EARLY IDENTIFICATION of autism spectrum disorder (ASD) is critical so that intervention demonstrated to be effective if delivered during the preschool years can be accessed. The American Academy of Pediatrics recommends screening for ASD at 18- and 24-month well child visits using a standardized screening instrument so that children at risk can be referred for an ASD evaluation and treatment. Studies of toddlers and preschool children with ASD show reduction in autistic symptoms and improvements in IQ, development, and adaptive behavior relative to controls in children receiving intensive early behavioral intervention/applied behavioral analysis (Birnbrauer & Leach, 1993 ; Dawson et al., 2010 ; Howard, Sparkman, Cohen, Green, & Stanislaw, 2005 ; Lovaas, 1987 ; Sheinkopf & Siegel, 1998 ; Smith, Groen, & Wynn, 2000 ; Warren et al., 2011) and that experimental children showed improvements relative to controls 4.5 years later (McEachin, Smith, & Lovaas, 1993). Research suggests that intensive early behavioral intervention is more effective if delivered before versus after the age of 5 years (Fenske, Zalenski, Krantz, & McClannahan, 1985) and before versus after the age of 4 years (Harris & Handleman, 2000). Meta-analyses reveal that early intensive behavioral intervention is a strong evidence-based practice for young children with ASD (Howlin, Magiati, & Charman, 2009 ; Reichow, 2012) and has stronger research support than other interventions (Eldevik et al., 2009 ; Foxx, 2008). However, a review published by the Cochrane Collaboration states that findings on early behavioral intervention should be interpreted with caution because of small sample sizes and lack of random assignment in all but one study (Reichow, Hyme, Barton, & Boyd, 2018). In addition, recent reviews have determined that other interventions meet criteria for evidence-based practice, especially for older children, including social skills groups (Reichow, Steiner, & Volkmar, 2012) and cognitive behavioral intervention (Wong et al., 2015).
Given the critical need to identify ASD early so that children can access evidence-based intervention, valid and clinically feasible diagnostic tools are necessary for use with toddlers and preschoolers. This is important because most of the DSM-5 “deficits in social communication and social interaction” ASD criteria descriptions (e.g., “failure of normal back-and-forth conversation” and “difficulties in ... making friends”) are more applicable to older than to very young children, making it difficult for clinicians to apply DSM-5 criteria to toddlers and preschoolers. The DSM-5 requires all three social communication/interaction symptoms to be present for a diagnosis, which is problematic for very young children for whom DSM-5 symptom descriptions are above their developmental level. According to Zwaigenbaum et al. (2009), “unique challenges exist in applying current diagnostic guidelines for autism spectrum disorders in children under the age of 2 years” (p. 1383).
ASD assessment instruments that have support for use with young children may be too time-consuming for routine clinical use. For example, the Autism Diagnostic Interview-Revised (ADI-R) requires 90–150 min for the parent interview and scoring. The Checklist for Autism Spectrum Disorder (CASD; Mayes, 2012) and the Checklist for Autism Spectrum Disorder-Short Form (CASD-SF; Mayes, 2017) are possible brief alternatives (the CASD parent interview and scoring can be completed in 15 min) that have normative and validity data for assessing ASD in children 1–17 years of age and differentiating ASD from other disorders. The purposes of this study were to determine (1) whether CASD and CASD-SF scores differed between toddlers and preschoolers versus older children with ASD in a sample of 1,266 children with ASD and (2) sensitivity and specificity of the CASD and CASD-SF in identifying toddlers and preschoolers with ASD and with diagnoses other than ASD.
This study is a retrospective analysis of existing clinical data from psychiatry and developmental pediatrics child diagnostic clinics and a secondary analysis of data collected by the Stoelting Company for the CASD standardization study. The study has institutional review board (IRB) approval. The IRB waived informed consent for children evaluated in the psychiatry child diagnostic clinic because analyses were conducted retrospectively on existing clinical data. Informed consent was obtained from parents of children evaluated in the developmental pediatrics clinic (which routinely consents patients) and in the CASD standardization study.
Autism spectrum disorder
All children in the sample of children with ASD were consecutive referrals diagnosed with ASD in a psychiatry diagnostic clinic. Diagnoses were made by two licensed PhD psychologists with extensive training and expertise in diagnosing ASD in toddlers, preschoolers, and older children. Diagnoses were based on a semistructured diagnostic interview with the parents, ASD and global behavior rating scales (Pediatric Behavior Scale and CASD completed by parents, teachers, childcare and early intervention providers, and therapists involved with the child), review of records, psychological testing (IQ, achievement, and neuropsychological), and clinical observations during the psychological evaluation. The licensed psychologists integrated data from all sources to make ASD diagnoses.
