R E S E A R C H

A R T I C L E

Relationships Among Age, Gender, Anthropometric Characteristics, and Dynamic Balance in Children 5 to 12 Years Old Sarah M. Butz, PT, PhD, PCS; Jane K. Sweeney, PT, PhD, PCS, FAPTA; Pamela L. Roberts, EdD, PT; Mitchell J. Rauh, PT, PhD, MPH, FACSM Pupil Personnel Services (Dr Butz), New Fairfield Public Schools, New Fairfield, Connecticut; Rocky Mountain University of Health Professions (Dr Sweeney), Provo, Utah; Department of Physical Therapy (Dr Roberts), University of Connecticut, Storrs, Connecticut; Doctor of Physical Therapy Program (Dr Rauh), San Diego State University, San Diego, California.

Purpose: To examine relationships among age, gender, anthropometrics, and dynamic balance. Methods: Height, weight, and arm and foot length were measured in 160 children with typical development aged 5 to 12 years. Dynamic balance was assessed using the Timed Up and Go (TUG) test, Pediatric Reach Test (PRT), and Pediatric Balance Scale (PBS). Results: Moderate to good positive relationships (r = 0.61 and r = 0.56) were found between increasing age and PRT and PBS scores. A fair negative relationship (r = −0.49) was observed between age and TUG test. No significant gender-by-age group difference was observed. Age had the strongest influence on TUG and PBS scores; arm length had the strongest influence on PRT scores. Conclusions: Dynamic balance ability is directly related to chronological age. Age and arm length have the strongest relationships with balance scores. These findings may assist pediatric therapists in selecting dynamic balance tests according to age rather than specific diagnosis. PowerPoint Slideshow Abstract: For more insights from the authors, access Supplemental Digital Content 1, at http://links.lww.com/PPT/A75. (Pediatr Phys Ther 2015;27:126–133) Key words: age, body measures, child, correlational study, gender, postural balance INTRODUCTION AND PURPOSE The development of postural control in children typically occurs in a stage-like progression, based on development of specific systems involved in postural control.1,2 0898-5669/2702-0126 Pediatric Physical Therapy C 2015 Wolters Kluwer Health, Inc. and Section on Copyright  Pediatrics of the American Physical Therapy Association

Correspondence: Sarah M. Butz, PT, PhD, PCS, 1 Evergreen Rd, Sandy Hook, CT 06482 ([email protected]). At the time this research was completed, Sarah M. Butz was a student in the Doctor of Philosophy program at Rocky Mountain University of Health Professions, Provo, Utah. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.pedpt.com). The authors declare no conflicts of interest. DOI: 10.1097/PEP.0000000000000128

126

Butz et al

Standing balance is regulated by a complex mix of systems, including the visual, vestibular, somatosensory, and musculoskeletal systems.3-5 Development of each system varies with age. The somatosensory system matures first and the vestibular system last.6 Postural control therefore develops as each system reaches the necessary threshold to support the associated behavior.5,7 Postural control is defined here as the ability to keep the center of mass over the base of support.8 Dynamic balance (DB) is operationally defined as the ability to maintain postural control during movements, such as reaching or walking.9 Dynamic balance is an essential physical therapy examination component for children. Measures of DB can be used before and after physical therapy interventions to determine the effect of specific treatment strategies on desired outcomes. Pediatric therapists need to understand DB measures and the relationship between age and DB. Various measures are available to assess DB including the Timed Up and Go (TUG) test,10 Pediatric

Pediatric Physical Therapy

Copyright © 2015 Wolters Kluwer Health, Inc. and the Section on Pediatrics of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.

