Adaptive behavior profiles in young children with autism spectrum disorder diagnosed under DSM-5 criteria

Highlights

• Adaptive behavior in children aged one to three with ASD versus non-spectrum developmental delays diagnosed under DSM-5 criteria were explored.

• An adaptive behavior profiles of motor skills > daily living skills > socialization > communication was found in both the ASD and comparison group.

• Cluster analyses separated children into two clusters characterized by mild versus borderline adaptive functioning.

• Findings support the view of ASD as a continuum of symptoms, as specified by the DSM-5.

Abstract

Background

This study explored adaptive behavior profiles in a clinical sample of well-characterized children aged one to three years with ASD.

Method

Profiles were compared to a sample of children with non-ASD developmental delays. Cluster analyses were performed to determine whether differences in adaptive skills effectively distinguished children with ASD from other young children presenting for assessment due to behavioral or other concerns, but who received other non-ASD diagnoses.

Results

A profile of motor > daily living > socialization > communication skills was found in both children with ASD and children with non-spectrum diagnoses, showing that this profile is not unique to young children with ASD. A two-group cluster solution was found which differentiated children by developmental functioning level rather than by diagnosis.

Discussion

The results of this study provide support for two developmental profiles for adaptive functioning in children with ASD: an average to borderline delayed profile and a borderline to more severely delayed profile that may remain stable or worsen over time. They additionally highlight the importance of delivering early targeted interventions to children with ASD who have greater deficits in adaptive functioning due to their association with poorer long-term outcomes.

Introduction

Adaptive behavior describes skills necessary for independent functioning within society, including practical, conceptual, social, and linguistic skills (Schalock & Borthwick, 2010; Sparrow, Cicchetti, & Balla, 2005). In individuals with developmental disabilities, level of adaptive functioning has been strongly linked with prognosis (Gillham, Carter, Volkmar, & Sparrow, 2000; Klin et al., 2007), and measurement of adaptive behavior is essential for tracking developmental gains and identifying appropriate educational and habilitative goals (Tassé et al., 2012). Children with Autism Spectrum Disorder (ASD) often demonstrate higher rates of deficits in adaptive than cognitive functioning, even when impairments in intellectual functioning are not present (Perry, Flanagan, Geier, & Freeman, 2009; Pugliese et al., 2015). Since adaptive skills are typically learned through interaction with one’s social environment, children with ASD are likely to have greater difficulties in this area due to the nature of their social deficits. For example, daily living skills such as personal hygiene are difficult for a child to learn if he or she does not attend to the adult modeling these skills. Initial difficulties in acquiring adaptive skills are compounded by evidence that the adaptive functioning of children with ASD does not increase at a rate proportional to their aging, resulting in relatively greater deficits over time (Green and Carter, 2014, Pugliese et al., 2015). Given that children can be reliably diagnosed with ASD under three years (Stone et al., 1999), having a clear understanding of adaptive behavior profiles associated with ASD in this age range can be pivotal in providing targeted interventions to facilitate developmental outcomes. The current paper adds to existing knowledge by examining adaptive behavior profiles in very young children at a pivotal age for early intervention, aged one to three years, who were diagnosed with ASD using Diagnostic and Statistical Manual – Fifth Edition (DSM-5) criteria (American Psychiatric Association, 2013).

Individuals with ASD of a broad age range may exhibit a unique adaptive behavior profile based on core adaptive areas: relative strengths in motor functioning, followed by daily living skills, communication, and a relative weakness in socialization (i.e., motor > daily living skills > communication > socialization) (Fenton et al., 2003; Mouga, Almeida, Café, Duque, & Oliveira, 2015; Paul, Loomis, & Chawarska, 2014; Perry et al., 2009; Yang, Paynter, & Gilmore, 2015); however, not all studies have shown this profile. A summary of studies published in the last 10 years reviewing adaptive behavior profiles and predictors of adaptive behavior in children with ASD can be seen in Table 1.

In summary, evidence for this profile has been found based on standard scores in under two-year-olds with moderate delays (Paul et al., 2014), and under six-year-olds with profound delays (Perry et al., 2009). Support has also been found based on age-equivalent scores in children aged one to five years (Yang et al., 2015), under six-year-olds with average to moderate delays (Perry et al., 2009), and in one to nine-year-olds with moderate to severe delays (Fenton et al., 2003). In contrast, a profile of daily living skills > socialization > communication has been found using standard scores in nonverbal three to 15-year-olds (Sparrow et al., 2005), in two to three-year-olds with low average cognitive functioning (Ray-Subramanian et al., 2011), and using standard and age equivalent scores in children aged one to four with moderate to severe delays (Fenton et al., 2003).

Inconsistencies in adaptive behavior profiles of children with versus without ASD may be attributed to ASD symptom severity, chronological age, intellectual quotient (IQ), language (Paul et al., 2014), and socio-cultural variables (Tassé et al., 2012). For example, young children with ASD have been found to be more impaired on the Receptive Language subdomain and Daily Living Skills domain of the Vineland Adaptive Behavior Scales-Second Edition (Vineland-II; Sparrow et al., 2005) than peers with non-spectrum developmental delay (Paul et al., 2014). Children with ASD aged six to 18 years were differentiated, regardless of cognitive functioning, by more significant adaptive socialization deficits when compared to children with other neurodevelopmental disabilities (Mouga et al., 2015). These findings may not represent the core adaptive profiles of ASD versus non-ASD as they were drawn from children with a wide age range who were likely to have experienced variability in adaptive outcomes because of the educational, intervention-based, and social influences they received up to that point.

