Informed consent was obtained from a legal guardian for all participants, and this prospective study was approved by the Nationwide Children's Hospital Institutional Review Board. In addition, assent was obtained for children 9 years and older.
To maximize feasibility for a variety of clinicians and researchers, the equipment used during the 50-FWT was commercially available and inexpensive. Duct tape was used to mark off the 50-ft distance for ambulation. A New-Lifestyles NL-1000 accelerometer (New Lifestyles Inc, Lees Summit, Missouri) was used to count the total number of steps taken during the 50-FWT. A previous study found that this accelerometer accurately counts steps over a 50-ft distance in children 6 to 13 years of age with ITW.15 No significant difference was found in the step counts recorded by the accelerometer and those recorded via videotape (W = 65.00, P = .24).15 The accelerometer and video step counts were highly correlated (ρ = 0.92) for this age group.15
To ensure consistency of accelerometer placement between participants and to reduce undercounting, a 2.5″×48″ Nylatex Wrap (DJO Global Inc, Chattanooga, Tennessee) was placed around each child's waist, with the top portion at the level of the umbilicus. The accelerometer was secured to the Nylatex Wrap posteriorly in midline. Attaching the accelerometer to a firm elastic belt improves the stability of the device, which reduces undercounting.16 If the child's shirt interfered with the accelerometer, a Goody Ouchless Elastic Hair Tie (Goody Products, Inc, Atlanta, Georgia) was used to prevent the shirt from contacting the device.
A GAITRite (CIR Systems, Inc, Franklin, New Jersey) was used as an objective measure of toe walking to validate the 50-FWT. It is an electronic walkway with an active recording area of 15 ft long and 2 ft wide. Sensors are located within the walkway system and are activated by pressure. The information recorded by the sensors is transferred to a computer and recorded. In children with motor disabilities, the GAITRite has good to excellent reliability when assessing tempospatial gait parameters.17 More than 80% of the intraclass correlation coefficient (ICC) values met or exceeded a reliability coefficient criterion of 0.80.17 Although not tested with children, the GAITRite has strong concurrent validity in healthy adults.18 Concurrent validity was tested by comparing the results of the GAITRite with the Clinical Stride Analyzer using a paired sample t test, as the participants walked at slow, preferred, or fast speeds.18 Speed, cadence, stride length, single-leg stance, and double-limb support as a percentage of the gait cycle were compared.18 No significant differences were found, with the exception of the speed recorded (P < .05), single-leg stance time (P ≤ .0001), and double-limb support as a percentage of the gait cycle when walking at a preferred pace (P ≤ .0001).18 Walking at a preferred pace is the condition that most closely matches the current study.
The physical therapist defined a toe-walking step as any step with initial contact on the toe. The GAITRite defined it as initial contact on the toe or less than 70% of the foot being in contact with the ground during stance phase. Adding the 70% foot contact was required because if the entire foot did not contact the surface on any of the steps, the GAITRite did not have a reference for the length of the child's foot. Thus, it could not interpret if the small area of pressure recorded was due to the child having very small feet or toe walking. Therefore, if the entire foot did not make contact with the GAITRite on any of the steps, it was unable to accurately identify a toe-walking step. To determine whether 70% of the foot was in contact with the ground, the length of the child's foot was measured in centimeters from the end of the first metatarsal to the end of the calcaneus before ambulating over the GAITRite.
Description of the 50-FWT.
The aim of the 50-FWT was to determine the percentage of toe walking performed by each child over a 50-ft distance. During the 50-FWT, the children were barefoot to ensure the physical therapist's view of their feet was not obstructed by shoes. The Nylatex Wrap, accelerometer, and child's shirt were prepared, as described in the Equipment section. To prevent the child from walking past the 50-ft distance, a parent or therapist stood at the end of the walk test to assist in stopping. This allowed the physical therapist to read the accelerometer value before the child began to move again.
The child stood at the start line and the accelerometer was set to 0. Each child ambulated 50 ft. The physical therapist walked at least 6 ft behind the child to assess toe walking while not interfering with gait. The accelerometer counted the total number of steps taken during the 50-FWT. The therapist counted the number of steps with initial contact on the toe. The number of toe-walking steps counted by the therapist was divided by the total number of steps counted by the accelerometer to get a percentage of toe walking.
Children with ITW modified their gait from a toe-to-toe pattern to a heel-to-toe pattern.3,19 In an effort to encourage the children to use the gait pattern they unconsciously select, a cognitive task was given to each child as a distraction method. Examples of cognitive tasks used included spelling, performing math problems, singing a song, counting by 7s, or discussing a favorite activity.
