- Email: [email protected]
Age, social engagement, and physical activity in children with autism spectrum disorders
Research in Autism Spectrum Disorders 3 (2009) 22–31
Contents lists available at ScienceDirect
Research in Autism Spectrum Disorders Journal homepage: http://ees.elsevier.com/RASD/default.asp
Age, social engagement, and physical activity in children with autism spectrum disorders Chien-Yu Pan * Department of Physical Education, National Kaohsiung Normal University, No. 116, He-Ping First Road, Kaohsiung 802, Taiwan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 March 2008 Accepted 10 March 2008
Although engagement in social interactions is one of the key diagnostic features of autism spectrum disorders (ASDs), few studies have examined if social engagement related to physical activity of children with ASD. Age is another variable of interest to researchers studying behaviors, but has not been explored in physical activity and social engagement in this population. The purpose of this study was to examine the associations of age, social engagement and physical activity in children with ASD. Twenty-five children with ASD participated. Each child’s physical activity and social engagement was assessed using a uniaxial accelerometer and the direct observational assessment. Pearson product-moment correlation coefficients and multiple regression analysis were used to evaluate the associations and influences of selected variables. Age had somewhat positive influences on both physical activity and social engagement, and children with frequent social engagement with adults had displayed higher levels of physical activity. No evidence was found to support the notion that children with ASD become more inactive and more isolate as they age; however, limitations and directions for future research in this area are discussed. ß 2008 Elsevier Ltd. All rights reserved.
Keywords: Age Social engagement Physical activity Autism
Infrequent or no engagement in social interactions and stereotypic behaviors are two overt and defining characteristics of autism spectrum disorder (ASD) (American Psychiatric Association, 1994). To date, motor impairments observed in individuals with ASD have been categorized as associated symptoms (Ming, Brimacombe, & Wagner, 2007). These characteristics create severe limitations. Stereotypic behavior, for some children, interferes with learning and may alienate peers and adults
* Tel.: +886 7 7172930x3531; fax: +886 7 7114633. E-mail address: [email protected]. 1750-9467/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.rasd.2008.03.002
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
23
because of its highly unusual and stigmatizing nature. Social isolation from peers prevents the formation of social relationships, which are essential for early social development. Poor motor coordination could limit opportunities for this population to successfully participate in physical activity and place them at risk for developing sedentary lifestyle associated diseases (U.S. Department of Health and Human Services, 1996). It is not surprising, therefore, that children with ASD tend to withdraw from participation in physical activity due to the negative social and behavioral outcomes associated with the symptom. While it appears that the presence of an ASD affects opportunities for physical activity participation, this issue has received relatively little attention in the literature. Since ASD is the fastest growing developmental disability and there has been an 85% increase in the number of school age Taiwanese children being diagnosed with ASD in the last 10 years (Ministry of Education, 2006a), there is a need to better understand how disability-related symptoms such as social engagement affect certain health behaviors, including physical activity. Studies that specifically investigate social engagement in children with ASD in inclusive and natural settings are still relatively sparse. McGee, Feldman, and Morrier (1997) studied the naturally occurring levels of social behavior of both children with ASD and typically developing children in inclusive preschool settings, and found that children with ASD spent less time in the proximity of other children, showed less focus on adults and peers as interactive partners, received fewer social initiations from peers, used less verbalization towards other children, and engaged in more atypical behavior. Sigman and Ruskin (1999) found that in comparison with children with other disabilities, children with ASD spent a larger proportion of time engaged in nonsocial play and a smaller proportion of time in direct social play with others. Children with ASD were also found to make less initiation and show less responsiveness to peer initiations compared to controls. In a recent study by Jahr, Eikeseth, Eldevik, and Aase (2007), the frequency and latency of social interaction with typically developing children and those with ASD in inclusive kindergarten settings were compared. The results showed a significant difference in frequency of social interaction between the typical children and those with ASD and no difference between the groups on latency to initiate interaction. There is relatively little information regarding accelerometer-determined physical activity in children with ASD. Rosser-Sandt and Frey (2005) found that physical activity levels were similar in children with and without ASD and both groups acquired a majority of daily moderate physical activity during recess. Pan and Frey (2006) examined physical activity patterns in youth with ASD and observed that this group met minimum activity recommendations, but were less active than previous reports on peers without ASD using similar methodology (Mota, Santos, Guerra, Ribeiro, & Duarte, 2003; Trost et al., 2002). In addition, physical activity levels were higher among participants in elementary school compared to those in middle and high school. Pan (in press) compared the percentage of time children with and without ASD spent in moderate-to-vigorous physical activity (MVPA) during inclusive recess settings in Taiwan, and found that children with ASD were less active during overall recess, lunchtime, first and second morning recess compared to those without disabilities. All children also fell below 40% of recess time engaged in physical activity. The inconsistent findings in the literature highlight the difficulty of generalization from the previous studies using different methodologies. Additional studies are need in various national and international populations to determine physical activity differs between children with and without ASD. Age is one variable of interest to researchers studying individual’s behavior. However, to date, very few studies have examined whether age exerts an influence on the impact of ASD in the daily life of children. Several researchers have suggested that children with movement problems will increase as children’s play becomes more complex and rule-bound (Cairney, Hay, Faught, Corna, & Flouris, 2006; Wall, 2004). This is of concern, particularly in children with ASD who have difficulties understanding social cues. They are unable to follow the typical course of development in the acquisition and application of complex physical skills as they age. They will not be able to learn the higher-order strategic skills required for participation in complex play activities when they are getting older. Despite the lack of research in the potential age influences on social engagement and physical activity of children with ASD, it is reasonable to assume that the general age impact would also extend to individuals with ASD since ASD is considered to be uncured and the long-term consequences of ASD may not be favorable.
