Why add motor to the definition of ASD: A response to Bishop et al.’s critique of Bhat (2021)

I thank Bishop et al. for the opportunity to have an open and cordial debate on the issue of including motor problems within the ASD definition in a future Diagnostic & Statistical Manual (DSM) revision. I encourage content experts and various stakeholders to weigh in on this issue as well. I will begin with the point that Bishop et al. was in agreement with, “recommending screening and assessment of motor problems in children with ASD.” About ~80% of children with ASD in the SPARK study received speech and occupational therapy (OT) interventions that mainly address their communication and fine motor/sensory processing problems (Bhat, 2020). Motor problems of children with ASD remain underdiagnosed (only ~15% have a co-occurring diagnosis of motor delay) and most likely gross motor issues remain undertreated (~13%–32% received recreational or physical therapy [PT] interventions, respectively) even when ~87% of the SPARK cohort was at-risk for motor impairment based on the Developmental Coordination Disorder Questionnaire (DCD-Q) (Bhat, 2020). ASD diagnosticians need a clear directive to screen for motor problems and must make referrals for assessment and treatment of fine and gross motor issues to movement clinicians (OTs and PTs), when appropriate. Adding motor to the definition of ASD will bring this issue on the radar of diagnosticians and stakeholders to better address these problems. Interventions for children with ASD must shift from sedentary play at the desk or on the floor to include everyday functional motor skills, complex movements, fine and gross motor play, and physical activity in small-to-large groups utilizing engaging creative and general movement contexts. Overall, I agree with and extend Bishop et al.’s statement that we need to improve motor screening and referrals to movement clinicians for further motor assessments and interventions of individuals with ASD throughout the lifespan. Next, I will address the points of disagreement one by one.

1. “Your statement that there is a need to recognize motor impairments as a diagnostic criterion or specifier for ASD is clearly overstated given that your findings are based on the DCD-Q.” “Relations between motor and other ASD-related impairments do not indicate a primary connection between motor problems and ASD.”

“Motor impairments are not specific to or universal in ASD.”

Response:

While writing the introduction for the three SPARK study papers (Bhat, 2020,2021; Bhat et al., 2022), I made conscious efforts to acknowledge the copious literature on motor difficulties in ASD that came before the SPARK study and confirmed the pervasive and specific nature of motor impairments in ASD based on multiple studies using standardized motor assessments including childhood assessments such as the Bruininks-Oseretsky Test of Motor Proficiency (BOT), Movement-Assessment Battery for Children (MABC), Test of Gross Motor Development, (TGMD) or Physical and Neurological Assessment of Subtle Signs (PANESS), and infancy/early childhood measures such as the Vineland Adaptive Behavioral Scales (VABS), Mullen Scales of Early Learning (MSEL), Alberta Infant Motor Scale (AIMS), or Peabody Developmental Motor Scales (PDMS) (refer to citations in the introduction sections of Bhat, 2021; Bhat et al., 2022). Hence, both, the SPARK study findings based on DCD-Q and extensive preexisting literature using various motor measures lend support to the notion of adding motor to the definition of ASD.

