The Impact of Aquatic Exercise on Sleep Behaviors in Children With Autism Spectrum Disorder

Abstract

The purpose of this pilot study was to determine if participation in an aquatic exercise program improves sleep in children with Autism Spectrum Disorder (ASD). Participants included 8 children. An A-B-A withdrawal design was utilized. Each phase lasted for 4 weeks. The treatment included 60 min of aquatic exercise 2X/week. Phone calls to parents of the participants were made throughout the duration of the study. Parents were asked questions related to sleep latency, nighttime wakenings, and sleep duration. A one way repeated-measures analysis of variance (ANOVA) was utilized to determine if differences existed between phases. Statistically significant difference existed for sleep latency (p < .001) and sleep duration (p < .001). These results suggest that participation in aquatic exercise may improve the sleep habits of children with ASD.

Keywords

sleep autism spectrum disorders exercise

Introduction

Sleep disturbances are a common clinical problem among children with Autism Spectrum Disorder (ASD). These disturbances can affect daily functioning and intellectual development, and add stress at home (Reynolds & Malow, 2011). There is an extensive amount of evidence documenting sleep disturbances among children with ASD. Souders et al. (2009) found the prevalence of sleep problems to be 66.1% in children with ASD compared with 45% in typically developing (TD) children. Similarly, Eggerding (2010) and Reynolds and Malow (2011) found the prevalence of sleep disturbances to be 80%. Although these statistics clearly indicate that a problem exists, Richdale and Schreck (2009) suggest this may be an underestimate given that most of the studies rely on parent reports or sleep journals. Richdale and Schreck described an actigraphy study, where sleep was measured using body movement, in which 62 children with ASD, who had no parental report of a sleep dysfunction, were found to have decreased quality of sleep. Whereas TD children commonly have temporary sleep dysfunctions, children with ASD demonstrate long-term sleep problems (Eggerding, 2010).

The specific types of sleep disturbances among children with ASD have been described in the literature. Souders et al. (2009) found a statistically significant difference between sleep latency, or the amount of time it takes to fall asleep, when comparing TD children with those with ASD. They also found a significant difference for amount of activity demonstrated during sleep and the length of time that night wakenings lasted (34 min in TD children versus 49 min in children with ASD). Richdale and Schreck (2009) found the number of wakenings were not different but rather the length of these wakings, which can be up to 2 to 3 hr.

Hoffman et al. (2005) studied 80 children with ASD and found a relationship between specific sleep problems and the severity of ASD symptoms as measured on the GARS (Gilliam Autism Rating Scale). They found that sleep disordered breathing, including obstructive sleep apnea, was able to predict stereotypical behavior, problems with social interaction, and severity of autism (AQ = Autism Quotient). In addition, parasomnias, including nightmares, sleep walking, waking up screaming, and enuresis, were the best predictor for developmental disturbances. The investigators further concluded that sleep disordered breathing and parasomnias may negatively affect sleep duration and affect daytime sleepiness. Similarly, Richdale and Schreck (2009) examined parent-reported sleep problems, actigraphy studies, and polysomnography studies and found that sleep duration negatively affected daytime symptoms.

Sleep disturbances in children with ASD affect not only the child, but the entire family unit. Sleep disorders in children with ASD may also result in sleep problems for parents and siblings as well (Eggerding, 2010). They may also increase parental stress levels and negatively affect caregiver health and quality of life (Reynolds & Malow, 2011).

The suggested causes of sleep dysfunctions among children with ASD can be classified into categories, including biological abnormalities, psychological factors, and environmental influences. Biological abnormalities include disrupted circadian rhythms, non-adherence to typical wake–sleep cycles, and decreased melatonin release, which helps to regulate circadian rhythms. Psychological factors include increased levels of anxiety and a higher likelihood to develop depression among children with ASD (Richdale & Schreck, 2009). Environmental influences may include too early of a bedtime, which may be due to parental need for rest in the evening (Allik, Larsson, & Smedje, 2006). Other environmental factors may be related to bedtime parenting practices, such as allowing their child to fall asleep on the couch and not maintaining a bedtime routine. In addition, difficulty with communication and not understanding social cues that imply bedtime and falling asleep; high sensitivity to light, touch, and sound; and perseverating on an idea that may hinder their ability to calm down, all may affect the sleep behaviors of children with ASD (Reynolds & Malow, 2011).

