical lens—believes that SLI is too exclusive, failing to account for children who have both language issues and nonverbal IQ scores below 85.
In this vein, a group of 57 researchers from six countries (half from the United Kingdom) recently issued a position statement asserting that children who used to fit criteria for SLI should now be said to have “developmental language disorder” (DLD). Nonverbal IQ score should not factor into whether or not a child has DLD, according to their statement, published in 2016 (see sources below). Under this view, DLD includes children who have a language impairment with and without an intellectual disability.
Leading this movement is Dorothy Bishop, professor of developmental neuropsychology at Oxford University, who, in a 2010 article in The Guardian, called SLI one of the “common childhood disorders that’s been left out in the cold.” Bishop maintains that SLI has an image problem: Though as common as dyslexia and far more common than autism spectrum disorder (ASD), it receives much less research funding and wields less advocacy power or name recognition, she says, as part of her argument for broadening the category.
Many SLI researchers, however, are concerned about removing the SLI category because it was explicitly defined to exclude children with other cognitive issues. They point out that the research cutoff for SLI (nonverbal IQ of 85 or higher) was meant to ensure that any studies on children with SLI were as free as possible from the effects of cognitive impairment other than their language impairment. They believe that removing that cutoff could introduce a lot of heterogeneity, which will cause confusion and slow research—thus thwarting development and testing of effective interventions for these children.
To better understand the arguments for and against SLI, it helps to consider some history.

Some researchers believe that removing the IQ cutoff could introduce a lot of heterogeneity, which will cause confusion and slow research—thus thwarting development and testing of effective interventions for these children.

The case for DLD
It was in the early 2000s that lecturers and publications began raising questions about the usefulness and validity of the term SLI. So in 2015, Bishop coordinated an international group of 57 experts from six countries in the CATALISE (Criteria and Terminology Applied to Language Impairments: Synthesizing the Evidence) project. Using the Delphi method, an iterative process of review using anonymous ratings and feedback, the group sought consensus regarding terminology and diagnostic criteria to identify children with language disorders (see sources).
The group reached at least 78 percent agreement (44 out of 57) on the recommended terminology and definitions. Regarding SLI, they suggested that:
  • SLI is an “artificial abstraction” invented by researchers; it is not designed for clinical purposes.

  • Nonverbal IQ does not seem to distinguish subtypes of children with language impairment.

  • The strict definition of SLI may exclude children who need services.

They decided that the umbrella term “language disorder” (LD) refers to children with language difficulties that create obstacles to communication or learning in everyday life and that are unlikely to resolve on their own, based on past research. Not included in the LD umbrella are late talkers (under age 5), uncomplicated phonological problems in preschoolers (speech sound disorder), or language limitations due to a lack of exposure to English (or the home language).
Included in the LD umbrella are two subgroups:
  • Children who have a biomedical condition that may affect their language development, such as Down syndrome, intellectual disability or sensorineural hearing loss. For these children, the panel recommends a diagnosis of “language disorder associated with X” (where X is the biomedical condition).

  • Children with “developmental language disorder” (DLD), in which language difficulties are not associated with a known biomedical condition, such as brain injury, cerebral palsy, sensorineural hearing loss, ASD or intellectual disability. DLD can co-occur with impairments in the areas of attention, motor coordination, literacy, speech, behavior or emotional problems, executive function, or auditory processing. A DLD diagnosis does not require a mismatch between verbal and nonverbal ability. Children with low nonverbal IQ scores who do not meet criteria for intellectual disability (generally with scores between 70 and 85) can be diagnosed with DLD.

To continue conducting SLI/DLD research, it is still imperative to collect IQ information, and to use inclusion and exclusion criteria relevant to the questions being investigated.

