The poor levels of fitness in children with autism are prompting concern for the children’s future health. This study looked to assess if a computer-based activity programme could improve fitness levels (as reflected in cardiopulmonary function) of these children, and achieve a reduction in their body mass index. In a randomised controlled trial, 50 children with autism (of which 33 were under the age of 11 years and 39 were boys) were allocated to an intervention group which encouraged them to use the Nintendo Wii and the software package “Mario and Sonics at the Olympics” in addition to their routine physical education classes. 50 children with autism (34 under the age of 11 years and 40 being boys) acted as controls. At the end of one year, analysis of the changes in scores using analysis of covariance (ANCOVA) on the Eurofit fitness tests showed that the intervention group had made statistically significant improvement on all tests other than flexibility. These improvements were also significantly better than controls. This type of intervention appears to be an effective addition to standard fitness training in order to help children with autism improve their fitness levels.
There is increasing concern about the decline in fitness among children generally and how this reduction increases the risk of ongoing health issues in these children. This fitness decline is associated with a dramatic rise of obesity among the young generally, and since many children and young people with autism have a relatively inactive lifestyle, it has been suggested that this decreased physical activity, together with other factors such as unusual dietary patterns, is the major factor in the relatively greater increased rate of obesity found in children with autism when compared to their typically developing peers [
Research has shown that increasing the physical activity of children with autism improves general fitness, has a positive impact upon cardiorespiratory function, and offers the potential to control weight gain [
To achieve this change is however not easy, with the difficulty of encouraging exercise in children with special needs being well discussed in the literature [
In order to evaluate the potential value of computer-based activity programmes to children with autism, a study was undertaken to compare one such programme to standard school-based physical education in the children attending schools with classes specifically for children with autism in the North East of England.
The aim of the project was to assess whether additional physical activity in the form a computer-based activity game could improve the cardiopulmonary fitness of children with autism and have a positive impact upon their body mass index (BMI) which also is used as a proxy for fitness. The project was a parallel design with stratification to control for age and gender, and then block randomisation to the intervention and nonintervention groups.
One element of the assessment chosen was the multistage progressive shuttle run test (known as the bleep test), which has been shown to be a good measure of cardiopulmonary function because
Having obtained ethical approval from academic bodies, educational gatekeepers, and parents, the study was carried out over the period of one academic school year (September to June 2011-2012). Three schools in the locale that had classes specific for the teaching of children with autism agreed to take part in the study. The children in these settings ranged in age from 5 to 15 years of age and none had any physical disorder or illness that would reduce their ability to participate in the fitness programme. The families of all the children were approached to participate in the study. After the details of the study had been explained all agreed. The Statements of Special Educational Need for all of the children indicated that they all had an IQ in the moderate/severe range of ability. The children in this pooled sample were then randomly allocated to intervention or nonintervention groups, controlling for age and gender. Selection was completed when 50 families had been allocated to each group. Background information was collected from the family concerning make-up, medical conditions, employment of parents, and so forth.
The physical fitness of the children was evaluated using elements of the Eurofit physical fitness test battery [
Fitness tests used in the study.
Test | Area of measurement/explanation | Equipment required |
---|---|---|
Body mass index | Body fat measurement | Height scale and weighing scales |
|
||
National Coaching Foundation (NCF), multistage fitness test-progressive shuttle run test known as the bleep or beep test. | Cardiorespiratory fitness (aerobic) |
Test CD and CD player or MP3 download, 20 m running space, marker cones. |
|
||
Standing long jump test (broad jump) | Measurement of explosive leg strength. A starting line is marked on the floor and the participants have to take a standing two-footed jump from this line with the aim to jump as far as they can. The arms may be used to aid the jump. | 2 m of space. Marker for starting line. Tape measure. Or Standing broad jump kit. |
|
||
|
Measurement of speed and agility and aerobic capacity. Subjects run between two markers of 5 m; there and back counts as 1 repeat and this is continued until 10 repeats have been carried out. | Cones, tape measure to mark out 5 m, stop watch. |
|
||
Partial curl up test |
Measurement of abdominal strength and endurance. The child lies on the floor with knees flexed. The subject curls up his/her trunk then lowers back to the floor, repeating this as many times as they can. The test lasts for duration of 30 seconds. | Mat for the child to lie on. Stop watch. Additional person to hold legs or time test. |
|
||
Sit and reach test | Test measures the flexibility of the lower back and hamstring muscles, that is, hip flexibility. The test involves sitting on the floor with legs out straight ahead and then reaching as far forward as possible. | Sit and reach board/equipment or tape measure. |
As described in Table
The testing was carried out in the schools’ usual PE areas so the children would be familiar with the spaces. All the schools carried out the tests using the same set procedure, as described in the protocols which accompany each of the tests, and were fully explained to the participating staff. To help the children understand the test elements visual modelling was used together with simplified instructions. To ensure that the children gave their best effort they were instructed to go on until they could go on no longer. In addition, since the teaching staff leading the testing knew the children well they were allowed to explain this requirement in a way they felt the children would best understand.