Management of sleep issues in children with ASD is imperative given the impact that such disturbances have on daytime behavior and parental stress. According to Johnson and Malow (2008), there has yet to be a Food and Drug Administration (FDA) approved medication to treat pediatric insomnia. However, the impact of melatonin, a nutritional supplement not FDA-regulated, on sleep problems in children with ASD has been explored. Malow et al. (2012) conducted a 17-week actigraphy study on 24 participants and found the use of melatonin significantly decreased sleep latency. The study did not specifically look at sleep duration, sleep efficiency, or night wakenings, although the researchers suggested none of these showed improvements with melatonin. While melatonin may appear to be a quick fix in the management of sleep issues, Johnson and Malow’s review of evidence on pharmacologic treatment indicated that at least minimal adverse effects, such as mild tiredness, difficulty awakening, and headaches, may occur.

Parents of children with ASD often prefer behavioral approaches to pharmacological interventions (Williams, Sears, & Allard, 2006). Behavioral approaches may increase parents’ sense of competence, control, and ability to cope (Wolfson, Lacks, & Futterman, 1992). These treatment options eliminate the concern for adverse events and specifically target the behavioral symptoms of the child. In a systematic review of behavioral treatment of sleep problems in children with ASD, Vriend, Corkum, Moon, and Smith (2011) analyzed 15 studies that involved a combination of sleep hygiene and at least one other behavioral intervention (such as standard extinction, scheduled awakenings, chronotherapy, and stimulus fading) for effects on sleep problems and effectiveness as an intervention. They determined that even though all 15 studies used establishing good sleep hygiene/parent education as an intervention, the interventions were not sufficient to eliminate sleep problems in children with ASD. Although some of the interventions appeared to be beneficial, none met the criteria to be classified as well-established or probably efficacious as defined by Chambless and Hollon (1998). Limitations of these studies included heterogeneous samples, limited sample size, and inclusion of children with comorbid conditions.

As described above, problems exist with both pharmacological and behavioral treatment approaches. While melatonin has risks of adverse side effects, behavioral interventions have not statistically been found to be effective in improving sleep habits. Furthermore, attempting to address behavioral symptoms requires one to determine the cause of each child’s sleep problems. In fact, many investigators focus on determining the cause of sleep problems to establish an effective treatment, although the cause is typically related to multiple issues (Reynolds & Malow, 2011). However, many of the causes listed are simply theorized as possibilities and very few studies exist that point to a definitive cause and effect. Even if there was an established cause, as many of these studies suggest, it appears to be multiple factors that disrupt sleep patterns in children with ASD, which would result in multiple interventions. As a result, non-specific treatment approaches may be considered, which can be beneficial in spite of multiple comorbidities. Exercise is one non-specific intervention of interest.

A healthy, safe, and relatively inexpensive means of improving sleep is exercise. In a systematic review examining the effects of exercise on sleep in adults, Youngstedt (2005) found that the majority of epidemiological studies indicated a significant positive relationship between self-reported exercise and self-reported sleep quality. An increase in total sleep time and decrease in wake time after sleep onset was found when individuals exercised within 4 hr of bedtime (Youngstedt, 2005). The impact of exercise on the sleep patterns of TD adolescents has also been explored. Brand et al. (2010) conducted a study investigating the effects of exercise on sleep patterns and psychological functioning in TD adolescents. They found that compared with controls, athletes who exercised at higher levels had shorter sleep onset latency, fewer awakenings, higher concentration during the day, and lower tiredness during the day.