Reach Test (PRT),11 and Pediatric Balance Scale (PBS).12 Unfortunately, the evidence for using many DB measures in the pediatric population is limited. Although measures of DB are important elements of examination, which measures correlate with specific ages or how anthropometric variables might influence balance is unclear. Important precursors for appropriate use of balance measures in the pediatric population include understanding the influence of anthropometrics during balance measurement as well as the interrelationships among age, gender, and DB. The purpose of this study was to examine relationships among age, gender, and DB in children aged 5 to 12 years during performance on the TUG test, PBS, and PRT. A secondary purpose was to examine the influence of various anthropometric variables, including height, weight, arm length, and foot length, on balance abilities. BACKGROUND Postural control and balance abilities vary across age levels due to the maturation of the systems that contribute to postural control, including the visual, somatosensory, and vestibular systems.1,2 Shumway-Cook and Woollacott1 reported a transition from immature to mature balance responses between the ages of 4 and 10 years. Children in the 4- to 6-year age group begin to use somatosensory information appropriately and develop balance strategies for control with altered balance conditions.1,13-16 Seven- to 10-year-old children use more mature strategies with altered conditions,1 developing better visual control and integrating vision with other sensory information.13 Largo et al17 found that motor performance, including balance abilities, was dependent on developmental status and age between 5 and 18 years, with balance leveling off in early adolescence. While some authors indicated that children as young as 9 years exhibited adult postural strategies,1 others have suggested that adult postural strategies are used near early adolescence.6 Many authors have discussed variations in postural control related to age and gender. Nolan et al4 found age and gender differences in standing balance in children aged 9 to 16 years. Odenrick and Sandstedt18 and Riach and Hayes19 found that movements of the center of pressure stabilized earlier in girls than in boys. Sellars20 found a significant gender difference on the 1-foot standing balance test in children with a mean age of 61 months with girls demonstrating higher quality static balance than boys. Donahoe et al21 reported that age, not gender, contributed to variance in balance measured by the Functional Reach Test (FRT). Habib et al22 examined balance abilities in children aged 5 to 13 years and found both agerelated and gender differences. All children demonstrated improved balance abilities with increasing age, and variation occurred between genders depending on socioeconomic status (SES). In the high SES group, boys demonstrated superior balance compared to girls; however, in the lower SES group, girls exceled in static balance activities.22 In contrast, Williams et al23 found no significant difference Pediatric Physical Therapy

in mean TUG scores between males and females aged 3 to 9 years. Various anthropometric variables may influence postural control or balance abilities. Habib and Westcott24 found base of support (length of foot) predicted FRT scores when examining all ages combined. They also reported height as a predictor of scores on the TUG test in all children and on the PRT in the younger children only. In a recent study by Norris et al,25 weight was the only anthropometric variable found to predict FRT scores in children aged 3 and 4 years. Although Donahoe et al21 did not find height, weight, or arm length predicted FRT scores, Duncan et al26 reported an association with arm length and functional reach distance, and found functional reach could be predicted using height. From these equivocal findings, the influence that anthropometric variables may have on balance abilities measured by the TUG test, PBS, and PRT in 5- to 12-year-old children remains unclear. Postural control is also dependent on task difficulty. Figura et al27 found task-dependent changes in standing balance in children 6 to 10 years of age. Centers of pressure were larger with more difficult tasks, specifically standing on one foot versus both feet close together. Streepey and Angulo-Kinzler28 reported that balance control was dependent on both age and reaching task difficulty. Significant differences were found between younger and older children with moderately difficult tasks. Because only reaching tasks were examined, the relationship between age and task difficulty with other DB activities is unclear. Balance or postural control is influenced by multiple variables, including task constraints, biological competence and age, and environmental affordances.29-31 Habib et al24 suggested some important considerations when examining postural control and balance abilities: (1) postural stability is task-specific; (2) different balance tests evaluate various age- and gender-related aspects of balance; (3) postural control improves with increasing age and then plateaus in adolescence; and (4) age and gender interactions need to be considered in different balance tests. For these reasons, postural control should be examined across ages and gender in children, using a variety of balance tasks. METHODS Design The study was exploratory, using a correlational design to examine DB in children using 3 balance scales: the PRT, PBS, and TUG test. The correlational design allowed the search for relationships between age, gender, and children’s balance abilities.32 Each participant performed the same 3 DB measures in random order. Subjects A convenience sample of children with typical development, aged 5 to 12 years, was recruited. A minimum of 5 years of age was established for participant selection

Relationships Among Age, Gender, Anthropometrics, and Dynamic Balance

127

Copyright © 2015 Wolters Kluwer Health, Inc. and the Section on Pediatrics of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.