Studying associations and predictions after controlling for covarying factors is a common research practice to explain connections between developmental domains and has been used by many others. In Bhat (2021) and Bhat et al. (2022), I have reviewed the extensive past evidence on how fine and gross motor impairments are associated with and predictive of performance in other developmental domains including cognitive, social communication/language, and adaptive functioning skills of young and older children with ASD, in many cases after controlling for age, sex, and non-verbal IQ/ability (refer to citations in the introduction sections of Bhat, 2021; Bhat et al., 2022). The SPARK study dataset only extends and validates past findings in a large sample of children with ASD using the DCD-Q screener because it is not feasible to administer a full motor assessment in such a large sample. However, we still need to conduct larger studies wherein movement experts objectively evaluate motor problems of children with ASD. The behavioral and neural evidence on specificity of motor impairments in ASD indicate that visuomotor (ball and balance), praxis/motor planning, and fine motor coordination skills are specifically affected in ASD (refer to citations in the introduction section of Bhat et al., 2022) and have been linked to clear neural abnormalities (see discussion sections of Bhat, 2021; Bhat et al., 2022). The factor analysis of the DCD-Q data from the SPARK cohort revealed unique factors compared to the general population with subdomains of visuomotor, multilimb coordination/planning, and fine motor skills being uniquely affected in children with ASD (Bhat et al., 2022). Future studies must examine whether these patterns of motor impairment are differentially affected in ASD versus other clinical populations (DCD, ADHD, etc.). About 35%–97% of children with ASD are reported to have motor impairments using various standardized measures (Bhat, 2020; Green et al., 2009; Licari et al., 2019; Miller et al., 2021). The reported prevalence may vary depending on whether the study includes children with moderate motor delays. In my opinion, recognizing moderate motor delay (more common than severe delay) when screening young children at risk for/with ASD will be important to initiate intervention and prevent further motor decline. In summary, the question of representing motor issues within the definition of ASD has been asked before (Hilton et al., 2012; Licari et al., 2019) and I cannot take credit for asking the question. Overall, ASD is a spectrum disorder with children having wide-ranging symptom severity. While there are a substantial number of children having motor problems based on symptom severity, not all children with ASD will have functionally significant motor impairments.

2. “We urge caution in using the DCD-Q screening tool because DCD-Q has not been validated for children with ASD” and mainly assesses adaptive motor impairment and not core motor difficulties.

Response:

The DCD-Q has been validated against the MABC in school-age children with ASD in two different studies conducted in UK and Belgium (N = 97 in Green et al., 2009; N = 115 in van Damme et al., 2021). These studies report 71%–87% accuracy in diagnosing motor issues using the DCD-Q, which was confirmed by a concurrent MABC assessment. The 15 DCD-Q items are very similar to items on other screening tools that ask parents to assess their child’s performance on a particular skill, for example, “Does your child throw a ball in a controlled and accurate fashion?” This is very similar to the question in the widely used Ages and Stages Questionnaire (ASQ), “Does your child throw a ball overhand in the direction of a person standing 6 feet away?” Generally, all screening tools have similar questions for various developmental domains including motor, personal-social, communication, and cognitive/problem solving and we take their results on face value once they have gone through the rigor of psychometric validation. The same standard should be and has been applied to the DCD-Q.

3. There are a variety of problems in children with ASD, why then include motor issues within the ASD definition.

Response:

ASD is fundamentally a neurodevelopmental disorder and nervous system domains include cognitive-attentional, social communication, affective, sensory-perceptual and motor. Historically, motor symptoms in various mental disorders (e.g., schizophrenia and ASD) have not received much attention (Mittal et al., 2017). For the diagnosis of ASD, data from large-scale cohorts from the SPARK and Western Australian Register studies (Bhat, 2020, 2021; Licari et al., 2019) and the growing evidence on abnormal neural correlates of motor function (see citations in the discussion sections of papers under question) directly link motor issues in ASD to ASD neuropathology. Recently, the National Institute of Mental Health modified its Research Domain Criteria (RDoC) matrix to include the Motor System Domain (RDoC Framework Development). Through this modification, the mental health community recognized the complex interactions between cognitive-attentional, social-affective, and sensori-motor symptoms as a result of mental disorders. Representing the motor system within the definition of ASD would only align the DSM with the more progressive RDoC framework. In fact, I would argue that the entire DSM could be reframed using the RDoC framework with better recognition of the transdiagnostic nature of co-occurring impairments that are generally ignored/untreated as comorbidities and are shared across various developmental disorders such as ASD, DCD, Attention Deficit Hyperactivity Disorder (ADHD), etc. as well as other neurological disorders.

4. Motor and social problems are present early on in life in children with co-occurring intellectual disability (ID) and should not be diagnosed separately in children with ASD + ID.

Response:

Regarding the statement, “Motor, language, and social delays in children with ASD + co-occurring ID should not be co-diagnosed due to their early presence and be explained by ID itself”, I think this rule is being inconsistently applied to motor issues because language impairments are already one of the ASD specifiers and motor impairments could follow this precedent and be included as an ASD specifier.