While land-based aerobic exercise may be beneficial to improve sleep patterns, aquatic exercise may be a better option than land-based aerobic exercise for children with ASD. Water provides relatively constant somatosensory input, which is an important feature for sensory integration (SI) techniques often used in treating children with ASD. It has been suggested that SI techniques moderate arousal and improve a child’s ability to interpret and use sensory input (Case-Smith & Miller, 1999). Although evidence on the benefits of aquatic exercise for children with ASD is limited, significant improvements have been found in social behaviors, physical fitness, and parent satisfaction. In two separate 10-week water exercise and swimming programs, Yilmaz, Yanardag, Birkan, and Bumin (2004) found reductions in stereotypical movements, and Pan (2010) identified decreases in antisocial behavior problems in children with ASD. Researchers have also identified increased physical fitness (including improved cardiorespiratory endurance; Fragala-Pinkham, Haley, & O’Neil, 2008), enhanced swimming skills (Fragala-Pinkham, Haley, & O’Neil, 2010; Huettig & Darden-Melton, 2004; Pan, 2011; Rogers, Hemmeter, & Wolery, 2010), improvements in motor performance (Best & Jones, 1974; Yilmaz et al., 2004), and better overall physical fitness (Pan, 2011; Yilmaz et al., 2004). In a preliminary study by Vonder Hulls, Walker, and Powell (2006), clinicians reported a substantial increase in swim skills, attention, muscle strength, balance, tolerating touch, initiating/maintaining eye contact, and water safety. Vonder Hulls et al. (2006) suggest that aquatic therapy improves physical function and has psychological benefits (sense of accomplishment, greater confidence, improved self-esteem).

By including aquatic exercise into the weekly routine of children with ASD, it may benefit their sleeping patterns and possibly improve daytime behavior, which may positively affect relationships with family and peers. The impact of exercise on sleep habits in children with ASD has not been explored. The purpose of this study was to explore the impact of aquatic exercise on sleep behaviors in children with ASD. It was hypothesized that children with ASD would fall asleep quicker and would have a longer sleep duration with fewer wakenings throughout the night following an aquatic exercise session.

Method

Participants

Participants were recruited through letters sent home through local ASD support classrooms. Inclusion criteria consisted of parent report of a formal diagnosis of ASD, ages 6 to 11, parent/guardian report of sleep dysfunction, the ability to be comfortable in the pool, and parent/guardian consent. A formal ASD scale was not used to quantify severity. Any individuals who had a complex medical comorbidity in addition to ASD, or any other developmental disorders, were excluded. The participants were not engaged in other after-school activities on the days they participated in the aquatics program.

The participants included eight children (five males, three females), with a mean age of 8.88 years. Six of the participants were taking at least one medication (M = 2.38), and one participant was non-verbal. Five participants were a “healthy weight,” one participant was “overweight,” and two participants were “obese” according to their body mass index.

Materials

The Children’s Sleep Habits Questionnaire (CSHQ), developed by clinical researchers at Brown University, was used to quantify sleep problems in the participants at the beginning of the study. It consisted of eight subscales (Bedtime Resistance, Sleep Onset Delay, Sleep Duration, Sleep Anxiety, Night Wakings, Parasomnias, Sleep Disordered Breathing, and Daytime Sleepiness) and 33 items for the total sleep disturbance score. Next to each item was a scale, a score of 1 indicated that the behavior “rarely” occurs (0–1 times per week), 2 indicated the behavior “sometimes” occurs (2–4 times per week), and 3 indicated the behavior “usually” occurs (5–7 times per week) and whether or not the item was a problem (Yes, No, N/A). The instructions asked the parent/guardian to answer the questions based on the most recent typical week. The highest possible score for the questionnaire was 99, indicative of severe sleep problems, whereas the lowest possible score was 33, indicative of minimal to no sleep problems. Owens, Spirito, and McGuinn (2000) determined the CSHQ instrument to have acceptable test–retest reliability with correlations for the subscales ranging from 0.62 to 0.79. The investigators also found the measure to be valid, 30 of the 33 items were statistically significant at p < .001. They concluded that a cut-off score of 41 properly identifies 80% of the children with sleep disorders, yielding a sensitivity of 0.80 and specificity of 0.72.