At the end of the first trial of the 50-FWT, the physical therapist and parent collaborated to determine whether the gait pattern displayed by the child was consistent with his/her unconsciously, self-selected pattern. The therapist observed the child's gait when he/she was unaware to assess their unconsciously, self-selected pattern. For example, gait was observed when the child walked from the lobby to the therapy room or when the therapist and parent were engaged in discussion. Parent input was also obtained. Each parent was asked after the first trial whether the gait pattern used was consistent with the pattern they observe most often when their child is not cued. If the gait pattern was deemed inconsistent with the child's usual performance by the therapist or parent, a second trial was conducted. A maximum of 2 trials were completed by each participant. Forty-three percent of the participants completed a second trial during reliability testing and 73% during validity testing.
Reliability of the 50-FWT.
Interrater reliability was assessed by the authors who had 9, 4, and 6 years of pediatric physical therapy experience, respectively. Table 2 includes demographic information for the raters. To assess interrater reliability, each participant had 2 of the 3 physical therapists observing gait during the 50-FWT. Different combinations of 2 raters were used throughout the trials. The 2 physical therapists, masked to each other's results, simultaneously counted the number of steps with initial contact on the toe as each subject performed the 50-FWT. Each therapist divided the number of toe-walking steps counted by the total steps recorded by the accelerometer. The percentages obtained by the 2 therapists were correlated.
Validity of the 50-FWT.
To determine concurrent validity, the child ambulated over the GAITRite. The GAITRite identified the number of toe-walking steps and a physical therapist, masked to the results of the GAITRite, simultaneously counted the number of steps with initial contact on the toe. The percentage of toe walking determined by the GAITRite was calculated by dividing the number of toe-walking steps by the total number of steps. Both values were acquired from the GAITRite results. The percentage of toe walking calculated by the therapist was determined by dividing the number of toe-walking steps counted by the therapist by the total step count recorded by the accelerometer. The percentage of toe walking obtained by the GAITRite was correlated to the percentage obtained by the therapist to assess the concurrent validity of the 50-FWT.
The children did not walk the full 50-ft distance during the validity testing. This was because the GAITRite only had 15 ft of active recording area. No warm-up or cool-down area was provided because an accelerometer was used to count the total number of steps taken by the child in order for the therapist to determine the percentage of toe walking. If the children were allowed to take additional steps outside the 15-ft area, the accelerometer would have counted steps that were not part of the trial. This would have artificially reduced the percentage of toe walking calculated by the therapist but not the GAITRite. Thus, the 2 values could not have been correlated accurately.
Given the shorter distance used during validity testing, there was a concern that the children may not have reached their typical gait speed. To evaluate this concern, each participant's gait speed was calculated from a video of him/her performing the 50-FWT as part of the previous accelerometer validity study.15 A stopwatch was used to measure the length of time it took each child to complete the 50-FWT. The distance was known to be 50 ft, and, therefore, a speed was able to be calculated. The speed calculated on the 50-FWT was compared with the speed recorded by the GAITRite.
Effect of Rater Experience Level.
The 3 raters had varying years of experience as pediatric physical therapists. Thus, to explore the role of experience level in accurate identification of a toe-walking step, the median difference score obtained between the GAITRite and each rater was compared. It was only possible to compare the results of the raters with 4 and 9 years of experience because the rater with 6 years of experience only completed 1 of the assessments using the GAITRite.
The data were not normally distributed. Therefore, when necessary, nonparametric statistics were used for data analysis. There were no missing data for this study. Descriptive statistics were used to report the difference in the number of steps counted as toe walking and the percentages of toe walking calculated between raters during reliability and validity testing. Interrater reliability was evaluated using an ICC3,1 and a 95% confidence interval. Concurrent validity was determined by correlating the results obtained via the GAITRite and the physical therapist using a Spearman ρ. Gait speeds obtained during the 50-FWT and during validity testing where a 15-ft distance was used were compared using a 2-tailed Wilcoxon signed rank test. Finally, the effect of rater experience level on their ability to identify a toe-walking step was evaluated using a 2-tailed Mann-Whitney U test.
A summary of the descriptive statistics can be found in the boxplots in Figures 2 and 3. Figure 2 graphs the difference in the number of steps counted as toe walking between the 2 physical therapists during reliability testing and between the GAITRite and 1 physical therapist during validity testing. Figure 3 graphs the variation in the percentage of toe walking calculated by each rater during the reliability and validity testing. Each boxplot graphs the medians, interquartile ranges, minimum, and maximum values.
Interrater reliability of the 50-FWT was excellent, with an ICC of 0.98 and a 95% confidence interval of 0.96 to 0.99. Concurrent validity was excellent with a Spearman ρ value of 0.95 (P < .001). No significant difference was found between gait speed during the 50-FWT and the 15-ft GAITRite walk (W = 160.00, P = .14) demonstrating that the children walked at similar speeds during both assessments. Finally, no significant difference was found between the physical therapists with 4 and 9 years of experience in the median difference score between the GAITRite and the rater (U = 17.03, P = .20).
