Held, Sharon L. PT, DPT, MS, PCS; Kott, Karen M. PhD, PT; Young, Brenda L. PhD
Examination is one cornerstone of client management in physical therapy. Judgments are made based on the examination results about the need for therapy services, what interventions to use, and what outcomes to expect. In recent years, the traditionally narrow framework for this decision making, the reduction or elimination of impairments to body function and structures, has been expanded. In the newer, broader view of health, the clinician is challenged to perform examinations that encompass multiple domains.1,2 Examination should include tests of activity/functional limitation and participation/disability. Function and functional outcomes are best defined in relation to the environment in which they occur and participation or disability is best measured in relation to daily activities within social roles.2 Achievement of optimal outcomes, the ultimate objective of the management process, may be facilitated when appropriate variables related to the outcomes are considered. Current research3,4 supports the need to examine function and establish outcomes in context-specific situations by showing that children perform the same task differently in various environments. The environment therefore is one variable to be considered when examination is performed.2
How might this factor, environment, be characterized for examination? Shumway-Cook et al5 suggest several physical features of environmental dimensions that influence physical performances including distance and the affects of ambient conditions (such as illumination), physical load (such as items carried), terrain (such as obstacles), and the need to make postural transitions. Objective and standardized measures of the influence of these environmental variables will help to focus interventions, make it easier to gauge progress, and enhance the likelihood of optimal outcomes.
Gait, locomotion, and balance encompass one category of interrelated skills that should be evaluated at multiple levels for indications of participation (mobility in social context), activity (ability to walk), and body function and structure (balance while walking) because these data help inform clinical decision making.1,2 While a review of recent research reveals standardized tools for pediatric measurement at the levels of function and impairment for these skills, few contain the physical features of the environmental dimensions, as suggested by Shumway-Cook et al,5 or have been demonstrated to be reliable, valid, easy to administer, and cost-effective.
In 1999, Knutson et al6 reported results for the distance walked in 30 seconds by children as a functional outcome measure. Although testing was completed in a school environment and included four 90-degree turns (containing physical features of distance and transitional movements), the scope of environmental dimensions was limited. In a more recent article, Zaino et al7 reported on a test for children, the Timed Up and Down Stairs, which was developed as a functional mobility outcome measure. While this test includes an element common in many environments (ie, stairs) and contains elements of transitional movements, in this test too the scope of environmental dimensions was limited.
Research demonstrates that obstacle courses are a valid and reliable way to expand the physical features of environmental dimensions into the evaluation of functional mobility (balance and gait movements used in daily life),8–10 to differentiate between groups of adults,11,12 and to observe the results of interventions.13–16 Obstacle courses have been recently introduced as a testing tool for pediatric populations.17,18
In the development of a standardized test with more environmental dimensions, such as an obstacle course, an analysis of the task or tasks performed within an environment is critical. For example, a routine walking task for a child in school could require getting up from a surface (ie, desk chair) and moving from one location to another (ie, walk from the desk to the front of the class) while staying upright (ie, avoiding the other desks and chairs or stepping over books or papers on the floor). This routine task could be further complicated by having the child carry an object while walking or having the child perform in a poorly lit environment. Examining the context in which the tasks occur assists the examiner in judging the capability of the child to move within various environments. Functioning ultimately is determined by the interactions of the individual and the physical features of the environment.19
By applying the different physical features of the environmental dimensions suggested by Shumway-Cook et al,5 an obstacle course, used as a test of locomotion and balance, presents the opportunity to take a simple walking task through a predictable environment (with distance and obstacles staying the same) and a less predictable environment (with ambient condition and physical loads changed by walking wearing shaded glasses or walking carrying a tray). These changes could be used to challenge the child to demonstrate the capability to perform variations of the task in a timely and safe manner.