Procedure

An A-B-A withdrawal design was utilized to measure changes within the participants. Each phase lasted for 4 weeks, with A being the control phase, and B being the treatment phase. The purpose of using this design was to show that the desired response, improved sleep patterns, only occurred in the presence of treatment (Portney & Watkins, 2000). If the second control phase showed that the children’s behaviors returned to baseline, one can conclude that exercise caused the behavioral response during the treatment phase. This design allowed control for extraneous variables as it was not likely that these factors would be present both prior to and after treatment.

Prior to the start of the study, the researchers met with parents to explain the study parameters and describe the information that they should be recording on days of data collection. During all three phases of the study, the researchers made phone calls to parents/guardians questioning them about their child’s previous night of sleep (2X per week). Calls were made on the same 2 days throughout the duration of the study, coinciding with days following exercise during the B phase. From this data, number of wakenings, sleep latency, and total time asleep were calculated for each child.

An interview protocol was established for the semiweekly telephone calls to the parent/guardian and consisted of the following questions: What time did your child go to bed last night? How many minutes did it take for him/her to fall asleep? How many times did your child wake up last night? On average how long were they awake each time? What time did your child wake up this morning? Is there anything else you would like to tell me about your child’s sleep last night?

The treatment phase (phase B) consisted of a pediatric aquatic exercise program that was consistent with a previously described aquatic program (Oriel et al., 2012). A resting heart rate (HR) was obtained from each participant prior to entering the pool, which was determined using a finger pulse oximeter. Each participant was then paired with one of the researchers or volunteers, which was kept as consistent as possible throughout the eight aquatic exercise sessions. If it was not possible to pair a child one-on-one, small groups with participants of similar level were formed. To maintain organization and consistency, one researcher remained out of the pool to lead the session.

The aquatic exercise program was 60 min in duration. The program began with a warm-up, which consisted of walking in a group circle clockwise and counterclockwise. After the warm-up, upper and lower extremity circuits were performed. Upper extremity exercises included punches, forward/backward arm circles, bicep curls, and shoulder abduction/adduction. Lower extremity exercises included hip abduction, marching in place, wall kicks, and rocket blasts off the wall. These circuits were followed by cardiovascular exercises, which included jumping jacks, jogging in place, and kick board relays. Next, the children participated in a game, which included red light–green light, keep away, or sharks and minnows. Every session ended with a free swim in which the participants were given the opportunity to play with toys (rubber ducks, balls, basketball hoop, boats, pool noodles, kick boards, and pool rings) and swim. The participants were continuously encouraged to remain active throughout the entire session, particularly the free swim. The cool-down was the same as the warm-up and continued as the participants got out of the pool one-by-one. As each participant exited the pool, his or her pulse was taken to obtain the exercise HR.

Data Analysis

Statistical analyses were performed using SAS statistical software, version 9. Descriptive statistics were utilized to analyze demographic data of all participants. A One-Sample Kolmogorov–Smirnov Test was used to determine if the data were normally distributed, given the small sample size. A one-way repeated measures analysis of variance (ANOVA) was used to determine if there were differences between the means of the treatment and control phases for each of the three variables (total sleep, night wakenings, and sleep latency). Post hoc testing, using Tukey’s Test, was performed if statistical significance was present to determine between which phases differences existed.

Results

CSHQ

According to the CSHQ, the participants in the study had an average score of 55.4, and ranged from 44 to 63. Of these factors, parents/guardians identified an average of 8 behaviors as problems for their child, with a range of 1 to 20 (out of a total possible 33).

One-Sample Kolmogorov–Smirnov Test

A One-Sample Kolmogorov–Smirnov Test was performed to determine if the data were normally distributed given the small sample size observed in this study. The distribution was found to be normal for total sleep, sleep latency, and night wakenings for all study phases.