Children with ASD were divided into five age groups: 1–2, 3–5, 6–9, 10–13, and 14–17 years. These age groups differed in IQ (F = 18.7, p < .0001) and sex (χ2 = 19.1, p = .001). The groups did not differ significantly (p > .01) in race (χ2 = 7.3) or parent occupation (χ2 = 12.5). Children 5 years or younger versus older than 5 years differed in IQ (t = 4.1, p < .0001), sex (χ2 = 7.1, p = .008), and parent occupation (χ2 = 11.9, p = .001). Race was nonsignificant (χ2 = 2.7, p = .10).
Nonautistic developmental disability
Toddlers and preschoolers with disorders other than ASD (i.e., intellectual disability, developmental delay, language disorder, and/or ADHD) were from the psychiatry child diagnostic clinic, a developmental pediatrics child diagnostic clinic at the same institution, or the CASD standardization sample (Mayes, 2012 , 2017). The CASD was completed by licensed psychologists in the psychiatry and standardization samples and by a board-certified developmental pediatrician in the developmental pediatrics sample.
The 65 typical toddlers and preschoolers were from the CASD standardization study. The CASD was completed by licensed psychologists.
Demographic data for all samples are presented in Table 1. Toddlers and preschoolers with ASD versus other disorders did not differ in IQ (t = 1.5, p = .14) or sex, race, and parent occupation (χ2 = 0.2–2.9, p > .09).
The 30-item CASD (Mayes, 2012) is completed by a clinician based on a semistructured 15-min diagnostic interview directly asking the parents about each of the 30 autism symptoms that comprise the CASD, observations of the child during the clinical evaluation of the child, and information from teachers, childcare providers, and therapists who complete the CASD prior to the diagnostic interview with the parents. CASD symptoms are scored as present or absent based on lifetime occurrence.
As reported in the CASD manual (Mayes, 2012), the CASD is normed and standardized on 2,469 children with ASD, other disorders, and typical development 1–18 years of age, with IQs ranging from 8 to 146. All children in the normative sample of children with ASD had CASD scores at or above the autism cutoff (≥15). In the independent standardization study conducted by the Stoelting Company, the CASD administered by a clinician differentiated children with ASD from children with other disorders (e.g., intellectual disability, language disorder, and attention-deficit/hyperactivity disorder [ADHD]) and typical development with 99.5% accuracy using the cutoff 15 or more. In other studies, the CASD distinguished ASD from apraxia of speech (Tierney et al., 2015), reactive attachment disorder (Mayes, Calhoun, Waschbusch, & Baweja, 2017), ADHD (Mayes, Calhoun, Mayes, & Molitoris, 2012), and intellectual disability, learning disability, and language disorder (Mayes, 2012). Concurrent validity is strong. Diagnostic agreement between the CASD and the Childhood Autism Rating Scale (CARS) was 98%, agreement with the Gilliam Asperger's Disorder Scale was 94%, and agreement with the ADI-R was 93% (Mayes et al., 2009 ; Murray, Mayes, & Smith, 2011).
The CASD-Short Form (CASD-SF) consists of the smallest subset of items from the 30-item CASD that differentiated referred children with and without ASD with 100% accuracy in the CASD-SF development sample (Mayes, 2017). The six-item subset comprising the CASD-SF was cross-validated on an independent sample of children referred to a child psychiatry diagnostic clinic with and without ASD, on data from the CASD standardization sample, and on data from the CASD ASD normative sample, resulting in 98.5%, 97.6%, and 99.8% diagnostic accuracy, respectively. Agreement between CASD-SF diagnoses and diagnoses made using the CARS and ADI-R was 97.9% and 96.3%, respectively.
For the 1,266 children with ASD, Pearson correlations were used to examine the relationship between age and CASD and CASD-SF scores, and analysis of covariance ([ANCOVA], controlling for group differences in IQ and sex, post hoc t tests with a Bonferroni correction for the number of comparisons, and Cohen's d were used to determine differences in CASD and CASD-SF scores between the five age groups. CASD and CASD-SF scores for children 5 years or younger versus older than 5 years were compared using ANCOVA, covarying group differences in IQ, sex, and parent occupation.
Sensitivity and specificity (percentage of children correctly identified as having and not having ASD) using CASD and CASD-SF scores were calculated. CASD and CASD-SF scores for toddlers and preschoolers with ASD versus other disorders were compared with independent-samples t tests and Cohen's d.