The Timed Up and Go Test (TUG)20 is a valid and reliable measure of functional mobility in pediatric21–23 and adult groups24–27 providing the examiner with an opportunity to judge stability and time walking forward over a given distance (including one turn and sit-to-stand and stand-to-sit transitions). The purpose of the TUG is to determine the relative risk of the performer falling or becoming functionally dependent, but it has also been shown to display changes in functional abilities in children.23 Similar to the Thirty-Second Walk Test,6 the TUG addresses the physical features of distance and transitions in a predictable context.
The purpose of the Standardized Walking Obstacle Course (SWOC) is to determine ambulation capacity by measuring stability and speed during gait under different circumstances in a safe, reproducible, and efficient way.28 The measures within the SWOC (time, number of steps, and observations of stability) are standard measures used in clinical practice for all patient populations.1 The SWOC was tested in 24 young adults with no known pathology and 13 older adults with arthritis and was reported to have excellent test-retest reliability (as determined by no significant differences in scores between sessions p = 0.38 to 0.97), high intertester reliability for the measurement of time (rs = 0.96 to 1.00) and the variables of steps off the path (determined as no significant difference in the frequency of observations between raters χ2df1 < 3.84 for both groups) and stumbles (p = 0.28 for the group with arthritis and p = 2.0 for the group of young adults).28 Concurrent validity has also been demonstrated between the SWOC and the timed Fifty Foot Walk Test (rs = 0.72 to 0.88).28 Based on these findings, the reliability and validity of the SWOC in testing of children is predicted.
Although supporting evidence for the usefulness of the SWOC has been reported,17 the tool's reliability and validity need to be formally tested in children. The purpose of this study was twofold. The first purpose was to determine the stability of the test and ability of the raters to be consistent on different occasions by determining interrater and intrarater (test-retest) reliability, respectively. For the purpose of this study design, the following definitions of these terms were used. Interrater reliability meant two raters measuring the same children simultaneously.29,30 “By strict definition intrarater reliability is the consistency with which one rater assigns scores to a single set of responses on two occasions.”30 p. 257 Often videotaping of performances is used to determine intrarater reliability. However, according to Domholdt, “For most of the measurements we take in rehabilitation... we do not have the ability to exactly reproduce the movement of interest.”30 p. 257 An alternative for examining intrarater reliability is to have a client perform the movement two or more times. The computed reliability coefficient is calculated on the scores from the two or more administrations. By definition, test-retest reliability is also calculated on scores from two or more administrations. The only limitation on the time interval for test-retest reliability is that it be “far enough apart to avoid fatigue or learning, but close enough to avoid genuine changes in the measured variable.”29 p. 67 In the opinion of Portney and Watkins, “In a test-retest situation, when a rater's skill is relevant to the accuracy of the test, intrarater reliability and test-retest reliability are essentially the same estimate.”29 p. 69
The second purpose of this study was to determine concurrent validity of the SWOC with the TUG in children with and without developmental disabilities. Concurrent validity, one type of validity evidence, is examined through comparison of the results of newer test (SWOC) with the results obtained from the known test (TUG). Results of the test of concurrent validity may then assist in the most appropriate selection of tests and measures for clinical populations. In this case, concurrent validity measured the degree of association between the measures of the SWOC and TUG scores of time and number of steps. Establishing concurrent validity of the SWOC and TUG should demonstrate how well the SWOC and TUG reflect similar aspects of gait, locomotion, and balance under standard and varied conditions.
Children with and without developmental disabilities were recruited from public schools and a private school for children with disabilities in Buffalo, NY, and after-school programs in Amherst, NY. Inclusion criteria for all participants were the ability to follow simple directions (three-stage directions in proper order), the ability to walk without any assistive devices and while carrying an object with two hands, and no record of cardiopulmonary comorbidity that would impede completion of the tasks. The parent or the physical therapist working with the children determined whether he or she met these criteria.
