A Comparative Study of Adaptive Behavior and Intellectual Functioning in Children with Autism Spectrum Disorder, Attention Deficit Hyperactivity Disorder, and Typically Developing Children

Introduction

Autism Spectrum Disorder (ASD), is a neurodevelopmental disorder characterized by impaired communication, social interaction, and restrictive, stereotypical behavior. Current diagnostic specifiers include accompanying intellectual impairment, language impairment, other neurodevelopmental or behavioral conditions, and known medical or genetic conditions.¹

Assessment of intellectual ability, which renders an intellectual quotient (IQ), has been a traditional method of assessing cognitive ability. Intelligence is one of the most closely studied variables of ASD, where IQ ranges from intellectual disability to extremely high IQ levels.² IQ is related to outcomes in academics and functioning in adulthood.³ The contribution of IQ is undeniable in planning individualized therapeutic or educational plans.⁴ However, there exist several challenges in obtaining IQ scores for children with ASD, due to symptoms like variable attention,⁵ sensory abnormalities,6 poor motor skills,⁷ executive dysfunction,⁸ and poor language processing.⁹

Along with intellectual ability, adaptive functioning is of equal importance, if not more. The Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) altered the diagnostic criteria for ASD placing a strong emphasis on adaptive behavior.¹ Adaptive behavior is the ability to lead a functional and independent life, coping with and overcoming various adaptive challenges, at various stages of life. It is the ability to translate their ‘cognitive capacities to real-life competencies’.¹⁰ It involves various skills like communication, practical skills of daily living and coping, social skills of play, and interpersonal interaction. It does not focus on ability, where assumptions are made based on how much the child should be capable of doing, but on what the child does.¹¹

Studies have shown a strong correlation between Vineland Adaptive Behavior Scale-II (VABS) and Wechsler’s scales.¹² The focus on IQ seems to be now diminishing, while adaptive behavior is seen as more illustrative of their abilities.¹³ Perry et al.,¹⁴ identified an “autistic profile” where high-functioning children tended to have higher IQ scores than adaptive behavior scores, and vice versa was true for low-functioning children.

The objective of the present study was to understand the relationship between adaptive behavior scores and the findings of IQ assessments in children with ASD. It would be beneficial if the interview-based adaptive behavior scale can replace the cognitive measure of IQ, in a clinical setting, and contribute to further treatment/ therapy plans.

Materials and Methods

The current study was part of a larger study that focused on the cognitive functions of children with ASD, attempting to ascertain their cognitive and neuropsychological profiles. The obtained profile was aimed to be compared with the profiles of those with Attention Deficit Hyperactivity Disorder (ADHD) and Typically Developing Children (TDC) groups. The study was approved by the Institutional Ethics Committee.

Research Design

A cross-sectional comparative design was used.

Participants

The children (ASD and ADHD groups) were recruited from the Child Psychiatry and Child Psychology OPD, of an urban tertiary hospital, for a duration of three years. All families were explained about the study, and participation in the study was strictly voluntary and based on informed consent.

Sample

The sample consisted of 110 children with a diagnosis of ASD, in the age range of 6.0 to 12.11 years (mean age = 7.9 years). Thirty-one children had a diagnosis of ADHD, in the age range of 7.0 to 12.11 years (mean age = 9.2 years). Convenient sampling methods were used for the clinical groups. Forty children from the community were recruited using purposive and snowball sampling methods, for the TDC group, in the age range of 7.0 to 12.11 years (mean age = 9.7 years).

Assessment Tools

Wechsler’s Intelligence Scale for Children – IV-India (WISC-IV): The Intelligence test used was the fourth edition of the WISC, with Indian norms. The WISC-IV is a multi-subtest assessment measuring various cognitive skills which are then grouped into four indices, and the Full-Scale IQ

(FSIQ). Ten subtests, with four optional substitute tests, yield the final scores. The four indices – Verbal Comprehension Index (VCI), Perceptual Reasoning Index (PRI), Perceptual Reasoning Index (WMI), and Processing Speed Index (PSI) are all interpreted in the same range as IQ (90 to 109 being average).¹⁵

Vineland Behavior Adaptive Scale – II (VABS-II): This was utilized as the adaptive behavior measure in the study. It has two forms- the parent/caregiver form and the survey interview form, both can be scored manually or using scoring software. The VABS measures adaptive behavior in four domains – Communication, Daily Living Skills, Socialization, and Motor Skills. The scale renders an Adaptive Behavior Composite (ABC) score, which is a total score of the child’s adaptive behavior.¹⁰ The current study utilized the parent/caregiver form.

