Sensory processing and its relationship to participation among childhood occupations in children with autism spectrum disorder: exploring the profile of differences

Abstract

Autism is a condition manifested by persistent deficits in social communication and social interaction across multiple contexts, and, sensory processing difficulties may further affect childhood-occupation and hinders their overall development. This study examines the sensory-processing and childhood participation of children with autism (6 to 10 years), and ‘age/gender-matched typical children as control. The Sensory Processing (SSP) and the Participation of childhood-occupation (PICO) measures were used to collect data from 93 parents of children with autism and 95 parents of typically-developing’ children, recruited from hospitals, school and care centres. This study had 74 percent (n = 72) male participants (i.e. 4:1 male to female ratio) in the autism group, and found sensory processing difficulties were associated with specific childhood-participation limitation. Children with autism had lower participation (across level, frequency and enjoyment) than the typical children. Sensory-processing’ difficulties for the children with autism (n = 93) were at 68.8 percent, compared to 21.5 percent in the typical group (n = 95), and were significantly different (p < 0.001). The differences in sensory-processing difficulties were significant (p < 0.05) between groups, across all domains, except for movement sensitivity (p > 0.28). Auditory filtering was the sole sensory-processing difficulty with significant correlations with all three dimensions of participation [i.e. difficulty (r = 0.36, p < 0.01), frequency (r = 0.22, p < 0.05) and enjoyment (r = 0.27, p < 0.01)]. Our finding highlighted that auditory-filtering difficulties is a key sensory processing difficulty, and it significantly lower the childhood-participation in the autism group. Adding to a growing body of literature on assessment in autism, there is a need for early screening, and to engage parents in early intervention to improve specific sensory processing deficits but also to focus on the child’s strengths, for a comprehensive supportive care.

Introduction

Autism spectrum disorders (ASD) is a condition manifested by persistent deficits in social communication and social interaction across multiple contexts (American Psychiatric Association 2000). An estimate of between 0.19/1000 to 11.6/1000 of population in the world is affected by ASD. (Chiarotti and Venerosi 2020, Center for Disease Control and Prevention 2014, Isaksen et al. 2013, Matson and Kozlowski 2011), and these incidences are rising steeply due to greater awareness, and earlier detection with more parents coming forth for assessment and treatment. Across Asia, the prevalence of ASD is estimated around 1.9 in every 10, 000 children (before 1980), and 14.8 in every 10,000 in 2007 (Sun and Allison 2010). In children with ASD, sensory-processing differences are common when compared with community controls (Ben-Sasson et al. 2009, Caron et al. 2012, Kern et al. 2008). A high percentage of between 69 to 95 percent of children with autism in Australia and United States displayed symptoms of sensory processing differences (Baranek et al. 2006, Leekam et al. 2007, Tomchek and Dunn 2007, Tomchek et al. (2014), with debilitating consequences on activity participations such as daily living skills and academic performance (Ashburner et al. 2008, Baker et al. 2008). These differences result in challenging distressing behaviour patterns (Adamson et al. 2006), and ultimately contributed to a poorer quality of life (Dunn 2001) for both the child and their family. Additionally, it increases the financial burden of families that try various therapies to overcome these challenges (Sharpe and Baker 2007, Kogan et al. 2008, Ou et al. 2015), resulting in mothers of children with autism reporting greater distress and depression than mothers of children without autism (Kiani et al. 2014). Therefore, the burden of care for children with autism should be given due attention, with more research data gathered to understand how and where an intervention can be targeted to reduce this public health burden. However, such studies are rather limited in developing countries around Asia. Therefore, this survey was conducted on Malaysian children with autism with age/gender-matched typically developing children recruited as controls. The aim of the study is to investigate the patterns and relationship between sensory processing profiles and child participation in childhood occupation. The findings will serve as the initial database for profiling differences in sensory processing for future in-depth research. The objectives of the study are: (1) to examine ‘sensory processing’ and ‘childhood participation’ patterns among children with autism; (2) to compare these sensory processing and participation patterns with controls (i.e. matched typically developing children) and finally, (3) to examine the relationship between the sensory processing pattern and child participation.