24
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
The purpose of this study was to examine the associations of age, social engagement and physical activity in children with ASD when participation in both structured (physical education) and unstructured (recess) play opportunities are considered. It was hypothesized that age would show a significant influence on both physical activity and social engagement participation in both free play and organized settings in children with ASD. In addition, more frequent social engagement behaviors would be observed in higher physical activity levels of children with ASD. 1. Methods 1.1. Participants and setting Twenty-five boys with ASD aged between 7 and 12 years (9.28 1.46) volunteered to participate and returned signed parental informed consent prior to study involvement. Twenty-two of the children had normal speech patterns, and the remaining three children made a few sentences, as based on parent reports and the researcher interaction with participants. All children with ASD met the criteria for autistic disorder of the DSM-IV-Revised system (American Psychiatric Association, 2000) and the Identification Standard for Students with Special Needs (Ministry of Education, 2006b), as assessed by trained and knowledgeable doctors through medical and psychological assessment in the public hospitals (Executive Yuan, 2006a). Level of severity (mild, moderate, severe, and very severe) is based on functioning in the social adaptive skill areas and language comprehension and expression (Executive Yuan, 2006b). Diagnoses included autism (mild or high-functioning, n = 11; moderate, n = 8) and Asperger’s syndrome (n = 6). None of the children had any additional medical conditions such as epilepsy, cerebral palsy or motor deficiencies. All the children reside in urban settings and 24 live in a two-parent household. None were enrolled in a segregated or inclusive school-based physical activity programs or sports teams. The children with ASD received part-time special education services which were individually tailored for each child and based on recommendations from multidisciplinary teams of professionals from regional specialist clinics, parents and educational authorities. No children with ASD had either full- or part-time staff assistant available throughout the day. Mean height was 136.50 10.79 cm, mean body mass was 35.38 8.55 kg, and mean body mass index (BMI) was 18.81 2.97 kg/m2. These values were within the normal ranges for Taiwanese children of this age (Ministry of Education, 2007). Participants’ schools were located in the same geographical area of high social and economic deprivation in a large urban city, and all participants received physical education and recess in inclusive settings without any level of supports. Each school had three morning recesses and three afternoon recesses, and each recess length was 10–20 min in duration. The area of the playground for all schools was very limited because they were located in a crowded urban city. There was no visible playground equipment or markings to stimulate play. Children were allowed to take their own balls, toys, etc. during recess, but there was little evidence of any other equipment being used or being made available. The physical education lesson was 40 min in duration and children in grades 1–2 and 3–6 were required to take 1 and 2 lessons each week, respectively. Twelve of children with ASD attended regular physical education taught by class teachers, and the remaining 13 children were taught by physical education specialists. 1.2. Assessment procedures The study protocol was reviewed and approved by the National Science Council, Taiwan. Study objectives and methods were individually explained to and written consent and assent obtained from parents and children, respectively. Stature and body mass measures were collected according to class teacher reports only if the child was measured at the beginning of the semester this study was conducted. Otherwise, these measures were collected in school health center with all children dressed in light clothing and shoes removed for calculation of body mass index (kg/m2) prior to study involvement. 1.2.1. Physical activity measurement The children’s physical activity levels and patterns were assessed by GT1M ActiGraph, a uniaxial accelerometer that measures vertical acceleration of human motion as well as how many steps they
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
25
take. The accelerometer has been used extensively and reported as a valid objective measure of physical activity in children (Trost et al., 1998). It has also been used in youth with ASD (Pan, in press; Pan & Frey, 2006; Rosser-Sandt & Frey, 2005) and found that this population can tolerate the instrument. Consistent with previous studies, the accelerometer was placed in a small nylon pouch and worn over the right hip with an elastic belt. Prior to data collection, verbal and written instructions on how to wear and care for the device were provided. All participants also engaged in practice trials wearing the device. Participants wore the accelerometer for five consecutive school days during their regularly school schedules. At the start of each school day, the accelerometers were attached by the research assistants to the children and participants were then asked to follow their regular daily routine. Accelerometers were removed at the end of the school day, and the data were immediately downloaded and re-initialized for the next day. Some participants wore the same monitor every day, and all occurring times for physical education class and each recess period were recorded by class teachers and research assistants. Physical activity data were reduced from one physical education class and one recess period each day for five consecutive school days in accordance with social engagement observation. Accelerometers were programmed to collect data in 1-min intervals. Activity counts were analyzed to determine counts per minute (CPM), time spent in MVPA (3.0 MET) and vigorous physical activity (VPA; 6.0 MET) using age-specific count cutoffs (Freedson et al., 1997; Trost, Pate, Freedson, Sallis, & Taylor, 2000). To control for the differences in the monitoring recess and physical education length, the relative (percentage) time spent in MVPA and VPA, and step counts per minute were calculated and used in the subsequent analyses. 1.2.2. Social engagement measurement Social engagement patterns were assessed using the Engagement Check (McWilliam, 1990). It is an observational tool that uses momentary time sampling procedure with a 15-s observe and 15-s record cycle. McWilliam and Bailey (1995) have reported suitable reliability and validity estimates on children with ASD. For the current study, child’s engaged behavior is divided into two types (with adults and peers) and two forms (interactive and noninteractive) of engagement. Interactive engagement was defined as the child’s focus on another person, and his/her behavior being aimed at producing a social effect. Interdependent play, mutual organization, gestures, and talking are examples of interactive engagement. Noninteractive engagement was defined as the child attending to another person or playing nearby with similar materials. Looking, orienting, tracking, and listening are examples of noninteractive engagement. Each child was systematically observed one time in physical education, and one time in recess each day for consecutive five school days. Observations of the two settings occurred on the same week. Informal observations on days prior to formal data collection helped reduce reactivity, and conversations with teachers after the formal observation verified that data were collected on typical social engagement behaviors. The percentage of each social engagement category and the percentage of total social engagement scores were used in the data analyses. 1.2.3. Observer training and interobserver reliability Interobserver agreement was calculated by dividing agreements by agreements plus disagreements and multiplying by 100 for each category. Before the study began, four observers were trained by the researcher to use systematic observation methods and this study’s social engagement instrument. Each observer was required to review the observation instrument and become familiar with the behavioral definitions. During the training sessions, the observers collected data on the Engagement Check by watching videotapes of children with ASD in physical education classes and recess periods previously made by the researcher. When a question arose about how to record a specific behavior, the researcher and the observers discussed the behavior until all five agreed on the appropriate recording procedure. Study observations began only after an interobserver agreement of 85% or greater was achieved for each variable across all four observers. Interobserver agreements that did not meet criterion were addressed by extending the observer training until the interobserver
26
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
agreement criterion was reached. The observers reached an overall agreement of at least 85% agreement for these data. 1.3. Statistical analyses Prior to analysis, physical activity data were examined separately at each setting for the assumptions of multivariate analysis. The residuals were normally distributed, and the assumption of homoscedasticity was met. None of the cases were identified through Mahalanobis distance as multivariate outliers, leaving all cases for the final analysis. Pearson product-moment correlation coefficients were calculated to evaluate the relationship between age, social engagement, and physical activity. When a variable was significant in the correlations, it was entered into the multiple regression analysis. For each variable, the partial correlation is reported, which reflects the contribution of that variable, adjusting for all other variables. Changes in multiple correlations squared (R2 change) are reported to demonstrate the amount of variance explained by each variable. The adjusted multiple correlations squared represent the full model. Because six physical activity and five social engagement variables are separate measures, prediction of these variables was examined in separate regression models. All statistical analyses were computed using SPSS Statistical Software Package, Version 13.0. Values are reported as mean S.D., with significance set at P < 0.05. 2. Results Descriptive statistics for physical activity and social engagement behaviors in children with ASD during physical education and recess are shown in Table 1. The physical education and recess monitoring time was 34.09 4.18 and 50.20 9.72 min, respectively. On average, children with ASD were more active physically and socially during physical education than recess. 2.1. Age and physical activity Table 2 shows bivariate correlations of each physical activity variable and children age. Children age was positively correlated with counts/min (r25 = 0.50, P < 0.01) and steps/min (r25 = 0.57, P < 0.01) in recess and positively correlated with 5-min MVPA during physical education (r25 = 0.43, P < 0.05). The results of the multiple regression analyses showed that children age was significant in the three regression models, explaining 22%, 29% and 15% of the variance in children overall physical activity (counts/min, F1,23 = 7.63, P < 0.05; R2 change = 0.25), steps/min (F1,23 = 10.76, P < 0.01; R2 change = 0.32) and 5-min MVPA (F1,23 = 5.12, P < 0.05; R2 change = 0.18), respectively.
Table 1 Physical activity and social engagement during PE and recess (n = 25)
Physical activity Counts/min MVPA VPA Steps/min 5-min MVPA 10-min MVPA Social engagement Adult-interactive Adult-noninteractive Peer-interactive Peer-noninteractive Total social engagement
PE
Recess
1538.35 862.38 54.86 24.72 8.30 9.16 38.34 20.27 5.25 11.03 0.75 2.05
1318.61 852.49 46.84 25.33 5.82 7.20 33.55 20.62 3.61 3.72 0.76 1.38
28.57 27.49 0.78 2.47 60.09 26.34 24.78 27.78 114.22 44.23
9.34 9.03 0.46 1.61 59.98 33.97 9.58 21.01 79.36 48.80
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
27
Table 2 Bivariate correlations for the physical activity variables and children age during PE and recess PE Age Counts/min MVPA VPA Steps/min 5-min MVPA 10-min MVPA
Recess
0.34 0.09 0.04 0.16 0.43* 0.29
0.50* 0.27 0.37 0.57** 0.16 0.01
*P < 0.05; **P < 0.01; two-tailed; n = 25.
Table 3 Bivariate correlations for the social engagement variables and children age during PE and recess PE Age Adult-interactive Adult-noninteractive Peer-interactive Peer-noninteractive Total social engagement
Recess
0.18 0.24 0.48* 0.17 0.48*
0.19 0.11 0.12 0.15 0.19
*P < 0.05; two-tailed; n = 25.