Letters were sent to administrators of area schools to obtain permission to recruit participants and conduct the study in their facility. Administrators or physical therapists subsequently sent out parent letters and consent forms. Recruitment from these areas allowed for increased ethnic diversity. Informed consent was obtained from a parent or guardian of each subject, and assent was obtained from all participants, regardless of age, prior to data collection. Participants and the parent or guardian had the right to decline or withdraw from participation at any time.
For the children without developmental disabilities, the age range for recruitment was three to 17 years inclusively. From pilot testing (authors' observations), children between three and four years of age were most consistently able to follow the directions. Children on average progress to independent walking between 12 and 15 months of age and changes in parameters of walking such as cadence and velocity, gradually occur from toddlerhood to adulthood.31 Walking velocity, however, is more dependent on height than age per se and children on average tend to reach adult height at or soon after finishing puberty.32 Girls tend to finish puberty around 15 years of age and boys around 16 or 17 years of age.
Children with developmental disabilities were tested if they met the inclusion criteria. Since these children remain in school through the age of 21 years in the United States and may need physical therapy services, the age range for children with developmental disabilities in this study was three to 21 years, inclusively.
This sample of convenience included 83 children who completed all conditions on the SWOC and TUG on day one, but only 73 of these children were present for data collection on day two. The data from the ten children absent from their programs (possibly due to illness) on day two were eliminated from the final analysis because we needed complete data to determine intrarater (test-retest) and interrater reliability in testing. The remaining 73 children ranged in age from six to 21 years for the children with developmental disabilities (n = 23) and four to 11 years of age for the children without developmental disabilities (n = 50). The type of developmental disability was not exclusionary because we were not seeking to compare performances between the groups on the items of the SWOC, but to determine its stability as a measurement. The children with developmental disabilities had medical and education classification including multiple handicap (n = 7), orthopedically impaired (n = 3), other health impaired (n = 3), cerebral palsy (n = 3), Down syndrome (n = 2), attention deficit disorder (n = 2), mental retardation (n = 2), and developmental delay (n = 1). The general demographic and descriptive data for this sample are presented in Table 1.
The SWOC and TUG were the two instruments used in this study. The SWOC requires walking on a designated path (12.2 m or 39.5 feet long and 0.9 m or 36 inches wide) and includes negotiating three directional turns (30 degrees right, 90 degrees left, and 70 degrees right), stepping over an axillary crutch, walking across a visually stimulating mat, stepping around a trash can, walking across a shag rug, and transitioning from sitting to standing and standing to sitting from a chair with armrests and a chair without armrests. Figure 1 shows the sequence of the obstacle course, with trial 1 proceeding from the chair with armrests to the chair without armrests and trial 2 proceeding in the reverse direction. The TUG requires standing up from a chair with armrests, walking forward 3 m to a line, turning, and returning to the chair to sit down. Both instruments are highly standardized in relation to their component parts, ie, distances walked and directions to complete the tests, and the placement of objects or obstacles is standardized for the SWOC. The benefit of this high standardization is that it places the sources of variability on the subjects' scores and not on the instruments.30
This research received ethical approval from the Human Subjects Research Review Committee at Daemen College, Amherst, NY. The procedures used were developed to address measures of reliability and validity. To test the stability of the scores over time, the subjects were evaluated twice at the same time of day and location, at intervals of exactly one week apart, with 32 children seen in the morning and 41 in the afternoon. The two examiners (S.L.H., K.M.K) also varied the role of test administrator to equalize possible examiner influences. The two examiners have tested over 300 children on the SWOC and have 20 and 29 years of experience in physical therapy all in pediatrics, respectively. The testing locations included gymnasiums, hallways, and designated therapy rooms and were kept constant for time one and time two. Besides the two examiners and the child being tested, others in the area while trials were being conducted varied. For example, the next child to be tested might be present, another adult might be waiting for the child, and individuals may have walked through the area. Although somewhat variable, the testing location and environment reflected typical situations in a school setting.
Testing was scheduled no longer than one week apart to eliminate potential influences on the scores such as perceptible development changes of the children and designed to counterbalance the potential sources of error such as day-to-day and time-of-day variability.32 This design also optimized the ability to use the data for analyses across and within raters.
