Physical Activity in Individuals with Autism Spectrum Disorders (ASD): A Review Physical Activity in Individuals with Autism Spectrum Disorders (ASD): A Review

Current recommendations indicate that children and youth ages 5–17 should partici- pate in 60 min and adults in 150 min of moderate-to-vigorous physical activity daily. Research suggests that physical activity levels of individuals with autism spectrum dis- order (ASD) are lower than typically developing and developed peers. Despite evidence for PA decreasing negative behaviors and promoting positive behaviors, individuals with ASD may be less motivated and less likely to participate. Individuals with ASD may be more likely to be overweight or obese than their typically developing counter- parts as a result of decreased activity levels. Conflicting findings regarding PA levels in individuals with ASD have been reported. Given mixed evidence, further inquiry is warranted. The present chapter provides a review of literature pertaining to PA in individuals with ASD. Four databases were searched. Predetermined search terms and inclusion/exclusion criteria were clearly outlined to identify relevant articles which were then critically appraised. This research provides a greater understanding of the status of PA participation of individuals with ASD. Four to remove irrelevant articles and duplicates. The first two authors subsequently appraised abstracts. Finally, full texts were assessed based on the specific inclusion and exclusion criteria outlined previously. A total of 69 articles were included in the final review. Articles were sorted into five categories: (1) levels of PA (n = 10); (2) predictors related to PA (n = 4); (3) PA related to other outcome variables (n = 4): (4) PA interventions leading to changes in other outcome variables (n = 30); and (5) interven tions that lead to changes in PA (n = 5). Categories 1 (levels of PA) and 2 (predictors related to PA) were combined in consideration of articles that assessed both variables (n = 16 for a total n = 30). Each article was critically analyzed based on the following components: descriptive information, research methodology, participant characteristics, physical activity measures Findings then synthesized.


Introduction
Numerous physical and mental health benefits have been attributed to regular participation in physical activity (PA) and limited sedentary behavior [1][2][3]. Nevertheless, global levels of insufficient PA are reported, and physical inactivity levels are rising [4,5]. The World Health Organization (WHO) [5] recommends that children and youth ages 5-17 should participate in 60 min and adults ages 18-64 should participate in 150 min of moderate-tovigorous PA (MVPA) daily. For individuals with autism spectrum disorder (ASD), recent reports indicate that the levels of PA are significantly lower than typically developing and developed peers [6].
The WHO estimates one person in 160 has an autism spectrum disorder [7]. A group of neurodevelopmental disorders diagnosed in childhood and persisting throughout life, ASD is characterized by varying challenges with communication, social interaction, and repetitive behaviors and movements [8]. Although not recognized as a formal diagnostic feature, sensorimotor impairments have also been identified as a cardinal feature of ASD [9,10]. Furthermore, comorbid conditions typically manifest in individuals with ASD, including attention-deficit hyperactivity disorder (ADHD), anxiety disorders, and chronic sleeping problems [11][12][13].
The aforementioned difficulties and comorbid conditions combined have been shown to significantly impact the quality of life for individuals with ASD [14]. Despite evidence for PA decreasing negative behaviors and promoting positive behaviors [15], individuals with ASD may be less motivated and less likely to participate in PA [16]. As a result of decreased activity levels, individuals with ASD are more likely to be overweight or obese than their typically developing counterparts [6], thus leading to further health-related challenges. Notwithstanding the previous literature, conflicting findings regarding physical activity in individuals with ASD have also been reported. In one recent example, Corvey et al. [17] identified no relationships between ASD and overweight or physical activity after controlling for comorbidities and medications. Tyler et al. [18] found that, despite being less active than their typically developing peers, children with ASD did meet physical activity guidelines set out by the US Department of Health and Human Services (i.e., 60 min of moderate-to-vigorous PA/ day). Clearly, further inquiry is warranted.
The present chapter provides a review of literature pertaining to physical activity in individuals with ASD. Four research questions were assessed as follows: (1) What is the status of PA participation; (2) Does PA decrease negative behaviors and/or promote positive behaviors; (3) What facilitators and barriers exist; and (4) What PA intervention programs have demonstrated effectiveness?