2.2. Age and social engagement Table 3 shows bivariate correlations of each social engagement variable and children age. Children age was positively correlated with peer-interactive (r25 = 0.48, P < 0.05) and total social engagement behaviors (r25 = 0.48, P < 0.05) during physical education. Children age was not correlated with any social engagement behaviors in recess. The results of the multiple regression analyses showed that children age was significant in the two regression models, explaining 19% and 20% of the variance in children peer-interactive (F1,23 = 6.68, P < 0.05; R2 change = 0.23) and total social engagement (F1,23 = 7.01, P < 0.05; R2 change = 0.23), respectively. 2.3. Social engagement and physical activity Table 4 shows bivariate correlations of each physical activity variable and children social engagement. Noninteractive engagement with adults during physical education was positively correlated with children VPA (r25 = 0.74, P < 0.01) and steps/min (r25 = 0.51, P < 0.01). None of the social engagement behaviors was correlated with any physical activity in recess.
Table 4 Bivariate correlations for the physical activity and social engagement variables during PE and recess
AI ANI PI PNI TSE
Counts/min
MVPA
PE
PE
0.18 0.33 0.28 0.03 0.28
Recess 0.12 0.15 0.10 0.10 0.04
**P < 0.01; two-tailed; n = 25.
0.18 0.33 0.04 0.15 0.06
VPA Recess 0.19 0.23 0.11 0.16 0.04
PE 0.23 0.74** 0.04 0.23 0.35
Recess 0.13 0.16 0.05 0.03 0.00
Steps/min
5-min MVPA
10-min MVPA
PE
PE
PE
0.08 0.51** 0.05 0.09 0.17
Recess 0.06 0.11 0.06 0.12 0.09
0.25 0.02 0.16 0.11 0.32
Recess 0.09 0.29 0.06 0.22 0.08
0.01 0.12 0.33 0.24 0.05
Recess 0.06 0.17 0.14 0.24 0.03
28
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
The results of the multiple regression analyses showed that noninteractive engagement with adults was significant in the two regression models, explaining 53% and 23% of the variance in children VPA (F1,23 = 27.86, P < 0.01; R2 change = 0.55) and steps/min (F1,23 = 8.26, P < 0.01; R2 change = 0.26), respectively. 3. Discussion Factors (age and social engagement) thought to be determinants for physical activity in children with ASD were partially found to be related to physical activity in this study. Age in particular did not support the notion that children with ASD are more likely to be inactive as they age. Infrequent social engagement rate did not widen with age as the activity-deficit hypothesis proposed by researchers (Bouffard, Watkinson, Thompson, Causgrove Dunn, & Romanow, 1996; Wall, 2004). Furthermore, the overall activity levels were not dependent on social engagement either during physical education or in recess setting, indicating that those who engaged more in social interactions were no more physically active than peers who engaged less socially. However, the noninteractive engagement with adults was positively correlated with children’s VPA and steps/min during physical education. An interesting note related to age and physical activity from the current sample was the positive relationships between age and both the overall physical activity and step counts in recess and the positive relationship between age and 5-min MVPA during physical education. This is not supportive of Kozub and Oh’s findings (2004) of negative relationships between age and total bouts of MVPA in individuals with visual impairments, using RT3 activity monitors. This is also not supportive of the same inverse relationships between age and minutes of moderate physical activity measured using RT3 monitors in individuals with intellectual disability (Kozub, 2003). However, in an earlier study of Longmuir and Bar-Or (2000) using a self-report questionnaire, physical activity estimates do not support a marked age-related decline in individuals with cerebral palsy, muscular dystrophy, or visual impairments, but a decline in physical activity with age was observed in youth with chronic medical conditions. In the general population, age-related declines are noted for both males and females between the ages of 8 and 18 years (Thompson, Baxter-Jones, Mirwald, & Bailey, 2003). The inconsistent findings in the literature highlight the importance of using objective assessment methods, support the examination of physical activity patterns within various environments, and confirm the need to measure the physical activity of specific subgroups to avoid generalizations across different disabling conditions. Most of previous studies from aforementioned above in youth with and without disabilities have documented a negative relationship between age and physical activity. Pan and Frey (2005) used accelerometer to assess the influence of age on physical activity in youth with ASD, and found a significant negative association between age and youth physical activity. Results of the current investigation confirm the importance of age as a significant physical activity correlate in children with ASD, but do not reinforce the magnitude of decline in physical activity with age. The failure to find a consistent age influence may be because the age span of the current study is not broad enough to include adolescents. Most youth in the Pan and Frey’s study (2005) were in middle or high school, and reported no recess time and decreased physical education requirements, while elementary school age children had access to these physical activity resources. It may be the case that greater decline in physical activity do not emerge until mid- to late adolescence. This may be particularly true for organized sports, which become increasingly competitive in high school and where children can exercise greater autonomy in their choices. To date, no published studies have examined whether Taiwanese adolescents with ASD participate in fewer activities than others of the same age. Further work should examine this question with this population. Children with ASD in the current study engaged in more peer-interactive interactions as they age, and the total social engagement rate was higher as children become older during physical education. This is not consistent with Oh, Mehmet, and Kozub’s (2004) study, indicating that there was no significant relationships between age and social engagement during physical education in students with visual impairments. The lack of age effect among individuals with visual impairments may have been due to vision and communication challenges resulting in initial overall low rate of social
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
29
engagement (Zanandra, 1998). It is likely that the factors influencing this behavior are different since the presence of a disability significantly alters life experience (Seligman, 1999). The symptoms of ASD are considered to be modifiable when favorable environmental interaction has been provided to make subsequent interactions easier, resulting in both improved skill and motivation to interact (Mundy, Henderson, Inge, & Coman, 2007). Although the impact of environment elements is beyond the scope of the current study, physical education is usually more structured and supervised than recess and student’s interaction is required, making it an ideal opportunity for the promotion of children’s social engagement. Participants in the current study were observed paying more attention to teacher/peer demonstration, feedback and learning outcomes during physical education; whereas the majority of the children with ASD were observed spending the most of time not interacting in recess. More research is needed to determine if the unique elemental characteristics play a role in social engagement of children with ASD. Age-related trends in physical activity and social engagement are an important concern for health professionals given that ASD was considered to be unmodifiable until recently. Compared to their typically developing peers, children with ASD may be more likely to perform poorly in school and are at greater risk for emotional and behavioral problems as they age. However, Mundy et al. (2007) shed new light on this concept, suggesting that favorable environmental and social interaction has the potential to make subsequent positive behaviors. Such favorable interactions can result in not only improved social skills (Thomas & Smith, 2004) and physical fitness (Lotan, Isakov, & Merrick, 2004), but also in reduction of stereotypic behavior (Prupas & Reid, 2001). The positive age-related trends found in physical activity and social engagement of the current participants is supportive of this notion. It appears from the results of the current study identified above that the divergence in activity with age hypothesis (Bouffard et al., 1996; Wall, 2004) is also not supported. Behaviors of children with ASD can be explained to some extent by the self-determination model (Wehmeyer & Garner, 2003), indicating that behavior is a function of environmental as well as personal factors, which in turn, both environmental and personal factors affect individual’s behaviors. Therefore, physical activity and social engagement behaviors of children with ASD may be more affected by social and environmental constraints than the actual impairment. Noninteractive engagement with adults during physical education is positively related to both VPA and steps/min, suggesting that as adult-noninteractive engagement behaviors increased, physical activity levels increased. This is expected because each participant received regular attention and encouragement from the teacher during physical education, which may have been responsible for some increase in social engagement and physical activity. Within the physical education class, participants were provided an opportunity to develop the necessary skills to interact with their teacher. The teacher’s presence and use of verbal and nonverbal curing have been sufficient to perpetuate the social engagement and physical activity behavior in physical education class. However, results of the current study do not support the earlier finding indicating that there was no significant relationships between social engagement and physical activity in individuals with visual impairments during physical education (Oh et al., 2004). This finding may be due to children with visual impairments in the previous study being from a segregated blind school where more competent children may not have opportunity to find highly skilled playing partners and friends of similar interests to engage physically and socially. More studies are needed to determine if setting is a determinant of social or activity patterns in individuals with ASD. These data represent an initial attempt to study age, social engagement and physical activity in children with ASD. However, the convenience sampling and the low sample size limit the inferences of these findings. Regardless, this study is important because it is the first attempt to identify relationships between age, social engagement and physical activity in children with ASD. Findings indicate differences in the relationships between age, social engagement and physical activity among children with various types of disabilities. Further work with a larger sample of children with ASD in a longitudinal design is necessary to properly identify the age and social engagement influences on physical activity. In addition, the impact of other symptoms influences such as language and motor coordination on physical activity of children with ASD is of interest. Additional information is needed to better understand the influences of physical activity for intervention in children with ASD.
