The effects of a contoured foam seat on postural alignment and upper-extremity function in infants with neuromotor impairments. (Research Report).

Children with neuromotor impairments frequently demonstrate difficulty balancing and stabilizing their body relative to the supporting surface. Adaptive seating, the customized prescription and application of sitting support devices based on therapeutic principles, (1) is frequently used by physical therapists and occupational therapists to improve the posture and stability of children with neuromotor impairments. Benefits of adaptive seating include improved postural alignment (2) and facilitation of upper-extremity function. (3) In addition, McEwen (4) found that the rate of adults' initiations of communication was higher when students with multiple disabilities were positioned in a wheelchair as opposed to a side-lyer or a mat on the floor.

Two of the motor control theories that have been applied to adaptive seating are neuromaturational theory (5) and dynamical systems theory. (6) According to neuromaturational theory, adaptive seating is aimed at decreasing the influence of primitive reflexes, normalizing the resistance of muscles to stretching, and providing proximal stability to promote distal mobility and function. Dynamical systems theory proposes that motor behaviors emerge from the interaction of a variety of neural, musculoskeletal, sensory, adaptive, and anticipatory mechanisms in task-specific contexts. (6) Adaptive seating from the dynamical systems perspective is aimed at influencing the starting conditions of the movement and at limiting the degrees of freedom or body segments free to move. (7) Adaptive seating devices influence the starting conditions of the movement by changing the alignment of body segments relative to each other and to the line of gravity. The number of body segments free to move also can be constrained by components of an adaptive seating device, such as the footrest on a wheelchair.

Rochat and Goubet (8) demonstrated the importance of a stable base of support as a foundation for accuracy in reaching in infants who are developing typically. When increased levels of pelvic support were provided using a pneumatic seat cushion, the infants modified their reaching and grasping patterns to resemble those of older infants. These modified patterns were characterized by coordination between reaching of the hand and forward leaning of the trunk and by a transition from bimanual to lateralized reaching. The belief that stability of proximal body parts (spine, shoulders, and pelvis) is a prerequisite for distal control and functional movement is frequently referred to in literature on treatment of children with cerebral palsy, (9) as well as in the adaptive seating literature. (10) Research supporting this hypothesis, however, is limited. Research on infants who are typically developing suggests that although proximal control is related to distal motor function, the relationship is not necessarily one of cause and effect. (11,12)

Findings from studies examining the effects of adaptive seating systems on a variety of dependent variables have been mixed. Improvements in postural alignment as a result of adaptive seating devices have been reported, (2,3,13,14) and some authors (3,15) have suggested that forward-tipped and upright seating systems improve performance of functional upper-extremity tasks for children with neuromotor impairments. Other researchers, however, reported no change in upper-extremity performance with independent variables such as the angle of hip flexion, (16,17) seat inclination, (18) anterior trunk supports, (19) and adaptive seating systems. (20,21) Benefits of adaptive seating reported by parents include increased ease in feeding, transporting, and caring for their children. (14,20,22)

We contend that findings from the adaptive seating literature are limited in their application to clinical practice for several reasons. Roxborough (1) indicated that the quality of the studies varies widely, with many studies demonstrating methodological problems relating to subject selection, lack of controlled conditions, choice of outcome measures, and potential confounds from other environmental variables. With few exceptions, (13,14,22) efficacy studies on adaptive seating systems have been limited to children 2 years of age and older. Feedback from caregivers regarding functional outcomes in natural settings related to the use of adaptive seating also has been limited. Finally, instrumentation in some studies (15,18,23) has required costly equipment or techniques such as electromyography that may limit direct replication in many clinical settings. As a result, additional studies examining adaptive seating devices are needed. (1,24,25)

Despite limited evidence, we have observed that pediatric therapists frequently use adaptive seating devices and anecdotally report improvements in posture, head control, and upper-extremity function. One type of adaptive seating device currently used by therapists is a contoured seat carved from medium-density foam (Fig. 1). This contoured foam seat (CFS) is intended to improve pelvic alignment, increase postural stability, and improve somatosensory feedback for children with a variety of impairments that limit postural control. Advantages of the CFS versus other types of adaptive seating devices include its cost-effectiveness, transportability, and ease of fabrication and modification.

[FIGURE 1 OMITTED]

The purpose of this study was to evaluate the effects of a CFS on postural alignment and functional upper-extremity activity for infants with neuromotor impairments. Feedback from parents following use of the CFS at home also was obtained. The selection of dependent variables was based on a review of the literature and input from clinicians who had used the CFS with a variety of children. The research questions were:

1. What are the effects of a CFS on postural alignment for infants with neuromotor impairments?

2. What are the effects of a CFS on the infants' ability to engage with toys?

3. How do parents perceive the use and effects of a CFS when used at home?

Method

Subjects

A convenience sample of 4 infants, ages 9 to 18 months (corrected for prematurity), was recruited by contacting therapists from 3 community-based early intervention centers. Including 4 subjects allowed replication of the study across infants with varied ages, diagnoses, and ethnic backgrounds. All subjects met the following inclusion criteria: (1) were receiving physical therapy or occupational therapy services, (2) did not have fixed hip deformities or evidence of visual impairments, (3) were recommended as appropriate candidates for a CFS by their treating therapists, (4) demonstrated the ability to reach out and grasp toys presented at midline with either hand when held in a supported sitting position on an adult's lap, (5) were unable to sit independently, and (6) demonstrated a level of sitting of 3 or 4 as assessed on the Level of Sitting Scale. (7) Subject characteristics and descriptors for the levels of sitting indicating the amount of support required for them to maintain a bench sitting position are provided in the Table. With the exception of subject 3, none of the infants had sat in a highchair or had used a CFS. Prior to enrollment, informed consent was obtained from all subjects' parents. Subjects' rights were protected following study approval by the Institutional Review Board at Children's Hospital and Regional Medical Center, Seattle, Wash.