To overcome the learning effect associated with field-based fitness tests, the children had two trial assessments, and then their scores were recorded at a further assessment, with one of the team (KD) in attendance. The study took place over three academic terms, with the same fitness level tests being repeated at the conclusion of the study period, again under supervision.
In addition to their standard school physical education programme, the children in the intervention group used a Nintendo Wii and the software package “Mario and Sonics at the Olympics.” The participants and staff received an induction in relation to the Wii and associated software games package at the beginning of the intervention. The package was chosen because it had various levels of physical interaction, and each pupil played the game in groups of 2 to 4, under supervision, for a period of 15 minutes per day, three times per week.
Both groups received the standard physical education programme which, in England, is statutorily defined in the National Curriculum for Physical Education [
To gain some measure of the dynamics of the families in each of the study groups, the general family functioning of the two groups was compared by asking the children’s main care giver to complete the Family Adaptation and Cohesion Evaluation Scales (FACES IV) [
As can be seen in Table
The demographic information of the sample.
Intervention group ( |
Control group ( |
|
---|---|---|
Boys | ||
≤10 yrs | 27 | 27 |
≥11 yrs | 12 | 13 |
Girls | ||
≤10 yrs | 6 | 7 |
≥11 yrs | 5 | 3 |
Only child | 9 | 25 |
Living with both parents | 35 | 40 |
Both parents working | 7 | 9 |
Both parents unemployed | 13 | 12 |
All statistical analyses were conducted with IBM-SPSS version 21.0. Normalcy of the fitness and family data was assessed using the Shapiro-Wilk test [
Testing of the family assessment data by the Shapiro-Wilk test confirmed it to be parametric, and the results from the FACES IV questionnaire showed no significant difference in family functioning between the two groups, with the control group mean for cohesion being 57.82 (std dev 12.77) and for the intervention group 60.54 (std dev 11.65,
The fitness results showed a wide variation, and the Shapiro-Wilk test confirmed that the results could not be considered to be normally distributed. As can be seen in Table
Initial and follow-up fitness results as medians and interquartile range (IQR) scores (
Control group | Intervention group | Difference between initial scores of control and intervention groups ( |
|||||
---|---|---|---|---|---|---|---|
Median (IRQ) | Difference between initial and follow-up scores ( |
Median (IRQ) | Difference between initial and follow-up scores ( |
||||
Initial score | Follow-up score | Initial score | Follow-up score | ||||
Body mass index (BMI) | 20.2 (5.6) | 21.5 (5.5) | 4.86*** | 20.1 (6.1) | 19.8 (5.6) | −5.37*** | 0.72 |
Bleep test | 2 (3) | 2 (2) | 1.61 | 3 (4) | 4.5 (5) | 5.92*** | 2.52* |
Shuttle run | 84.5 (49) | 90 (42) | 4.04*** | 92 (27) | 68 (43) | −4.89*** | 2.33* |
Broad jump | 54.5 (23) | 52.5 (20) | −0.77 | 80 (41) | 92.5 (50) | 5.12*** | 4.59*** |
Sit-ups | 7.5 (5) | 7.5 (4) | 1.94 | 10 (10) | 13 (7) | 5.08*** | 2.27* |
Flexibility | 10 (6) | 10 (8) | 2.48* | 8 (13) | 9 (13) | 0.09 | 2.05* |
After the study period the children in the control group had shown minor improvements in flexibility, as assessed using the sit and reach test, over the time of the study, but the sit-up score showed little change, as did the broad jump score and bleep test. The control group actually showed a statistically significant increase in their body mass index (BMI), with only four children showing improvement. Also the group’s shuttle run times showed an increase (deterioration) with only eight of the fifty improving their scores.