30
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
Acknowledgments This research was supported by grant NSC 95-2413-H-017-010, National Science Council, Taiwan. The author would like to thank Yu-Feng Wang, Ching-Yi Hsu, Hsiang-Chun Huang, and Chi-Hung Tien for their assistance with data collection. Special thanks to all children who participated in this study and parents of children for supporting this project.
References American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders: Text revision (4th TR ed.). Washington, DC: American Psychiatric Association. Bouffard, M., Watkinson, E. J., Thompson, L. P., Causgrove Dunn, J. L., & Romanow, S. K. E. (1996). A test of the activity deficit hypothesis with children with movement difficulties. Adapted Physical Activity Quarterly, 13, 61–73. Cairney, J., Hay, J., Faught, B. E., Corna, L. M., & Flouris, A. D. (2006). Developmental coordination disorder, age, and play: A test of the divergence in activity-deficit with age hypothesis. Adapted Physical Activity Quarterly, 23, 261–276. Executive Yuan. (2006a). Hospitals for the identification of disabilities. Retrieved July 11, 2007, from http://www.doh.gov.tw/cht/ content.aspx?doc_no=45921. Executive Yuan. (2006b). Levels of disabilities. Retrieved July 9, 2007, from http://www.doh.gov.tw/cht/content.aspx?dept=R&class_ no=211&now_fod_list_no=6753&array_fod_list_no=6580&level_no=3&doc_no=41058&show=. Freedson, P. S., Sirard, J., Debold, E. P., Pate, R. R., Dowda, M., Trost, S. G., et al. (1997). Calibration of the computer science and applications, Inc. (CSA) accelerometer. Medicine and Science in Sports and Exercise, 29, S45. Jahr, E., Eikeseth, S., Eldevik, S., & Aase, H. (2007). Frequency and latency of social interaction in an inclusive kindergarten setting. Autism, 11, 349–363. Kozub, F. M. (2003). Explaining physical activity in individuals with mental retardation: An exploratory study. Education and Training in Developmental Disabilities, 38, 302–313. Kozub, F. M., & Oh, H. K. (2004). An exploratory study of physical activity levels in children and adolescents with visual impairments. Clinical Kinesiology, 58, 1–7. Longmuir, P. E., & Bar-Or, O. (2000). Factors influencing the physical activity levels of youths with physical and sensory disabilities. Adapted Physical Activity Quarterly, 17, 40–53. Lotan, M., Isakov, E., & Merrick, J. (2004). Improving functional skills and physical fitness in children with Rett syndrome. Journal of Intellectual Disability Research, 48, 730–735. McGee, G. G., Feldman, R. S., & Morrier, M. J. (1997). Benchmarks of social treatment for children with autism. Journal of Autism and Developmental Disorders, 27, 353–364. McWilliam, R. A. (1990). The engagement check. Chapel Hill, NC: Frank Porter Graham Child Development Center. McWilliam, R. A., & Bailey, D. B. (1995). Effects of classroom social structure and disability on engagement. Topics in Early Childhood Special Education, 15, 123–147. Ming, X., Brimacombe, M., & Wagner, G. C. (2007). Prevalence of motor impairment in autism spectrum disorders. Brain and Development, 29, 565–570. Ministry of Education. (2006a). Annual statistics in special education. Retrieved July 09, 2007, from http://163.21.111.100/tlearn/book/ BookAll.asp?BookMainID=7. Ministry of Education. (2006b). Identification standard for students with special needs. Retrieved May 11, 2007, from http://law.moj.gov.tw/Scripts/Query4A.asp?FullDoc=all&Fcode=H0080065. Ministry of Education. (2007). Physical fitness. Retrieved March 1, 2007, from http://www.fitness.org.tw/TW/index.html. Mota, J., Santos, P., Guerra, S., Ribeiro, J. C., & Duarte, J. A. (2003). Patterns of daily physical activity during school days in children and adolescents. American Journal of Human Biology, 15, 547–553. Mundy, P. C., Henderson, H. A., Inge, A. P., & Coman, D. C. (2007). The modifier model of autism and social development in higher functioning children with autism. Research and Practice for Persons with Severe Disabilities, 32, 124–129. Oh, H. K., Mehmet, A. O., & Kozub, F. M. (2004). Physical activity and social engagement patterns during physical education of youth with visual impairments. Review: Rehabilitation Education for Blindness and Visual Impairment, 36, 39–48. Pan, C. Y. (in press). Objectively measured physical activity between children with autism spectrum disorders and children without disabilities during inclusive recess settings in Taiwan. Journal of Autism and Developmental Disorders. Pan, C. Y., & Frey, G. C. (2005). Identifying physical activity determinants in youth with autistic spectrum disorders. Journal of Physical Activity and Health, 2, 412–422. Pan, C. Y., & Frey, G. C. (2006). Physical activity patterns in youth with autism spectrum disorders. Journal of Autism and Developmental Disorders, 36, 597–606. Prupas, A., & Reid, G. (2001). Effects of exercise frequency on stereotypic behaviors of children with developmental disorders. Education and Training in Mental Retardation and Developmental Disorders, 36, 196–206. Rosser-Sandt, D., & Frey, G. C. (2005). Comparison of physical activity levels between children with and without autistic spectrum disorders. Adapted Physical Activity Quarterly, 22, 146–159. Seligman, M. (1999). Childhood disability and the family. In V. L. Schwean & D. H. Saklofske (Eds.), Handbook of psychosocial characteristics of exceptional children (pp. 111–131). New York: Kluwer Academic/Plenum Publishers. Sigman, M., & Ruskin, E. (1999). Continuity and change in the social competence of children with autism, down syndrome, and developmental delays. Monographs of the Society for Research in Child Development, 64, 11–30. Thomas, N., & Smith, C. (2004). Developing play skills in children with autistic spectrum disorders. Educational Psychology in Practice, 20, 195–206.
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
31
Thompson, A. M., Baxter-Jones, A. D. G., Mirwald, R. L., & Bailey, D. A. (2003). Comparison of physical activity in male and female children: Does maturation matter? Medicine and Science in Sports and Exercise, 35, 1684–1690. Trost, S. G., Pate, R. R., Freedson, P. S., Sallis, J. F., & Taylor, W. C. (2000). Using objective physical activity measures with youth: How many days of monitoring are needed? Medicine and Science in Sports and Exercise, 32, 426–431. Trost, S. G., Pate, R. R., Sallis, J. F., Freedson, P. S., Taylor, W. C., Dowda, M., et al. (2002). Age and gender differences in objectively measured physical activity in youth. Medicine and Science in Sports and Exercise, 34, 350–355. Trost, S. G., Ward, D. S., Moorehead, S. M., Watson, P. D., Riner, W., & Burke, J. R. (1998). Validity of the computer science and applications (CSA) activity monitor in children. Medicine and Science in Sports and Exercise, 30, 629–633. U.S. Department of Health and Human Services. (1996). Physical activity and health: A report of the surgeon general. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion. Wall, A. E. (2004). The developmental skill-learning gap hypothesis: Implication of children with movement difficulties. Adapted Physical Activity Quarterly, 21, 197–218. Wehmeyer, M. L., & Garner, N. W. (2003). The impact of personal characteristics of people with intellectual and developmental disability on self-determination and autonomous functioning. Journal of Applied Research in Intellectual Disabilities, 16, 255–265. Zanandra, M. (1998). Play, social interaction and motor development: Practical activities for preschoolers with visual impairments. Journal of Visual Impairment and Blindness, 92, 176–189.