Research Design

We used both quantitative and qualitative research methods. A time-series, alternating-treatments design with a baseline was used to evaluate the effects of a CFS on postural alignment and engagement with toys. (26) When rating engagement with toys, raters recorded 2 types of data: (1) the number of hands in contact with a toy and (2) the number of hands or forearms in contact with the highchair tray. We reasoned that bimanual play may be compromised in infants who require arm support to stabilize themselves when sitting. Therefore, we hypothesized that the ability of the infants to free their hands from the support of the highchair tray reflected improved postural control and that this would increase the infants' repertoire of interactions with toys.

Data were collected under 3 seating conditions: (1) a regular highchair (Graco model 3306L *), (2) a regular highchair with a thin (0.32-cm [1/8 in]) foam seat liner, and (3) a regular highchair with a CFS insert. During the baseline period (phase 1), for each infant, data were collected for two 5-minute periods with the infant seated in the regular highchair. Our original intent was to collect baseline data for 3 days. However, because of the high variability in subject 1's data for midline positioning, we decided to extend his baseline period to 4 days. In addition, on one randomly determined occasion during the last 3 sessions of phase 2, we used the regular highchair. This was done to provide data on the potential influence of maturation. During the intervention period (phase 2), data for 2 conditions (foam liner and CFS) were collected daily for 8 days. Within each day, these conditions were randomly applied as determined by a coin toss.

The foam liner provided a nonslip surface in contrast to the vinyl surface of the highchair and was included to control for the effects of the texture of the seating surface. All infants used the same highchair during data collection. Subjects served as their own controls, and we believe the relatively rapid comparison of the alternative treatments provided control for maturation and behavioral state fluctuations. Qualitative methods also were used, with the infants' mothers participating in semistructured interviews following 4 weeks of CFS use at home.

For the purposes of our study, postural alignment was defined as the arrangement of upper body segments over the pelvis to allow midline positioning in the frontal plane while sitting. Measurement of postural alignment relative to midline was chosen because midline is the preferred position in the seating literature (9) as well as a recommended position of function for children with neuromotor impairments, (10) Postural alignment was rated from videotapes using anatomical markers and visual guides placed on the highchair back. Engagement with toys was defined as manipulation of a toy with either or both hands. Examples of manipulation included instances when an infant was playing with a toy while resting his or her hands or forearms on the highchair tray, banging a toy on the highchair tray, and bringing a toy to the mouth. Because tray height may have an influence on an infant's ability to manipulate toys, the highchair tray height relative to each infant's body was held constant across conditions.

Fabrication and Measurement Procedures

A customized CFS was fabricated for each infant by a pediatric physical therapist who designs CFSs and has used them for 12 years. The infant was held in a supported sitting position on a block of medium-density foam that was 10.2 cm (4 in) in depth and approximately 38.1 cm (15 in) square. A tracing was made with a marker around the infant's hips and thighs, and an electric knife was used to carve out the depressions of the seat. For carving the tracing around the hips and thighs, the knife was held at an approximately 45-degree angle to create a sloped, nested effect around the pelvis. The depression was shaped in an effort to keep the pelvis in neutral relative to the highchair seat. The area for the hips was slightly deeper than the area for the thighs, placing the infant in approximately 95 degrees of hip flexion as determined visually by 2 experienced pediatric therapists. In all seats, a foam pommel held the legs in abduction and provided additional stability. Lateral trunk support was provided by adding 10.2-cm foam blocks to both sides of the CFS. The foam side supports extended to just below the level of the highchair tray so that they were hidden from the raters' view.

Prior to data collection, the infant's acromial process on each shoulder was marked with an indelible marker. The distance between the 2 points was measured, and this distance was defined as the infant's trunk width. A vertical dowel was secured to the back of the chair to mark the midline of the chair. The highchair back was divided into 5 sections marked with colored tape. The middle section of the highchair back (left untaped) was referred to as "midline," and the width of this section equaled the infant's trunk width plus 5.08 cm (2 in) on both sides. The other 4 sections were marked on the sides of the highchair back with 3.8-cm-wide (1.5-in-wide) colored tape. During videotaping, adhesive circular markers were placed over the marks on the infant's acromial processes, and raters scored the position of the markers relative to the 5 sections on the highchair back (Fig. 2). When both of the acromial process markers were in the middle section, a score of 0 was given. If the infant moved so that a marker was in front of one of the side sections, a score of 1 or 2 was given, depending on the deviation from midline. Deviation to the left or right also was noted. A fabric drape was hung from the outer lip of the highchair tray so that raters were unaware of the seating intervention. For each infant, 6 toys that we considered visually appealing were selected based on parental input, the infant's developmental level, and potential for bimanual manipulation. A sequence of toy use was determined using simple random sampling to ensure that the infants had an equal chance to play with each toy.