This study is based on a conceptual framework (Figure 1) informed by the Miller’s Sensory Processing Disorder (Kilroy et al. 2019), which extends the work of Ayres’s Sensory integration theory (Ayres 1972). Sensory processing is part of normal development whereby one interprets and responds to daily sensory experiences as reflected in one’s ability to self-regulate, interact socially, and develop adaptive behavioural skills (Ahn et al. 2004, Miller 2014, Bar-Shalita et al. 2008). About 5 to 16.5 percent of the general population (Ben-Sasson et al. 2009) have symptoms associated with sensory processing challenges, but it is only considered a problem when it interferes with functioning in daily life (Miller et al. 2017). Sensory-processing disorder refers to a set of impairments where sensory information is not adequately processed, resulting in functional differences in the activity participation of children with autism (Miller et al. 2009). Sensory processing differences include a heterogeneous set of symptoms that affects the way in which individuals use sensory information for emotion regulation, motor performance, social interaction, and daily life functioning at school, home, and in the community (Miller et al. 2017). Miller’s hypothesis is that children with autism have a sensory processing difference in any of the three components: I) Sensory-based motor disorder (dyspraxia or postural disorders), II) sensory discrimination (exhibiting problems in registration via signal detection and interpretation), and III) sensory modulation (signalling over-responsivity, under responsivity or sensory craver). (Miller et al. 2009). These components interact between one’s innate capabilities and environmental opportunities. Genetic factors may also make one more vulnerable to environmental toxins/stressors which interfere with sensory integrative processing (Ayres, 1972). Any limitations in any of the components of these sensory processing abilities can affect the level of difficulty experienced when performing any activity, the frequency of performance, and/or the enjoyment of activity. Participation promotes learning and development, but the ability to participate successfully in daily activities are influenced by one’s sensory integration abilities (Schaaf et al. 2010). Participation in childhood occupation is an important outcome of the intervention (Coster and Khetani 2008), and a significant indicator of quality of life (World Health Organization 2007). Childhood occupation is defined as the occupations (activities) in which children are involved, which include daily personal care, play, socialization, and education (Bar-Shalita et al. 2009).

Figure 1.

Conceptual framework with Sensory processing (independent variables) influences participation.

QOL = Quality of Life

Genetic, cultural, and environmental factors may influence the sensory processing abilities. Yet, most research on sensory integration are mostly from developed countries, and very few from developing countries, and existing data does not allow comparison studies. There are still many gaps in research pertaining to databases needed to understanding how these sensory processing profiles vary to increase understanding and for targeting specific underlying factors for more effective clinical interventions. The need to study sensory processing differences and to measure its relationship with patient-orientated outcomes such as participation in childhood occupation is still emerging.

Method

This is a case-control study with parents of children with autism and parents with typical sensory processing abilities from schools and centres around Kuala Lumpur. There is no specific community involvement, but permissions to recruit patients were obtained from the directors of five large public hospitals (i.e. Kuala Lumpur, Selayang, Serdang, and Tengku Ampuan Rahimah, Klang Hospitals) were approached. Ethical approvals were obtained from the Medical Ethics Committee of University Malaya Medical Centre and Research (962.25), the Ministry of Education (KP-BPPDP-603/5/JLD.10) to access and recruit the parents of ‘typical’ children attending school, and from the Ministry of Health (NMRR-12-1214-14435)/MREC of hospitals to access children with autism.

Study sample and recruitment

All participants were Malaysian children between the ages of 6 years and 10 years, matched for age, and gender. The inclusion criteria for children with autism is a diagnosis using either the ICD 10 (World Health Organization, 2007) or DSM-IV-TR (American Psychiatric Association 2000), confirmed by the attending paediatrician; 6 to 10 years old; and their parents must be able to speak and read the Malay language. The control group is the normative sample from the community with the inclusion criteria of i) never were diagnosed medical conditions that might compromise development (e.g. attention deficit hyperactivity disorder, down syndrome, cerebral palsy, significant prematurity at 28 weeks gestation or less, very low birth weight 2000 gram or below), 6 to 10 years old and, have no sibling/s with an autism spectrum disorder that might also compromise development, and parents must be able to speak, and read.

The parents of children with autism (case group), and parents of age/gender matched ‘typical’ children (community control group) were recruited using a snowball sampling strategy. Parents of children with autism (6-10 years old) were recruited from the Occupational therapy departments at five public hospitals (University Malaya Medical Centre, Kuala Lumpur Hospital, Serdang Hospital, Tengku Ampuan Rahimah Klang Hospital, Selayang Hospital) in the greater Kuala Lumpur region. Parents for the control group were recruited from tuition centres/care centres and from public schools.

Sample size

G*power calculation (Faul et al. 2009) was performed based on the differences between the two groups (case; control) at p ≤ 0.05, 95 percent powers and Cohen’s d of 1.08 (Bar-Shalita et al. 2008), on the three scales: (1) level of activity performance, (2) level of enjoyment of the activity, and (3) frequency of performance of the activity. Based on this previous study (Bar-Shalita et al. 2008), the mean (±SD) computed for participation in childhood occupation for both groups were reported as 125.3 (±16.94) and 148.5 (±10.04) for ‘level of activity performance’, 105.1 (±26.27) and 127.2 (±12.11) for ‘level of enjoyment of the activity’ and for the frequency of performance 59.30 (±13.95) and 67.60 (±12.5) respectively. Thus, a minimum of 21 participants was required per study group to detect a difference of the defined magnitude for the scale with the highest SD [i.e. level of enjoyment as measured with the Short Sensory Profile (SSP)]. A second calculation was performed based on the correlation between sensory processing patterns and participation in childhood occupation for children with autism. In this calculation, the conventional effect size was used since there was no previous study that examined this specific association for autism children. Assuming an effect size of 0.3 at 80 percent power, the sample size required for this study was 84 participants. With an added 10 percent attrition rate, the required sample size for children with autism would be at least 92 participants. Therefore, the higher sample size was followed for this study.