Total Sleep

The one-way repeated measures ANOVA revealed that a statistically significant difference (p < .001) existed between treatment phases for total sleep. The means for each phase were as follows: A1 = 493.59 min (8.23 hr), B = 576.91 min (9.62 hr), and A2 = 549.05 min (9.15 hr). The Tukey post hoc test indicated the statistically significant differences were between A1 and B (p < .001) and A1 and A2 (p < .001). Figure 1 shows the average number of hours of sleep for each participant during each of the study phases.

Figure 1.

Average number of hours of sleep for each participant during each study phase.

Sleep Latency

The one-way repeated measures ANOVA revealed that a statistically significant difference (p < .001) existed between treatment phases for sleep latency. The means for each phase were as follows: A1 = 38.95 min, B = 21.76 min, and A2 = 25.91 min. The Tukey post hoc test indicated that the statistically significant differences were between A1 and B (p < .001) and A1 and A2 (p < .004). Figure 2 shows the average number of minutes it took each participant to fall asleep during each of the study phases.

Figure 2.

Average number of minutes to fall asleep for each participant during each study phase.

Night Wakenings

The one-way repeated measures ANOVA revealed that a statistically significant difference did not exist between treatment phases for night wakenings. The means for each phase were A1 = 0.99, B = 0.37, and A2 = 0.64.

Table 1 provides a summary of the statistical results (minimum, maximum, mean, standard deviation) for each variable throughout the study phases.

Table 1.

Summary of Results (N = 8).

Sleep parameter	Minimum	Maximum	M	SD
Total sleep (min)
A1	375.00	576.40	493.59	60.51
B^a	523.00	637.30	576.91	37.48
A2^b	443.80	635.30	549.05	54.67
Number of wakenings
A1	0.25	2.75	0.99	0.77
B	0.00	1.13	0.37	0.41
A2	0.00	1.80	0.64	0.64
Sleep latency (min)
A1^a	12.50	75.00	38.95	21.19
B	6.13	54.38	21.76	15.94
A2^b	10.00	60.00	25.91	17.78

Indicates a statistically significant difference A1 to B. ^bIndicates a statistically significant difference A1 to A2.

HR

The percentage of maximum HR that each participant was working at was calculated. The participants’ averages ranged from 58% to 71% with a group mean of 64%.

Discussion

This study examined whether children with ASD would fall asleep faster and would have longer sleep duration with fewer wakenings throughout the night following participation in an aquatic exercise session. The results suggested sleep latency decreased and sleep duration increased as a result of the aquatic exercise sessions.

Previous research has found that children with ASD are more likely to have sleep problems than TD children. The prevalence has been estimated to range from 66% (Souders et al., 2009) to 80% (Eggerding, 2010; Reynolds & Malow, 2011) in children with ASD. The results of the CSHQ support those findings as all of the participants in this study reported difficulty with sleep patterns. The mean score on the CSHQ was 55.4, ranging from 44 to 63, indicating that their sleep problems were substantial given that a score of 41 identifies the majority of children with sleep disturbances according to Owens, Spirito, and McGuinn. Parents/guardians also identified numerous components of the questionnaire as problematic. The mean score for identified specific issues was 8, ranging from 1 to 20, indicating that not only did the child exhibit this behavior but that the parent/guardian felt burdened by it. Issues identified by 50% of parents/guardians included that the child does not fall asleep within 20 min of going to bed and the child sleeps too little. Thirty-eight percent of parents/guardians reported the child needs the parent in the room to fall asleep, the child is not ready to go to bed at bedtime, the child is afraid of sleeping in the dark, the child is afraid of sleeping alone, the child does not sleep the same amount each day, the child is restless and moves a lot during sleep, and the child wakes up (at least) once during the night.