[FIGURE 2 OMITTED]

Data Collection and Videotaping Procedures

All data collection sessions were conducted at approximately the same time of day, with the highchair tray and seatbelt always in place. If, at any time during phase 1 or 2, an infant appeared uncomfortable as evidenced by fussing or crying, the infant was repositioned. For each session, the first author kept a written log on the infant's behavioral state, number of repositionings, order of seating interventions and toy presentations, and general observations regarding the infant's posture and reaching skills. Infants were videotaped from an anterior view with a camera (Sony model CCD-F33 ([dagger])) positioned approximately 3 m (10 ft) in front of the highchair. All intervals during videotaping were timed by the first author.

Data were collected during two 5-minute periods on each day of data collection. During the baseline period, data were collected in the highchair condition for both 5-minute periods. During the intervention period, the foam liner and the CFS were alternated randomly.

We attempted to keep procedures consistent throughout the study. Each infant was given a 2-minute free play period in the highchair to accommodate to the seating condition. The first author then positioned the infant in midline by aligning the infant's nose with the vertical dowel.

Videotaping of the first 5-minute observation period began immediately following presentation of a toy placed at midline. If an infant did not visually attend to a toy or touch it within 20 seconds, the toy was replaced with the next toy in the randomly ordered sequence. If the infant dropped the toy on the floor, a new toy was presented. After the first 5-minute data collection period, the infant was removed from the highchair for approximately 2 minutes and then seated in the highchair for a second 2-minute accommodation period. A toy was again presented at midline on the tray, and videotaping of the second 5-minute data collection period began.

After all videotaping sessions were completed, each mother was asked to use the CFS in her infant's highchair for 4 weeks at home. During this time, mothers were contacted by the first author to answer questions or address any problems that might have arisen from use of the CFS. None of the mothers reported difficulties. After 4 weeks, a semistructured interview was conducted with each mother and audiotaped. Using 8 open-ended questions, each mother was asked to comment on positive and negative effects of the CFS relative to her infant's positioning, play, and behavior. Frequency of use and effects of the CFS on caregiving responsibilities also were studied.

Videotape Rating Procedures

Rating of videotapes occurred throughout data collection and was performed by 3 volunteer physical therapists. Prior to data collection and following a training session, each rater achieved a minimum level of 85% agreement with the first author for each of the 3 dependent measures. Interrater agreement for postural alignment and both measures of engagement with toys was examined using the point-by-point percentage of agreement method. (27) Scores were compared during each 10-second interval and needed to be identical for both raters to be considered an agreement.

Momentary time sampling, involving the recording of behaviors at specific moments in time, was used to score each 5-minute videotape segment. (27) With this technique, raters were cued by an audiotape to record data every 10 seconds, yielding 30 data points for each segment. Raters viewed each 5-minute videotape twice, first to record postural alignment and second to rate engagement with toys. Postural alignment ratings were based on the position of the anatomical markers relative to the 5 marked areas of the highchair back. Raters were instructed to score an "omit" for any interval where the anatomical markers were obscured by a toy or a body part.

Data Analysis

Data for the variables of postural alignment and engagement with toys were analyzed separately for each infant, with all raw data initially prepared in table form. Because there were minimal data for deviation into the colored sections closest to midline for 3 infants, data for both sections of lateral deviation from midline were combined. Postural alignment data were presented as percentage of intervals in midline. Data for engagement with toys were presented as the percentage of intervals with 2 hands on a toy and as the percentage of intervals when the infant had no hands or forearms on the tray while manipulating a toy. All data were graphed and analyzed visually for level and trend differences across treatments.

We also applied 2-tailed randomization tests to all intervention data to evaluate the statistical reliability of observed differences. Randomization tests essentially develop a distribution of all possible study outcomes, based on the actual data obtained, and then compare the actual outcome with those possible outcomes to determine the statistical reliability of findings. (28) We chose randomization tests because traditional inferential statistics are generally not recommended when data are collected repeatedly from a single individual. Such data often display a serial dependency in which earlier data can predict later data, and most traditional statistics require that each data point be statistically independent of every other data point. The tests were applied to the randomized trial data during the intervention, excluding the data for the one day when the regular highchair was used.

Semistructured interviews with parents were audiotaped and transcribed. All transcripts were reviewed by the first author while listening to the audiotapes to check accuracy. Analysis of interview data was done using the constant-comparative method as described by Glaser and Strauss (29) to establish coding categories. Interview data were reduced to text segments that were analyzed and sorted by the first author. Using this method, the first text segment became the first entry into the first coding category. Subsequent text segments were compared with the one existing category and judged to fit or to be the first member of a second coding category. This process was repeated until all text segments were sorted into major coding categories. Text segments were then reviewed and assigned subcodes, using the same method. as above.

Reliability

Interrater reliability sessions between the first author and each rater were conducted once during the baseline phase and twice during the intervention phase for each subject. Point-by-point percentage of agreement for all subjects ranged from .87 to 1.00 (median=.97) for postural alignment, from .87 to 1.00 (median=.93) for number of hands on toy, and from .83 to .97 (median=.87) for number of hands in contact with the highchair tray. In an effort to determine whether the same procedures were consistently used, (30) checks were conducted during 12 of the 45 data collection sessions using a checklist of the 21 procedures the first author was to perform. There was 100% replicability for 44 sessions, and 95% replicability for 1 session.