Data collection

Parents were provided with a written information sheet about the study and asked to sign a consent form. They were assured of confidentiality, anonymity, and freedom to withdraw at any time from the study with no bias to any future medical care.

A parent demographic (age, education, parent etc) and child profile (age, gender, and diagnosis) questionnaire were used to gather data about the participants (children and parents of both groups). Two measures were used to collect data – the Short Sensory Profile (SSP) and the Participation in Childhood Occupation (PICO). Permission to use and translate the SSP measure (McIntosh et al. 1999) into Malay was granted from Pearson Company (Pearson 2013). Permission for the use and translation of the PICO-Q was granted by its authors (Bar-Shalita et al. 2009). Parents completed the Short Sensory Profile (SSP) and Participation in Childhood Occupations Questionnaire (PICO). The SSP and PICO measures are discussed below.

Short Sensory Profile (SSP): The short sensory profile (SSP) is the most common measure of response to sensory features in children (3-10 years old), with autism spectrum disorder (ASD), being the short version of the original Sensory Profile (McIntosh et al. 1999, Dunn 1999), to measure sensory processing differences in children. It has seven subscales: Tactile sensitivity, Taste/Smell Sensitivity, Movement Sensitivity, Under-responsive/Seeking Sensation, Auditory Filtering, Low Energy/weak, and Visual/Auditory Sensitivity, and responses are on a 5-point Likert scale (frequently = 1 to Never = 5). Higher scores indicate more impairments. The total score is calculated by the sum of the section score. Interpretation of the total score is as follows: Definite difference indicates sensory processing differences, probable difference indicates questionable areas of sensory processing ability, and typical performance indicates typical sensory processing ability. The internal reliability (for the total and section scores) of SSP is of Cronbach’s alpha ranging from α  = 0.70 to 0.90. The internal validity correlations for the total and section ranged from 0.25 to 0.76. The findings from the short SSP are similar to the long version of the sensory profile (Bar-Shalita et al. 2009). SSP has been translated into Spanish (Román-Oyola and Reynolds 2013), Hebrew (Engel-Yeger 2010), Tamil (Sankar and Priyadarshini 2014) and in the Malay language (Ee et al. 2016).

Participation in childhood occupation

The Participation in Childhood-occupation is a 30-item caregiver questionnaire with five areas of functional activities: personal activities of daily living (13 items), academic activities (5 items), play and leisure (4 items), social skills (4 items), habits and routines (2 items), with two additional questions to measure the general level of participation in activities the child must perform (1 item) and activities the child chooses to perform (1 item) (Bar-Shalita et al. 2009). This 30-item PICO 2^nd edition has an extra 8 items from the initial 22-item PICO measure. It measures five areas of functional activities (instead of four) with the additional category of social skills. Every item describes an activity that is scored on three different dimensions: i) level of activity performance, ii) level of enjoyment of the activity, and iii) frequency of performance of the activity. Each of the three dimensions has a score for every performance area with 15 sub-scores on this PICO-Q 2^nd edition. For each item from each of the three dimensions, parents are asked to rate using a 5-point Likert scale for i) level of activity performance, ii) level of frequency, and iii) level of enjoyment of the activity; while a 4-point Likert scale was used for the frequency of performance of activity. The PICO tool was tested on children with and without sensory modulation difficulty and showed good internal reliability with Cronbach’s alpha (alpha= 0.86 to 0.89) and acceptable test–retest reliability index (r ≥ 0.69 and 0.86).

Data analyses

The demographic profiles of the children and their caregivers were described using mean, standard deviation, frequency, and percentages - to describe the sensory processing pattern and the participation pattern in children. The continuous data in the Short Sensory Processing (SSP) measure were transformed into categorical data according to the classification of “Definite Difference, Probable Difference, and Typical Performance” (McIntosh et al. 1999). The Participation of Childhood Occupation measures’ scores were transformed into dichotomous scales of “Not difficult, difficult”; “infrequent, frequent”; “not enjoyable and enjoyable” for its three aspects (quality) of participation. Chi-square, cross tabs were used to compare sensory processing and participation between children with autism and typically developing children with categorical variables; and to examine the association between sensory processing and participation of children with autism, Spearman’s Rank-Order correlation was used to correlate the nonparametric data. All tests were analysed as two-tailed and using the alpha level of 0.05 as the criterion for statistical significance.