because 5 years is considered a transition period in the development of postural control,1 and children younger than 5 years have greater difficulty following directions for the balance tests. For a medium effect size between 0.30 and 0.40, a sample of 100 children was needed to provide a power of 0.86 to 0.99.32 A selected study size of 160 met the medium effect size. The 160 participants included 20 subjects at each age from 5 to 12 years, with equal numbers of boys and girls in each group. Selection criteria for study participants were (1) age 5 to 12 years, (2) absence of neurological or orthopedic diagnoses, (3) no history of developmental delay or balance impairments, and (4) no orthopedic surgeries within the past 6 months. All diagnostic conditions and surgical histories were identified through parent report. The Rocky Mountain University of Health Professions Institutional Review Board approved the study. Parental permission and subject assent were obtained for all subjects. The study took place in multiple settings: (1) elementary and middle schools in Newtown, Connecticut; (2) private schools in New Milford, West Haven, and Danbury, Connecticut; and (3) an outpatient rehabilitation facility in Stratford, Connecticut. Each school had a multiple-purpose room available for data collection, and a conference room was used in the outpatient rehabilitation facility. Tests and Measures Timed Up and Go. The time (seconds) for participants to stand up, walk 3 m, turn around, walk back, and sit in a chair is measured using the TUG test. The TUG test is a reliable and valid test of balance and functional mobility,22,23 demonstrating high interrater reliability with intraclass correlation coefficient (ICC) values of 0.8122 to 0.8923 and high test-retest reliability with an ICC of 0.83.23 The TUG test has also shown moderate to high validity with the Berg Balance Scale (BBS) (r = −0.72) and gait speed (r = −0.55).10 Pediatric Balance Scale. The PBS is a modified version of the BBS and includes 14 balance items, each scored from 0 to 4, with a total possible score of 56.12 In children the PBS is a reliable test of balance with high interrater (ICC = 0.99) and test-retest (ICC = 0.99) reliability.12 The validity of the PBS has not been reported. Pediatric Reach Test. The PRT is a valid, reliable measure of postural control modified from the FRT, with the addition of reaching to the side and reaching in both sitting and standing.11 The PRT included 6 reaching activities with reach distance measured in centimeters or inches, with a sum of all distances as the final score.11 Moderate to high interrater reliability (ICC = 0.50-0.93) and test-retest reliability (ICC = 0.54-0.88) were demonstrated with the PRT. High construct validity has been found between the PRT and the measure of steadiness during standing (r = −0.79)11 and in children with cerebral palsy (CP), Gross Motor Function Classification System levels I-IV (rs = −0.88).11 128

Butz et al

Personnel A board-certified pediatric physical therapist (PCS) conducted training for research assistants (a physical therapist student and a pre–physical therapist student). Each research assistant, to ensure full understanding of the procedures for completing the balance measures, documenting data, and using the video camera, completed trials of the 3 balance measures. Training was completed after each assistant met appropriate levels of reliability with the PCS. The PCS assessed intrarater reliability of the 3 tests with the first 10 participants to ensure acceptable reliability (ICC ≥ 0.75) before testing the 160 participants in the main study. Each research assistant also completed the intrareliability testing (5 participants per assistant). Pilot study intrarater reliability was 0.96 for the TUG test (P < .001), 1.0 for the PBS, and 0.81 for the PRT (P = .001). The pilot study was videotaped to examine procedural and reliability testing.

Procedures Participants removed their shoes before height, weight, and foot length were measured by the principal investigator. Each child was weighed in kilograms by using a portable digital scale and then moved to stand against the wall face out with both feet on a pad of paper. Height was marked on the wall, and foot length was marked on the paper on the floor. Height was measured in centimeters using a measuring tape from the floor to a mark placed on the wall. Foot length was measured in centimeters by using a measuring tape as the perpendicular distance from the back of the heel to the end of the big toe. Participant arm length was measured in centimeters from the base of the dominant middle finger to the acromion. All participants then donned shoes for a practice trial of the 3 balance assessments before testing began. Test administration and verbal directions were adapted for children. Modifications to the TUG test were similar to the protocol used by Williams et al23 : (1) children were given a concrete task of touching a star on the wall; (2) a seat without arms was used; (3) no qualitative instructions were used (eg, “walk as fast as you can”); and (4) timing began when the child left the seat instead of at the word “go.” The protocol for the PBS followed instructions developed by Franjoine et al12 for use with the modified version of the BBS. The procedures for the PRT followed the detailed instructions described by Bartlett and Birmingham.11 The TUG test, PBS, and PRT were performed in a computer-generated random order. Each subject’s performance was videotaped to check scoring reliability. Two assistants rescored 5 different participants from the videotapes. The videotapes were randomly selected between the 80th and 100th participants. The primary investigator also scored the same participants from the videotapes to examine reliability and minimize examiner drift. Because examiner drift was not found (intrarater reliability: r = 0.98-1.0; Pediatric Physical Therapy

Copyright © 2015 Wolters Kluwer Health, Inc. and the Section on Pediatrics of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.

interrater reliability: r = 0.86-1.0), additional training was not conducted. To maintain consistency in measurement, the examiner read all directions and started and stopped each assessment. To minimize bias, the examiner reviewed the data only after all testing sessions were complete, except for the midstudy reliability data.

Balance Tests

n

Intraratera

Interraterb

TUG PBS PRT

10 10 10

0.998 1.0 0.98

0.999 1.0 0.86

Abbreviations: ICC, intraclass correlational coefficient; PBS, Pediatric Balance Scale; PRT, Pediatric Reach Test; TUG, Timed Up and Go. a ICC (3,1) for PBS and PRT and ICC (3,2) for TUG (P < .05). b ICC (2,1) for PBS and PRT and ICC (2,2) for TUG (P < .05).