Using the codes established by the first author, a second rater independently coded 43 text segments (approximately 90% of the coded data) into the 5 major codes. This was done in an effort to ensure consistency in coding. A coding segment was considered an agreement only in cases where both raters assigned a single, identical code. Interrater agreement using the point-by-point method was .93.

Results

Postural Alignment

Figure 3 shows the results for postural alignment for all infants. In the baseline phases, data were collected during two 5-minute periods each day of data collection. During the second 5-minute period in session 2 (baseline), subject 1 maintained a midline position for 97% of the intervals by rounding his upper body forward and leaning on the highchair with both elbows. During the majority of sessions, however, he maintained a midline position during 30% of the intervals or less. At the beginning of the intervention phase, there was a change in level (ie, change in the magnitude of the data) with the introduction of the CFS. Throughout the intervention phase, consistent differences in level were maintained, with no overlapping data points across treatments. The randomization test comparing the CFS and foam liner in the intervention phase resulted in a probability value of .008.

[FIGURE 3 OMITTED]

Baseline data for subject 2 were fairly stable, with no marked variability or trends. Due to the subject's discomfort after quickly slumping to his left side, however, he required 6 repositionings during session 1 and 4 repositionings during session 2. With the introduction of the CFS during session 4, there was an immediate change in level, with the subject maintaining a midline position during 100% of intervals on all but one day. Data when using the foam liner were variable, with a higher mean percentage of intervals in midline as compared with baseline (49% versus 28%). Throughout the intervention phase, the infant did not require repositioning when using the CFS and required only one repositioning when using the foam liner. The lowest data point occurred during session 10, during which the infant sat in the regular highchair and was repositioned once. The randomization test comparing the CFS and foam liner in the intervention phase resulted in a probability value of .008.

Baseline data for subject 3 were unstable, with a sharp decrease in midline positioning during session 2. This infant's typical pattern was to slump to his left side, with his left arm behind the highchair tray and his head close to the tray. Due to his discomfort in this position, as evidenced by his fussing and crying, he needed to be repositioned once during session 1 and 6 times during session 3, inflating the percentage of time in midline during these sessions. An improvement in his midline positioning was noted with the introduction of the CFS, with percentage of intervals in midline ranging from 93% to 100%. Data for the foam liner revealed no trends. The randomization test comparing the CFS and foam liner in the intervention phase resulted in a probability value of .008. When the regular highchair was used during session 10, there was a noticeable change in level, with sitting posture in midline improved, but still below that of the CFS. However, during this session, the infant was fussy as he pushed back into trunk extension and shoulder retraction, and he required 3 repositionings after slumping to his left side.

As shown in Figure 3 (panel D), baseline data for subject 4 were fairly stable. This infant's typical pattern was to quickly slump to her left side after being positioned in midline. During session 3, she required a total of 4 repositionings. With introduction of the CFS, there was a change in level that was maintained throughout data collection. Data when using the foam liner were fairly stable with the exception of a noticeable improvement in midline positioning during session 7, when the subject required one repositioning. As with the other subjects, the randomization test comparing the CFS and foam liner in the intervention phase resulted in a probability value of .008.

Engagement With Toys

AS shown in Figure 4, there were no clear effects of either the CFS or the foam liner on bimanual play. Findings were mixed related to effects of the CFS on increasing ability to play with toys with arms free from support of the highchair tray (Fig. 5). For subjects 1 and 2, there was no reliable effect of the CFS on the ability to play with toys with arms free from support (Fig. 5, panels A and B). AS shown in Figure 5 (panel C), the mean percentage of intervals with arms free from support for subject 3 during baseline was 18% (range=0%-37%). This infant's ability to free his arms from support, however, may have been facilitated by the first author repositioning him once during session 1 and 6 times during session 3. While using the CFS, the mean percentage of intervals with arms free from support was 39%. Data while using the foam liner were stable, but lower, with a mean percentage of intervals with arms free from support of 3%. Subject 3 showed a higher frequency of playing with toys with arms free from support when using the CFS versus the foam liner, with the randomization test resulting in a probability value of .008.

[FIGURES 4-5 OMITTED]

For subject 4, baseline data were stable at a low level and remained low throughout the intervention phase when she was using the foam liner (Fig. 5, panel D). Data when using the CFS were quite variable, with some data points substantially higher than those in the baseline phase. The mean percentage of intervals with arms free from support was 20% when subject 4 was using the CFS, as compared with 3% when using the foam liner and 4% during the baseline phase.

Parental Perceptions

After coding of semistructured interviews, 2 major themes--acceptability of the CFS for daily use and greater independence for mothers and infants--were identified. All 4 mothers reported that they consistently used the CFS as an insert in their infants' highchairs. In addition to feeding their infants in the CFS, some mothers also used the CFS in other ways, such as placing it on the floor where their infants could play. All mothers expressed their satisfaction with the CFS, as opposed to previous positioning strategies, as exemplified by the following quotation:

   It (CFS) definitely helped give him the support like nothing else really
   would because it's customized and it just hugs him very nicely ... there's
   nothing else that we have that makes it so that he can't scoot forward by
   sliding his butt forward ... there are other things but they don't give him
   the support that he needs right now. (Mother of subject 2)

All mothers reported benefits for themselves as a result of using the CFS, including increased independence when performing caregiving responsibilities and household tasks.