Result

A total of 186 parents (93 children with autism and 93 typically developing children, age/gender-matched) participated. Most (77.4 percent/n = 72) of the children with autism in this study were male - an approximate ratio of 4:1 male to female (see Table 1).

Table 1.

Demographic data of children with autism and typically developing children and their caregivers.

Characteristics	Autism (n = 93)		Typically developing (n = 93)
	n	(%)	n	(%)	between groups
CHILD AG E
6 years old	27	29.0	27	29.0	p > 0.05
7 years old	31	33.3	31	33.3
8 years old	18	19.4	18	19.4
9 years old	11	11.8	11	11.8
10 years old	6	6.5	6	6.5
CHILD’S GENDER
Boy	72	77.4	72	77.4	p > 0.05
Girl	21	22.6	21	22.6
CHILD’S ETHNICITY
Malay	55	59.1	73	78.5	p > 0.05
Chinese	28	30.1	13	14
Indian	10	10.8	7	7.5
PARENT’S AGE
Below 21	4	4.3	5	5.4	p > 0.05
21-30 years old	58	62.4	53	57.0
31-40 years old	30	32.2	30	32.2
41-50 years old	1	1.1	5	5.4
RESPONDENT
Father	29	31.2	37	39.8	p > 0.05
Mother	64	68.8	56	60.2
Parent’s Highest education
Primary School	2	2.2	1	1.1	p > 0.05
Lower Certificate of Education	5	5.4	3	3.2
Certificate/Diploma	55	59.1	52	55.9
Degree/Master/PhD	31	33.3	37	39.8

A high percentage of children with autism (68.8%) compared to typically developing children (21.5%) experienced sensory processing differences, where total scores and all subsection scores across the short sensory processing scores measured were significantly different between the two groups. Children with autism experienced significantly greater sensory-processing differences, and the two most difficult categories were ‘Under-responsive/Seeks-sensation’ (77.4%) and ‘Auditory-filtering’ (61.3%); while the least difficult category was ‘movement sensitivity’ (33.3%). Across typical performance, the sensory-processing’ differences for children with autism (n = 93) were at 68.8 percent while in the typically developing children group (n = 93) it was at 21.5 percent. The differences were significant (p < 0.001).

The between-group differences for total scores measured with the Short sensory processing tool, and the subsection scores of Malaysian children with autism and that of the typically developing children group were significant for all, except for movement sensitivity (p ≤ 0.280). Levels of childhood participation were examined for - i) differences, ii) frequency, and iii) enjoyment. Children with autism have significantly higher levels of difficulty with social integration (p ≤ 0.05); lower frequency of participation in meal preparation (p ≤ 0.05) (meal preparation is considered a higher cognitive task requiring planning, sequencing and follow-through). A high frequency of participation (p < 0.05) was analysed in watching videos/DVD, eating, and drinking (which we considered as routine daily activities). For enjoyment in participation, children with autism did not enjoy social-interaction, but instead showed high enjoyment in solitary activities or daily routines such as watching videos/DVD on the computer or television, eating and drinking and meal time. (see Table 2)

Table 2.

Sensory processing in children with ASD (n = 93) and typically developing (n = 93).

SSP-M Group	Classification	n (%)			P value
		TP	PD	DD
Tactile sensitivity	ASD	28(30.1)	26(28.0)	39(41.9)	0.001
	Typical Developing	49(52.7)	24(25.8)	20(21.5)
Taste Sensitivity	ASD	25(26.9)	15(16.1)	53(57.0)	0.001
	Typical Developing	58(62.4)	19(20.4)	16(17.2)
Movement Sensitivity	ASD	42(45.2)	20(21.5)	31(33.3)	0.120
	Typical Developing	53(57.0)	21(22.6)	19(20.4)
Underresponssive Seeks Sensation	ASD	7(7.5)	14(15.1)	72(77.4)	0.001
	Typical Developing	45(48.4)	18(19.4)	30(32.3)
Auditory filtering	ASD	17(18.3)	19(20.4)	57(61.3)	0.001
	Typical Developing	58(62.4)	21(22.6)	14(15.1)
Low Energy/weak	ASD	44(47.3)	11(11.8)	38(40.9)	0.001
	Typical Developing	80(86.0)	6(6.5)	7(7.5)
Visual/auditory sensitivity	ASD	41(44.1)	14(15.1)	38(40.9)	0.001
	Typical Developing	70(75.3)	13(14.0)	10(10.8)
Total Score	ASD	10(10.8)	19(20.4)	64(68.8)	0.001
	Typical Developing	46(49.0)	27(29.0)	20(21.5)

Note: DD = Definite Difference PD = Probable Difference TP = Typical Performance.