SLI’s role in research
The very aspect of SLI that prompted objections from the CATALISE group—its restrictive definition—is exactly what some CSD researchers value about it, including some who participated in CATALISE. For decades, the designation has allowed researchers to study language impairment while minimizing confounding variables. The gap between intellectual disability (IQ below 70) and the research cutoff for SLI (85 or higher) is used in research to account for variability in individual IQ scores from time point to time point, says Rice, and to factor out the effects of other cognitive impairments.
The term SLI has strong roots in the research literature, Rice notes. “The science is very sound on SLI. We have robust findings from one lab to another, a documentation of the long-term trajectory of what happens to these children,” she says. “I’m not opposed to an overarching way to describe all kids with language impairments, because there are many kids with language impairments who don’t have SLI. There’s a need for an overarching term, but I don’t know why we need DLD when we have others” [for example, “language disorders” is an already-existing overarching term used in DSM-5]. For more on other terminology, see part two of this series on SLI versus DLD, page 54.
Another argument for SLI’s usefulness in research, says Sean Redmond, professor in the Department of Communication Sciences and Disorders at the University of Utah, is it can help answer questions about underlying mechanisms and heritability. “I don’t think we would have gotten where we are with well-vetted clinical markers that appear to be unique to language disorder without the SLI designation,” he says. “If you open it up to a broader phenotype, you’re likely to get a lot of noise, especially in the genetic work.”
Not only could DLD introduce unwanted heterogeneity into SLI studies, it could possibly threaten the long and rich history of SLI findings—along with a robust evidence base for the disorder, some researchers worry. “We’ve had hundreds of studies that use SLI and we don’t want to lose that,” says Laurence B. Leonard, Rachel E. Stark Distinguished Professor in the Department of Speech, Language, and Hearing Sciences at Purdue University. “For every DLD study, there are eight that use SLI.”

A countering group of clinicians and researchers—looking at SLI through more of a clinical lens—believes that SLI is too exclusive, failing to account for children who have both language issues and nonverbal IQ scores below 85.

Not a threat?
Not all researchers, however, see a DLD expansion as threatening to the SLI database. “We do not have to throw away the existing literature on SLI,” Bishop says. “Because nonverbal IQ does not seem to distinguish subtypes of children with language problems, the conclusions in the literature to date are likely to continue to apply to the broader category of children with DLD.”
Some researchers are comfortable with DLD. “We haven’t used SLI for some time; we have used ‘language impairment,’” says Bonnie Brinton, professor in the Department of Communication Disorders at Brigham Young University. “We study social and emotional competence in children with language impairment, and we’re not finding this category is as specific as SLI would imply. There’s a lot of variability among children.”
But in the view of Margaret Rogers, ASHA’s chief staff officer for science and research, eliminating the SLI category would run counter to evidence-based practice. “There’s a lot of pressure to establish evidence-based clinical practice guidelines to justify reimbursement and funding for speech-language pathology services,” she says, “and the unfortunate thing is that SLI is one of the most well-researched conditions in CSD. The literature on SLI has some of the best diagnostic and treatment studies in the discipline—well-done controlled investigations that could contribute strongly to the development of evidence-based systematic reviews and clinical practice guidelines. Depth and strength of evidence are what we need to survive as a field. We need to show that we’re making a difference, and we can do that with SLI.”

Not all researchers see a DLD expansion as threatening to the SLI database. “We do not have to throw away the existing literature on SLI,” says Dorothy Bishop.

The IQ question
At the core of the CATALISE panel’s work—and arguably what’s most controversial about it—is eliminating the requirement that a language impairment diagnosis depends on a discrepancy between nonverbal IQ and language. The CATALISE group cites several lines of reasoning:
  • Children with language impairment and IQs both above and below 85 have similar linguistic profiles. There is no evidence to show that children with an IQ at or above 85 respond better to intervention than those with a lower IQ.

  • There is a movement away from sole reliance on IQ tests for diagnosing intellectual disability, and toward accounting for the ability to function adaptively in everyday life.

  • Children with language impairment and IQs both above and below 85 have similar linguistic profiles. There is no evidence to show that children with an IQ at or above 85 respond better to intervention than those with lower IQ.