   In terms of feeding, it was a lot easier because I wasn't having to deal
   with towels and everything, and he was just sitting right up. (Mother of
   subject 3)

   It's (CFS) been a huge relief, and just time-wise it's allowed me to do
   things while he's doing things. I mean, he's doing his work, his play,
   while I'm getting things done, and we're both happier actually. Less
   anxiety and he's just real happy. So it's made things a lot easier for both
   my husband and myself. (Mother of subject 2)

   When she's in the highchair, I can do the housework and see her, and when I
   am in the kitchen, I can see her. (Mother of subject 4)

In addition to positive effects of the CFS related to their caregiving responsibilities, mothers perceived multiple benefits of the CFS for their infants. All respondents reported greater independence for their infants in functional skills such as play, hand use, and social interaction. Two mothers indicated that their infants were better able to use their hands for play, as follows:

   It (CFS) makes it easier for him. Where before he would have to use his
   hands to keep himself up ... now it's more that he can just play ... he
   doesn't have to worry about that really. He's been able to use his hands to
   push his pop-up toys back down, and he's basically mobile with his hands.
   He's not having to support himself in sitting up or anything, so he can
   play real easily. (Mother of subject 1)

   I guess this is kind of an assumption, but it's pretty coincidental with
   putting him in the foam seat that he did start transferring toys and
   pulling things over his head with 2 hands ... he just has a lot of
   opportunity without trying to concentrate on using one hand to hold himself
   up. (Mother of subject 2)

Prior to receiving the CFSs, one mother had been feeding her infant in a highchair, and one mother had been using a walker for mealtime positioning. These mothers perceived that their infants' postural alignment was improved when they used the CFS, as described below:

   When I had him in the highchair before the study, I would roll towels and
   stuff them in the sides and behind him or beneath him or whatever to try to
   get him to stay up because he was completely flexible ... but with the foam
   seat, he was able to sit up, and there was no problem. (Mother of subject
   3)

   The foam seat is OK for her. She doesn't fall to the side. She keeps up
   straight in it. (Mother of subject 4)

Another perceived benefit of the CFS reported by some mothers was improved social interaction for their infants. The mothers described new opportunities for their infants to engage with other family members, as follows:

   He clearly likes dinnertime when we're all home ... he's looking at us, and
   we give him crackers to play around with ... he loves being at the table
   with all of us.... It (CFS) enables me to face him instead of being behind
   him.... I can face him and play with him, and that's a nice thing, too,
   because then he can learn facial expressions. Actually, that's very
   valuable. (Mother of subject 2)

   Using the foam seat, if there's a group of us sitting at the table, he
   doesn't feel like he's out of what's going on, because he's sitting right
   up there. (Mother of subject 1)

In addition to improved performance of specific skills such as hand use, some mothers reported that their infants had greater independence in the following ways:

   It's (CFS) helping him do things by himself and just exploring his little
   world. And he clearly enjoys being able to do that. (Mother of subject 2)

   I think it (CFS) freed him up to not have to concentrate so much on sitting
   up and to be able to put his energy into playing or eating or grasping ...
   or anything than just having to concentrate on keeping his head above the
   tray, basically. (Mother of subject 3)

Discussion

All 4 infants with neuromotor impairments had better postural alignment when using the CFS in a highchair as compared with when using either no adaptation or a foam liner. These findings are consistent with the seating literature (31,32) that suggests that the position of the pelvis dictates posture in the rest of the body, and with experimental studies (2,3,13) supporting pelvic positioning as a determinant of postural alignment for children with neuromotor impairments.

There are several hypotheses that may explain why the CFS improved postural alignment. The CFS may offer a biomechanical advantage by keeping the pelvis in a neutral position and by limiting the degrees of freedom, thus enabling the infant to control other body segments. In contrast to the planar surface of the highchair seat, the contoured CFS provided increased body contact with the seating surface, which may have provided increased support and control. (31) Postural control also may have been assisted by the nonslip foam surface of the CFS. The vinyl surface of the highchair allowed lateral or forward pelvic movement, leading to pelvic obliquity and asymmetrical weight bearing. Although not as dramatic or stable as the changes noted while in the CFS, subject 2's ability to maintain a midline position was improved when seated on the thin foam liner (Fig. 3, panel B), suggesting increased postural stability. We believe it is also possible that the relatively compliant texture of the CFS versus the harder surface of the highchair seat helped improve postural alignment. Green et al (33) observed that children with cerebral palsy performed at higher levels of ability when lying on a carpeted floor versus the harder surface of an acrylic tabletop. Those authors speculated that the children sank into the more compliant floor surface, which in turn provided greater load bearing and a reduction in degrees of freedom.

Campbell (34) argued that research involving children with neuromotor impairments should include measures of functional performance. We chose toy use as a dependent variable to reflect function. Based on the recommendation that midline orientation and proximal stability facilitate movement and function of the upper extremities (35) and on the proposed relationship between upright sitting and reaching ability in infants who are typically developing, (36) we hypothesized that use of the CFS would promote upright sitting, with concomitant improvement in reaching and manipulation skills. We contend that intervals when an infant had both hands in contact with a toy and when an infant was able to play with hands and forearms free from support of the highchair tray reflected improved function and freedom of movement of the upper extremities.