In Bhat (2021) and Bhat et al. (2022), I have reviewed the literature on how motor impairments vary in children with ASD with and without ID. While motor problems increase in severity in children with ASD as a function of cognitive delay, multiple studies confirm that motor incoordination/dyspraxia is frequent in high-functioning children with ASD and is associated with and predictive of ASD severity even after controlling for IQ (see citations in the introduction sections of Bhat, 2021; Bhat et al., 2022). The recent Ketcheson et al. (2021) paper also confirms this in the SPARK study cohort.

Many recent studies in my lab and others have made conscious efforts to ensure that during motor assessment standard ASD treatment principles are followed (e.g., providing visual, verbal, and hand-on-hand models, breakdown of task, picture schedules, and appropriate reinforcements including rewards and breaks) to confirm a child’s understanding of the steps in the motor task and to sustain their interest in activities (Kaur et al., 2018; Liu & Breslin, 2013). This approach is consistent with other psychological testing that assesses a child’s social interaction and communication skills. Based on my 13 years of experience testing motor skills in children with ASD and given the abundant current evidence for motor issues in ASD, I can assure you that the motor issues in many children are genuine and reflective of their core motor challenges. Visuomotor and dynamic balance/multilimb skills such as throwing a ball to a target, dribbling a ball, jumping on one leg or side to side, etc., fine motor skills such as copying shapes, cutting shapes, writing legibly, etc., as well as everyday functional motor skills like dressing, tying shoe laces, climbing stairs, transitions from one location to another, and outdoor play/sports are challenging motor skills that cannot be simply ignored and attributed to disinterest and lack of opportunity. When parents first encounter their child’s challenges during motor testing, they are surprised and ask why no one has addressed these problems all these years. At the same time, I will also concede that not all children with ASD will have functionally significant motor impairments that warrant intervention.

5. Children with ASD without ID are perhaps disinterested in motor activities, lack opportunity for motor participation or fear of social interactions, and in fact core motor competence is not affected in these children.

Response:

As mentioned earlier, recent studies confirm motor problems in high-functioning children with ASD while ensuring task understanding and engagement using common ASD treatment principles. In spite of these efforts, children with ASD continue to demonstrate motor challenges so we cannot ignore their motor challenges under the veil of disinterest and social difficulties. They continue to show motor challenges even when working with familiar individuals (their caregivers) indicating genuine motor difficulties. If we trust parental opinions on other screening tools such as the Social Communication Questionnaire (SCQ) and Repetitive Behaviors Scale-Revised (RBS-R) then we cannot single out the findings on the DCD-Q as being inaccurate and reflective of disinterest and lack of opportunity. I would argue that a variety of social skills are in fact motoric in nature and require the child to turn their head to a caregiver, show, give, take objects from them, and physically approach them, and these skills are also goal-directed actions that could be impacted by disinterest and lack of opportunity. If we trust the reports of social communication challenges faced by children with ASD, then why do we not trust parental reports of their motor challenges?

In conclusion, based on the discussions at the INSAR 2022 panel on where motor fits within the broader framework of ASD, I am in agreement with Bishop et al. that motor impairments can follow the precedent of language impairments (Rosen et al., 2021) and be included as a specifier of ASD because motor problems co-occur in other diagnoses and including them as a specifier allows them to be independently addressed by movement experts. Adding motor as a specifier to the definition of ASD will help recognize motor issues that have been historically ignored by the ASD clinical community and bring them on the radar of diagnosticians who will be encouraged to systematically screen for motor problems throughout the lifespan using screening tools such as the little DCD-Q, DCD-Q, and adult dyspraxia screeners. When an individual with ASD fails on the age-appropriate DCD-Q and there are caregiver concerns about motor challenges, further motor assessment and interventions should be triggered through referral to movement clinicians (OTs and PTs). In the long-term, caregivers will be able to advocate for motor assessments and interventions as well as greater physical activity and social participation for their children by requesting OT, PT, adaptive physical education, and/or creative movement/recreational therapies. Motor, social communication, cognitive, and functional challenges are intimately linked in ASD and promoting motor skills through play, motor practice, and individual and community-based physical activity throughout the lifespan will facilitate functional independence, social participation, as well as improved mental well-being and quality of life in individuals with ASD with unmet motor needs.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.