A general information sheet was completed before the data were collected on the first visit. Total time for the visit on day one was 15 minutes and for the visit on day two was 10 minutes. When using the SWOC on day one, the subjects were randomly assigned to begin under one of the three conditions: arms down at sides (walk); carrying a lunch tray with a place setting of plastic cup, plate, and utensils to block the view of the feet (walk with tray); or wearing shaded glasses (walk with glasses, ie, walking with arms down at sides but in a simulated dimly lit environment). The task of carrying a lunch tray was introduced in place of the more standard task of carrying a laundry basket with weights, which is used in adult testing. Carrying the lunch tray is a more age-appropriate task and had previously been tested33 and shown to be valid and reliable in children with Down syndrome. The initial condition and sequence of testing was repeated on day two. For each visit, the subject was given verbal directions and a demonstration of how to negotiate the obstacle course. The directions included the following: use your typical walking, that is, how you move from one room to another at home or school and try to keep your feet on the carpeted path at all times. When the tray was carried, additional directions included picking up the lunch tray from the stool after standing up, walking through the course holding it level (horizontally) so the items would stay on it, and placing it on the stool at the other end before sitting down. Subsequent to pilot testing, in which significant trial differences were noted from trial one to trial two, each subject was given a practice trial of walking with arms down at side and then walking carrying the lunch tray prior to initiation of the six trials for data collection. The practice ensured the child's understanding of the different tasks and eliminated the significant trial differences.
For the SWOC, a trial consisted of standing up, walking the course from the beginning to the end in one direction only, and sitting down. A second trial was not begun until the subject rested as long as needed. Walking through the course in the opposite direction was the second trial, after which a new condition of the SWOC began. Each child completed a total of six trials. To determine consistency of test scores by the same and different examiners on each of the test days and through each pass of the obstacle course, two examiners determined and recorded the time required using a digital stopwatch. They also counted the number of steps, the number of stumbles (any loss of balance or body contact with an obstacle along the course), and number of steps off the path (all or part of the foot touched the floor along side the path).
To control for variability of test dates and conditions, children performed the SWOC and TUG tests in the same sessions. Children were given verbal directions and demonstration of the TUG. A trial for the TUG was defined as completing the full circuit. The children then performed three trials of the TUG, the first was a practice trial. When the TUG was performed, the time and number of steps were recorded simultaneously by two examiners. The number of steps was counted to have the exact same measures as the SWOC, but the five-point ordinate scale typically used with the TUG was not scored because there is no similar measure on the SWOC. As previously stated, the TUG was found valid and reliable in pediatric and adult populations and requires no special training to administer.
On each testing date, following practice trials, subjects performed two trials for each condition of the SWOC and two trials for the TUG. No trial differences were noted (paired t test, p > 0.05). Therefore, time, number of steps, stumbles, and steps off the path were averaged for each subject. To interpret the strength of the correlations in this study the following scale was used: <0.49 is low, 0.50 to 0.69 is moderate, 0.70 to 0.89 is high, and 0.90 to 1.00 is very high.30 Interrater and intrarater (test-retest) reliability were calculated for time and number of steps using intraclass correlation coefficients (ICC) models (ICC models 2,2 and 3,2 respectively),29 which reflect the degree of association and agreement among ratings. It was not appropriate to calculate ICCs for the numbers of stumbles and steps off the path because of low variation among and between subjects in these measures, particularly for children without developmental disabilities. However, Spearman rank correlation coefficients were completed for these data to examine reliability. Pearson product-moment coefficient of correlations29 were used to calculate concurrent validity for time and number of steps between each of the SWOC conditions and the TUG. In all cases, the correlation coefficients were considered significant to the p < 0.05 level. Due to the numbers of correlations made to protect against type I error, Bonferroni correction29 was used to establish the α value of 0.005. The power29 associated with these correlations ranged from 0.91 to 0.99.