Contents lists available at ScienceDirect
Research in Autism Spectrum Disorders Journal homepage: http://ees.elsevier.com/RASD/default.asp
Age, social engagement, and physical activity in children with autism spectrum disorders Chien-Yu Pan * Department of Physical Education, National Kaohsiung Normal University, No. 116, He-Ping First Road, Kaohsiung 802, Taiwan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 March 2008 Accepted 10 March 2008
Although engagement in social interactions is one of the key diagnostic features of autism spectrum disorders (ASDs), few studies have examined if social engagement related to physical activity of children with ASD. Age is another variable of interest to researchers studying behaviors, but has not been explored in physical activity and social engagement in this population. The purpose of this study was to examine the associations of age, social engagement and physical activity in children with ASD. Twenty-five children with ASD participated. Each child’s physical activity and social engagement was assessed using a uniaxial accelerometer and the direct observational assessment. Pearson product-moment correlation coefficients and multiple regression analysis were used to evaluate the associations and influences of selected variables. Age had somewhat positive influences on both physical activity and social engagement, and children with frequent social engagement with adults had displayed higher levels of physical activity. No evidence was found to support the notion that children with ASD become more inactive and more isolate as they age; however, limitations and directions for future research in this area are discussed. ß 2008 Elsevier Ltd. All rights reserved.
Keywords: Age Social engagement Physical activity Autism
Infrequent or no engagement in social interactions and stereotypic behaviors are two overt and defining characteristics of autism spectrum disorder (ASD) (American Psychiatric Association, 1994). To date, motor impairments observed in individuals with ASD have been categorized as associated symptoms (Ming, Brimacombe, & Wagner, 2007). These characteristics create severe limitations. Stereotypic behavior, for some children, interferes with learning and may alienate peers and adults
* Tel.: +886 7 7172930x3531; fax: +886 7 7114633. E-mail address: [email protected]. 1750-9467/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.rasd.2008.03.002
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
23
because of its highly unusual and stigmatizing nature. Social isolation from peers prevents the formation of social relationships, which are essential for early social development. Poor motor coordination could limit opportunities for this population to successfully participate in physical activity and place them at risk for developing sedentary lifestyle associated diseases (U.S. Department of Health and Human Services, 1996). It is not surprising, therefore, that children with ASD tend to withdraw from participation in physical activity due to the negative social and behavioral outcomes associated with the symptom. While it appears that the presence of an ASD affects opportunities for physical activity participation, this issue has received relatively little attention in the literature. Since ASD is the fastest growing developmental disability and there has been an 85% increase in the number of school age Taiwanese children being diagnosed with ASD in the last 10 years (Ministry of Education, 2006a), there is a need to better understand how disability-related symptoms such as social engagement affect certain health behaviors, including physical activity. Studies that specifically investigate social engagement in children with ASD in inclusive and natural settings are still relatively sparse. McGee, Feldman, and Morrier (1997) studied the naturally occurring levels of social behavior of both children with ASD and typically developing children in inclusive preschool settings, and found that children with ASD spent less time in the proximity of other children, showed less focus on adults and peers as interactive partners, received fewer social initiations from peers, used less verbalization towards other children, and engaged in more atypical behavior. Sigman and Ruskin (1999) found that in comparison with children with other disabilities, children with ASD spent a larger proportion of time engaged in nonsocial play and a smaller proportion of time in direct social play with others. Children with ASD were also found to make less initiation and show less responsiveness to peer initiations compared to controls. In a recent study by Jahr, Eikeseth, Eldevik, and Aase (2007), the frequency and latency of social interaction with typically developing children and those with ASD in inclusive kindergarten settings were compared. The results showed a significant difference in frequency of social interaction between the typical children and those with ASD and no difference between the groups on latency to initiate interaction. There is relatively little information regarding accelerometer-determined physical activity in children with ASD. Rosser-Sandt and Frey (2005) found that physical activity levels were similar in children with and without ASD and both groups acquired a majority of daily moderate physical activity during recess. Pan and Frey (2006) examined physical activity patterns in youth with ASD and observed that this group met minimum activity recommendations, but were less active than previous reports on peers without ASD using similar methodology (Mota, Santos, Guerra, Ribeiro, & Duarte, 2003; Trost et al., 2002). In addition, physical activity levels were higher among participants in elementary school compared to those in middle and high school. Pan (in press) compared the percentage of time children with and without ASD spent in moderate-to-vigorous physical activity (MVPA) during inclusive recess settings in Taiwan, and found that children with ASD were less active during overall recess, lunchtime, first and second morning recess compared to those without disabilities. All children also fell below 40% of recess time engaged in physical activity. The inconsistent findings in the literature highlight the difficulty of generalization from the previous studies using different methodologies. Additional studies are need in various national and international populations to determine physical activity differs between children with and without ASD. Age is one variable of interest to researchers studying individual’s behavior. However, to date, very few studies have examined whether age exerts an influence on the impact of ASD in the daily life of children. Several researchers have suggested that children with movement problems will increase as children’s play becomes more complex and rule-bound (Cairney, Hay, Faught, Corna, & Flouris, 2006; Wall, 2004). This is of concern, particularly in children with ASD who have difficulties understanding social cues. They are unable to follow the typical course of development in the acquisition and application of complex physical skills as they age. They will not be able to learn the higher-order strategic skills required for participation in complex play activities when they are getting older. Despite the lack of research in the potential age influences on social engagement and physical activity of children with ASD, it is reasonable to assume that the general age impact would also extend to individuals with ASD since ASD is considered to be uncured and the long-term consequences of ASD may not be favorable.
24
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
The purpose of this study was to examine the associations of age, social engagement and physical activity in children with ASD when participation in both structured (physical education) and unstructured (recess) play opportunities are considered. It was hypothesized that age would show a significant influence on both physical activity and social engagement participation in both free play and organized settings in children with ASD. In addition, more frequent social engagement behaviors would be observed in higher physical activity levels of children with ASD. 1. Methods 1.1. Participants and setting Twenty-five boys with ASD aged between 7 and 12 years (9.28 1.46) volunteered to participate and returned signed parental informed consent prior to study involvement. Twenty-two of the children had normal speech patterns, and the remaining three children made a few sentences, as based on parent reports and the researcher interaction with participants. All children with ASD met the criteria for autistic disorder of the DSM-IV-Revised system (American Psychiatric Association, 2000) and the Identification Standard for Students with Special Needs (Ministry of Education, 2006b), as assessed by trained and knowledgeable doctors through medical and psychological assessment in the public hospitals (Executive Yuan, 2006a). Level of severity (mild, moderate, severe, and very severe) is based on functioning in the social adaptive skill areas and language comprehension and expression (Executive Yuan, 2006b). Diagnoses included autism (mild or high-functioning, n = 11; moderate, n = 8) and Asperger’s syndrome (n = 6). None of the children had any additional medical conditions such as epilepsy, cerebral palsy or motor deficiencies. All the children reside in urban settings and 24 live in a two-parent household. None were enrolled in a segregated or inclusive school-based physical activity programs or sports teams. The children with ASD received part-time special education services which were individually tailored for each child and based on recommendations from multidisciplinary teams of professionals from regional specialist clinics, parents and educational authorities. No children with ASD had either full- or part-time staff assistant available throughout the day. Mean height was 136.50 10.79 cm, mean body mass was 35.38 8.55 kg, and mean body mass index (BMI) was 18.81 2.97 kg/m2. These values were within the normal ranges for Taiwanese children of this age (Ministry of Education, 2007). Participants’ schools were located in the same geographical area of high social and economic deprivation in a large urban city, and all participants received physical education and recess in inclusive settings without any level of supports. Each school had three morning recesses and three afternoon recesses, and each recess length was 10–20 min in duration. The area of the playground for all schools was very limited because they were located in a crowded urban city. There was no visible playground equipment or markings to stimulate play. Children were allowed to take their own balls, toys, etc. during recess, but there was little evidence of any other equipment being used or being made available. The physical education lesson was 40 min in duration and children in grades 1–2 and 3–6 were required to take 1 and 2 lessons each week, respectively. Twelve of children with ASD attended regular physical education taught by class teachers, and the remaining 13 children were taught by physical education specialists. 