ASD = Autism spectrum disorder.

With childhood participation, the differences in participation for most children with autism were in social integration participation and participation related to homework, organizing the study environment, and food preparation. There were lower frequencies of participation in food preparation, extra-curricular activities outside of school, playing with friends, and integration in social situations, but a higher frequency in solitary tasks like watching video/television. Children with autism had higher enjoyment in solitary and basic activities like video watching and meal time. Children with autism have greater differences, lower frequencies, and less enjoyments (compared to typically developing children) across many activities. More differences in sensory processing correlate with more differences in participation (see Table 3 below and Tables 4a and 4b in the appendices).

Table 3.

Difficulties in participation in Children with autism (n 93) and Children of typical development (n = 93).

Item	Group	Difficulty in participation		p
		Very difficult	difficult
Washing	Autism	51(54.8)	42(45.2)	<0.001
	Typically dev	82(88.2)	11(11.8)
Toileting	Autism	54(58.1)	39(41.9)	<0.001
	Typically dev	83(89.2)	10(10.8)
Dressing	Autism	66(71.0)	27(29.0)	<0.001
	Typically dev	80(86.0)	13(14.0)
Eating and drinking	Autism	34(36.6)	59(63.4)	<0.001
	Typically dev	78(83.9)	15(16.1)
Meal time	Autism	56(60.2)	37(39.8)	<0.001
	Typically dev	82(88.2)	11(11.8)
Simple food preparation	Autism	26(28.0)	67(72.0)	<0.001
	Typically dev	76(81.7)	17(18.3)
Meal outside the house	Autism	47(50.5)	46(49.5)	<0.001
	Typically dev	83(89.2)	10(10.8)
Mobility in immediate environment	Autism	30(32.3)	63(67.7)	<0.001
	Typically dev	82(88.2)	11(11.8)
Sleep	Autism	66(71.0)	27(29.0)	<0.001
	Typically dev	86(92.5)	7(7.5)
Assisting In chores at home	Autism	32(34.4)	61(65.6)	<0.001
	Typically dev	77(82.8)	16(17.2)
Abiding by school regulations	Autism	52(55.9)	41(44.1)	<0.001
	Typically dev	85(91.4)	8 (8.6)

Item	Group	Difficulty in Participation		P value
		n (%)	n (%)
		Not Difficult	Difficult
Being responsible for homework	Autism	20(21.5)	73(78.5)	<0.001
	Typcally dev	75(80.6)	18(19.4)
Organizing study environment	Autism	22(23.7)	71(76.3)	<0.001
	Typically dev	75(80.6)	18(19.4)
Working with different mediums and materials	Autism Typically dev	42(45.2) 80(86.0)	51(54.8) 13(14.0)	<0.001
Integrating during recess and other school activities	Autism Typically dev	39(41.9)	54(58.1)	<0.001
		83(89.2)	10(10.8)
Participating in extra-curriculum outside school setting	Autism Typically dev.	31(33.3)	62(66.6)	<0.001
		75(80.6)	18(19.4)
Recreational activities	Autism	52(55.9)	41(44.1)	<0.001
	Typically dev	83(89.2)	10(10.8)
Watching videos/DVD's on computer/television	Autism Typically dev	54(58.1)	39(41.9)	<0.001
		82(88.2)	11(11.8)
Being able to keep oneself occupied	Autism	41(44.1)	52(55.9)	<0.001
	Typically dev	84(90.3)	9(9.7)
Integrates in social situations	Autism	20(21.5)	73(78.5)	<0.001
	Typically dev	81(87.1)	12(12.9)
Plays with friends in afternoon	Autism	22(23.7)	71(76.3)	<0.001
	Typically dev	81(87.1)	12(12.9)
Attends parties and events of school friends	Autism Typically dev	27(29.0)	66(71.0)	<0.001
		83(89.2)	10(10.8)
Goes out with extended family/ family friends	Autism	57(61.3)	36(38.7)	<0.001
	Typically dev	84(90.3)	9(9.7)
Being flexible during transition between activities	Autism	51(54.8)	42(45.2)	<0.001
	Typically dev.	79(84.9)	14(15.1
Being flexible during unexpected changes	Autism	42(45.2)	51(54.8)	<0.001
	Typically dev.	78(83.9)	15(16.1)

Table 4a.

Frequency of participation -No significant differences between children with autism (n = 93) and typically developing children (n = 93).