The means and standard deviations for all variables assessed on the SWOC and the TUG are provided in Table 2. Time for course completion and the number of steps taken were greater and had more variability for children with developmental disabilities across all conditions of the SWOC. Less time and fewer steps indicate better functional ability.
The TUG scores in this study were found to be very highly reliable (interrater) for time (ICC 2,2 0.99) and number of steps (ICC 2,2 0.94 to 0.98). Intrarater (test-retest) reliability of the TUG scores was found to be highly or very highly reliable for time (ICC 3,2 0.86 to 0.95) and number of steps (ICC 3,2 0.87 to 0.91). The results of the SWOC reliability tests are presented in Table 3. Interrater reliability of the SWOC scores was also found to be very highly reliable for time and number of steps. Intrarater (test-retest) reliability for the SWOC was found to be highly or very highly reliable for time and number of steps.
Due to the lack of variation among and between children without developmental disabilities (Table 2), no consistent pattern of significant correlations was found for the Spearman rank correlation coefficients for stumbles or steps off the path on the SWOC. However, for children with developmental disabilities, the Spearman rank correlation coefficients were moderate to very high and significant (0.60 to 0.99, p < 0.005) except for intrarater (test-retest) reliability when measuring stumbles for the condition of carrying the tray.
The concurrent validity of the SWOC was tested by examining correlations between time and number of steps on the conditions of the SWOC and on the TUG. All Pearson product- moment coefficients of correlation had p values less than 0.003 and were considered significant. Correlations between the TUG and SWOC walk condition were high to very high for time in both groups of children with r = 0.79 to 0.88 for those with developmental disabilities; and r = 0.82 to 0.90 for those without developmental disabilities. The number of steps taken for both tests were highly or very highly correlated for children with developmental disabilities (r = 0.85 to 0.92) and highly correlated for children without developmental disabilities (r = 0.77 to 0.81). Correlations between the TUG and the other SWOC test conditions (carrying the tray or wearing the shaded glasses) are reported in Table 4. They ranged from moderate to very high for all children for time and number of steps.
The SWOC is a new tool for use with children in any setting. When establishing a new tool for use reliability and validity of the measures must be determined.
The robust correlations between raters and within raters indicate that the measures of the variables are consistent over the conditions of the SWOC despite the variability within the participant groups. The ICC values support the SWOC as a reliable tool for use with all children who can follow the directions and walk without an assistive device. Measures of time and number of steps, stumbles, and steps off the path are easy to obtain for any rater.
All the correlations between the SWOC conditions and the TUG for both groups of children were high or very high for time and moderate to very high for number of steps. Consistent high to very high correlations for time and number of steps occurred between the SWOC walk condition and the TUG. All the correlations support the similarities between the SWOC and the TUG, which both involve transitions, turns, and walking. However, the SWOC was designed to consider assessment of movement that combines ambulation and manipulation tasks in an environment that can be predictable, but allows for intertrial variability.19 Shumway-Cook and Jette34 advocate consideration of the physical aspects of the environment during examination of ambulation because patients rarely walk in clear, flat, well-lit environments. In the conceptual framework of the physical environment of Shumway-Cook and Jette, the elements of the environment include distance to be covered, speed needed to negotiate, terrain variations, influence of the ambience, if loads need to be carried, the concentration of objects or people, and the postural transitions that need to be made.34 The SWOC incorporates many of these elements such as providing a variety of turns, obstacles to step over and around, and a diversity of surfaces, thus testing walking under various environmental situations (ie, carrying a tray, wearing shaded glasses, and hands free). This makes the SWOC different from the TUG. A closer analysis of the correlations reveals only one correlation that is very high with all others falling into the moderate to high range under conditions of carrying the tray and wearing shaded glasses, suggesting test differences.
