Abstract
This study examined the new parent-report assessment’s reliability and validity, the Eyuboglu Sensory Reactivity Scale (ESRS), in children with ASD. The ESRS was designed to assess the frequency of sensory behaviors in children with ASD. A total of 167 (age 2–18, M = 6.4±4.1 years, 79.6 % male) children and their families were included. Parents were asked to fill in the ESRS and the Autism Behavior Checklist (ABC). In addition, the Childhood Autism Rating Scale (CARS) was administered by the clinician. The psychometric properties were examined through reliability and validity analyses. According to explorative factor analysis, the ESRS consists of a 15 five-point Likert-type that has acceptable internal consistency and good validity for children aged 2 through 18. Cronbach’s alpha of the scale was 0.85. The scale also showed good concurrent and convergent validity. The ESRS has the potential to measure unusual sensory reactions and can be used to follow up sensory interventions in children with ASD.
Autism Spectrum Disorder (ASD) is an early-onset neurodevelopmental disorder characterized by a deficit in social communication and interaction and restricted, repetitive patterns of behavior, interests, or activities (American Psychiatric Association, 2013). In addition to the social and communication characteristics of autism, sensory processing difficulties are common features that might be frequently overlooked. In the latest version of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) sensory issues have been defined as one of the core symptoms of autism (American Psychiatric Association, 2013). In DSM-5, hyper-reactivity and hypo-reactivity, including sensory seeking behaviors, were added to the restricted and repetitive behavior domain and described as “Hyper- or hypo-reactivity to sensory input or unusual interests in sensory aspects of the environment” (American Psychiatric Association, 2013).
As Kanner’s first description of autism includes some unusual reactions to the sensory stimulus (Kanner, 1943), many studies have demonstrated that children with ASD mostly show atypical sensory reactions (Eyuboglu & Eyuboglu, 2020; Lane et al., 2014; Tomchek & Dunn, 2007; Zachor & Ben-Itzchak, 2013). Although the nomenclature of sensory-related symptoms varies among researchers, studies reported up to 90–95% of sensory-related symptoms in individuals with autism (Leekam et al., 2007; Tomchek & Dunn, 2007). The estimated sensory features have been frequently described in three main patterns: hyperreactivity (an overreaction to sensory stimuli, i.e., covering ears to sounds), hyporeactivity (a decreased response to sensory stimuli, i.e., no response to painful stimuli), and sensory seeking behavior (unusual sensory interests, i.e., licking or smelling object) (Ausderau et al., 2014). Sensory experiences may be related to auditory, visual, tactile, gustatory, kinesthetic, or proprioceptive systems. These unusual reactions may vary between children with ASD by age and autism severity but usually impact children’s daily life (Zachor & Ben-Itzchak, 2013) and may disrupt family functioning. It also caused substantial impairment in social functioning and adaptive behavior (Boyd et al., 2009; Hilton et al., 2010; Pfeiffer et al., 2005). Increased sensory reactions may affect occupational performance, social participation, and daily family routine (Dickie et al., 2009). In addition to the nature of the disorder, unusual sensory behavior may limit children’s participation in community activities. A relationship between sensory reactivity and autism severity has also been shown (Boyd et al., 2010), and unusual sensory behaviors have been demonstrated to be a risk factor for self-injurious and repetitive and restrictive behaviors (Duerden et al., 2012; Lidstone et al., 2014). Moreover, researchers suggest that unusual sensory reactions are among the earliest markers of autism (Clifford et al., 2013; Germani et al., 2014). Though the causal relationship between sensory reactions and autism is not fully understood, sensory features are important to understanding the child’s emotional and behavioral attitudes, as shown in the literature. Therefore, proper identification of specific patterns of sensory reactions that could be correlated with emotional, behavioral, and cognitive skills may help professionals choose appropriate interventions. Modifying the sensory components could enhance the family’s well-being and improve the child’s maladaptive behavior and life quality. It is also important to note that the inclusion of unusual sensory behaviors into DSM-5 has become a key component of ASD diagnosis and emphasizes the importance of developing suitable measures.
There are some measures designed to assess the sensory symptoms in children with autism, such as the Sensory Profile (Dunn, 1999), the Sensory Behaviour Questionnaire (Neil et al., 2017), the Sensory Experiences Questionnaire (Baranek et al., 2006), and the Sensory Processing Measure (Boyd et al., 2009). These parent-report instruments are valid for obtaining information about children’s behavior, but one of the major limitations of current scales is measuring sensory behaviors with social and emotional domains. However, some scales were developed to gather general sensory-related symptoms rather than autism-specific behaviors. Further, several observational tools also exist. The Sensory Processing Scale Assessment (Schoen et al., 2014) and the Sensory Processing Assessment for Young Children (Patten et al., 2013) are among the clinician-administrated observational tools to assess sensory behaviors in children. The long implementation duration and narrow age range make it hard to apply these observational tools. Besides, children with ASD, particularly young children, are unlikely to report their sensory experiences due to impairment in language and cognitive skills (Dickie et al., 2009). Therefore, parent-rated scales are mainly used to assess sensory reactions. Despite there is no consensus on how to best measure sensory behaviors, it is necessary for reliable and valid scales that optimally assess the full range of sensory-related symptoms, including hyper-reactivity, hypo-reactivity, and sensory seeking behavior as delineated in the DSM-5.
