Play is an essential part of childhood. Play is the activity by which children explore their environment and learn how to deal effectively with their surroundings. Exploratory play may be difficult in an upright posture when balance control is poor, as is the case in infants with Down syndrome (DS).1 However, supramalleolar orthoses (SMOs) can improve balance in children with DS.2 Orthoses may improve the ability of children with DS to engage in upright exploratory play by providing foot and ankle stability. If the ability to control the base of support increases, children with DS may be better able to use their hands to interact with objects during upright play, rather than using them for stability purposes.
Hand use is vital to play and exploratory behavior. When children are able to sit independently, they are able to use their hands to explore the area and objects immediately around them. When children can support themselves in an upright position, they can explore a new visual perspective. Initially, though, the need to use their hands for support while upright may limit manual exploration. In addition, infants with DS may not quickly learn to perceive the various uses of objects available within reach.3,4 Children with DS may not understand that they can let go of a support surface and explore with their hands. There are also organismic constraints that limit infants' ability to let go of a support surface to explore with their hands.5 These include impairments such as decreased muscle tone, ligamentous laxity, and slow reaction times.6,7 These constraints may lead to decreased postural control in children with DS, which affects their ability to play in an upright position and requires them to use their hands to support their body, when upright, rather than using them to explore.
In children with DS, abnormal postural control may contribute to atypical exploratory play behavior. Poor postural development leads to a decreased set of movement possibilities within a given environment.8 Although children with DS develop upright balance, they have delayed and atypical postural control when compared to children with typical development (TD). Kubo and Ulrich1 found that new walkers with DS take longer to learn to couple trunk oscillations in the sagittal and frontal planes during gait. In addition, Shumway-Cook and Woollacott7 hypothesize that children with DS not only are delayed but also show a different developmental trend in their postural development when compared to children with TD. They found that children with DS between the ages of 1 and 3 years have sway responses that are slow, poorly organized, and inconsistent compared to the consistent sway response seen in children with TD in the same age range. This atypical postural response may lead to different solutions for maintaining upright position or loss of balance and may influence a child's ability to explore multiple environmental contexts.
An intervention that may help children with DS maintain balance is lower extremity orthotic use. Supramalleolar orthoses are external devices that are placed in shoes to provide support to the foot and ankle. They help prevent calcaneal eversion by aligning the subtalar joint in a neutral position, thus improving the bony alignment of the foot and ankle and influencing postural and gait characteristics.9 Supramalleolar orthoses lead to improved postural stability in children with DS. Martin2 studied SMO use in children with DS and found that the children, aged 3 years 6 months through 8 years, showed immediate and long-term improvements in postural stability, as measured by standardized test scores, when they used flexible SMOs. The orthoses provided foot and ankle stability that led to improved balance. These results are limited to older children, but infants with DS may also benefit from the stability that SMOs provide. It should be noted, however, that the use of orthoses in infancy leads to decreased motor skill development shortly after walking onset.10 Despite this, the improved balance and stability gained from SMOs may lead to an increase in hands-on exploratory play that may justify their use.
Because children with DS have decreased postural control, they may maintain upper extremity support in standing for a longer period than their peers with TD. Orthoses may help improve their lower extremity stability leading to less need for upper extremity support. The purpose of this study is to determine whether SMO use during an upright play situation in children with DS who are not yet walking contributes to a decrease in upper extremity support and allows for an increase in hands-on exploratory play.