Currently, the SWOC provides a measure of the child's capability of walking a given number of steps in a given time with stability. Since the SWOC is well correlated with the TUG in the walk condition, the child's functionally mobility can be established with either test. However, the changing conditions on the SWOC may provide enhanced information about the interaction of the environment and the individual. Varying the physical features in testing affords a novel and enhanced way to measure change and retention of function.35 Evidence of the child's ability to participate may be reflected in the measure of ability to negotiate around obstacles as suggested by Lepage et al.36 The SWOC's value as a tool perhaps is this added complexity, which more adequately describes the desired behaviors for children within the environment.
As might be expected, performance scores differed between children with and without developmental disabilities on both tools. This was most notable in the increased average time and number of steps and the observation of one or more steps off the path by the children with developmental disabilities. In the future, the responsiveness to change of the SWOC in children with disabilities might be seen as the absence of steps off the path or the change scores on each condition of the SWOC indicating improvement in function. At this time, no further inferences are drawn from these numbers because our purpose was only to determine measurement reliability and validity of the SWOC.
Measures on the SWOC demonstrated high interrater and intrarater (test-retest) reliability in children with and without developmental disabilities. Statistically significant correlations were found for measures of time and number of steps for the SWOC walk condition with the TUG, suggesting that the SWOC is a useful measure of mobility for children. However, the clinician could use the additional conditions of the SWOC to assess potential environmental influences. Additional work is needed to support the SWOC as a tool to measure the extent of the effect of environment on activity and participation for different populations of children and its impact on intervention planning.
The authors acknowledge the assistance of Kristen Barbour in developing the SWOC diagram, and the administrators, parents, teachers and the children from public schools 78 and 84 and Cantalician Center for Learning in Buffalo, NY, and St. Benedicts' After School Program and Town of Amherst Youth Program in Amherst, NY.
1. American Physical Therapy Association. Guide to Physical Therapist Practice, 2nd ed. Phys Ther. 2001;81:S40, S72–73.
3. Tieman BL, Palisano RJ, Gracely EJ, et al. Gross motor capability and performance of mobility in children with cerebral palsy: A comparison across home, school, and outdoors/community settings. Phys Ther. 2004;84:419–429.
4. Doty AK, McEwen IR, Parker D, et al. Effect of testing context on ball skill performance in 5-year-old children with and without developmental delay. Phys Ther. 1999;79:818–826.
5. Shumway-Cook A, Patla AE, Stewart A, et al. Environmental demands associated with community mobility in older adults with and without mobility disabilities. Phys Ther. 2002;82:670–681.
6. Knutson LM, Schimmel PA, Ruff A. Standard task measurement for mobility: Thirty-second walk test. Pediatr Phys Ther. 1999;11:183–190.
7. Zaino CA, Marchese VG, Westcott SL. Timed up and down stairs test: Preliminary reliability and validity of a new measure of functional mobility. Pediatr Phys Ther. 2004;16:90–98.
8. Means KM. The obstacle course: a tool for the assessment of functional balance and mobility in the elderly. J Rehabil Res Dev. 1996;33:413–429.
9. Rubenstein LZ, Josephson KR, Trueblood PR, et al. The reliability and validity of an obstacle course as a measure of gait and balance in older adults. Aging Clin Exp Res. 1997;9:127–135.
10. Means KM, O'Sullivan PS. Modifying a functional obstacle course to test balance and mobility in the community. J Rehabil Res Dev. 2000;37:621–632.
11. Imms FJ, Edholm OG. Studies of gait and mobility in the elderly. Age Ageing. 1981;10:147–156.
12. Means KM, Rodell DE, O'Sullivan PS. Use of an obstacle course to assess balance and mobility in the elderly: A validation study. Am J Phys Med Rehabil. 1996;75:88–95.
13. Means KM, O'Sullivan PS, Rodell DE. Balance, mobility, and falls among the elderly African American women. Am J Phys Med Rehabil. 2000;79:30–39.