Item	Group	Frequency of Participation n (%)		P value
		Not Frequent	Frequent
Eating and drinking	Autism	12(12.9)	81(87.1)	0.650
	Typically developing	10(10.8)	83(89.2)
Watching videos/DVD's on the computer or television	Autism Typically developing	11(11.8)	82(88.2)	0.468
		8(8.6)	85(91.4)
Meal time	Autism	15(16.1)	78(83.9)	0.189
	Typically developing	9(9.7)	84(90.3)
Abiding by school regulations	Autism	20(21.5)	73(78.5)	0.179
	Typically developing	13(14.0)	80(86.0)
Sleep	Autism	21(22.6)	72(77.4)	0.129
	Typically developing	13(14.0)	80(86.0)
Being flexible during unexpected changes	Autism Typically developing	29(31.2)	64(68.8)	0.063
		18(19.4)	75(80.6)

Table 4b.

Enjoyment in participation - No significant differences between children with autism (n = 93) and typically developing children (n = 93).

Item	Group	Enjoyment in Participation n (%)		P Value
		Do Not Enjoy	Enjoy
Recreational activities	Autism	12(12.9)	81(87.1)	<0.137
	Typically developing	6(6.5)	87(93.5)
Meal time	Autism	12(12.9)	81(87.1)	<0.137
	Typically developing	6(6.5)	87(93.5)
Watching videos/DVD's on the computer or television	Autism	11(11.8)	82(88.2)	<0.117
	Typically developing	5(5.4)	88(94.6)
Sleep	Autism	14(15.1)	79(84.9)	<0.105
	Typically developing	7(7.5)	86(92.5)
Eating and drinking	Autism	14(15.1)	79(84.9)	<0.105
	Typically developing	7(7.5)	86(92.5)

Table 5 shows that auditory filtering had a significant correlation with all three sections of PICO (difficulty in participation, frequency of participation, and enjoyment during participation), the strongest correlation being differences in participation. A significant positive correlation was found between sensory processing (total scores measured with the Short sensory processing tool) and participation (total score of each section in PICO), where auditory filtering is the only section in the sensory processing scale (SSP) that significantly correlates with all three sections of childhood participation (PICO) i.e. difficulty in level of participation (r = 0.36, p ≤ 0.01), frequency of participation (r = 0.22, p ≤ 0.05) and enjoyment in participation (r = 0.27, p ≤ 0.01).

Table 5.

Correlation between each section of SSP and sections of difficulties in performing activities in PICO (Spearman’s Rho) for children with autism.

Scale	PICO Difficulties	PICO Frequency	PICO Enjoyment
Tactile sensitivity	-.260**	.107	.240*
Taste Sensitivity	-.236*	.111	.154
Movement Sensitivity	-.066	.030	.051
Under responsive/Seeks Sensation	-.208*	.069	.024
Auditory filtering	-.361**	.222*	.265**
Low Energy/weak	-.270**	.094	.223*
Visual/auditory sensitivity	-.208*	.038	.157
SSP-M Total Score	-.385**	.146	.274

Note: Correlation is significant at ** p < 0.01 and * p < 0.05 level two tailed.

SSP = Short sensory processing scale.

PICO = Participation in childhood Occupation scale.

Discussion

This is the first study in Malaysia to compare sensory processing and participation in children with autism with age/gender matched typical children. There were more males than females in the autism group, which is consistent with published gender ratios of 4.5:1 (Lai et al. 2015). Typically developing children were age/gender matched to decrease confounding factors such as patterns of participation in activities which vary with age and gender (Little et al. 2014).

Sensory processing differences

In children with autism, ‘under-responsive/seeks-sensation’ (77.4%) and ‘Auditory-filtering’ (61.3%) categories were the most difficult for sensory processing, while movement sensitivity (33.3%) was the least difficult. Published studies on this area showed conflicting results. Some studies (Baker et al. 2008, Lane et al. 2010, Tomchek and Dunn 2007, Tomchek et al. 2014) are in support of this finding but others like Ermer and Dunn (1998) found lower incidence in the under-responsive/seeks sensation category. A high 68.8 percent of children with autism in this cohort experienced sensory-processing differences, and about 21.5 percent of typically developing children had sensory differences. This finding in children with autism with greater sensory processing differences is in line with other studies (Brockevelt et al. 2013, Lane et al. 2010) although the percentage of sensory processing differences in our study is slightly lower than the findings from USA and Australia (Baranek et al. 2006, Tomchek and Dunn 2007, Baker et al. 2008, Tomcheck, et al.,2014). More studies are needed across Asia as published data on the prevalence of sensory processing differences are relatively few (Chow 2005, Satiansukpong 2002, Sankar and Priyadarshini 2014). These deficits make comparison difficult. More studies are warranted to help investigate cultural-environmental related factors.

