Age Differences in Movement Patterns Used by Children ... - CiteSeerX

Age Differences in Movement Patterns Used by Children to Rise from a Supine Position to Erect Stance ANN

F. V A N S A N T

From the time upright locomotor ability is acquired until the end of the human life span, standing up from the floor is a skill important to a person's physical independence. This study was designed 1) to determine whether within the rising task the movement patterns of different regions of the body vary with age and 2) to describe movements used by children to perform this task. One hundred twenty children, ages 4 through 7 years, were filmed while rising from a supine position. Movement patterns were classified using categorical descriptions of the action of three body regions: the upper extremities, lower extremities, and axial region. The incidence of each movement pattern was calculated and graphed with respect to age. Age differences were found in the incidence of movement patterns of each body region. A trend toward increased symmetry of movement with increasing age was noted. The oldest subjects, however, did not commonly use symmetric form in rising. Developmental change in movement patterns used in the rising task likely continues beyond early childhood. Key Words: Functional training and activities; Kinesiology/biomechanics, general; Movement; Pediatrics, development.

The ability to rise from a supine position is an important component of physical independence. Physical therapists evaluate their patients' abilities to rise from the floor and include instruction in how to perform this task if a patient is unable to rise to a standing position. Little research, however, exists describing the precise form of the movement patterns used to rise to a standing position.1-3 From a developmental perspective, the task of rising is of theoretical importance. According to a traditional viewpoint, the process of normal motor development is postulated to be evidenced through the emergence of righting abilities1 such as rolling, sitting up, and eventually rising to erect stance. After the ability to rise is acquired, body movements used to accomplish this task are known to vary with age in very young children.1,2 Schaltenbrand has postulated that mature form in rising is achieved by the age of 4 or 5 years.1 In contrast, life-span developmental theory proposes that age-related developmental change in performance of motor tasks may occur throughout the human life span, not just during infancy and very early childhood.4 This theory challenges the traditional view that motor development is characterized solely by processes that lead to the acquisition of motor abilities, such as standing and walking. The life-span developmental viewpoint has important implications for the A. VanSant, PhD, is Associate Professor, Department of Physical Therapy, Medical College of Virginia, Virginia Commonwealth University, PO Box 224, MCV Station, Richmond, VA 23298-0224 (USA). She was a doctoral candidate, Department of Physical Education and Dance, School of Education, University of Wisconsin-Madison, Madison, WI, when this study was conducted. This study was completed in partial fulfillment of the requirements for Dr. VanSant's doctoral degree, University of Wisconsin-Madison. This study was supported in part by a grant from the Foundation for Physical Therapy and was presented at the Sixty-First Annual Conference of the American Physical Therapy Association, New Orleans, LA, June 16-20, 1985. This article was submitted August 3, 1987; was with the author for revision 14 weeks; and was accepted February 25, 1988. Potential Conflict of Interest: 4.

1330

physical therapist. If movement patterns do vary with age, then it is not sufficient for therapists to apply a single model of motor performance when teaching individuals of varying ages to perform a task. Ideally, physical therapists might be able to select the most age-appropriate patterns when instructing patients in the movements to be used in performance of a functional task such as rising to a standing position or rolling. Under the life-span developmental theory, the movement patterns used by adults when rising from a supine position were recently described, and specific hypotheses concerning development of movement patterns within the rising task were presented.3 The hypotheses were 1) that the movement patterns of body regions vary with age and 2) that the different movement patterns demonstrated by adults represent different developmental steps for the rising task. A specific developmental sequence comprising a series of movement patterns was hypothesized for each of three body regions. These hypothesized sequences are listed in Table 1. This study represents the next step in the validation of the theory that developmental change in movement patterns occurs across the human life span. A very important by-product of this study is the description of movement patterns used by young children when rising from a supine position.

BACKGROUND General trends in the development of body movement for the task of rising from a supine position have been described for the period of late infancy and very early childhood.1,3 A trend toward less body rotation was described, and an adult form ofrising,characterized by symmetry of body action, was reported to appear by the age of 4 to 5 years.1 A recent study of young adults revealed that this age group used various different upper extremity (UE), axial (AX) region, and lower extremity (LE) movement patterns when PHYSICAL THERAPY