Social Communication Scale (SCQ): It is a widely used screening tool for ASD for those above the age of four years, consisting of 40-item questionnaire with yes/no responses. It has statements based on the diagnostic criteria for ASD and has two forms – The Current version which takes into account the behavior for the past three months, and the Lifetime version.16 The present study utilized the Lifetime version which was answered by the parents.

Conners 3: The Conners Rating Scale assesses problem behavior in children in the age range of 6 to 18 years. The current edition has an updated scoring option based on the DSM 5 diagnostic criteria. There are several versions of the scale. The Conners Parent Rating Scale – Parent Short, is answered by parents and renders scores in six domains – Inattention, Hyperactivity, Learning problems, Executive function, Aggression, and Peer relations, with a cut-off of 60 for each domain.17 The current study utilized the Parent Short Version.

Procedure

For the ASD and ADHD groups, consultant evaluation and diagnosis at the OPD setting were completed, following which the child was referred for a detailed assessment. After consent was obtained to participate in the study, a detailed clinical interview was conducted with the parents and the child. Additionally, the SCQ and Conners Rating Scale – Parent Short were administered to confirm inclusion for study groups. The children who scored above the cut-off of 15 on SCQ for the ASD group, and below the cut-off for ADHD were considered for the study. The ADHD group required crossing the cut-off of 60 on the Inattention or Hyperactivity scales of Conners Rating Scale – Parent Short. The other domains of the Conners were not considered in this study.

The TDC group was sourced through snowball sampling from the community. A clinical interview confirmed the absence of any diagnostic neurodevelopmental, emotional, or behavioral disorder. They were required to score below the cut-off for both SCQ and Inattention and Hyperactivity domains of the Conners Rating Scale – Parent Short.

The IQ assessments were administered by a Clinical Psychologist. The VABS was given to the parents to answer, and a follow-up interview was conducted to ensure the questionnaires were answered adequately. The duration of the assessment was an average of two sessions per child.

Out of a total of 283 children recruited for the study, across the three groups, data from 181 children was taken for the final analysis after excluding the data that was not viable (rejected due to exclusion criteria, and drop-outs). The group-wise details of the study samples are depicted in Table 1.

Table 2 shows the number of children for each of the variables considered in the study. In the ASD group, 54 children could complete all the required subtests of the WISC IV, hence obtaining an FSIQ. Forty ASD children were unable to complete any of the four indices, while 16 children could complete at least one index, but not all four indices required to calculate FSIQ. The table also reveals how many of the 16 children were able to complete one of the four indices.

In the final ASD group, 54 children completed the IQ test, obtaining a full-scale IQ score, while 40 children could not complete the IQ test, where all 10 subtests of the WISC IV were administered, and they could not perform on subtests that were minimally required to calculate at least one index score, henceforth referred to the No IQ group. Sixteen children completed the IQ test partially, where they could complete enough subtests to calculate one or more indices but not all four required to calculate an FSIQ. IQ partial henceforth refers to the group where all 10 subtests of the WISC IV were administered, and only scores on a few subtests could be obtained that allowed for the calculation of one or more index scores and not the full-scale IQ. The VABS II was administered to all the children for obtaining adaptive behavior scores. The VABS II yielded the domain scores, for Communication, Daily Living Skills, and Socialization Skills, along with the Adaptive Behaviour Composite (ABC).

Results

The obtained data was analyzed using the SPSS software package. The data was initially subjected to tests of normalcy to understand the distribution. The ShapiroWilks tests indicated that for the entire sample, VABS had a normal distribution, while IQ had a non-normal distribution. Since one of the variables under study was found to be not normally distributed, further analysis was carried out using non-parametric statistics.