Twenty-two children with DS were enrolled in this study. Over time, 5 children dropped out of the study (1 had emerging medical problems, 1 did not tolerate the treadmill, and 3 received orthoses prior to the end of the study making them ineligible). The study was limited to infants with trisomy 21 and onset of participation occurred when infants were able to pull to stand using furniture but not yet able to cruise. Children were excluded from this study if they had a history of other developmental disabilities, uncorrected visual or hearing impairment, previous orthotic use, orthotic use other than the ones provided for this study, or were unable to tolerate the orthoses. The 22 children were randomly assigned into a control (n = 11) or an experimental group (n = 11). As a result of dropouts, the final distribution of subjects was 7 in the control group and 10 in the experimental group. Researchers measured the children in the experimental group for SMOs (SureStep, South Bend, Indiana) on their first visit and the children received them before the second visit. Researchers asked the families to put the orthoses on their children for 8 h/d and keep a log of the wearing time in a notebook. The children in the control group were measured for SMOs when they could take 3 independent steps and received the SMOs in 1 to 2 weeks. In addition to the orthoses, children also received treadmill training (TT) for 8 min/d, 5 d/wk while they participated in the study.10 Each child participated in this study from the time the child could pull to stand until 3 steps were accomplished independently. A parent of each child provided written informed consent prior to study entry. The institutional review board at the University of Michigan approved this study.
Researchers visited families in the study monthly. Play sessions were recorded at every other visit beginning at visit 1. Upright play activity was recorded with the following setup: A Leap Frog Learning Table (Leap Frog Enterprises, Emeryville, CA) was placed next to the family's couch. A mirror was placed behind the table and couch (opposite the camera) at a slight angle to allow researchers to view the side of the child that faced away from the video camera. A digital video camera was placed so that the infant's legs and trunk were within the frame. The video camera recorded at 30 Hz. We placed each child in an upright position between the table and couch, facing the table. The 20 minutes of recorded playtime began when the primary researcher took her hands off of the child. Researchers and family members sat on the other side of the table and encouraged the child to stand and play. If the child sat or fell down, the parent and researchers encouraged the child to pull to stand at either the table or the couch but did not assist the child in standing up. A child who moved out of view of the camera was placed in the sitting position between the table and the couch and then encouraged to pull to stand using activities on the table or toys that the child liked. During the play sessions, children in the experimental group wore SMOs and shoes, and children in the control group wore only shoes.
Each recorded 20-minute session was behavior coded in the laboratory. Behavior coders determined the amount of time spent in the upright position and, when in the upright position, how children supported their weight and played. The length of these bouts in the upright position was recorded in seconds via the time code embedded in the digital videotape. Coders counted only those behaviors that comprised 30 consecutive frames (eg, 1 second) or longer. When children were in an upright position, coders determined whether the trunk was leaning on the support surface or not and whether 0, 1, or 2 hands were used for support. Coders considered the children to be in the upright position when their hips were no longer moving in an upward motion. The children were no longer considered upright when a surface of the body, other than the feet, touched the ground or when another person touched them. Following contact, the child could be considered upright again. For each behavior, coders recorded time of onset and end, from which durations were calculated.
Seven coders participated in the upright behavior coding described earlier. They all coded a random selection of play sessions and showed high interrater reliability; intraclass correlation coefficient (ICC) (3,1) = .998 for time in leaning, ICC (3,1) = .990 for time in 2-handed support, and ICC (3,1) = .969 for time in 1-handed support).
The dependent measure in the analyses was the duration of 2-, 1-, or 0-hand support during the 20-minute play sessions. Linear mixed models were used to analyze the data. This statistical analysis allowed us to determine group differences and developmental changes over time and develop predictions on the basis of the data. First, we analyzed the duration of hand support as a percentage of total time in the upright position. To gain a greater understanding of the effect of leaning on hand support, we separately analyzed the duration of hand support while leaning and the duration of hand support while not leaning. The models included a term for the linear effects of development (changes in hand support over the course of the study), the quadratic effects of development, group effect, and both possible group by development (linear or quadratic) interactions. The linear effect of development showed changes that occurred in a progression from the beginning of the study to the end of the study, whereas the quadratic effect of development indicated developmental change in shorter rapid bursts. Because participation in the study was based on developmental skills, each child was in the study for a varying length of time. Because of this variability across subjects, the statistical models allowed time to vary randomly by subject. The alpha value was set at 0.05 for a statistical difference and at 0.10 for a statistical trend.