RESEARCH TABLE 1 Hypothesized Developmental Sequences for Three Body Regions Proposed upper extremity sequence Step 1 —Push and reach to symmetrical push Step 2—Asymmetrical push Step 3—Symmetrical push Step 4—Symmetrical reach Proposed axial region sequence Step 1—Full rotation, abdomen up Step 2—Partial rotation Step 3—Symmetrical, interrupted by rotation Step 4—Symmetrical Proposed lower extremity sequence Stepl—Half kneel Step 2—Asymmetrical squat Step 3—Symmetrical squat with balance step Step 4—Symmetrical squat

standing up from a supine position.3 Although the most common form of rising involved symmetrical movement patterns of each body region, this form of rising was predominant in only 25% of the adults studied. The various movement patterns demonstrated by the adults were proposed to represent different steps of developmental sequences for the three body regions. Such a hypothesis is consistent with recent research describing motor developmental sequences.5 Table 1 lists the sequences of development of movement patterns for each body region hypothesized from the study of adults.3 Having proposed those developmental sequences of movement patterns, this study was designed to determine whether these sequences are valid descriptions of age-related, developmental differences in movement patterns used to rise to a standing position. Only longitudinal study will fully validate a developmental sequence. Before expecting a long-term commitment of both subjects and researchers, however, a reasonable hypothesis regarding developmental change is necessary. Roberton and colleagues described a method of preliminary validation of motor developmental sequences before investing in longitudinal study.6 This procedure is termed prelongitudinal screening. The process first involves a thorough review of the literature describing motor performance in the task of interest and a pilot study, if necessary, to further describe motor performance in the task. The researcher proposes a developmental sequence for the task and then constructs a graph illustrating how the incidence of each step would be expected to vary with respect to time. A cross-sectional study of different age groups is conducted, and the graph depicting the hypothesized sequence is then used in the analysis of data gathered in the cross-sectional study. Specifically, the observed incidence of each developmental step is calculated and graphed with respect to age. A graph of the observed incidence of each developmental step at each age is compared to the graph depicting the hypothesized sequence. The order in which movement patterns are found to predominate with respect to age is compared to the order in which they are predicted to vary with age. The relative increase and decrease in incidence of the patterns is also examined to determine whether some movement patterns might predominate in younger or older age groups than those included in the crosssectional sample. A four-step sequence of movement patterns had been proposed for each of three body regions—UE, AX, and LE—for Volume 68 / Number 9, September 1988

the rising task.3 These proposed sequences represent the period in the human life span leading to predominance of "advanced" symmetrical movement patterns in this task. A graph depicting how the incidence of steps might vary with age in a four-step developmental sequence is illustrated in Figure 1. A more exact estimation of the rate of rise and decline for any step of UE, AX, or LE sequences was not predicted. For that reason, the graph shown in Figure 1 was originally hypothesized to represent the proposed form of the sequences for each of the three body regions. The age range selected for this study represents the period in the life span when advanced symmetrical rising action was reported to be achieved.1 Completely symmetrical form in rising, although common, had not been observed in a majority of young adults.3 I, therefore, anticipated that study of a younger age group might assist with the interpretation of why many adults demonstrated an asymmetrical form in rising. For example, if young children were found to commonly use symmetrical movements to rise, then the young adults using asymmetrical movement patterns might be demonstrating regression to patterns dominant earlier in the life span. If young children did not commonly demonstrate symmetrical movement patterns, one might then hypothesize that symmetry becomes predominant in latter childhood, adolescence, or at some time beyond young adulthood, if at any time in the life span. METHOD Subjects One hundred twenty children, ages 4 through 7 years, participated in the study. Each of four age groups (4, 5, 6, and 7 years) contained 30 subjects and equal numbers of boys and girls. Table 2 presents the mean age and standard deviation for each group. The sample was one of convenience with subjects recruited from day care centers, day camps, and primary schools in the Richmond, Va, metropolitan area. I eliminated any subject who by parent report had any physical or medical condition that limited physical activity. The protocol for this study was approved by the human subjects committees of the University of Wisconsin-Madison and Virginia Commonwealth University. Informed consent was obtained in writing from a parent of each subject, and each child's verbal assent was obtained before beginning data collection. Research Design The study was designed as a prelongitudinal screening of hypothesized developmental sequences. The research is a

Fig. 1. Hypothesized four-step developmental sequence if graphed across time. Steps 1 through 4, representing different movement patterns of a body region, are depicted as they would be expected to rise and fall in frequency of occurrence over time.