Sensory processing differences and participation differences- Autism and matched controls:

Figure 2 showed significant differences across all domains of sensory processing – which were lower in children with autism, compared to children with typical development. Definite differences in sensory processing differences were found at a total score of 68.7 percent (in children with autism) and 21.5 percent (in typical developing children), and these differences were statistically significant (p < 0.001). The findings of 21.5 percent prevalence rate of ‘sensory-processing’ differences in typically developing children were compared to studies from USA which reported a range between five percent to 19.9 percent (Ahn et al. 2004, Reynolds et al. 2008, Román-Oyola and Reynolds 2013), and at 15 percent in Israel (Engel-Yeger 2010). These differences may be due to cultural issues (Engel-Yeger 2010) which are believed to contribute towards the prevalence, risk, and protective factors in children with disorders (Glorisa and Alegría 2008). In addition, our study was conducted in an urban setting where a large proportion of families live in apartments with limited opportunities for children to play outside and explore the environment. This may account for the slightly higher percentage of typically developing children with sensory processing differences. On total scores and all subsection scores of Short sensory processing (SSP tool) between groups: the greatest disparity between both groups were in the auditory filtering section where children with autism were more affected. For the under responsive/seeking sensation category, both groups experienced high differences. This calls for clinicians to assess other factors (e.g. cultural-geographical variables) which may have contributed to these deficits.

Figure 2.

Sensory processing profile: autism (n = 93) vs typically developing (n = 95) Children.

Patterns of participation among children with autism

The patterns of participation in this cohort of children with autism were examined in terms of i) level of difference, ii) frequency of participation, and iii) level of enjoyment. We found a higher level of differences in participation, lower frequency of participation, and lower enjoyment of activity participation in children with autism across all domains.

1. Differences in participation: Our study found that children with autism have more differences in social integration skills. Almost half of these children also had difficulty watching videos/DVD on the computer or television, and a few had differences in daily routines like sleeping, dressing, having meals, and toileting. Video activities are highly prevalent among children with autism and one reason may be its solitary nature, since these children have differences in socialising (Hilton et al. 2008). Many studies have reported that children with autism have serious differences in social participation (Shattuck et al. 2011). For tasks that require higher cognitive abilities (such as homework, organizing the study environment, and food preparation), Jasmin et al. (2009) reported severe differences in self-care activities, but the findings were in contrast with findings by Anderson et al. (2007). More studies are needed to discern what factor influences these differences, with cognition levels adjusted to rule out the influence of cognitive levels.

2. Frequency of participation in childhood activities: Children with autism had lower frequencies of activity participation, especially in food preparation, extra-curricular activities outside of school, playing with friends, and/or integration in social situations. These are expected as most of these are complex tasks and require social skills, which is difficult for children with autism. Factors such as parental involvement and access to opportunities have been associated with increased frequency, even in typically difficult areas that have had prearranged training by professionals or family members (Orsmond et al. 2004, Orsmond et al. 2013). Van Eylen et al. (2011) suggested that proper activity instruction with careful integration of the amount of disengagement may help improve the frequency of activity participation.

3. Enjoyment in childhood activity participation: Children with autism showed low enjoyment in social interaction and high enjoyment in solitary activities or daily routines (such as watching videos/DVD on the computer or television, eating and drinking, and meal time). However, confounding factors such as family disengagement, quality of social-interaction, sensory-behavioural challenges, and interpersonal relationship problems have been found to affect their enjoyment (Askari et al. 2015). Future studies should examine these factors which were not explored in this study.

‘Childhood participation’: Comparing between autism and typically developing children

Overall, children with autism have a lower level of participation, lower frequency, and less enjoyment in most activities, but there are varying (and contrasting) differences across frequencies of participation in basic activities and in enjoyment of recreation activities. We compared the two groups on three levels of - differences, frequencies, and enjoyment of participation. Our study found lower involvement across all three levels of participation (Figure 3), which is consistent with findings from other researchers (LaVesser and Berg 2011, Potvin et al. 2013, Rodger and Umaibalan 2011). This cohort of children with autism had significantly lower participation, lower frequency of participation across many activities when compared to typically developing children, which is also consistent with many other studies comparing children with autism and typically developing children (Hilton et al. 2008, LaVesser and Berg 2011, Rodger and Umaibalan 2011). However, there were no significant differences between the two groups in the frequency of participation for ‘basic daily’ activities (such as eating and drinking, meal time and sleep, or in solitary tasks like watching videos). For the level of enjoyment in participation, we found that recreational activities were enjoyed by both autistic (87.1%) and typically developing (93.5%) children, with no significant difference between groups. One study reported no significant difference in enjoyment of recreational activities between groups (Potvin et al. 2013), but another found that children with autism have significantly less enjoyment in participating in recreational activities (Hochhauser and Engel-Yeger 2010, Solish et al. 2010). These findings suggest that consistent daily training may improve their participation in those activities found difficult. However, there may be other confounding factors such as cognitive ability and/or siblings/carers encouraging interactions with these children, which should be explored in future studies.