  • Discrepancy scores are unstable and cannot provide a reliable basis for classification or diagnosis. IQ scores can change over time.

  • Many children with low nonverbal IQ have adequate language function; therefore, low nonverbal ability does not set a limit on the rate of language development.

“It’s reasonable to consider DLD as the uber-category of childhood language impairments, with SLI as a subcategory, but not to get rid of SLI altogether.”

Indeed, in Redmond’s view, children with low nonverbal IQ and typical language skills are a relatively understudied group. “We’ve been focused primarily on kids with normal IQ but poor language, and haven’t really asked what about the kids with lower IQs but good language,” Redmond says. “We’ve missed an opportunity to know what good language does for you. I think the shift to the broader DLD designation may bring more attention to that.”
Leonard concurs that IQ scores can be a moving target: Children initially diagnosed with SLI show a trend of progressively lower IQ scores over 5 to 10 years, he notes. “The best guess as to why this happens is that the farther a child goes in school, the more knowledge of the world is transmitted by language,” Leonard says. “If a child starts out with language problems, and his or her access to knowledge becomes increasingly dependent on language, and if IQ tests tap into that type of knowledge, then it would be no surprise to see IQ scores drop.” (See “Language Impairment or Learning Disability?”.)
Still, says Leonard, even if you’re researching DLD, it’s necessary to measure IQ. “As one colleague pointed out, if there’s research on kids with IQs above 85 and someone tries to replicate it using kids with slightly lower IQs and can’t, does that mean that the first study’s findings are unreliable, or is it due to the IQ differences?” Leonard asks.
Pamela Hadley, associate professor in the Department of Speech and Hearing Science at the University of Illinois, agrees. To continue conducting SLI/DLD research, it is still imperative to collect IQ information, and to use inclusion and exclusion criteria relevant to the questions being investigated, she says.
“When you’re doing [SLI] genotype/phenotype studies, it makes sense to narrow your focus,” says Hadley. “That’s how dyslexia research moved forward. And if you’re asking a question about an intervention for kids with low language, we want to have our recruitment reflect the caseloads of clinicians, so we’d have a broader group. But we’ll also ask, did the subtype influence the response to the intervention, or did the intervention work better for one subtype versus another?”
Rogers backs these researchers’ position that it isn’t problematic to add DLD to the mix, as long as SLI doesn’t disappear. “I think it’s reasonable to consider DLD as the uber-category of childhood language impairments, with SLI as a subcategory,” Rogers says, “but not to get rid of SLI altogether.”
The Life of a Child With SLI

Children with specific language impairment (SLI) can be identified starting at about age 4, say SLI researchers. But they often are not identified until much later, if at all.

Some characteristics, according to SLI research (see sources), include:

  • They may have better receptive language than spoken language, but closer inspection often reveals language comprehension problems as well.

  • They have difficulty with verb endings and syntax. They might leave out “is” and “are,” or not put an “s” on the end of a verb when they should.

  • They often misuse pronouns—using “her” instead of “she,” for example.

  • They typically spend a prolonged period in a stage of sentence development in which words are left out.

  • As they get older, they may avoid complex sentences. They may not use abstract language. Many have trouble with reading.

  • They may have problems finding the right word.

  • They can be passive in conversations, particularly faster back-and-forth discussions in groups.

  • They may have difficulty with the language expectations in college classes or employment settings.

  • As adults, their employment options may be diminished. They may avoid highly verbal fields. They may have difficulty negotiating at work or speaking extemporaneously in meetings or discussions.