Furthermore, analysis of the demographic data revealed that two of the eight participants were taking anti-anxiety and/or depression prescription medication(s) at the time of the study. This observation is consistent with Richdale and Schreck’s (2009) finding that potential psychological causes of sleep dysfunction in children with ASD can be attributed to their increased anxiety and likelihood to develop depression. One of the participants was also receiving melatonin supplements to improve her sleep patterns. However, as the participant continued to exhibit sleep problems, evidenced by the CSHQ, it is plausible to conclude that her sleep symptoms did not dramatically improve through this method as the parents/guardians sought additional treatment options. Similarly, Malow et al. (2012) proposed that sleep duration and night wakenings did not improve with the use of melatonin, although they found sleep latency to decrease significantly.

The results of this study support the stated hypothesis in that the participants’ sleep latency would decrease during the treatment phase of this study. The participants fell asleep an average of 17.25 min faster in the treatment phase (B) than in the first control phase (A1) which was a significant finding. This change may be attributed to the fact that children with ASD are less active than their TD peers and were not likely to be engaging in any physical activity during the A1 phase. With the initiation of the treatment phase, the participants were involved in moderate-intensity exercise resulting in an increased HR which could have influenced their ability to fall asleep quicker at night as they were likely more tired. This is consistent with the findings of Brand et al. (2010) who found that athletes who exercised at high levels had shorter sleep latency than their peers.

There was also a significant difference in sleep latency between the first control phase (A1) and second control phase (A2). This is likely due to extraneous variables such as warmer weather and longer daylight allowing the participants to be more active in the evenings. In addition, parents may have observed the beneficial effects of the study and continued with exercise. There was no significant difference between B and A2, however on average, the participants did fall asleep 4.21 min faster during the aquatic exercise phase.

The results for sleep duration supported the hypothesis as participants slept for a longer duration during the treatment phase. The participants slept an average of 83.2 min longer in B than in A1. According to Sadeh, Gruber, and Raviv (2003) modest improvements in sleep duration may lead to improved neurobehavioral function. The improved sleep duration found was likely attributable to the participants’ increased activity leading to faster sleep onset and fewer wakenings resulting in longer sleep duration. Previous research has shown that when individuals exercise within 4 hr of bedtime they have an increase in sleep duration (Youngstedt, 2005). There was also a significant difference in sleep duration between A1 and A2. This may be due to improved bedtime routines maintained from the treatment phase. It was not expected to see carryover effects from the exercise phase into the second control phase, but it may be attributed to learned behavior from consistency during the 4 weeks of treatment rather than increased activity and HR. There was no significant difference between B and A2; however, the participants did sleep an average of 27.84 min more during the aquatic exercise phase.

The results for number of wakenings did not support the hypothesis that children with ASD would have fewer wakenings, although the mean number for the treatment phase was lower than either of the control phases. This was likely due to the low number of sleep wakenings for the group as a whole. However, for some participants there were more noticeable changes. For example, Participant #4 had an average of 2.8 wakenings during A1, 0.3 during B, and 1.1 during A2 (see Figure 3), representing that if wakenings are a problem for a child, participating in aquatic exercise can be beneficial. Similar findings were also described by Brand et al. (2010) as participants who exercised at higher levels had fewer night wakenings.

Figure 3.

Participant #4’s average number of wakenings in each phase of the study.

A person’s target HR for moderate-intensity physical activity should be between 50% and 70% of his or her maximum HR (Centers for Disease Control and Prevention, 2011). The estimated maximum age-related HR (220 − age = max HR) was calculated for each participant and then the percentage of the maximum HR was found using the post-exercise HR. The group mean HR was 64%, which was within the spectrum for moderate-intensity exercise.

A main strength of this study was the A-B-A design. It allowed the researchers to gather baseline data, implement the treatment, and then remove the experimental variable to see if the participants returned to baseline, which demonstrates that the changes were due to the treatment. A common problem of within-subjects design studies is that the behavior must be reversible to show that the treatment was effective. This problem was avoided as the sleep patterns did trend back toward baseline in A2. The A-B-A design also controlled for extraneous variables as it was not likely they would be present both before and after treatment.