Participant Characteristics at Study Onset
The participants' characteristics at the study onset are reported in Table 1. The control group had a longer length at birth than the experimental group. However, length was not different at study onset. The experimental group wore the SMOs for 6.25 (±4) h/d. There were no group differences for time in the upright position and leaning at the study onset (Table 2).
Duration of Hand Support as a Percentage of Total Time in the Upright Position
Hand use during leaning as percentage of total time spent in the upright position
The average percentage of the duration of 2-hand support while leaning on the table was 32.50% (±27.69) for the experimental group and 26.90% (±18.68) for the control group. The predicted values for 2-hand support are shown in Figure 1. As with all of the figures, the predicted values are based on each individual data point. The predictions are less stable on the later visits because they were based on fewer observations (some children did not require that many visits). The prediction lines, though helpful, do not always clearly show the significant differences between the groups. Over the course of the intervention, a significant linear decrease was found in the duration of 2-hand support (P = .05). No significant group effect was found. There were no significant group-by-development interactions.
There were no significant effects or interactions during 1-hand and 0-hand support in leaning.
Hand use during nonleaning as percentage of total time spent in the upright position
There were no significant effects or interactions during 2-hand support on the table without leaning.
The average percentage of the duration of 1 hand on the play surface without leaning was 19.13% (±19.99) for the experimental group and 13.84% (±14.43) for the control group. The predicted values for 1-hand support are shown in Figure 1. There was a trend toward a linear increase with development (P = .07). There were no significant group effects or group-by-development interactions.
The average percentage of the duration of 0 hands on the play surface without leaning (standing) was 1.48% (±5.20) for the experimental group and 0.40% (±1.22) for the control group. The predicted values for 0-hand support are shown in Figure 1. There was a significant quadratic increase over time (P = .03). There were no statistical differences between the groups and no significant group-by-development interactions.
Duration of Hand Support While Leaning
The average percentage of the duration of 2 hands on the table while leaning on the support surface was 47.59% (±27.65) for the experimental group and 41.87% (±23.76) for the control group. The predicted durations for 2-hand support while leaning over developmental time are shown in Figure 2. There was a trend toward a group difference indicating that the experimental group may have spent more time with 2 hands on the support surface while leaning (P = .10). Over the course of the intervention, the percentage of the duration of 2-hand support while leaning showed a significant linear decrease for both groups (P = .01). There was a trend toward a quadratic change over time (P = .08). There was a significant linear group-by-development interaction (P = .05) and a trend toward a significant group by development (quadratic) interaction (P = .07) indicating that the experimental group showed a greater decrease in the duration of time spent with 2 hands on the table while leaning.
The average percentage of the duration of 1 hand on the play surface while leaning was 34.60% (±21.92) for the experimental group and 42.57% (±15.93) for the control group. Over developmental time, the predicted durations for 1-hand support while leaning are shown in Figure 2. This shows a trend in favor of the control group (P = .06). Over the course of the intervention, the percentage of the duration of 1 hand on the play surface while also leaning significantly increased linearly (P < .01) and quadratically (P = .01). In addition, there were trends toward group-by-development interactions both linearly and quadratically (P = .09, P = .09, respectively).
There were no significant effects or interactions during 0-hand support on the table while leaning.
Duration of Hand Support While Not Leaning
The average percentage of the duration of 2-hand support while not leaning was 43.70% (±29.01) for the experimental group and 54.36% (±28.14) for the control group over the course of the intervention. The predicted durations for 2-hand support while not leaning over the course of the study are shown in Figure 3. The control group spent significantly more time with 2 hands on the support surface when not leaning than the experimental group (P = .01). There were no development effects or group-by-development interactions.