CASD scores and age in children with ASD
For the 1,266 children with ASD, there was a significant negative correlation between CASD scores and age (r = −.20, p < .0001) but explained variance was only 3.8%. Mean CASD scores were lower for each subsequent age group (Table 2), but differences were only statistically significant for some age group comparisons. CASD scores did not differ significantly between toddlers and preschoolers (p = 1.0, d = 0.3), but toddler scores were higher than childhood, early adolescence, and late adolescence scores (F = 8.4, p < .0001, post hoc p < .03, d = 0.5–1.3). Preschool scores were higher than early adolescence scores only (F = 8.4, p < .0001, d = 0.4). CASD scores for the three older age groups did not differ from each other (d = 0.0–0.4). Mean scores for the five age groups were consistent with the normative mean of 21.7 (d = 0.0–0.4). CASD scores for children 5 years or younger (M = 22.4, SD = 3.4) were higher than scores for older children (M = 21.4, SD = 3.3), F = 19.8, p < .0001, but the effect size was small (d = 0.3). All children with ASD had a CASD score in the ASD range.
CASD-SF scores for children with ASD were not significantly correlated with age (r = −.05, p = .06, explained variance 0.3%) and did not differ between the five age groups (F = 1.1, p = .36). Mean CASD-SF scores for the five age groups were similar to the normative mean of 5.3 (d = 0.0–0.2). CASD-SF scores did not differ between children 5 years or younger (M = 5.4, SD = 0.8) and older children (M = 5.3, SD = 0.8), F = 1.8, p = .18, d = 0.1. For children with ASD, 99.7% had a CASD-SF score in the ASD range.
Sensitivity and specificity for toddlers and preschoolers
For toddlers and preschoolers with ASD, 100% and 99.7% had CASD and CASD-SF scores, respectively, in the ASD range (sensitivity). In the typical toddler/preschool sample, CASD and CASD-SF scores correctly identified 100% and 96.9%, respectively, as not having ASD (specificity). For toddlers and preschoolers with disorders other than ASD, 100% and 96.2%, respectively, were correctly classified without ASD. Toddlers and preschoolers with ASD had much higher (p < .0001) CASD and CASD-SF scores than toddlers and preschoolers with disorders other than ASD (t = 50.3 and 46.6, d = 5.2 and 5.4).
The study purposes were to compare scores on the CASD and CASD-SF for toddlers and preschoolers versus older children with ASD to determine whether scores differed by age and to determine sensitivity and specificity for the CASD and CASD-SF for toddlers and preschoolers. Results demonstrated high sensitivity and specificity (96%–100%) for the CASD and CASD-SF in identifying and ruling out ASD in toddlers and preschoolers. Toddlers and preschoolers with ASD had mean CASD and CASD-SF scores slightly higher than the mean for older children with ASD. Although statistically significant, the influence of age was clinically minimal (with small to medium effect sizes). Because toddlers and preschoolers had somewhat higher CASD and CASD-SF scores than older children, young age did not place children with ASD at an increased risk for ASD being missed by the CASD and CASD-SF.
A study limitation was that CASD and CASD-SF specificity for older children was not investigated in this study, which needs to be done in future research. A significant limitation is that the same licensed psychologists or developmental pediatrician conducted the diagnostic interview with the parents and administered and interpreted the evaluation instruments and integrated information from multiple sources and measures, including the CASD, parent and teacher rating scales, and other instruments, to make diagnoses. Therefore, the clinicians were not blind to diagnoses or the results of the various instruments and diagnoses were not independent of the instrument results. Because of this, results should be interpreted with caution and future research needs to replicate the findings.
The finding that the CASD and CASD-SF differentiated very young children with and without ASD is particularly important because DSM-5 social communication/interaction symptom descriptions are more appropriate for older children than for younger children, making it difficult to apply DSM-5 criteria to toddlers and preschoolers. The CASD and CASD-SF are brief and clinically feasible instruments with established validity that can be incorporated into a comprehensive evaluation to assist in assessing ASD in very young children. The CASD and CASD-SF can be completed by parents, teachers, therapists, and childcare providers prior to the diagnostic evaluation to provide the clinician with information to aid in conducting the diagnostic interview with the parents and integrating information from multiple sources in the diagnostic process. Identifying toddlers and preschoolers with ASD is critical so that they can receive early evidence-based intervention.
Birnbrauer J. S., Leach D. J. (1993). The Murdoch early intervention program after 2 years. Behaviour Change, 10, 63–74.
Dawson G., Rogers S., Munson J., Smith M., Winter J., Greenson J., Varley J. (2010). Randomized, controlled trial of an intervention for toddlers with autism: The Early Start Denver Model. Pediatrics, 125, 17–23.
Eldevik S., Hastings R. P., Hughes J. C., Jahr E., Eikeseth S., Cross S. (2009). Meta-analysis of early intensive behavioral intervention for children with autism. Journal of Clinical Child and Adolescent Psychology, 38, 439–450.