Table 3 shows the descriptive analysis of VABS scores for the groups in the study. Out of a total of 110 children in the ASD group, 54 had completed IQ scores, 16 had partial IQ, and 40 had No IQ. However, VABS questionnaires were not returned for four children. The ASD group had 106 children with a median of 66.50 (59, 74), ADHD had 31 children with a median of 77 (71, 87), and TDC had 40 children with a median of 86.50 (80.25, 97). The ASD group was divided into three groups based on their completion of IQ. The IQ completed group had 54 children with a median of 73 (66,79), IQ partial group had 16 children with a median of 68 (61, 73), and the No IQ had 40 children with a median of 59 (54.25,66).

The average of VABS scores across the three groups varied, ASD had the lowest average (66.5), ADHD (77), and TDC had the highest average (86.5). The averages varied within the ASD group with IQ completed having the highest average (73) followed by IQ partial (68) and No IQ (59).

To understand the relationship between the VABS and the WISC IV, along with their domains and indices, Spearman’s rho was done. The following tables and discussion will only focus on the relationship of the domains or indices of the other measure and not within the same measure, as it is understood that the measure will have a strong relationship, among its own subscales. 

Table 4 shows the correlation between the VABS and WISC IV and their respective domains for the entire sample being studied. There was a significantly strong positive correlation between IQ and VABS, for the entire sample. Even the domains of the VABS correlated well with the indices of the WISC IV. Overall, there was a significant correlation between the domains of the two scales and the ABC and FSIQ.

The correlation between the two measures was understood based on the clinical group. Table 5 depicts the correlation between the VABS and WISC IV, and their domains for the ASD group. The Motor Skills domain of the VABS did not share a significant correlation with the FSIQ or any of the indices. The Processing Speed index of the WISC IV also did not share significance with the ABC or any domain of the VABS.

Table 6 depicts the correlation between the VABS and WISC IV, and their domains for the ADHD group. The Motor Skills domain of the VABS did not share a significant correlation with the Processing Speed index. The Processing Speed index of the WISC IV did not share a significance with the ABC or any domain of the VABS. The Working Memory index of the WISC IV shared a significant relationship only with Motor Skills, among all the domains of the VABS.

Table 7 shows the relationship between the two measures in the TDC group. VCI of the WISC IV was strongly correlated with DLS, Socialization, and the ABC of the VABS. There was no correlation between the other indices of the WISC IV and the VABS, including FSIQ and ABC for the TDC group.

Discussion

The present study was designed to understand the relationship between adaptive behavior functioning and cognitive functioning. The objectives were to illustrate the relationship between VABS and the FSIQ scores for the entire sample and the study groups – ASD, ADHD, and TDC. We also sought to understand the relationship between VABS and IQ in terms of their domain scores and index scores.

There was a difference in the average VABS scores of the three groups, where the ASD group obtained the least ABC scores, the TDC group had the highest and ADHD group was in the mid-range. These findings are similar to clinical expectations and the existing literature, where children with neurodevelopmental disorders have lower adaptive functioning than their typically developing peers.^18,19 Children with ASD exhibit poorer adaptive skills when compared to TDC or other neurodevelopmental disorders, in the case of our study – ADHD group.^14,20-22

The current findings showed that VABS scores also strongly correlated with the IQ scores of the entire sample. Studies have shown adaptive behavior scores have a positive association with IQ^14,23 while some other studies have shown an inverse relationship between IQ and VABS for ASD.24 The domain-wise analysis of our data has shown some strong correlations between domains and indices, with some exceptions. The Motor skills of VABS had a non-significant relationship with all the indices of the WISC-IV. This is due to the nature of the items in the MS scale which measures gross and fine motor skills. The child’s motor skills have minimal impact on subtests of the WISC IV. In the ASD group, VCI had a negative non-significant relationship, where it is purported that children with higher verbal comprehension abilities are more involved with academics, reading, and hobbies that involve tabletop or linguistic-based activities, especially in the ASD group. Similarly, PSI also had a negative non-significant correlation with motor skills. Despite fine motor skills being required for both subtests of the PSI, we believe that children with lower PSI scores rush through the activity with lesser concern for accuracy. The presence of higher motor skills may have them hurry through the activity which results in poor accuracy and hence poorer PSI scores.