These findings regarding adaptive profiles contrast to results from Balboni, Tasso, Muratori, & Cubelli (2016), who found significantly lower adaptive communication scores in six-year-old children with ASD compared to children without ASD, while scores for other domains were comparable. Samples were matched on age, gender, cognitive functioning, language abilities, and parents’ educational level. Receiver Operating Characteristic (ROC) and linear regression analyses were conducted with individual Vineland-II items to identify a) which Vineland-II item subsets from each domain would classify children with ASD versus without and b) the proportion of the ASD sample that would be correctly classified. Only items from the Communication and Socialization domains (i.e., Playing and Imitating skills, Following Instructions, Beginning to Talk, and Interactive Speech) significantly differentiated children, correctly categorizing 73% of the sample overall and 75% of children with ASD.

In other studies, adaptive functioning in ASD has been linked more strongly to child cognitive functioning or developmental level, rather than ASD severity (Klin et al., 2007, Mouga et al., 2015, Paul et al., 2014, Ray-Subramanian et al., 2011, Yang et al., 2015). For example, while adaptive communication or socialization was found to be unrelated to ASD symptom severity measured via Autism Diagnostic Observation Schedule-Generic (ADOS-G) (Lord, Rutter, DiLavore, & Risi, 2002) Comparison Scores (Gotham, Pickles, & Lord, 2009) in a sample of two-year-olds, it was related to cognitive functioning (Ray-Subramanian et al., 2011). In another large sample of children with ASD aged four to 17 years, Socialization, Communication, and Daily Living Skills were shown to be positively associated with IQ, and Communication and Socialization were poorer in older children, suggesting that adaptive functioning may show a relative decline with age (Kanne et al., 2011). This study additionally found that the Receptive Language subscale of the Communication domain and the Interpersonal Skills subscale of the Socialization domain were less strongly correlated with IQ than remaining subscales, suggesting that these elements of socio-communication, which are areas of deficit associated with ASD, were less strongly tied to cognitive development than other adaptive areas. Significant negative relationships were found between measures of Socialization and Communication on the Vineland-II and ASD severity based on the ADI-R (Rutter, Le Couteur, & Lord, 2003) and Social Responsiveness Scale (Constantino, 2012), whereas no significant relationships were found between adaptive behavior and measures of ASD severity using ADOS Social Comparison Scores.

Inconsistencies in adaptive behavior profiles may also be attributed to the format in which IQ, ASD severity, and adaptive behavior were measured and reported (i.e., standard scores, ratio IQs, age equivalent scores) (Kanne et al., 2011, Ray-Subramanian et al., 2011). Current knowledge on adaptive behavior profiles in young children with ASD is limited in that most published studies have used now out-of-date diagnostic criteria and assessment tools, children were assessed using measures that are not sensitive to within-group differences in early ages, and ASD profiles were explored across large age ranges in which children typically experience substantial growth and variability in adaptive skills. Best-practice recommendations from the Diagnostic and Statistical Manual – Fifth Edition (DSM-5; American Psychiatric Association, 2013) Committee on Intellectual Disability stated that adaptive behavior should be measured using assessments that are individualized, standardized, and culturally-appropriate. While many studies have implemented adaptive behavior assessments meeting these criteria to identify adaptive behavior profiles in children with ASD, few studies (Balboni et al., 2016, Kanne et al., 2011, Paul et al., 2014, Ray-Subramanian et al., 2011, Yang et al., 2015) have used the second edition of the Vineland (Sparrow et al., 2005). To our knowledge, only one study (Mouga et al., 2015) has also examined adaptive behavior profiles in children with ASD diagnosed under updated diagnostic criteria from the DSM-5 (American Psychiatric Association, 2013) or International Statistical Classification of Disease and Related Health Problems – 10th Revision (ICD-10, World Health Organization, 2010). Studies have also used a wide range of developmental assessments: the Bayley Scales of Infant Development: Second Edition, (Bayley, 1993); Mullen Scales for Early Learning (Mullen, 1995); Stanford Binet, various editions (Roid, 2005, Thorndike, 1972; Thorndike, Hagen, & Sattler, 1986); and Wechsler Intelligence Scales for Children: Third Edition (Wechsler, 1991), as well as score formats (full scale IQ, verbal or nonverbal IQ, verbal or nonverbal developmental quotients) to represent developmental functioning, which makes it challenging to compare outcomes across studies and draw clear conclusions. Given the wide age ranges examined to date and the importance of age in predicting adaptive functioning, there is a need for research on adaptive behavior profiles in younger children, focusing on narrower age ranges (e.g., Perry et al., 2009, Ray-Subramanian et al., 2011). In addition to identifying clear adaptive profiles in young children with ASD, it is necessary to determine how such profiles differentiate children with ASD from those with other neurodevelopmental disorders or delays.

To our knowledge, this is one of the first studies (e.g., Mouga et al., 2015) to explore how adaptive behavior profiles in early childhood are associated with an ASD diagnosis based on DSM-5 criteria (American Psychiatric Association, 2013). It extends upon previous studies exploring adaptive behavior profiles in children under three years of age (Paul et al., 2014, Ray-Subramanian et al., 2011, Yang et al., 2015) by presenting a clinical sample of children with ASD compared to children with non-ASD neurodevelopmental delays as they presented for their initial diagnostic evaluation. We provide a clinical characterization of our sample using measures considered sensitive to between and within-group differences in early childhood (Akshoomoff, 2006, Esler et al., 2015, Gotham et al., 2009; Hus, Gotham, & Lord, 2012). To further investigate the relationship between adaptive behavior and ASD symptoms, we adopted an individual differences approach by using cluster analysis procedures to determine how variability in adaptive profiles predicted diagnostic status and differentiated early profiles of functioning in ASD that were separated across clusters.