In summary, even several instruments have been designed internationally to gauge parents’ perception and description of sensory behaviors in autism; none has been mainly developed for the Turkish sample. Hence, to address this gap, the current study investigated a new parent-rated scale’s psychometric properties to assess sensory reactivity in children with ASD. The Eyuboglu Sensory Reactivity Scale (ESRS) was developed to estimate various sensory behaviors in children and adolescents with ASD. This purely sensory scale will allow professionals to observe multiple sensory domains such as hyperreactivity, hyporeactivity, and sensory-seeking behavior. Finally, the current study aimed to investigate the factor structure of the new ESRS and its reliability and concurrent validity with previously validated measures.
Methods
Participants and Procedure
The current study sample consisted of 167 children and adolescents with ASD (male = 133, female = 34) and their parents. The sample size was determined by at least 5 participants per item to conduct factor analysis. The participants’ age ranged from 2 to 18 years, with an average of 6.4 years (SD = 4.1). Participants were recruited from a pediatric psychiatry outpatient clinic randomly. All children have been diagnosed by an experienced child and adolescent psychiatrist according to the DSM-5 (American Psychiatric Association, 2013) with a detailed clinical assessment. Diagnosis of autism was carefully made based on clinical assessment, behavioral observation, caregiver and teacher report, and psychological evaluation. Before the data collection, the current study’s aim and procedures were described to parents and participants. All participants and their parents were informed that participation was voluntary. Written and informed consent was obtained from parents who agreed to participate in the study. Consent was also sought from participants who were able to do so. Then, the clinician collected sociodemographic data such as age, gender, parents’ age, and level of education. All parents completed the Autism Behaviour Checklist and the Eyuboglu Sensory Reactivity Scale. The clinician administrated the Childhood Autism Rating Scale to evaluate autism severity. In the participation process, six families refused to participate in the study.
The study was approved by the local Ethics Committee.
Measures
Eyuboglu Sensory Reactivity Scale: Scale Development and Establishing Content Validity
The Eyuboglu Sensory Reactivity Scale (ESRS) is a parent-report instrument designed to evaluate unusual sensory reactions. The ESRS is intended to be used either for research purposes or clinical assessment. The scale development took place in several phases. First, the item pool was drafted with 48 items by the first author based on the current literature on sensory behaviors in children with ASD and the author’s experiences. Subsequently, the authors examined the items, and six items were found to be similar and removed from the items pool. Then, a group of five professionals (academicians from psychiatry, pediatrics, public health, and educational science) reviewed the items to assess content validity and coded each item as “appropriate,” “appropriate but needs revision,” and “not appropriate.” The examiners were anonymized to each other during the entire scale modification process. Based on all reviewers’ suggestions, the items coded “not appropriate” were removed, and corrections were made on items coded as “appropriate but needed revision.” Then, nine items were removed, and 12 were revised following the authors’ discussion on considering experts’ comments. The revised scale was sent back to the experts to assess each item again. When all items were rated “appropriate” by three experts, we considered a match to have been reached. Eight experts evaluated the suitability and comprehensibility of each scale item. The scale’s content validity rates ranged from 0.8 to 1.0, and the content validity index was 0.94. Finally, the draft of the scale consisted of 33-items, to be answered by parents, and was entered into quantitative analysis.
The final version of the draft ESRS covered items for auditory, tactile, visual, taste/smell, somatosensory, and kinesthetic systems and measured different sensory presentations (hyper-reactivity, hypo-reactivity, and sensory seeking). The items assessed the existence and frequency of unusual responses to various sensory stimuli or environments. Parents were asked to score each item on a five-point Likert scale (0 = never, 1 = rarely, 2 = occasionally, 3 = usually, and 4 = always). The scores are summed to create a total score in which higher scores indicate greater sensory behaviors levels.
Childhood Autism Rating Scale (CARS)
CARS is a behavioral observation scale consisting of 15 items developed to assess autism severity (Schopler et al., 2007). It is mainly used in the assessment process of autism and in differentiating autism from other neurodevelopmental disorders. The total score ranges from 15 to 60. Turkish validity and reliability studies of the CARS were performed, and the cut-off score was 30 (Incekas Gassaloglu et al., 2016). The scores between 30 and 36.5 indicate mild to moderate autism, and scores from 37 to 60 indicate severe autism.