The average percentage of the duration of time spent with 1-hand on the table while not leaning was 38.28% (±26.47) for the experimental group and 40.97% (±26.33) for the control group over the course of the intervention. The predicted durations for 1-hand support while not leaning over developmental time are shown in Figure 3. There was a significant linear increase with development (P = .03). There were no significant group effects or group-by-development interactions.
The average percentage of the duration of time spent with 0 hands on the table while not leaning (standing) was 2.43% (±8.57) for the experimental group and 1.82% (±5.18) for the control group over the course of the intervention. The predicted durations for standing over developmental time are shown in Figure 3. There was a significant quadratic effect of time (P = .04) indicating a sudden shift to standing independently. There were no significant group effects or group-by-development interactions.
The purpose of this study was to determine whether SMO use in children with DS who were not yet walking improved upright balance during a play situation leading to decreased upper extremity support in the upright position. Although it was hypothesized that orthotic use would decrease upper extremity support, we found no positive effect of orthotic use on this behavior. In fact, children with DS who were not yet walking and who did not use SMOs showed a trend toward a larger percentage of the duration in leaning with 1 hand and no hands on the play surface than did children who were not yet walking and who wore SMOs. Although exploratory play is an important part of development and all the infants in this study participated in hands-on exploratory play to some extent, the need to remain stable in the upright position appeared to trump the ability to free up their hands for play and exploration.
Total Hand Use While in the Upright Position
The orthoses do not seem to affect hand use while in the upright position. There were no group differences in 2-, 1-, and 0-hand support, whether leaning or not leaning as a percentage of duration in the upright position (Figure 1). Also, no group-by-development interactions were found. The passage of developmental time appears to have the greatest effect on hand use while in the upright position. The duration in 2-handed support while leaning as a percentage of time in the upright position significantly decreased during the study. In addition, duration in standing increased rapidly at the end of the study just before walking onset. Both groups spent the majority of their time in the upright position in a leaning posture with 2 hands on the support surface. As the infants developed, they were able to free up 1 or 2 hands both in leaning and nonleaning to explore and play.
Hand Use During Nonleaning and Leaning Portions of the Upright Position
Hand use was also analyzed separately during nonleaning and leaning portions of the period in the upright position. During the nonleaning portion, the control group spent a significantly larger percentage of time with 2 hands on the support surface. However, during the leaning portion, the statistical trend suggests that the control group may spend a longer duration in 1-hand support. There was also a significant group-by-development interaction indicating that the groups were developing in a different manner (Figure 2). The children in the control group appeared to use their upper extremities less for support but also appeared to lean more to explore. When infants in both groups moved their trunks away from the support surface, they spent more time with both hands in a supported, stabilizing position.
Although the control group spent a larger percentage of the duration of 2-hand support during nonleaning, there was no overall difference between the groups over the course of the study in 2-handed support while leaning. However, all infants decreased the time that they spent with 2 hands on the support surface in leaning and nonleaning. The 2-handed leaning position is very stable; it was the dominant posture at the beginning of the study, when infants were just learning how to stand. At that point, they used their trunk and both hands to stabilize the body. As children learn to control their bodies better, they use their trunks to support the body in a more adaptive manner, first by leaning against the support surface and then by supporting the trunk in the upright position. This led to the observed decrease in 2-hand support as the study progressed and allowed infants to begin playing and exploring with their hands.
Although no difference was found between the groups in the percentage of time in 1-hand support during nonleaning, the control group performed better than the experimental group over the course of the study in the percentage of time spent in 1-hand support while leaning (Figure 3). Over developmental time, all the infants in the study increased 1-hand support in both leaning and nonleaning. In leaning, the results showed that although there was a nonlinear effect over the time of the study, there was a quadratic effect. This is reflected in the “inverted U”–shaped pattern in the statistical predictions (Figure 3). One-hand support peaked at visits 5 and 7 and then began to decline as 0-hand support began to increase. The trend toward significant group-by-development interactions suggests that these 2 groups are developing in a slightly different manner. The control group began with a larger percentage of development in 1-handed support while leaning and peaked on visit 5, whereas the experimental group built up to a peak at visit 7.