14. Means KM, Rodell DE, O'Sullivan PS, et al. Rehabilitation of elderly fallers: Pilot study of a low to moderate intensity exercise program. Arch Phys Med Rehabil. 1996;77:1030–1036.
15. Means KM, Rodell DE, O'Sullivan PS. Obstacle course performance and risk of falling in community-dwelling elderly persons. Arch Phys Med Rehabil. 1998;79:1570–1576.
16. Lamoureux EL, Murphy A, Sparrow A, et al. The effects of progressive resistance training on obstructed-gait tasks in community-living older adults. J Aging Phys Activity. 2003;11:98–110.
17. Kott KM, Held SL. Effects of orthoses on upright functional skills of children and adolescents with cerebral palsy. Pediatr Phys Ther. 2002;14:199–207.
18. Benedetto M, Thawinchai N, Prasertsukdee S, et al. Reliability and validity of a new assessment tool to measure pediatric functional mobility: A pilot study. Pediatr Phys Ther. 1999;11:214–215.
19. Shumway-Cook A, Woollacott M. (2001). Motor Control Theory and Practical Applications, 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2001:7.
20. Podsiadlo D, Richardson S. The Timed Up and Go: A basic functional mobility test for frail elderly persons. J Am Geriatr Soc. 1991;39:142–148.
21. Lowes LP. Evaluation of the standing balance of children with cerebral palsy and the tools for assessment (dissertation). Allegheny University of Health Sciences, Philadelphia, 1997.
22. Habib Z, Westcott S. Assessment of anthropometric factors on balance tests in children. Pediatr Phys Ther. 1998;10:101–109.
23. Habib Z, Westcott S, Valvano J. Assessment of balance abilities in Pakistani children: a cultural perspective. Pediatr Phys Ther. 1999;11:73–82.
24. Nair B, Dobson A, O'Dea I, et al. Further validation of “Timed Up and Go” in stroke patients. Australasian J Ageing. 1999;18:98–99.
25. Shumway-Cook A, Brauer S, Woollacott M. Predicting the probability for falls in community-dwelling older adults using the Timed Up & Go Test. Phys Ther. 2000;80:896–903.
26. Morris S, Morris ME, Iansek R. Reliability of measurements obtained with the Timed “Up & Go” Test in people with Parkinson disease. Phys Ther. 2001;81:810–818.
27. Steffen TM, Hacker TA, Mollinger L. Age-and gender-related test performance in community-dwelling elderly people: Six-Minute Walk Test, Berg Balance Test, Timed Up & Go Test, and Gait Speeds. Phys Ther. 2002;82:128–137.
28. Taylor MJ, Gunther J. Standardized walking obstacle course: Preliminary reliability and validity of a functional measurement tool. J Rehabil Outcomes Meas. 1998;2:15–25.
29. Portney LG, Watkins MP. Foundations of Clinical Research Applications to Practice, 2nd ed. Upper Saddle River, NJ: Prentice Hall Health; 2000.
30. Domholdt E. Physical Therapy Research Principles and Applications, 2nd ed. Philadelphia: WB Saunders; 2000.
31. Long T, Toscano K. Handbook of Pediatric Physical Therapy, 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2002:15–17, 20.
33. Kokubun M. Are children with Down syndrome less careful in performing a tray-carrying task than children with other types of mental retardation? Percept Mot Skills. 1999;88:1173–1176.
34. Shumway-Cook A, Jette A. Environmental influences on motor function: theoretical and practical considerations. Presented at American Physical Therapy Association Combined Sections Meeting, February 6, 2004, Nashville, TN.
35. Schmidt RA, Lee TD. (1999) Motor Control and Learning a Behavioral Emphasis, 3rd ed. Champaign, IL: Human Kinetics; 1999:266–271.
36. Lepage C, Noreau L, Bernard PM. Association between characteristics of locomotion and accomplishment of life habits in children with cerebral palsy. Phys Ther. 1998;78:458–470.
© 2006 Lippincott Williams & Wilkins, Inc.