Methods
In July of 2016, a computerized search of four electronic databases (PubMed, PSYCHINFO, Web of Science, and EBSCOhost) was conducted. Two sets of key words were used in the search strategy to identify articles that included participants with ASD (Autism Spectrum Disorder, ASD, Autism, Autistic disorder, Pervasive Developmental Disorder Not Otherwise Figure 1 depicts a summary of the phases of the review process. The initial search produced 1823 articles. Titles were screened to remove irrelevant articles and duplicates. The first two authors subsequently appraised abstracts. Finally, full texts were assessed based on the specific inclusion and exclusion criteria outlined previously. A total of 69 articles were included in the final review. Articles were sorted into five categories: (1) levels of PA (n = 10); (2) predictors related to PA (n = 4); (3) PA related to other outcome variables (n = 4): (4) PA interventions leading to changes in other outcome variables (n = 30); and (5) interventions that lead to changes in PA (n = 5). Categories 1 (levels of PA) and 2 (predictors related to PA) were combined in consideration of articles that assessed both variables (n = 16 for a total n = 30). Each article was critically analyzed based on the following components: descriptive information, research methodology, participant characteristics, physical activity measures and/or intervention, outcome measures, and overall findings. Findings were then synthesized.

PA interventions leading to change in other outcome variables
Thirty studies  (see Table 3 Table 3), or null effects (n = 9; indicated by † in Table 3) following the PA interventions were reported. Taken together, there is no evidence to suggest that PA interventions have negative effects, nor is there evidence to show one PA intervention is superior to others, likely attributed in part to the multiple outcome measures.

Interventions that lead to changes in PA
Five studies [81][82][83][84][85] were included (see Table 4 and included walking, jogging, snowshoeing, and cycling. One study investigated a motor skills intervention. All participant groups included over 50% males, and only two articles included participants over the age of 18 [83,84]. Four studies [81,[83][84][85] found an increase in participation and overall levels of PA, whereas one study, focusing on a motor skills intervention, found no difference in PA levels [82]. Together, findings revealed PA and/or health interventions can influence sustained PA levels post-intervention; however, there is insufficient evidence to conclude whether interventions that are not PA-based influence PA levels.

Discussion
Taken together, findings revealed lower levels of PA in individuals with ASD [ 24,27,29,33,37,38,44]. Nevertheless, studies that report no difference were also common [e.g., 17,34]. Barriers to PA include, but are not limited to, finances, lack of resources and opportunities, poor motor skills, behavioral and learning problems, the need for supervision, family time constraints, lack of a partner, and lack of available transportation. Must et al. [39] reported a positive relationship between age and the total number of barriers. Furthermore, the number of barriers was inversely related to the number of PA hours and total number and types of activities per year. Facilitators to PA included good equipment and community programs.
There was evidence that PA interventions can improve certain outcome measures, such as communication, balance, and fitness levels [e.g., 54,56,71]; however, it is also important to note that others observed no effect [e.g., 62,70]. Importantly, there was no evidence to suggest that PA interventions cause negative effects. Interventions that aimed to address levels of PA specifically found that PA interventions lead to increased PA levels, while one motor skill intervention [82] was not effective. Overall, no one intervention was suggested as optimal for decreasing negative and/or promoting positive behaviors.
Common limitations included small sample sizes with little ethnic and socioeconomic diversity that limited generalizability and underpowered analyses. Unequal sex distributions were repeatedly observed, as many participant groups were comprised of mainly males. It is important to consider that this may be a result of the intrinsic property of ASD being five times more prevalent in males than in females (CDC, 2014). Assessments of PA levels were limited, in some cases by parent-report assessments, where more objective assessments (i.e., accelerometer data) were limited by compliance, and the inability of the tool to assess all PAs (e.g., water activities). With respect to interventions, short durations were commonly reported. Furthermore, studies investigating a change in PA as the outcome variable were limited. Finally, most studies included children that were high functioning on the spectrum. Methodologically this review was limited to four search engines, and papers published within the last decade. Unpublished studies and studies published in languages other than English were not included. The quality of the studies was also not evaluated. These may have biased the results.
Future research of PA interventions should investigate the legitimacy and benefits of specific PA interventions, which may help determine the effects of distinct outcome measures. Furthermore, research on interventions leading to a change in PA should investigate non-PA interventions in the future to determine the plausibility of changing PA levels through other intervention methods (i.e., motor skill interventions). In addition, it would be beneficial to investigate the long-term changes in PA following these interventions to determine whether this effect is sustained over time. Overall, research investigating physical activity for individuals with ASD should be explored with larger sample sizes, over longer time periods and across the spectrum. This would provide more comprehensive information on the pros and cons of physical activity for this vulnerable population.