Fenske E. C., Zalenski S., Krantz P. J., McClannahan L. E. (1985). Age at intervention and treatment outcome for autistic children in a comprehensive intervention program. Analysis and Intervention in Developmental Disabilities, 5, 49–58.
Foxx R. M. (2008). Applied behavior analysis treatment of autism: The state of the art. Child and Adolescent Psychiatric Clinics of North America, 17, 821–834.
Harris S. L., Handleman J. S. (2000). Age and IQ at intake as predictors of placement for young children with autism: A four- to six-year follow-up. Journal of Autism and Developmental Disorders, 30, 137–142.
Howard J. S., Sparkman C. R., Cohen H. G., Green G., Stanislaw H. (2005). A comparison of intensive behavior analytic and eclectic treatments for young children with autism. Research in Developmental Disabilities, 26, 359–383.
Howlin P., Magiati I., Charman T. (2009). Systematic review of early intensive behavioral interventions for children with autism. American Journal on Intellectual and Developmental Disabilities, 114, 23–41.
Lovaas O. I. (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55, 3–9.
Mayes S. D. (2012). Checklist for Autism Spectrum Disorder
. Wood Dale, IL: Stoelting.
Mayes S. D. (2017). Checklist for Autism Spectrum Disorder
-Short Form. Wood Dale, IL: Stoelting.
Mayes S. D., Calhoun S. L., Mayes R. D., Molitoris S. (2012). Autism and ADHD: Overlapping and discriminating symptoms. Research in Autism Spectrum Disorders, 6, 277–285.
Mayes S. D., Calhoun S. L., Murray M. J., Morrow J. D., Yurich K. K. L., Mahr F., Petersen C. (2009). Comparison of scores on the Checklist for Autism Spectrum Disorder
, Childhood Autism Rating Scale (CARS), and Gilliam Asperger's Disorder Scale (GADS) for children with low functioning autism, high functioning autism or Asperger's disorder, ADHD, and typical development. Journal of Autism and Developmental Disorders, 39, 1682–1693.
Mayes S. D., Calhoun S. L., Waschbusch D., Baweja R. (2017). Autism and reactive attachment/disinhibited social engagement disorders: Co-occurrence and differentiation. Clinical Child Psychology and Psychiatry, 22(4), 620–631. doi:10.1177/1359104516678039
McEachin J. J., Smith T., Lovaas O. I. (1993). Long-term outcome for children with autism who have received early intensive behavioral intervention. American Journal on Mental Retardation, 97, 359–372.
Murray M. J., Mayes S. D., Smith L. A. (2011). Brief report: Excellent agreement between two brief autism scales (Checklist for Autism Spectrum Disorder
and Social Responsiveness Scale) completed independently by parents and the Autism Diagnostic Interview-Revised. Journal of Autism and Developmental Disorders, 41, 1586–1590.
Reichow B. (2012). Overview of meta-analyses on early intensive behavioral intervention for young children with autism spectrum disorders. Journal of Autism and Developmental Disorders, 42, 512–520.
Reichow B., Hyme K., Barton E. E., Boyd B. A. (2018). Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD). Cochrane Database of Systematic Reviews, 5, CD009260. doi:10.1002/14651858.CD009260.pub3
Reichow B., Steiner A. M., Volkmar F. (2012). Social skills groups for people aged 6 to 21 with autism spectrum disorders. Evidence-Based Child Health, 7, 266–315.
Sheinkopf S. J., Siegel B. (1998). Home-based behavioral treatment of young children with autism. Journal of Autism and Developmental Disorders, 28, 15–23.
Smith T., Groen A. D., Wynn J. W. (2000). Randomized trial of intensive early intervention for children with pervasive developmental disorder. American Journal on Mental Retardation, 105, 269–285.
Tierney C., Mayes S. D., Lohs S. R., Black A., Gisin E., Veglia M. (2015). How valid is the Checklist for Autism Spectrum Disorder
when a child has apraxia of speech? Journal of Developmental and Behavioral Pediatrics, 36, 569–574.
Warren Z., McPheeters M. L., Sathe N., Foss-Feig J. H., Glasser A., Veenstra-VanderWeele J. (2011). A systematic review of early intervention for autism spectrum disorders. Pediatrics, 127, 1303–1311.
Wong C., Odom S. L., Hume K. A., Cox A. W., Fettig A., Kucharczyk S., Schultz T. R. (2015). Evidence-based practices for children, youth, and young adults with autism spectrum disorder: A comprehensive review. Journal of Autism and Developmental Disorders, 45, 1951–1966.
Zwaigenbaum L., Bryson S., Lord C., Rogers S., Carter A., Carver L., Yirmiya N. (2009). Clinical assessment and management of toddlers with suspected autism spectrum disorder: Insights from studies of high-risk infants. Pediatrics, 123, 1383–1391.