—Nancy Volkers

Sources
Bishop, D. V. M., Snowling, M. J., Thompson, P. A., & Greenhalgh, T. (2017). Phase 2 of CATALISE: A multinational and multidisciplinary Delphi consensus study of problems with language development: terminology. Journal of Child Psychology and Psychiatry, 58(10), 58(10),1068–1080. [Article]
Bishop, D. V. M., Snowling, M. J., Thompson, P. A., & Greenhalgh, T. (2017). Phase 2 of CATALISE: A multinational and multidisciplinary Delphi consensus study of problems with language development: terminology. Journal of Child Psychology and Psychiatry, 58(10), 58(10),1068–1080. [Article] ×
Bishop, D. V. M., Snowling, M., Thompson, P.A., & Greenhalgh, T. (2016). CATALISE: A multinational and multidisciplinary Delphi consensus study. Identifying language impairments in children. PLoS One, 11(7), e0158753. [Article] [PubMed]
Bishop, D. V. M., Snowling, M., Thompson, P.A., & Greenhalgh, T. (2016). CATALISE: A multinational and multidisciplinary Delphi consensus study. Identifying language impairments in children. PLoS One, 11(7), e0158753. [Article] [PubMed]×
Leonard, L. B. (1989). Language learnability and specific language impairment in children. Applied Psycholinguistics, 10(2), 179–202. [Article]
Leonard, L. B. (1989). Language learnability and specific language impairment in children. Applied Psycholinguistics, 10(2), 179–202. [Article] ×
Leonard, L. B. (1987). Is specific language impairment a useful construct? In Rosenberg S. (Ed.),. Cambridge monographs and texts in applied psycholinguistics. Advances in applied psycholinguistics, Vol. 1. Disorders of first-language development; Vol. 2. Reading, writing, and language learning, (1–39). New York: CambridgeUniversityPress.
Leonard, L. B. (1987). Is specific language impairment a useful construct? In Rosenberg S. (Ed.),. Cambridge monographs and texts in applied psycholinguistics. Advances in applied psycholinguistics, Vol. 1. Disorders of first-language development; Vol. 2. Reading, writing, and language learning, (1–39). New York: CambridgeUniversityPress.×
Rice M. L. (2014). Advocating for SLI. Commentary on Bishop, D.V.M., 2014. Ten questions about terminology for children with unexplained language problems. International Journal of Language and Communication Disorders, 49, 381–415. [Article] [PubMed]
Rice M. L. (2014). Advocating for SLI. Commentary on Bishop, D.V.M., 2014. Ten questions about terminology for children with unexplained language problems. International Journal of Language and Communication Disorders, 49, 381–415. [Article] [PubMed]×
Rice, M. L., Redmond, S. M., & Hoffman, L. (2006). Mean length of utterance in children with specific language impairment and in younger control children shows concurrent validity and stable and parallel growth trajectories. Journal of Speech, Language, and Hearing Research, 49(4), 793–808. [Article]
Rice, M. L., Redmond, S. M., & Hoffman, L. (2006). Mean length of utterance in children with specific language impairment and in younger control children shows concurrent validity and stable and parallel growth trajectories. Journal of Speech, Language, and Hearing Research, 49(4), 793–808. [Article] ×
2 Comments
December 6, 2018
Lisa Achelis
Thank you!
This article is so well thought out and organized; after attending a session at ASHA I was left baffled and felt as though the gap between the Ivory Tower and the "trenches" was growing even more rapidly. You've really helped to close the gap a bit here. There's a lot of work left to do to translate this into the field, especially as it pertains to qualifying diagnoses and ICD-9 codes, codes but I really appreciate you laying such a clear foundation for the topic. Thank you!
December 7, 2018
Ann Kummer
Reimbursement for "develpmental" language disorder?
I wonder if you have considered the effect of the word "developmental" on reimbursement for language disorders. Historically, the use of that word is an automatic denial from third party payers. I think there is a general believe that if the problem is development, than the child will "grow out of it." As the senior director of speech-language pathology at Cincinnati Children's, my team and I have fought this battle with payers for decades. We developed a brochure (that has been the basis of the ASHA brochure) on insurance issues for pediatric patients to be used to educate consumers and payers. However, it remains a very big issue.
Submit a Comment
Submit A Comment
Name
Comment Title
Comment


This feature is available to Subscribers Only
Sign In or Create an Account ×
FROM THIS ISSUE
December 2018
Volume 23, Issue 12