1331

cross-sectional survey describing age differences in motor performance. Equipment Two 16-mm spring-wound cinematographic cameras were used to film the children as they performed the rising task. Each camera was equipped with a 26-mm lens. Filming speed was 24 frames/sec. Each camera was placed on a tripod with the optical axis of the lens about 1 m above the floor. Both cameras were located 7.92 m from the center of a 1.22- x 1.83-m exercise mat. One camera was sighted perpendicular to the length of the mat and provided a side view of the subject at the start of each trial. The other camera was positioned perpendicular to the width of the mat and allowed a foot view. Data Collection After I introduced filming assistants and explained the equipment and procedures, each subject was asked to lie supine on the mat and on the signal "Go" to stand up as quickly as possible. I used the preliminary instruction to stand quickly to facilitate automaticity in the subjects' movements. I gave no other instructions concerning how to perform the task. These instructions to the subjects were consistent with those used in my previous study of adults.3 Each child participated in a practice trial to ensure that they understood the directions given. An assistant and I filmed each subject performing 10 consecutive trials of rising from a supine position. Indiscriminate praise such as "Good standing" or "Nice" was given throughout the 10 trials. Data Reduction The films were viewed using a motion analyzer projector, enabling variable frame-rate and stop-action viewing. Movement patterns observed within a body region were classified using descriptive categories developed to describe adult rising movements. Concentrating first on the AX region, I classified the movement patterns demonstrated on the first trial across a block of at least 30 subjects and repeated this procedure for successive trials until all 10 trials had been reduced. I primarily used the side-view films for data reduction. When a movement pattern was observed that could not be easily classified, that movement was described in writing. I used both the footand side-view films when describing movements to ensure accuracy. After reducing all trials, I reviewed written descriptions and as necessary modified the descriptive categories developed for adults or developed new categories to describe the children's AX movement patterns. I then repeated this same procedure for the other two body regions.

TABLE 2 Mean Age and Standard Deviation for Each Age Group

a

Age Group

X

s

(yr)a

(yr/mo)

(mo)

4 5 6 7

4/6 5/5 6/6 7/4

3.3 3.4 3.3 2.9

n = 30 in each age group.

1332

Data Analysis Reliability of categorical descriptions. I trained another rater to analyze righting movements with reference to the categorical descriptions of AX, UE, and LE movement patterns. After training, the rater independently classified 50 randomly selected trials of the subjects' performances. I compared this rater's classifications to my original classification of these trials by calculating percentages of exact agreement. If less than 90% of exact agreement was obtained for any body region, I met with the rater to clarify the reasons for disagreement. We worked together refining the categorical descriptions or generating decision rules to improve the consistency of the classification process. I then randomly selected another set of 50 trials, and we repeated the process until 90% or greater of exact agreement was obtained for all three components. I also reclassified a randomly selected set of 50 trials and determined intrarater percentage of agreement. Developmental sequences. For each body region, I graphed the observed incidence of each movement pattern as a percentage of trials for an age group. I then compared these graphs with the graphs constructed to portray how I expected the incidence of each pattern to vary with age. I was particularly concerned with the order in which movement patterns became predominant and the relative increase and decrease in frequency of each pattern with age. For example, movement patterns predicted to occur early in a developmental sequence should be predominant in younger subjects. Conversely, movement patterns predicted to develop later should be more common in older subjects, and the incidence of early-appearing behaviors should decline. Description of righting in children. The percentage of trials classified in each category was calculated to portray the frequency of occurrence of each type of movement pattern at each age. "Profiles," or the combinations of UE, AX, and LE movement patterns displayed by each subject on each trial, were also examined. The frequency with which different profiles occurred within each group was determined, and these data were used to describe children's body action in rising. RESULTS Movement Pattern Categories An assumption central to this study was that movement pattern categories formed to describe adults' rising movements were also descriptive of children's rising movements. Before presenting the results of the analysis of age differences in movement patterns used in rising, the results of using the original categories to describe children's movements are presented. Upper extremity movement patterns. No new UE movement patterns were identified in this sample of children. Minor revisions were made to the descriptive categories that had been used to characterize adult UE action.3 The "push and reach to symmetrical push" category was changed to "push and reach to bilateral push" because the children frequently used their UEs to push synchronously but with some asymmetry. The original "push and reach" pattern was retitled "asymmetrical push." The primary characteristic of this movement pattern category is a period of pushing with one UE. Some children, like the adults in the previous study,3 began by using both UEs to push, but then resorted to a single-limb push. Some children included in this category demonstrated an alternating push with first one and then the PHYSICAL THERAPY