As shown in Figure 4, there were no effects of the CFS on increasing bimanual play for any subjects, regardless of their postural alignment. One possible reason for this is the nature of the task. Although toys were selected to encourage bimanual play, they did not require 2 hands to lift or explore. A second possibility is that the achievement of independent sitting facilitated lateralized versus bimanual upper-extremity movements, as proposed by Rochat and Goubet, (8) who found that providing external postural support at the pelvis enabled infants to modify their reaching patterns and transition from bimanual to lateralized reaching. Finally, motivation to use both hands could not be controlled, as the infants could not be instructed to "use both hands."

The effects of the CFS on increasing arm use free from support varied across subjects, with improvement noted for subject 3 (Fig. 5, panel C). This infant's higher level of postural control at enrollment into the study may have been a factor in his improved freedom of upper-extremity movement relative to the other subjects. For subject 4, the mean percentage of time with arms free from support (Fig. 5, panel D) increased when using the CFS (20% versus 4% during the baseline phase). Although she did not lean on her left forearm as frequently when in the CFS as compared with during the baseline phase, she did not demonstrate increased bimanual play during the intervention phase. This was most likely due to the position of marked forearm pronation with elbow, wrist, and finger flexion of her right upper extremity. We did not examine the distinction between weight bearing in the upper extremities for support (leaning) versus light contact (resting) of the upper extremities on the highchair tray. There may be differences in functional upper-extremity use if postural instability requires an individual to keep one hand positioned for support versus the serendipitous placement of a hand or forearm on a convenient surface. These variables, however, were not evaluated in this study.

Shumway-Cook and Woollacott (37) proposed that there are multiple systems contributing to postural control, including musculoskeletal components, sensory systems and strategies, neuromuscular synergies, anticipatory and adaptive mechanisms, and internal representations, organized according to the functional task and environmental constraints. The CFS may have a role in modifying some of the systems, such as musculoskeletal components and sensory systems, but a complex interaction of multiple systems may be required for functional movement to emerge.

Findings from our study support the systems theory that the demands of the motor task and the environment are critical components in the emergence of postural control. (38) According to the mothers' accounts, some infants demonstrated new motor skills while using the CFS such as stabilizing the base of a pop-up toy with one hand and activating the pop-up figures with the other hand (subject 1), and lifting both hands overhead to remove a hat (subject 2). Although quantitative data from our study do not reflect these abilities, these skills reportedly emerged when the infants were seated in the CFS and were presented with tasks requiring these skills.

Acceptability of the CFS by caregivers was high, with all mothers indicating that the CFS was easy to use and that they used it consistently. All mothers corroborated the observational measures that demonstrated the effectiveness of the CFS in promoting improved postural alignment. Furthermore, feedback from the mothers suggested that their infants demonstrated improved manipulation of toys when they were seated in the CFS at home. Used as part of their daily routines, the mothers identified several perceived benefits of the CFS not examined in this study. An unexpected finding was the reported increase in social interaction for infants as a result of sitting in the CFS. Three of 4 mothers described the new opportunities their infants had to engage with family members at the dinner table and how the face-to-face position was beneficial for imitating vocalizations and facial expressions.

All mothers also described perceived benefits for themselves as a result of using the CFS, including more time for household tasks because the CFS provided their infants with a comfortable position for play where they were contented. Two mothers stated that they experienced less anxiety because they knew the CFS was a safe, comfortable, positioning device that increased their infants' independence. These parents discussed family resources such as time and energy, outcomes that typically have been overlooked in seating studies. These findings corroborate those of Pain et al, (38) who found that ease of use and comfort for their child were important factors affecting the usefulness of adaptive chairs for parents.

The mothers in this study, in our view, provided a holistic view of their infants' lives within the context of the family as well as insights into potential outcome measures that are functional for infants and meaningful to parents. Mothers' concerns were focused on how the CFS affected their infants' abilities to fulfill their expected developmental roles such as self-feeding, playing, and engaging with family members. Although reducing impairments is a valid concern of physical therapists, (39) we also recommend that therapists should incorporate the perspectives of the family in goal setting and address functional skills that children typically demonstrate in natural settings such as home and school.

Limitations

There were several limitations of this study that need to be considered. The first limitation relates to the generalizability of the CFS fabrication. Because the clinician who fabricated the CFSs had 12 years of experience in their fabrication and use, her level of experience with CFSs was not similar to that of the typical clinician. A second limitation was the selection of the dependent measure, engagement with toys, and its relationship to the confounding variables of the infants' motivation and their developmental levels. The subjects appeared to be highly motivated to engage with the toys, regardless of the seating intervention. However, the task (engage with toys) the infants were presented with did not require them to demonstrate the dependent measures to be successful, and they could not be coaxed to demonstrate their "best performance." Unilateral toy play may have been more developmentally appropriate for the subjects than the dependent measure of bimanual play. The use of alternative dependent measures that are highly motivating and a measurement system that examines the variety, quantity, and quality of upper-extremity function should be explored in future research. A more sophisticated motion analysis system for data recording would provide additional data for analyzing functional manipulation skills. Finally, the small sample size limits generalizability of findings.