For the ADHD group, there was a positive correlation in our sample between the VABS and FSIQ. However, there was a non-significant relationship between the WMI and the PSI with the domain of the VABS and the ABC. Both the WMI and the PSI are domains that children with ADHD traditionally score poorly on, as indicated by lower median sores on these indices, due to which they have a non-significant relationship with the VABS domains and ABC. Among ADHD, the VABS scores are lower than neurotypical peers.²⁵ ADHD symptomatology, especially inattention has been linked to poorer adaptive functioning.²⁶

Interestingly the present study did not demonstrate a significant relationship between the VABS and IQ for the TDC group. As the VABS is a parent-rated tool, only the evaluation of the parents is taken into consideration. Parents of children with a neurodevelopmental disorder like ASD or ADHD are more aware of the children’s development, or the lack of it when compared to the children from the community.

IQ and VABS scores have been documented to have an association where lower IQ groups have lower adaptive scores.²⁷ The poorer correlation for TDC was due to them scoring higher on the VABS, closer to the ceiling level of the domains, as indicated by higher medians of the VABS domains for TDC in our sample. VABS also tends to have a lower correlation for above-average IQ groups, where the median FSIQ score for the TDC is in the above-average range.²⁸

VABS tends to measure skills more widely than those measured by IQ tests. Due to this, more children have higher VABS scores than IQ scores in the neurodevelopmental group.²⁹ Children with average and above average IQs have VABS scores lower than the IQ score, while children with low IQ scores have VABS scores higher than IQ.^14,23 Also, older children tend to have lower VABS scores. Although the average VABS scores were significantly higher among the lower IQ groups.^27,30

Cognitive assessment is an integral part of a child’s profile of skills,¹³ while the DSM-5 decrees adaptive functioning to be part of the diagnostic process for ASD and Intellectual developmental disabilities (IDD).¹ Very young children, presenting for an initial diagnosis most often have challenges in completing a cognitive evaluation.³¹

Research thus far, and the outcome of our study indicates that VABS has a good correlation with IQ in the neurodevelopmental disorder group, especially among those with lower IQs. Most of the children deemed uncooperative for assessment, actually have autistic symptomatology masking their skills and abilities. In the absence of being able to conduct an IQ test for young children with suspected neurodevelopmental disorders, the VABS should be used as a primary assessment tool to plan treatment and therapy, until the child acquires pre-requisite skills for a formal assessment (receptive skills, management of RRBs). This will ensure clinicians be more efficient, and decrease the pressure on the child and the parents who want to know the ‘IQ’ of the child.

Limitations

Based on the exclusion criteria of the study, the current study excluded children with comorbidities. Hence, the findings of this paper are restricted only to children with ASD with no comorbidities. It would be interesting to understand the association between VABS and IQ among other neurodevelopmental disorders, along with various combinations of their presentation. Secondly, our analyses did not include the demographic variables of age and gender, along with levels of IQ for the ASD group. Further research can ensure the above variables are considered. Lastly, the obtained sample for the ASD group was mixed in terms of age of diagnosis, firsttime diagnosis, and those with an existing diagnosis for repeat assessments. They also varied on the presence and amount of therapeutic intervention and medication.

Conclusion

The utility of adaptive behavior scales is paramount for children with ASD. The relationship between VABS scores and IQ has been important. Ideally, the adaptive and cognitive profiles together, help in building a robust treatment plan. While it is not always possible to complete an IQ test for children with ASD, the parent-rated adaptive behavior questionnaire is easier to complete. Our data establishes a relationship between IQ and VABS for the ASD group. We believe that in addition to adding a valuable facet to the child’s repertoire of skills, the VABS scores replace the IQ scores until the behavioral and symptomatic challenges are dealt with, to complete a detailed cognitive evaluation. The VABS scores can be seen as a function of IQ, particularly for young newly diagnosed children who are intervention-naïve, and treatment-naïve.

Ethics Statement:

The study received ethical clearance from the Institute Ethics Committee. All the children assessed and their families volunteered to be part of the research study. Informed consent was obtained from all parents.

Source of Support

There is no declaration of financial assistance or grants.

Declaration

The authors declare that there is no conflict of interest or community involvement and contribution in this article.