1.2. Assessment procedures The study protocol was reviewed and approved by the National Science Council, Taiwan. Study objectives and methods were individually explained to and written consent and assent obtained from parents and children, respectively. Stature and body mass measures were collected according to class teacher reports only if the child was measured at the beginning of the semester this study was conducted. Otherwise, these measures were collected in school health center with all children dressed in light clothing and shoes removed for calculation of body mass index (kg/m2) prior to study involvement. 1.2.1. Physical activity measurement The children’s physical activity levels and patterns were assessed by GT1M ActiGraph, a uniaxial accelerometer that measures vertical acceleration of human motion as well as how many steps they
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
25
take. The accelerometer has been used extensively and reported as a valid objective measure of physical activity in children (Trost et al., 1998). It has also been used in youth with ASD (Pan, in press; Pan & Frey, 2006; Rosser-Sandt & Frey, 2005) and found that this population can tolerate the instrument. Consistent with previous studies, the accelerometer was placed in a small nylon pouch and worn over the right hip with an elastic belt. Prior to data collection, verbal and written instructions on how to wear and care for the device were provided. All participants also engaged in practice trials wearing the device. Participants wore the accelerometer for five consecutive school days during their regularly school schedules. At the start of each school day, the accelerometers were attached by the research assistants to the children and participants were then asked to follow their regular daily routine. Accelerometers were removed at the end of the school day, and the data were immediately downloaded and re-initialized for the next day. Some participants wore the same monitor every day, and all occurring times for physical education class and each recess period were recorded by class teachers and research assistants. Physical activity data were reduced from one physical education class and one recess period each day for five consecutive school days in accordance with social engagement observation. Accelerometers were programmed to collect data in 1-min intervals. Activity counts were analyzed to determine counts per minute (CPM), time spent in MVPA (3.0 MET) and vigorous physical activity (VPA; 6.0 MET) using age-specific count cutoffs (Freedson et al., 1997; Trost, Pate, Freedson, Sallis, & Taylor, 2000). To control for the differences in the monitoring recess and physical education length, the relative (percentage) time spent in MVPA and VPA, and step counts per minute were calculated and used in the subsequent analyses. 1.2.2. Social engagement measurement Social engagement patterns were assessed using the Engagement Check (McWilliam, 1990). It is an observational tool that uses momentary time sampling procedure with a 15-s observe and 15-s record cycle. McWilliam and Bailey (1995) have reported suitable reliability and validity estimates on children with ASD. For the current study, child’s engaged behavior is divided into two types (with adults and peers) and two forms (interactive and noninteractive) of engagement. Interactive engagement was defined as the child’s focus on another person, and his/her behavior being aimed at producing a social effect. Interdependent play, mutual organization, gestures, and talking are examples of interactive engagement. Noninteractive engagement was defined as the child attending to another person or playing nearby with similar materials. Looking, orienting, tracking, and listening are examples of noninteractive engagement. Each child was systematically observed one time in physical education, and one time in recess each day for consecutive five school days. Observations of the two settings occurred on the same week. Informal observations on days prior to formal data collection helped reduce reactivity, and conversations with teachers after the formal observation verified that data were collected on typical social engagement behaviors. The percentage of each social engagement category and the percentage of total social engagement scores were used in the data analyses. 1.2.3. Observer training and interobserver reliability Interobserver agreement was calculated by dividing agreements by agreements plus disagreements and multiplying by 100 for each category. Before the study began, four observers were trained by the researcher to use systematic observation methods and this study’s social engagement instrument. Each observer was required to review the observation instrument and become familiar with the behavioral definitions. During the training sessions, the observers collected data on the Engagement Check by watching videotapes of children with ASD in physical education classes and recess periods previously made by the researcher. When a question arose about how to record a specific behavior, the researcher and the observers discussed the behavior until all five agreed on the appropriate recording procedure. Study observations began only after an interobserver agreement of 85% or greater was achieved for each variable across all four observers. Interobserver agreements that did not meet criterion were addressed by extending the observer training until the interobserver
26
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
agreement criterion was reached. The observers reached an overall agreement of at least 85% agreement for these data. 1.3. Statistical analyses Prior to analysis, physical activity data were examined separately at each setting for the assumptions of multivariate analysis. The residuals were normally distributed, and the assumption of homoscedasticity was met. None of the cases were identified through Mahalanobis distance as multivariate outliers, leaving all cases for the final analysis. Pearson product-moment correlation coefficients were calculated to evaluate the relationship between age, social engagement, and physical activity. When a variable was significant in the correlations, it was entered into the multiple regression analysis. For each variable, the partial correlation is reported, which reflects the contribution of that variable, adjusting for all other variables. Changes in multiple correlations squared (R2 change) are reported to demonstrate the amount of variance explained by each variable. The adjusted multiple correlations squared represent the full model. Because six physical activity and five social engagement variables are separate measures, prediction of these variables was examined in separate regression models. All statistical analyses were computed using SPSS Statistical Software Package, Version 13.0. Values are reported as mean S.D., with significance set at P < 0.05. 2. Results Descriptive statistics for physical activity and social engagement behaviors in children with ASD during physical education and recess are shown in Table 1. The physical education and recess monitoring time was 34.09 4.18 and 50.20 9.72 min, respectively. On average, children with ASD were more active physically and socially during physical education than recess. 2.1. Age and physical activity Table 2 shows bivariate correlations of each physical activity variable and children age. Children age was positively correlated with counts/min (r25 = 0.50, P < 0.01) and steps/min (r25 = 0.57, P < 0.01) in recess and positively correlated with 5-min MVPA during physical education (r25 = 0.43, P < 0.05). The results of the multiple regression analyses showed that children age was significant in the three regression models, explaining 22%, 29% and 15% of the variance in children overall physical activity (counts/min, F1,23 = 7.63, P < 0.05; R2 change = 0.25), steps/min (F1,23 = 10.76, P < 0.01; R2 change = 0.32) and 5-min MVPA (F1,23 = 5.12, P < 0.05; R2 change = 0.18), respectively.
Table 1 Physical activity and social engagement during PE and recess (n = 25)
Physical activity Counts/min MVPA VPA Steps/min 5-min MVPA 10-min MVPA Social engagement Adult-interactive Adult-noninteractive Peer-interactive Peer-noninteractive Total social engagement
PE
Recess
1538.35 862.38 54.86 24.72 8.30 9.16 38.34 20.27 5.25 11.03 0.75 2.05
1318.61 852.49 46.84 25.33 5.82 7.20 33.55 20.62 3.61 3.72 0.76 1.38
28.57 27.49 0.78 2.47 60.09 26.34 24.78 27.78 114.22 44.23
9.34 9.03 0.46 1.61 59.98 33.97 9.58 21.01 79.36 48.80
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
27
Table 2 Bivariate correlations for the physical activity variables and children age during PE and recess PE Age Counts/min MVPA VPA Steps/min 5-min MVPA 10-min MVPA
Recess
0.34 0.09 0.04 0.16 0.43* 0.29
0.50* 0.27 0.37 0.57** 0.16 0.01
*P < 0.05; **P < 0.01; two-tailed; n = 25.
Table 3 Bivariate correlations for the social engagement variables and children age during PE and recess PE Age Adult-interactive Adult-noninteractive Peer-interactive Peer-noninteractive Total social engagement
Recess
0.18 0.24 0.48* 0.17 0.48*
0.19 0.11 0.12 0.15 0.19
*P < 0.05; two-tailed; n = 25.