Areas for Future Research

Our experiences suggested several directions for future research. First, after viewing the videotapes, raters reported that infants were looking at toys more frequently when they were erect and in midline. These observations are consistent with the work of Bertenthal and von Hofsten, (40) who emphasized the importance of gaze stability in mastering reaching and manipulation. Thus, we believe some future research should examine the effect of adaptive seating on gaze stability. Second, the measurement of variables necessary for postural control in sitting could be refined by use of a motion analysis system. Third, future research would be strengthened by involving parents in the determination of outcome variables. Fourth, including the home and community settings in research efforts is recommended to strengthen the validity and generalizability of findings. In addition, research with greater numbers of subjects is needed.

Conclusion

In this study, we used a reliable, low-cost measurement system (videography) that clinicians can use to examine the usefulness of seating interventions with individual children. The findings support the efficacy of a CFS in promoting improved postural alignment in midline for infants with neuromotor impairments. Advantages of the CFS as compared with other adaptive seating devices include cost-effectiveness, ease of fabrication, and its ability to be modified to accommodate growth or changing postural support needs. Benefits of the CFS reported by the mothers included increased independence in play and social interaction for their infants and increased maternal freedom in performing household tasks and caregiving. These findings, in our view, highlight the importance of therapists focusing on interventions designed to help children participate in activities that are functionally relevant to them and their families.

Table.
Characteristics of Subjects

Subject   Age
No.       (mo)   Ethnicity    Diagnosis

1         18     African      Spastic
                   American     quadriplegia
2         12     Caucasian    Down
                                syndrome
3          9     Caucasian    Congenital
                                hypotonia
4         12     Pacific      Asymmetrical
                   Islander     spastic
                                quadriplegia

                                                   Data
Subject                    Level of                Collection
No.       Diagnosis        Sitting (7)             Site

1         Spastic          3--required support     Day care
            quadriplegia     at trunk and pelvis
2         Down             3--required support     Home
            syndrome         at trunk and pelvis
3         Congenital       4--required support     Home
            hypotonia        at pelvis
4         Asymmetrical     3--required support     Home
            spastic          at trunk and pelvis
            quadriplegia

* Graco Children's Products, PO Box 100, Elverson, PA 19520.

([dagger]) Sony Corp of America, 550 Madison Ave, New York, NY 10022.

References

(1) Roxborough L. Review of the efficacy and effectiveness of adaptive seating for children with cerebral palsy. Assist Technol. 1995;7:17-25.

(2) Miedaner JA. The effects of sitting positions on trunk extension for children with motor impairment. Pediatric Physical Therapy. 1990;2: 11-14.

(3) Myhr U, von Wendt L. Improvement of functional sitting position for children with cerebral palsy. Dev Med Child Neurol. 1991;33:246-256.

(4) McEwen IR. Assistive positioning as a control parameter of social-communicative interactions between students with profound multiple disabilities and classroom staff. Phys Ther. 1992;72:634-644.

(5) McGraw MB. The Neuromuscular Maturation of the Human Infant. New York, NY: Columbia University Press; 1943.

(6) Kamm K, Thelen E, Jensen JL. A dynamical systems approach to motor development. Phys Ther. 1990;70:763-775.

(7) Roxborough L, Fife S, Story M, Armstrong R. Seated Postural Control Measure Manual: Research Version. Vancouver, British Columbia, Canada: Sunny Hill Health Centre for Children; 1994.

(8) Rochat P, Goubet N. Development of sitting and reaching in 5- to 6-month-old infants. Infant Behavior and Development. 1995;18:53-68.

(9) Trefler E, Hobson DA, Taylor SJ, et al. Seating and Mobility for Persons With Physical Disabilities. Tucson, Ariz: Therapy Skill Builders; 1993.

(10) Scherzer AL, Tscharnuter I. Early Diagnosis and Therapy in Cerebral Palsy: A Primer on Infant Developmental Problems. 2nd ed. New York, NY: Marcel Dekker Inc; 1990.

(11) Case-Smith J, Fisher AG, Bauer D. An analysis of the relationship between proximal and distal motor control. Am J Occup Ther. 1989;43: 657-662.

(12) Fetters L, Fernandez B, Cermack S. The relationship of proximal and distal components in the development of reaching. Journal of Human Movement Studies. 1989;17:283-297.

(13) Dilger NJ, Ling W. The influence of inclined wedge sitting on infantile postural kyphosis [abstract]. Dev Med Child Neurol. 1986; 28(suppl 53):23-24.

(14) Hulme JB, Gallacher K, Walsh J, et al. Behavioral and postural changes observed with the use of adaptive seating by clients with multiple handicaps. Phys Ther. 1987;67:1060-1067.

(15) Nwaobi OM. Seating orientations and upper extremity function in children with cerebral palsy. Phys Ther. 1987;67:1209-1212.

(16) McCormack DJ. The effects of keyguard use and pelvic positioning on typing speed and accuracy in a boy with cerebral palsy. Am J Occup Ther. 1990;44:312-315.

(17) Seeger BR, Caudrey DJ, O'Mara NA. Hand function in cerebral palsy: the effect of hip-flexion angle. Dev Med Child Neurol. 1984;26: 601-606.