Data Analysis The average of 2 test trials was used to calculate the final score in seconds for the TUG test. One test trial per item occurred for the PRT with the total score for the PRT equaling the sum of all 6 items in centimeters. The PBS final score was the combined score of the 14 items, each worth between 0 and 4 points. For the pilot study and midpoint reliability check, the ICC was used. The ICC (3,1) was used to determine the intrarater reliability for the PBS and PRT, and the ICC (3,2) was used to analyze the intrarater reliability for the TUG test.32 An ICC was also used to assess the interrater reliability during the midpoint check. The ICC (2,1) was implemented to explore the interrater reliability for the PBS and the PRT, and the ICC (2,2) was used to examine the interrater reliability for the TUG test.32 Means and standard deviations were calculated for the dependent variable TUG test, PRT and PBS scores and age, gender, height, weight, arm length, and foot length, the independent variables. The Pearson product moment correlation was used to analyze the relationships between age and the TUG test and between age and the PRT. The Spearman rank correlation coefficient was used to examine the relationship between age and the PBS. Point biserial correlations were implemented to investigate the correlations between gender and the TUG test, and gender and the PRT. Rank biserial correlation was used to assess the correlation between gender and the PBS. Finally, multiple linear regression was used to examine the associations between height, weight, arm and foot lengths, and the TUG test, PRT and PBS scores. For this study, correlations from 0.00 to 0.25 indicated little to no relationship, 0.26 to 0.50 indicated fair relationship, 0.51 to 0.75 indicated moderate to good relationship, and greater than 0.75 indicated good to excellent relationship.32 An α level of 0.05 or less determined a statistically significant relationship. Interpretation of the significance of the correlation included the value of the correlation and the statistical significance. SPSS 13.0 (SPSS Inc, Chicago, Illinois) and Microsoft Excel 2007 (Microsoft Office) software were used for all data analyses.

RESULTS Results of reliability testing completed mid-way through the study are included in Table 1. Intrarater and interrater reliability were high for the TUG test, the PRT, and the PBS. Pediatric Physical Therapy

TABLE 1 Intrarater and Interrater Reliability

Means and standard deviations for each balance measure by age are shown in Table 2. The mean TUG test values decreased with increasing age. The mean PBS scores increased with increasing age for years 5 to 7 and then plateaued for years 8 to 12 at the maximum possible score of 56. The mean PRT scores also increased with increasing age, except for years 9 and 10. A significant negative correlation was found between age and balance scores on the TUG test (Table 3). Children aged 5 years took longer to complete the TUG test, whereas children aged 12 years took less time. A significant positive correlation was found between age and balance scores on the PRT (see Table 3). Children aged 5 years reached a shorter distance than children aged 12 years. Significant positive correlations were found between age and balance scores on the PBS for all children and for children aged 5 to 7 years (Table 3). Children aged 8 to 10 years and 11 to 12 years all scored the same value on the PBS, a maximum score of 56, leading to correlation values of 1.0. Although a negative correlation between gender and DB was observed using the TUG test, no significant difference in scores was found between girls and boys (Table 4). A negative, nonsignificant correlation between gender and DB measured by the PRT was observed, indicating no significant difference in scores between girls and boys (Table 4). A positive but extremely low correlation between gender and DB was observed when using the PBS. No statistic has been recommended to determine significance32 (Table 4). Means and standard deviations for each anthropometric variable by age category are presented in Table 5. For ages 5 to 12 years, the mean height increased with each year of age. The average weight increased consecutively with each year except for the 12-year-old group. The standard deviation for the 11-year-old group mean was high compared with the other standard deviations. Arm length also increased consecutively with each year except for the 12-year-old group. The mean foot length increased with each year of age. Multiple linear regression analyses (Table 6) indicated that arm length was the strongest predictor of balance abilities on the PRT. Age was identified as the strongest predictor of balance abilities on the TUG test. Age was also determined to be the strongest predictor of balance

Relationships Among Age, Gender, Anthropometrics, and Dynamic Balance

129

Copyright © 2015 Wolters Kluwer Health, Inc. and the Section on Pediatrics of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.

55.10 (0.88) 55.80 (0.42) 55.90 (0.32) 56.00 (0) 56.00 (0) 56.00 (0) 56.00 (0) 56.00 (0) 5.06 (0.33) 4.78 (0.47) 4.58 (0.44) 4.63 (0.78) 4.29 (0.47) 4.29 (0.55) 4.59 (0.58) 3.64 (0.72)

147.62 (23.48) 152.58 (20.31) 147.64 (26.93) 168.25 (29.05) 177.73 (39.49) 176.03 (29.77) 191.01 (22.41) 202.32 (24.58)

PBS, points TUG, s

Mean (SD)

Girls

PRT, cm

TABLE 3 Correlations Between Age and Dynamic Balance Balance Measure

P

Relationships among age, gender, anthropometric characteristics, and dynamic balance in children 5 to 12 years old.

To examine relationships among age, gender, anthropometrics, and dynamic balance...
109KB Sizes 0 Downloads 6 Views