2.2. Age and social engagement Table 3 shows bivariate correlations of each social engagement variable and children age. Children age was positively correlated with peer-interactive (r25 = 0.48, P < 0.05) and total social engagement behaviors (r25 = 0.48, P < 0.05) during physical education. Children age was not correlated with any social engagement behaviors in recess. The results of the multiple regression analyses showed that children age was significant in the two regression models, explaining 19% and 20% of the variance in children peer-interactive (F1,23 = 6.68, P < 0.05; R2 change = 0.23) and total social engagement (F1,23 = 7.01, P < 0.05; R2 change = 0.23), respectively. 2.3. Social engagement and physical activity Table 4 shows bivariate correlations of each physical activity variable and children social engagement. Noninteractive engagement with adults during physical education was positively correlated with children VPA (r25 = 0.74, P < 0.01) and steps/min (r25 = 0.51, P < 0.01). None of the social engagement behaviors was correlated with any physical activity in recess.
Table 4 Bivariate correlations for the physical activity and social engagement variables during PE and recess
AI ANI PI PNI TSE
Counts/min
MVPA
PE
PE
0.18 0.33 0.28 0.03 0.28
Recess 0.12 0.15 0.10 0.10 0.04
**P < 0.01; two-tailed; n = 25.
0.18 0.33 0.04 0.15 0.06
VPA Recess 0.19 0.23 0.11 0.16 0.04
PE 0.23 0.74** 0.04 0.23 0.35
Recess 0.13 0.16 0.05 0.03 0.00
Steps/min
5-min MVPA
10-min MVPA
PE
PE
PE
0.08 0.51** 0.05 0.09 0.17
Recess 0.06 0.11 0.06 0.12 0.09
0.25 0.02 0.16 0.11 0.32
Recess 0.09 0.29 0.06 0.22 0.08
0.01 0.12 0.33 0.24 0.05
Recess 0.06 0.17 0.14 0.24 0.03
28
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
The results of the multiple regression analyses showed that noninteractive engagement with adults was significant in the two regression models, explaining 53% and 23% of the variance in children VPA (F1,23 = 27.86, P < 0.01; R2 change = 0.55) and steps/min (F1,23 = 8.26, P < 0.01; R2 change = 0.26), respectively. 3. Discussion Factors (age and social engagement) thought to be determinants for physical activity in children with ASD were partially found to be related to physical activity in this study. Age in particular did not support the notion that children with ASD are more likely to be inactive as they age. Infrequent social engagement rate did not widen with age as the activity-deficit hypothesis proposed by researchers (Bouffard, Watkinson, Thompson, Causgrove Dunn, & Romanow, 1996; Wall, 2004). Furthermore, the overall activity levels were not dependent on social engagement either during physical education or in recess setting, indicating that those who engaged more in social interactions were no more physically active than peers who engaged less socially. However, the noninteractive engagement with adults was positively correlated with children’s VPA and steps/min during physical education. An interesting note related to age and physical activity from the current sample was the positive relationships between age and both the overall physical activity and step counts in recess and the positive relationship between age and 5-min MVPA during physical education. This is not supportive of Kozub and Oh’s findings (2004) of negative relationships between age and total bouts of MVPA in individuals with visual impairments, using RT3 activity monitors. This is also not supportive of the same inverse relationships between age and minutes of moderate physical activity measured using RT3 monitors in individuals with intellectual disability (Kozub, 2003). However, in an earlier study of Longmuir and Bar-Or (2000) using a self-report questionnaire, physical activity estimates do not support a marked age-related decline in individuals with cerebral palsy, muscular dystrophy, or visual impairments, but a decline in physical activity with age was observed in youth with chronic medical conditions. In the general population, age-related declines are noted for both males and females between the ages of 8 and 18 years (Thompson, Baxter-Jones, Mirwald, & Bailey, 2003). The inconsistent findings in the literature highlight the importance of using objective assessment methods, support the examination of physical activity patterns within various environments, and confirm the need to measure the physical activity of specific subgroups to avoid generalizations across different disabling conditions. Most of previous studies from aforementioned above in youth with and without disabilities have documented a negative relationship between age and physical activity. Pan and Frey (2005) used accelerometer to assess the influence of age on physical activity in youth with ASD, and found a significant negative association between age and youth physical activity. Results of the current investigation confirm the importance of age as a significant physical activity correlate in children with ASD, but do not reinforce the magnitude of decline in physical activity with age. The failure to find a consistent age influence may be because the age span of the current study is not broad enough to include adolescents. Most youth in the Pan and Frey’s study (2005) were in middle or high school, and reported no recess time and decreased physical education requirements, while elementary school age children had access to these physical activity resources. It may be the case that greater decline in physical activity do not emerge until mid- to late adolescence. This may be particularly true for organized sports, which become increasingly competitive in high school and where children can exercise greater autonomy in their choices. To date, no published studies have examined whether Taiwanese adolescents with ASD participate in fewer activities than others of the same age. Further work should examine this question with this population. Children with ASD in the current study engaged in more peer-interactive interactions as they age, and the total social engagement rate was higher as children become older during physical education. This is not consistent with Oh, Mehmet, and Kozub’s (2004) study, indicating that there was no significant relationships between age and social engagement during physical education in students with visual impairments. The lack of age effect among individuals with visual impairments may have been due to vision and communication challenges resulting in initial overall low rate of social
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
29
engagement (Zanandra, 1998). It is likely that the factors influencing this behavior are different since the presence of a disability significantly alters life experience (Seligman, 1999). The symptoms of ASD are considered to be modifiable when favorable environmental interaction has been provided to make subsequent interactions easier, resulting in both improved skill and motivation to interact (Mundy, Henderson, Inge, & Coman, 2007). Although the impact of environment elements is beyond the scope of the current study, physical education is usually more structured and supervised than recess and student’s interaction is required, making it an ideal opportunity for the promotion of children’s social engagement. Participants in the current study were observed paying more attention to teacher/peer demonstration, feedback and learning outcomes during physical education; whereas the majority of the children with ASD were observed spending the most of time not interacting in recess. More research is needed to determine if the unique elemental characteristics play a role in social engagement of children with ASD. Age-related trends in physical activity and social engagement are an important concern for health professionals given that ASD was considered to be unmodifiable until recently. Compared to their typically developing peers, children with ASD may be more likely to perform poorly in school and are at greater risk for emotional and behavioral problems as they age. However, Mundy et al. (2007) shed new light on this concept, suggesting that favorable environmental and social interaction has the potential to make subsequent positive behaviors. Such favorable interactions can result in not only improved social skills (Thomas & Smith, 2004) and physical fitness (Lotan, Isakov, & Merrick, 2004), but also in reduction of stereotypic behavior (Prupas & Reid, 2001). The positive age-related trends found in physical activity and social engagement of the current participants is supportive of this notion. It appears from the results of the current study identified above that the divergence in activity with age hypothesis (Bouffard et al., 1996; Wall, 2004) is also not supported. Behaviors of children with ASD can be explained to some extent by the self-determination model (Wehmeyer & Garner, 2003), indicating that behavior is a function of environmental as well as personal factors, which in turn, both environmental and personal factors affect individual’s behaviors. Therefore, physical activity and social engagement behaviors of children with ASD may be more affected by social and environmental constraints than the actual impairment. Noninteractive engagement with adults during physical education is positively related to both VPA and steps/min, suggesting that as adult-noninteractive engagement behaviors increased, physical activity levels increased. This is expected because each participant received regular attention and encouragement from the teacher during physical education, which may have been responsible for some increase in social engagement and physical activity. Within the physical education class, participants were provided an opportunity to develop the necessary skills to interact with their teacher. The teacher’s presence and use of verbal and nonverbal curing have been sufficient to perpetuate the social engagement and physical activity behavior in physical education class. However, results of the current study do not support the earlier finding indicating that there was no significant relationships between social engagement and physical activity in individuals with visual impairments during physical education (Oh et al., 2004). This finding may be due to children with visual impairments in the previous study being from a segregated blind school where more competent children may not have opportunity to find highly skilled playing partners and friends of similar interests to engage physically and socially. More studies are needed to determine if setting is a determinant of social or activity patterns in individuals with ASD. These data represent an initial attempt to study age, social engagement and physical activity in children with ASD. However, the convenience sampling and the low sample size limit the inferences of these findings. Regardless, this study is important because it is the first attempt to identify relationships between age, social engagement and physical activity in children with ASD. Findings indicate differences in the relationships between age, social engagement and physical activity among children with various types of disabilities. Further work with a larger sample of children with ASD in a longitudinal design is necessary to properly identify the age and social engagement influences on physical activity. In addition, the impact of other symptoms influences such as language and motor coordination on physical activity of children with ASD is of interest. Additional information is needed to better understand the influences of physical activity for intervention in children with ASD.