RESEARCH other UE. The "symmetrical push" category was modified only slightly; adults had placed their hands beside their pelvis at the start of the movement, whereas children often placed their hands alongside their trunk. Movement pattern category "symmetrical reach" was not revised. The revised categorical descriptions are presented in the Appendix. Axial region movement patterns. A new movement pattern of the AX region was identified and entitled "full rotation, abdomen down." This newly identified form of movement was not observed in adult subjects.3 Some revision of the original categorical descriptions was necessary. The "full rotation, abdomen up" action category was expanded to accommodate subjects who used predominantly neck and trunk flexion to bring their abdomen to a position facing the support surface. The "partial rotation" category was edited to include information previously included in a decision rule enabling discrimination between it and the "full rotation, abdomen up" pattern. The "symmetrical, interrupted by rotation" description was modified to portray a common movement characteristic of children. Although some of the young subjects began the task with a symmetrical pattern, more commonly they began flexing with a slight degree of rotation. In the latter instance, the direction of rotation reversed back toward a front-facing position. In some subjects, a double reversal of the rotational element was seen. In all instances, the tendency was to bring the head and trunk predominantly forward. The category, therefore, was retitled "forward with rotation." The Appendix presents the revised categorical descriptions of the AX movement patterns. Because each distinct movement pattern is interpreted to be a developmental step for AX movement within the task of rising, the discovery of an additional movement pattern affected the original hypothesis of a four-step developmental sequence. The newly identified movement pattern was hypothesized to appear earlier in the life span than the patterns observed in adult subjects.3 For this reason, the "full rotation, abdomen down" pattern was hypothesized to be Step 1 of the sequence and inserted to create a five-step sequence of AX region movement pattern development for this task. Lower extremity movement patterns. A new LE movement pattern category—"jump to squat"—was identified. This movement pattern had been observed in an adult subject, but had previously been considered idiosyncratic.3 Observation of the same movement pattern in several children made it obvious that a separate categorical description was warranted. The "half kneel" description3 was expanded to include bilateral kneeling action that preceded moving into half kneeling. The description of the "asymmetrical squat" pattern was modified to emphasize the wide base of support some children demonstrated when assuming a squat posture. These children's hips often flexed and internally rotated with the knees flexing, bringing the feet into position beside the pelvis. In this pattern, asymmetries of hip rotation were common. After assuming this squat position, the children extended their LEs ending the task in a wide-based stance. The LE symmetrical squat patterns, which included symmetrical narrow-based squat patterns with and without balance steps, were collapsed into one category. In the adult study,3 a discrimination had been made between those subjects who demonstrated a narrow-based squat and balance steps at the end of rising and those who took no balance steps. It became apparent during the study of children that the only difference between the two action patterns might be the amount of force generated, resulting in some instances in a loss of balance that required Volume 68 / Number 9, September 1988

stepping action. The two original categories, therefore, were merged, and the term "narrow-based" was added to the title to more clearly portray the distinguishing characteristic of this movement pattern. The identification of an additional movement pattern and the merging of two categories from the original descriptions of LE action resulted in a revision of the original four-step developmental sequence. The revised sequence also comprised four steps, but the first step was proposed to be the newly identified "jump to squat" pattern and the final step represented the merging of the original symmetrical squat patterns (with and without balance steps). Objectivity of Categorical Descriptions Percentages of exact agreement between myself and the trained rater ranged from 92% to 98% for classifying movement patterns of the three body regions. My repeated classifications of a randomly selected set of 50 trials also resulted in percentages of exact agreement of 92% to 98% across the three body regions. Developmental Sequences—Age Differences in Movement Patterns Used in Rising Upper extremity. The frequency with which each UE movement pattern appeared across trials for each age group is presented in Table 3 and portrayed graphically in Figure 2. The pattern predicted to be the first step in the sequence— push and reach to bilateral push—was observed most frequently in the youngest subjects and decreased in incidence with increasing age. The movement pattern hypothesized to be the second step of the sequence—asymmetrical push— showed a peak incidence in the 5- and 6-year-old children, being less common in both younger and older subjects. The proposed third step in the sequence—symmetrical push—was relatively rare in younger children but was seen with increasing frequency in the older children. The last step in the hypothesized UE developmental sequence—symmetrical TABLE 3 Incidence of Each Movement Pattern in Each Age Group of Children Expressed as Percentage of Trialsa Age Group (yr)

Movement Pattern Upper extremity Push and reach to bilateral push Asymmetrical push Symmetrical push Symmetrical reach Axial region Full rotation, abdomen down Full rotation, abdomen up Partial rotation Forward with rotation Symmetrical Lower extremity Jump to squat Half kneel Asymmetrical wide-based squat Symmetrical narrow-based squat with balance step a

4

5

6

7

40 57 3 0

12 73 15 0

5 71 24 0

8 44 47 1

1 25 26 47 1

0 3 22 68 7

0 4 6 76 14

0