The A-B-A design does present ethical issues, given that an effective intervention may be withdrawn. In this case, parents could have chosen to begin aquatic programming with their child at the completion of the study. Parents did inquire about such programming, although the number of children who actually began a community-based aquatics program was not tracked. The study design could also have been strengthened by adding an additional B phase, and should be considered in future studies.

The strict exclusion criteria were another strength of the study. By not including children with comorbid developmental disabilities, the researchers were able to see changes in a more homogeneous group. Another positive was the biweekly questions during the calls to the parents/guardians remained consistent due to the script that was developed. During the treatment phase, the exercises performed remained fairly consistent but that was not of as much importance as the participants’ activity levels. The researchers were able to keep the participants physically active at a moderate-intensity level as measured by their HR. Using HR to quantify the intensity at which the participants exercised was also a strength as it is a standard and acceptable outcome measure.

Limitations of this study included extraneous variables such as bedroom setup, season, and participants’ activity level. For example, one of the participants had a routine that he would wake up in the middle of the night and move to a different bed. The parent/guardian thought that this was due to the small size of his bed and not having enough space to move around so he would switch to the larger bed. At the initiation of the A2 phase, the parent/guardian moved the larger bed into his room and he stopped waking up in the middle of the night, hence fewer night wakenings.

In addition, researchers were unable to stop the children from engaging in aquatic exercise in the control phases. Researchers could encourage parents to avoid these activities, but had no control over what participants actually did.

Number of participants was another major limitation of this study. Although statistically significant results were obtained from eight children with ASD, these data would have been stronger with a larger group. Future studies should focus on more reliable methods to recruit participants. Some suggestions include communication through ASD support groups as well as starting the process early enough to increase parent/guardian awareness of the study.

Finally, the current study focused on the use of subjective measures for means to collect data. Parents/guardians were expected to monitor their child’s sleep as best as they could; however, it is unrealistic for them to know precisely what time their child fell asleep at night and awoke in the morning or the frequency of wakenings during the night. Future studies should consider usage of an actigraphy wristband to monitor sleep and movement throughout the night. This device would detect body motion, and make quantification of sleep habits more precise.

Future research related to the impact of aquatic exercise on sleep habits in children with ASD should consider social validity and treatment fidelity. It is important to determine how important such interventions are to families, and if such interventions can be reproduced in community-based settings.

The results of this study suggest that participation in an aquatic exercise program may lead to improved sleep habits in children with ASD. The participants in this study slept longer and fell asleep faster following aquatic exercise. Although the resultant impact of improved sleep was not explored, it is our hope that daytime behaviors and relationships with family were positively influenced. Overall, these findings provide a solid foundation for future research in this area.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Allik

Larsson

Smedje

(2006). Sleep patterns of school-age children with Asperger syndrome or high-functioning autism. Journal of Autism and Developmental Disorders, 36, 585–595.

Best

J. F.

Jones

J. G.

(1974). Movement therapy in the treatment of autistic children. Australian Occupational Therapy Journal, 21, 72–86.

Brand

Gerber

Beck

Hatzinger

Pühse

Holsboer-Trachsler

(2010). High exercise levels are related to favorable sleep patterns and psychological functioning in adolescents: A comparison of athletes and controls. Journal of Adolescent Health, 46, 133–141.

Case-Smith

Miller

(1999). Occupational therapy with children with pervasive developmental disorders. American Journal of Occupational Therapy, 53, 506–551.

Centers for Disease Control and Prevention. (2011, March 30). Target heart rate and estimated maximum heart rate. Retrieved from http://www.cdc.gov/physicalactivity/everyone/measuring/heartrate.html

Chambless

D. L.

Hollon

S. D.

(1998). Defining empirically supported therapies. Journal of Consulting and Clinical Psychology, 66, 7–18.

Eggerding

(2010, April 1). Put sleep difficulties to bed: Advice for parents of children with autism. Exceptional Parent Magazine, 22, 40.

Fragala-Pinkham

Haley

S. M.

O’Neil

M. E.

(2008). Group aquatic aerobic exercise for children with disabilities. Developmental Medicine & Child Neurology, 50, 822–827.