Autism Behavior Checklist (ABC)
ABC consists of 57 items that assess autism symptoms and general behaviors often seen in children with autism (Krug et al., 1980). The ABC has questions grouped into five subscales: Sensory, Relating, Body and Object Use, Language, and Social and Self-Help skills. The scale scores range from 0 to 159, and higher scores indicate more symptoms (Krug et al., 1980). The Turkish validity and reliability studies were performed by Irmak et al. (2007).
Data Analysis
The Statistical Package for Social Sciences (IBM Corporation, Armonk, NY, USA) version 23 was used to analyze the data. The distribution of data was explored by the Kolmogorov-Smirnov test. The continuous data were summarized, and mean, SD, and categorical data were shown as frequency and percentages. Data was put into data analysis if missing items did not exceed 10% of items on a particular scale (7 cases were removed due to missing items exceeding 10%). The psychometric properties of the ESRS were interpreted with reliability and validity analyses. Cronbach’s alpha (1951) was used to calculate internal consistency, and the correlation coefficient was performed to assess concurrent and discriminant validity. Bivariate analysis was performed to explore the relation between ESRS, ABC, and CARS. Statistical significance was set to 0.05.
Results
Description of Sample
A total of 167 parents (1 parent per child) completed the assessment. The sample of children was composed of 79.6% male and 20.4 % female participants. Mean age of children with ASD was 6.4±4.1 years. The mean CARS score was found to be 40.7±5.6, indicating that most children with ASD were classified as having severe autism. According to the CARS scores, 31.1% of children with ASD were classified as mild-moderate, and 68.9% % of them with severe autism. The mean score of ABC was found to be 69.5±26.7. Additional information on participants and their family characteristics are presented in Table 1.
Description of Sample
Psychometric Properties
Factor Analysis
Bartlett’s Sphericity test and Kaiser-Meyer-Olkin (KMO) were calculated to assess the data’s suitability for structure detection. KMO value was 0.834, and Bartlett’s test was significant (p < 0.001), which shows the data is considered suitable for factor analysis. Exploratory factor analysis (EFA) was performed on the total sample of N = 167. According to the results of factor analysis, 18 items with a factor loading below the cut-off point of 0.40 were removed from the scale (items 4, 5, 6, 7, 10, 13, 14, 19, 20, 21, 22, 24, 25, 26, 27, 30, 31, and 32 of the initial item pool). As a result of the factor analysis, the unidimensional scale explained 32.8% of the total variance. The correlation value of items ranges from 35.6 to 58.8%. The factor loading range from 0.42 to 0.68.
Internal Consistency
The internal consistency of ESRS was adequate, with Cronbach’s alpha of 0.85 (95% CI: 0.83–0.87). Cronbach’s alpha values with items removed ranged from 0.83 to 0.84 (Table 2).
Factor Loadings, Adjusted Mean Score Correlation and Cronbach’s Alpha Values for Items of the Eyuboglu Sensory Reactivity Scale
There were no significant associations between total ESRS scores and children’s gender (r = –0.008, p = 0.915) and age (r = –0.127 p = 0.103). There was also no significant difference in ESRS total score by gender (t = 0.154, p = 0.877).
Concurrent Validity
To test the concurrent validity of the ESRS, bivariate analysis was performed between the ABC-sensory subscale and ESRS. High scores on both scales represent more sensory reactivity scores. The total ESRS and ABS-Sensory subscale correlated moderately (rho = 0.54, p < 0.001).
External Validity
The ESRS total scores were positively correlated with ABC and CARS scores confirming convergent validity (Table 3).
Associations Between ESRS, ABC, and CARS Scores
*p < 0.05 (2-sided). **p < 0.01 (2-sided).
The final version of the Eyuboglu Sensory Reactivity scale consisted of 15 items (see Table 2).
Discussion
There is a lack of accepted instruments to assess sensory features in children with ASD, particularly for Turkish populations, in Turkish language respectively. Further, since these behaviors are now defined as being the core symptoms of autism, reliable sensory reactivity measures are necessary. Detecting the sensory symptoms and their effect on mental health in children with ASD will provide valuable information to identify appropriate interventions. The present study tested a newly developed parent-rated scale, the Eyuboglu Sensory Reactivity Scale (ESRS), developed based on the DSM-5 criteria for ASD. The scale’s factorial structur was analyzed using Exploratory Factor Analysis. Results presented indicated an acceptable model. Our results suggest that ESRS consists of 15 items, rated on a 5-point Likert-type, has acceptable internal consistency, and has good validity for children aged 2 through 18. In addition, consistent with DSM-5, the ESRS measures sensory symptoms related to hypo-reactivity (item 1, 5, 6, 7, 13, 14), hyper-reactivity (item 3, 4, 9, 15), and sensory seeking behavior (item 2, 8, 10, 11, 12) and provides a quantitative measurement of sensory reaction that is commonly described by children with autism. The scale can be used as a sum score of all items.