There was no difference between the groups for 0-handed support during the period in leaning and nonleaning. There was a significant quadratic effect of development on 0-hand support during the period of nonleaning. This indicates an abrupt appearance of unsupported stance just before walking onset (Figure 4). The use of 2 hands for play (0 hands for support) while leaning signals a switch from trunk leaning as necessary to maintain the upright position to trunk leaning as an adaptive skill that allows 2-handed exploration. Although there were no group differences over the course of the study, the predicted means show that this shift occurred earlier in the control group. The control group began to play and explore using 2 hands while leaning at the 5-month visit, whereas the experimental group achieved this at the 7-month visit (Figure 3). This increased amount of time to explore may have given the children in the control group the opportunity to develop a deeper understanding of and experience with their surrounding.
When looking at how hand support changes over developmental time, it is important to note that both groups spent the longest time in 2-hand support while leaning in this play context. However, the control group appeared to have an advantage when not in 2-hand support during leaning. At the point when 1-hand support during leaning appeared higher in the experimental group, the control group was more likely to be spending more time leaning with no hands on the support surface. Interestingly, the predicted percentages indicate that the experimental group steadily decreased the time that they spend leaning, whereas the control group remained somewhat constant. However, although both groups showed an increase in the amount of 1- and 0-handed support, the control group appeared to spend more time in 1- and 0-handed support though they tended to be leaning more. Perhaps the orthoses provided enough stability for the infants to begin exploring their trunk stability while exploring and playing with toys in their environment less. As the infants' ability to control their legs and trunk increased, their ability to explore with 1 hand while not leaning also increased. Overall, the infants who did not wear orthoses spent a longer time using their hands to explore because they relied more on their trunk for support.
Engaging in and learning about the world is implicit in the act of playing. In addition, hand use is a key component to exploratory behavior. The children in this study displayed a progression from limited ability to free their hands from a support role to decreased hand support while in the upright position. Although the use of SMOs in infants with DS seems to improve stability enough to decrease the time spent leaning in this prewalking population, it inadvertently may further limit the ability of infants with DS to use their hands to explore while in the upright position. Because the infants are actively exploring their trunk movements, they are relying on upper extremity support for stability. This increased stability comes at the cost of hands-on exploration.
There are a few limitations that should be kept in mind when considering the results and conclusions of this study. This study did not have an orthoses-only group. Both groups received TT. It is possible that TT sufficiently improves postural control in infants with DS to allow for less support in the upright position. Although orthoses, in addition to TT, appear to lead to a different developmental pattern than TT alone, orthoses alone may lead to a yet different developmental pattern. In future studies, an orthoses-only group should be included. This study did not consider individual differences in terms of orthotic type. Less supportive orthoses may be more beneficial to some children with DS and may have fewer developmental implications. Future studies are needed to determine how to choose the type of orthosis and whether other forms of orthoses may be more beneficial. In addition, the sample size in this study is small due to a high number of dropouts in the control group whose physical therapists recommended orthotic use before the child could walk. This resulted in lower statistical power. Further studies should include a larger sample size.
Orthoses are often recommended for children with DS who are not yet walking despite little to no empirical evidence to support the practice. The results of this study and recently published data on the same cohort of children suggesting that early use of orthoses leads to decreased motor skill development shortly after walking onset10 bring this practice under question. Although the use of SMOs may improve upright stability, it does not appear to decrease upper extremity support in children with DS who are not yet walking. In fact, the early use of SMOs may inhibit this behavior and could have a negative effect the children's ability to manually explore their environment. On the basis of this information, health care professionals may want to refrain from using SMOs in children with DS who are not yet walking.
The authors thank the children who participated in this study and their families.
© 2011 Lippincott Williams & Wilkins, Inc.