Fragala-Pinkham

Haley

S. M.

O’Neil

M. E.

(2010). Summative evaluation of a pilot aquatic exercise program for children with disabilities. Disability and Health Journal, 3, 162–170.

10.

Hoffman

Sweeney

Gilliam

Apodaca

Lopez-Wagner

Castillo

(2005). Sleep problems and symptomology in children with autism. Focus on Autism and Other Developmental Disabilities, 20, 194–200.

11.

Huettig

Darden-Melton

(2004). Acquisition of aquatic skills by children with autism. Palaestra, 20, 20–25, 45–46.

12.

Johnson

Malow

(2008). Assessment and pharmacologic treatment of sleep disturbance in autism. Child and Adolescent Psychiatric Clinics of North America, 17, 773–785.

13.

Malow

Adkins

McGrew

Wang

Goldman

Fawkes

Burnette

(2012). Melatonin for sleep in children with autism: A controlled trial examining dose, tolerability, and outcomes. Journal of Autism and Developmental Disorders, 42, 1729–1737.

14.

Oriel

Marchese

Shirk

Wagner

Young

Miller

(2012). The psychosocial benefits of an inclusive community-based aquatics program. Pediatric Physical Therapy, 24, 361–368.

15.

Owens

J. A.

Spirito

McGuinn

(2000). The Children’s Sleep Habits Questionnaire (CSHQ): Psychometric properties of a survey instrument for school-aged children. Sleep, 23, 1043–1051.

16.

Pan

C. Y.

(2010). Effects of water exercise swimming program on aquatic skills and social behaviors in children with autism spectrum disorders. SAGE Publications and The National Autistic Society, 14, 9–28.

17.

Portney

L. G.

Watkins

M. P.

(2000). Foundations of clinical research: Applications to practice (2nd ed.). Upper Saddle River, NJ: Prentice Hall.

18.

Reynolds

Malow

(2011). Sleep and autism spectrum disorders. Pediatric Clinics of North America, 58, 685–698.

19.

Richdale

Schreck

(2009). Sleep problems in autism spectrum disorders: Prevalence, nature, and possible biopsychosocial aetiologies. Sleep Medicine Reviews, 13, 403–411.

20.

Rogers

Hemmeter

M. L.

Wolery

(2010). Using a constant time delay procedure to teach foundational swimming skills to children with autism. Topics in Early Childhood Special Education, 30, 102–111.

21.

Sadeh

Gruber

Raviv

(2003). The effects of sleep restriction and extension on school-age children: What a difference an hour makes. Child Development, 74, 444–455.

22.

Souders

Mason

Valladares

Bucan

Levy

Mandell

. . .Pinto-Martin

(2009). Sleep behaviors and sleep quality in children with autism spectrum disorders. Sleep, 32, 1566–1578.

23.

Vonder Hulls

D. S.

Walker

L. K.

Powell

J. M.

(2006). Clinicians’ perceptions of the benefits of aquatic therapy for young children with autism: A preliminary study. Physical & Occupational Therapy in Pediatrics, 26, 13–22.

24.

Vriend

J. L.

Corkum

P. V.

Moon

E. C.

Smith

I. M.

(2011). Behavioral interventions for sleep problems in children with autism spectrum disorders: Current findings and future directions. Journal of Pediatric Psychology, 36(9), 1017–1029.

25.

Williams

P. G.

Sears

L. L.

Allard

(2006). Sleep problems in children with autism. Journal of Sleep Research, 13, 265–268.

26.

Wolfson

Lacks

Futterman

(1992). Effects of parent training on infant sleeping patterns, parents’ stress, and perceived parental competence. Journal of Consulting and Clinical Psychology, 60, 41–48.

27.

Yilmaz

Yanardag

Birkan

Bumin

(2004). Effects of swimming training on physical fitness and water orientation in autism. Pediatrics International, 46, 624–626.

28.

Youngstedt

S. D.

(2005). Effects of exercise on sleep. Clinics in Sports Medicine, 24, 355–365.