(18) McClenaghan BA, Thombs L, Milner M. Effects of seat-surface inclination on postural stability and function of the upper extremities of children with cerebral palsy. Dev Med Child Neurol. 1992;34:40-48.

(19) Trefler E, Angelo J. Comparison of anterior trunk supports for children with cerebral palsy. Assist Technol. 1997;9:15-21.

(20) Pope PM, Bowes CE, Booth E. Postural control in sitting the SAM system: evaluation of use over three years. Dev Med Child Neurol. 1994;36:241-252.

(21) Gross A. The effects of a supported seating position on upper extremity control in children with cerebral palsy [abstract]. Physical & Occupational Therapy in Pediatrics. 1989;9:143.

(22) Colbert AP, Doyle KM, Webb WE. ESEMO seats for young children with cerebral palsy. Arch Phys Med Rehabil. 1986;67:484-486.

(23) Bablich K, Sochaniwskyj A, Koheil R. Positional and electromyographic investigation of sitting posture of children with cerebral palsy [abstract]. Dev Med Child Neurol. 1986;28(suppl 53):25.

(24) Harris SR. Efficacy of physical therapy in promoting family functioning and functional independence for children with cerebral palsy. Pediatric Physical Therapy. 1990;2:160-164.

(25) Mac Neela JC. An overview of therapeutic positioning for multiply-handicapped persons, including augmentative communication users. Physical & Occupational Therapy in Pediatrics. 1987;7:39-60.

(26) Ottenbacher KJ. Evaluating Clinical Change: Strategies for Occupational and Physical Therapists. Baltimore, Md: Williams & Wilkins; 1986: 111-136.

(27) Kazdin AE. Single-Case Research Designs: Methods for Clinical and Applied Settings. New York, NY: Oxford University Press; 1982.

(28) Edgington ES. Single-subject randomization tests. In: Edgington ES. Randomization Tests. 3rd ed. New York, NY: Marcel Dekker Inc; 1995: 263-301.

(29) Glaser B, Strauss AL. The Discovery of Grounded Theory: Strategies for Quantitative Research. Chicago, Ill: Aldine; 1967.

(30) Billingsley FF, White OR, Munson R. Procedural reliability: a rationale and an example. Behav Assess. 1980;2:229-241.

(31) Cook AM, Hussey SM. Assistive Technologies: Principles and Practice. St Louis, Mo: Mosby; 1995.

(32) Reid DT. Development and preliminary validation of an instrument to assess quality of sitting of children with neuromotor dysfunction. Physical & Occupational Therapy in Pediatrics. 1995;15:53-61.

(33) Green AM, Mulcahy CM, Pountney TE. An investigation into the development of early postural control. Dev Med Child Neurol. 1995;37: 437-448.

(34) Campbell SK. Efficacy of physical therapy in improving postural control in cerebral palsy. Pediatric Physical Therapy. 1990;2:135-140.

(35) Boehme R. Improving Upper Body Control: An Approach to Assessment and Treatment of Tonal Dysfunction. Tucson, Ariz: Therapy Skill Builders; 1988.

(36) Rochat P. Self-sitting and reaching in 5- to 8-month-old infants: the impact of posture and its development on early eye-hand coordination. J Mot Behav. 1992;24:210-220.

(37) Shumway-Cook A, Woollacott MH. Motor Control: Theory and Practical Applications. 2nd ed. Baltimore, Md: Lippincott Williams & Wilkins; 2001.

(38) Pain H, Gore S, McLellan DL. Parents' and therapists' opinion on features that make a chair useful for a young disabled child. Int J Rehab Res. 2000;23:75-80.

(39) Palisano RJ, Campbell SK, Harris SR. Decision making in pediatric physical therapy. In: Campbell SK, Vander Linden D, Palisano RJ, eds. Physical Therapy for Children. 2nd ed. Philadelphia, Pa: WB Saunders Co; 2000:198-224.

(40) Bertenthal B, von Hofsten C. Eye, head and trunk control: the foundation for manual development. Neurosci Biobehav Rev. 1998;22:515-520.

K Washington, PT, PhD, is Clinical Assistant Professor, Division of Physical Therapy, Department of Rehabilitation Medicine, Box 356490, University of Washington, Seattle, WA 98195 (USA) ([e-mail]). Address all correspondence to Dr Washington.

JC Deitz, OTR/L, PhD, FAOTA, is Professor, Division of Occupational Therapy, Department of Rehabilitation Medicine, University of Washington.

OR White, PhD, is Professor, Special Education Area, College of Education, University of Washington.

IS Schwartz, PhD, is Professor, Special Education Area, College of Education, University of Washington.

All authors provided concept/research design and data analysis. Dr Washington and Dr Deitz provided writing. Dr Washington provided data collection and project management. The authors gratefully acknowledge the community clinicians who provided their expertise and support for this study. Very special thanks are extended to the parents of the infants for their time, trust, and insights. The contoured foam seat used in this study was developed at Children's Therapy Center of Kent by Nancy Hylton, PT, and Gay Lloyd Pinder, PhD, CCC-SLP.

This study was supported, in part, by US Department of Education grant H029D20081.

This study was approved by the Institutional Review Board of Children's Hospital and Regional Medical Center, Seattle, Wash.

This article was submitted September 21, 2001, and was accepted May 16, 2002.