Figure 3.

Difficulties in participation in children with autism (n 93) and typical development (n = 93).

Association between sensory processing (using short sensory profile) and participation in childhood occupation (using PICO)

A significant positive correlation was found between sensory processing [using the Short Sensory Profile (SSP) total scores] and childhood participation (using PICO total scores) on each of its three sections [i.e. I) difficulty, 2) frequency and 3) enjoyment]. Differences in sensory processing correlate significantly and positively with differences in these levels of participation in occupation. There were also significantly positive but weak correlation between enjoyment of participation (indicated by higher scores in PICO) and SSP total scores – where higher sensory processing ability correlates with higher enjoyment. A key finding was that auditory filtering is the only section in SSP-M that significantly correlates with all three sections of PICO - with the difficulty in participation (r = 0.36, p ≤ 0.01), frequency of participation (r = 0.22, p ≤ 0.05) and enjoyment of participation(r = 0.27, p ≤ 0.01).

Auditory filtering differences have been reported to be strongly associated with problems in learning and poor attention, and poorer academic performance in children with autism (Ashburner et al. 2008). Low energy and tactile sensitivity were significantly correlated with differences and enjoyment in participation but not with frequency of participation. Previous studies also suggested that tactile sensitivity affects daily activities (Baranek et al. 1997) and academic and social activities (Ben-Sasson et al. 2009). Therefore, these components of sensory processing are critical factors impacting the level of difficulty and the level of enjoyments in childhood participation.

Occupational therapists intervene to enable children with autism to engage in activity participation by addressing person-environmental-occupation related barriers without giving too much stress or burden to the family. Studies on correlations between sensory processing differences and patterns of participation in childhood occupation are important and coincide with the concept of the World Health Organization (World Health Organization 2007), which emphasises the importance of the relationship between disability, performance and participation in daily living activities as outlined in this study’s conceptual framework.

Strength and limitation

This study is the first large-scale case-control study on a cohort of urban Malaysian children with autism with an age/gender-matched community control. It contributes to a database on the profile of sensory integration differences and the profile of childhood participation, and on the relationship between these two variables. However, there is a strong reliance on the data gathered from the parents, where the validity of the data would depend on their observations with sufficient interaction with the child. In addition, we rely on parents to inform us of their child’s diagnosis. Although we made careful consideration to exclude potential disorders in the typical group by ensuring the children do not have any disorders (and do not have sibling/s with an autism spectrum disorder). We rely on the parents to inform us if the child is having any illnesses or are being followed up and we exclude those that have any of these indications. However, we recognise that there may be some children with ‘undiagnosed co-existing conditions that may affect the study findings. Future studies should include reports from teachers and clinical observations for a more balanced picture. The current study did not include a robust examination of coexisting conditions which could influence the results as these samples were not matched based on the severity of intellectual disability. The identification of any other coexisting disabilities [e.g. cerebral palsy, visual/hearing impairment, communication difficulties, neurodevelopmental disorder, e.g. ADHD; medical (e.g. epilepsy or heart disease)] was verified by the parents and were not confirmed with other diagnostic tools or screening. There is also a need to consider a greater representative sample from both urban and rural settings with a multi ethnic sample. With wider and better representations, data analyses can include multiple regression to determine which sensory components predicted childhood participation (and using cognitive abilities as covariates), as well as to examine the overall impact on the quality of life of the child and their parents.

Conclusion

This is the first large-scale case-control study conducted on a cohort of Malaysian children. Significant differences in the pattern of sensory processing and in the pattern of participation were found between children with autism and ‘typically developing’ children. Children with autism showed major differences in seeking sensation, auditory filtering, and social situations (e.g. playing with friends), being responsible for homework, organizing the study environment, and preparing simple food. Activities which children with autism had less difficulty in were those with routine steps and with less demand on cognition.

Auditory filtering (of sensory processing) is the only section in SSP that shows correlation with all three sections of PICO (participation). The correlation is highest with the difficulty in participation (total score). A moderate to weak, but positive correlation between sensory processing (SSP total score) and childhood participation (PICO total score) suggests that addressing sensory processing alone may not necessarily result in an improvement in the participation of childhood occupation. As the first large-scale study exploring a cohort of children with autism, the findings from this study provided much needed baseline data on sensory processing differences and on the participation profile of children with autism and their matched typically developing peers. Future studies across other Asian countries to examine its cultural-geographical related factors may add to a growing body of literature pertaining to children with autism spectrum disorder.

Funding Statement

This article was funded by Umrg20/190