30
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
Acknowledgments This research was supported by grant NSC 95-2413-H-017-010, National Science Council, Taiwan. The author would like to thank Yu-Feng Wang, Ching-Yi Hsu, Hsiang-Chun Huang, and Chi-Hung Tien for their assistance with data collection. Special thanks to all children who participated in this study and parents of children for supporting this project.
References American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders: Text revision (4th TR ed.). Washington, DC: American Psychiatric Association. Bouffard, M., Watkinson, E. J., Thompson, L. P., Causgrove Dunn, J. L., & Romanow, S. K. E. (1996). A test of the activity deficit hypothesis with children with movement difficulties. Adapted Physical Activity Quarterly, 13, 61–73. Cairney, J., Hay, J., Faught, B. E., Corna, L. M., & Flouris, A. D. (2006). Developmental coordination disorder, age, and play: A test of the divergence in activity-deficit with age hypothesis. Adapted Physical Activity Quarterly, 23, 261–276. Executive Yuan. (2006a). Hospitals for the identification of disabilities. Retrieved July 11, 2007, from http://www.doh.gov.tw/cht/ content.aspx?doc_no=45921. Executive Yuan. (2006b). Levels of disabilities. Retrieved July 9, 2007, from http://www.doh.gov.tw/cht/content.aspx?dept=R&class_ no=211&now_fod_list_no=6753&array_fod_list_no=6580&level_no=3&doc_no=41058&show=. Freedson, P. S., Sirard, J., Debold, E. P., Pate, R. R., Dowda, M., Trost, S. G., et al. (1997). Calibration of the computer science and applications, Inc. (CSA) accelerometer. Medicine and Science in Sports and Exercise, 29, S45. Jahr, E., Eikeseth, S., Eldevik, S., & Aase, H. (2007). Frequency and latency of social interaction in an inclusive kindergarten setting. Autism, 11, 349–363. Kozub, F. M. (2003). Explaining physical activity in individuals with mental retardation: An exploratory study. Education and Training in Developmental Disabilities, 38, 302–313. Kozub, F. M., & Oh, H. K. (2004). An exploratory study of physical activity levels in children and adolescents with visual impairments. Clinical Kinesiology, 58, 1–7. Longmuir, P. E., & Bar-Or, O. (2000). Factors influencing the physical activity levels of youths with physical and sensory disabilities. Adapted Physical Activity Quarterly, 17, 40–53. Lotan, M., Isakov, E., & Merrick, J. (2004). Improving functional skills and physical fitness in children with Rett syndrome. Journal of Intellectual Disability Research, 48, 730–735. McGee, G. G., Feldman, R. S., & Morrier, M. J. (1997). Benchmarks of social treatment for children with autism. Journal of Autism and Developmental Disorders, 27, 353–364. McWilliam, R. A. (1990). The engagement check. Chapel Hill, NC: Frank Porter Graham Child Development Center. McWilliam, R. A., & Bailey, D. B. (1995). Effects of classroom social structure and disability on engagement. Topics in Early Childhood Special Education, 15, 123–147. Ming, X., Brimacombe, M., & Wagner, G. C. (2007). Prevalence of motor impairment in autism spectrum disorders. Brain and Development, 29, 565–570. Ministry of Education. (2006a). Annual statistics in special education. Retrieved July 09, 2007, from http://163.21.111.100/tlearn/book/ BookAll.asp?BookMainID=7. Ministry of Education. (2006b). Identification standard for students with special needs. Retrieved May 11, 2007, from http://law.moj.gov.tw/Scripts/Query4A.asp?FullDoc=all&Fcode=H0080065. Ministry of Education. (2007). Physical fitness. Retrieved March 1, 2007, from http://www.fitness.org.tw/TW/index.html. Mota, J., Santos, P., Guerra, S., Ribeiro, J. C., & Duarte, J. A. (2003). Patterns of daily physical activity during school days in children and adolescents. American Journal of Human Biology, 15, 547–553. Mundy, P. C., Henderson, H. A., Inge, A. P., & Coman, D. C. (2007). The modifier model of autism and social development in higher functioning children with autism. Research and Practice for Persons with Severe Disabilities, 32, 124–129. Oh, H. K., Mehmet, A. O., & Kozub, F. M. (2004). Physical activity and social engagement patterns during physical education of youth with visual impairments. Review: Rehabilitation Education for Blindness and Visual Impairment, 36, 39–48. Pan, C. Y. (in press). Objectively measured physical activity between children with autism spectrum disorders and children without disabilities during inclusive recess settings in Taiwan. Journal of Autism and Developmental Disorders. Pan, C. Y., & Frey, G. C. (2005). Identifying physical activity determinants in youth with autistic spectrum disorders. Journal of Physical Activity and Health, 2, 412–422. Pan, C. Y., & Frey, G. C. (2006). Physical activity patterns in youth with autism spectrum disorders. Journal of Autism and Developmental Disorders, 36, 597–606. Prupas, A., & Reid, G. (2001). Effects of exercise frequency on stereotypic behaviors of children with developmental disorders. Education and Training in Mental Retardation and Developmental Disorders, 36, 196–206. Rosser-Sandt, D., & Frey, G. C. (2005). Comparison of physical activity levels between children with and without autistic spectrum disorders. Adapted Physical Activity Quarterly, 22, 146–159. Seligman, M. (1999). Childhood disability and the family. In V. L. Schwean & D. H. Saklofske (Eds.), Handbook of psychosocial characteristics of exceptional children (pp. 111–131). New York: Kluwer Academic/Plenum Publishers. Sigman, M., & Ruskin, E. (1999). Continuity and change in the social competence of children with autism, down syndrome, and developmental delays. Monographs of the Society for Research in Child Development, 64, 11–30. Thomas, N., & Smith, C. (2004). Developing play skills in children with autistic spectrum disorders. Educational Psychology in Practice, 20, 195–206.
C.-Y. Pan / Research in Autism Spectrum Disorders 3 (2009) 22–31
31
Thompson, A. M., Baxter-Jones, A. D. G., Mirwald, R. L., & Bailey, D. A. (2003). Comparison of physical activity in male and female children: Does maturation matter? Medicine and Science in Sports and Exercise, 35, 1684–1690. Trost, S. G., Pate, R. R., Freedson, P. S., Sallis, J. F., & Taylor, W. C. (2000). Using objective physical activity measures with youth: How many days of monitoring are needed? Medicine and Science in Sports and Exercise, 32, 426–431. Trost, S. G., Pate, R. R., Sallis, J. F., Freedson, P. S., Taylor, W. C., Dowda, M., et al. (2002). Age and gender differences in objectively measured physical activity in youth. Medicine and Science in Sports and Exercise, 34, 350–355. Trost, S. G., Ward, D. S., Moorehead, S. M., Watson, P. D., Riner, W., & Burke, J. R. (1998). Validity of the computer science and applications (CSA) activity monitor in children. Medicine and Science in Sports and Exercise, 30, 629–633. U.S. Department of Health and Human Services. (1996). Physical activity and health: A report of the surgeon general. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion. Wall, A. E. (2004). The developmental skill-learning gap hypothesis: Implication of children with movement difficulties. Adapted Physical Activity Quarterly, 21, 197–218. Wehmeyer, M. L., & Garner, N. W. (2003). The impact of personal characteristics of people with intellectual and developmental disability on self-determination and autonomous functioning. Journal of Applied Research in Intellectual Disabilities, 16, 255–265. Zanandra, M. (1998). Play, social interaction and motor development: Practical activities for preschoolers with visual impairments. Journal of Visual Impairment and Blindness, 92, 176–189.