By contrast, the intervention group showed highly significant improvement in all of the fitness measures with, for instance, 46 showing improvement in their shuttle run times, (Table
When the degree of change seen in control and intervention groups over the time of the study is compared (Table
Comparison of changes to fitness scores between the control and intervention groups (
Number of children showing improvement in score | Significance in difference of degree of change between control and intervention groups ( |
||
---|---|---|---|
Control group ( |
Intervention group ( |
||
VO2 max | 13 | 43 | 16.13*** |
Body mass index (BMI) | 4 | 39 | 30.06*** |
Bleep test | 13 | 43 | 13.15*** |
Shuttle run | 8 | 46 | 41.35*** |
Broad jump | 18 | 39 | 40.30*** |
Sit-ups | 9 | 38 | 25.72*** |
Flexibility | 4 | 4 | 1.20 |
To offer some insight as to whether out-of-school activity might bias the fitness results, the parents were asked about the index child’s other activities. Only 10 children had any out-of-school activities, four in the intervention group and 6 in the controls.
The potential benefits from using computer-based activity games has been demonstrated in adults [
The importance of improving physical fitness is well recognised given its strong association with physical health. Physically active children and adolescents have lower blood pressure levels, more favourable lipoprotein levels, higher bone density, and decreased adiposity compared to their sedentary counterparts [
However in recent years there has been a growing concern about the reduction in physical exercise amongst children [
Children with autism have greater difficulties than typically developing children with participating in physical activity because of the nature of their difficulties. Although the literature on physical activity levels in children with autism is sparse [
As with typically developing children, regular physical activity has been shown to reduce the BMI of children with autism [
These findings strengthen the conclusion that it is important for these children to have a continuous programme of exercise, with a method of delivery that maintains the child’s engagement. There is evidence that children with autism have difficulty understanding the goals of physical exercise [
While there are an increasing number of computer-based programmes that offer guided fitness activities, programmes that are game-based have been viewed as being of little value to health and well-being. Indeed video games have often been blamed as contributing factors to the increasing obesity levels [
In any such study there are limitations to the conclusions that can be drawn. In this case the sample was drawn from a small geographical area in the North of England which is urban in nature, making it difficult to be confident about how much the findings would represent the impact upon children with autism in other areas. Also, although efforts were made to match the two groups, the relatively small population from which to draw meant that the matching process could only use a limited number of parameters. By choosing age and gender the possibility of differences in initial fitness levels between the groups was accepted, and in the event this proved to be the case. However it was anticipated that the potential impact of this upon the results could be reduced by using nonparametric statistical analyses and the degree of change in scores as the main outcome measure. Using a pooled sample from three schools gave rise to a potential contamination between control and intervention children, but it was decided to proceed with this method because it gave a degree of control about wider school issues and reduced the extra sessions that the staff in each school had to undertake. The staff had been asked not to alter the standard PE lessons that all the children provided, but the project resources did not permit the content of these to be monitored. It was recognised that the ongoing research may have prompted some change in content of the routine PE sessions, but it was hoped that the fact that both control and intervention children were sharing these lessons would minimise the impact.
In addition, it is known that a child’s level of physical activity is strongly influenced by family issues such as parental expectations and support, the modelling of siblings, and the opportunities that exist within their community to participate in physical exercise [
The growing concern about deteriorating fitness levels and their impact upon children’s health has a special significance for children with autism because of the way their patterns of functioning makes it difficult to encourage their participation in standard school physical education activities and how prone they are to become obese, further reducing their capacity to exercise. The results from this study indicate that an improvement in specific fitness measurements and BMI can be achieved by children with autism if they follow a computer-based activity programme (in this particular study the “Mario and Sonic at the Olympics” for the Nintendo Wii). Such an intervention is a practical option for schools, but the finding that some aspects of fitness did not improve emphasises the importance of offering such computer-based activities as part of a wider physical fitness regime so that physical functions that such programmes do not address can be tackled through other activities.
The authors declare that they have no conflict of interests regarding the publication of this paper.
The authors gratefully thank all of the schools, staff, children and their parents who took part in this study. This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.