Findings presented indicate that the ESRS can be used as a psychometrically valid measure to assess sensory-related behaviors in children with ASD. Besides, all parents participating in the current study completed the scale, suggesting that the ESRS is a feasible measure.
The ESRS provides additional information related to autism severity, which indicates convergent validity. The scale showed a significant association with both a parent-report measure (ABC) and a clinical observational tool (CARS), which assess autism symptomatology, meaning greater sensory reactivity is associated with higher autism severity. This result may show the increased ability of the existing scale to evaluate sensory behaviors in individuals with ASD. The significant associations between the ESRS, CARS, and ABC total scores show that the proposed scale has sufficient external validity. The ESRS was moderately correlated with the sensory subscale of ABC, which measures sensory behaviors in children with autism. Indeed, it would be better to correlate the ESRS with the Turkish version of sensory measures. Unfortunately, there is no specific scale translated to Turkish language and validated for children with ASD. Two scales in Turkish language assess sensory features: Adolescent/adult sensory profile (AASP) was adopted to Turkish language (Üçgül et al., 2017), and the Sensory Sensitivity Scale (SeSS) was developed and validated with a Turkish Sample (Aykan et al., 2020). However, The AASP has not been validated in children, and the study sample consisted of cases with schizophrenia. Another limitation of AASP was that the items do not merely focus sensory behavior but also cognitive and social abilities, which may result in inaccurate measurements of sensory behavior. Further, the SeSS was developed and validated for Turkish adults only. Another limitation of these two scales is that both scales are self-report instruments.
It is noteworthy that there was no effect of sex and age on ESRS. Although some studies showed that girls are more likely to show sensory-related symptoms than boys (e.g. self-report-study by Lai et al., 2011), we did not find any differences by gender. Contrary to studies suggesting a reduction in sensory reactivity with age (Humes, 2015), we could not confirm these results. This result may be attributed to the relatively small number of items in the ESRS.
Limitations
Several limitations need to be considered. First, although the sample size recommendations for EFA may vary, our sample size is below the frequently used ten respondents per item (Nunnally, 1967). Second, we did not have typically developing children for a comparison group. Third, test-retest reliability was not investigated. Besides, the scale’s ability to discriminate sensory behaviors between autism and other neurodevelopmental disorders such as intellectual disability and attention-deficit/hyperactivity disorder was not evaluated. Fourth, in the future, a subscale analysis of the ESRS should be performed. Finally, reversed items were not used because it is easier to complete the items in the same direction (Elwin et al., 2016).
Conclusion
In conclusion, sensory behavior is mostly assessed through questionnaires completed by parents or caregivers, particularly of minimally verbal or cognitive delayed children because of their communication difficulties. Other advantages of caregiver-report instruments may be easy implementation and short administration time (DuBois et al., 2017). This study demonstrated that the parent-rated ESRS is a psychometrically valid tool to assess sensory reactivity in children with autism. The ESRS is intended to be a part of the formal evaluation of diagnostic processes in the clinical settings and in catamnestic examinations. Additionally, the instrument can be used for scientific research purposes.
Footnotes
Bio Sketches
Murat Eyuboglu is an Associate Professor at the Department of Child and Adolescent Psychiatry, Eskisehir Osmangazi University, Turkey. His research focuses on neurodevelopmental disorders, particularly autism spectrum disorder, early identification, and intervention.
Damla Eyuboglu is an Associate Professor at the Department of Child and Adolescent Psychiatry, Eskisehir Osmangazi University, Turkey. Her research interests include infants and their learning environments and early intervention in autism spectrum disorder.
Nazlı Ece Karzan is a Child and Adolescent Psychiatrist at Adana Dr. Ekrem Tok Mental Health and Diseases Hospital, Turkey. Her research topics primarily include learning disorders and autism.
Ferhat Yaylaci is a Child and Adolescent Psychiatrist at Bursa Dortcelik Children’s Hospital. His primary research interests include autism spectrum disorder and pharmacotherapy in child and adolescent psychiatry.
Engin Karadag is a Professor at Akdeniz University Faculty of Education, Turkey. His research interests are behavioral sciences, educational research, social work, and sociology.
Didem Arslantas is a Professor at the Department of Public Health, Eskisehir Osmangazi University, Turkey. Her research interests include mental health, epidemiology, and health education.