The 50-FWT has excellent interrater reliability and concurrent validity for children 6 to 13 years of age with ITW. In addition, it may be accurate when used by experienced and inexperienced physical therapists. To the authors' knowledge, this is the first simple, cost-effective, commercially available, objective measure of the percentage of toe walking a child with ITW is performing. Although other objective measures of toe walking are available, they can be difficult to implement because of limited availability and increased cost, as discussed in the Introduction section.5–8 A method to measure the percentage of toe walking that is feasible in a variety of settings and for a diverse group of clinicians and researchers allows for objective measurement of the effect of treatment for ITW.
This study had a number of limitations. First, during the evaluation of interrater reliability, the same accelerometer step count was used by both raters because the number of toe-walking steps had to be assessed during a single trial. The 50-FWT determines the percentage of toe walking by dividing the number of toe-walking steps counted by the clinician by the total number of steps recorded by the accelerometer. Therefore, accuracy of the accelerometer is important. A previous study found excellent, but not perfect, accelerometer accuracy for children 6 to 13 years of age with ITW.15 Therefore, the interrater reliability value reported in this study may be slightly inflated, as the accuracy of the accelerometer is not accounted for within this measure.
In addition, concurrent validity was assessed by comparing the percentage of toe walking obtained by the GAITRite with the percentage obtained during the 50-FWT, which led to some limitations. Although the GAITRite has been found to be reliable in children with motor disabilities,17 validity has not been confirmed in children developing typically or those with ITW. Furthermore, the GAITRite only had 15 ft of active recording length; thus, validity was assessed over a shorter distance than the full 50-FWT. It is unknown whether the shorter distance impacted spatial-temporal gait parameters. However, the participants' gait speeds did not significantly differ when performing the full 50-FWT and the 15-ft GAITRite assessment.
Furthermore, the assessment of the effect of experience level on a physical therapist's ability to identify a toe-walking step was limited because only 2 distinct experience levels and 2 therapists were evaluated. It is unknown whether a larger sample size or more varied experience levels may have altered the result.
Finally, a potential limitation was the small sample size. However, the smaller a sample size, the more difficult it is to find a high correlation. Therefore, the high reliability and validity correlations measured in this study lend confidence that the sample size was sufficient.
In addition, to the study limitations, the 50-FWT has a number of inherent limitations. The first is that it is subject to underreporting of the percentage of toe walking a child is performing in his/her natural environment. Children with ITW are able to modify their gait from their unconsciously, self-selected toe-to-toe pattern to a heel-to-toe pattern.3 Because the 50-FWT is performed in a clinical setting where the children are aware of observation, they may self-correct their gait pattern. Although a cognitive task was used in an attempt to encourage each child's unconsciously, self-selected gait pattern, the effect of this is unknown. Evaluating test-retest reliability may be important to determine whether the 50-FWT is accurate over multiple trials given the variations in gait pattern that can occur in children with ITW.
In addition, the 50-FWT assesses a limited sample of the child's gait and is unable to be completed with children younger than 6 years. Fifty feet was chosen because it was felt that this distance was clinically feasible. However, a longer distance may provide children with more time to get into their unconsciously, self-selected gait pattern. Children must be at least 6 years old to accurately use the 50-FWT. This is because accelerometers are inaccurate in children 2 to 5 years of age with ITW. Therefore, they are unable to be used as part of the 50-FWT for this age group.15
The 50-FWT is a reliable and valid method to evaluate the percentage of toe walking performed by children 6 to 13 years of age with ITW. It is simple, cost-effective, commercially available, and objective, making it feasible in a variety of clinical and research settings. Given the limitations in the use of the GAITRite to assess validity, further investigation into the validity of this assessment may be beneficial to confirm the results. Investigation into test-retest reliability of the 50-FWT would be helpful because of the ability of children with ITW to modify their gait.3 Finally, further examination of the effect of health care professional experience level on reliability and validity of the 50-FWT would be valuable.
The authors thank Dr Elaine Clark, PT, DSc, PCS, for communication about the methods of this study.
1. Oetgen ME, Peden S. Idiopathic toe walking
. J Am Acad Orthop Surg. 2012;20:292–300.
2. Engstrom P, Tedroff K. The prevalence and course of idiopathic toe-walking
in 5-year-old children. Pediatrics
3. Hicks R, Durinick N, Gage JR. Differentiation of idiopathic toe-walking
and cerebral palsy. J Pediatr Orthop. 1988;8:160–163.
4. Van Kuijk AA, Kosters R, Vugts M, Geurts AC. Treatment for idiopathic toe walking
: a systematic review of the literature. J Rehabil Med. 2014;46:945–957.
5. Clark E, Sweeney JK, Yocum A, McCoy SW. Effects of motor control intervention for children with idiopathic toe walking
: a 5-case series. Pediatr Phys Ther. 2010;22:417–426.
6. Engstrom P, Bartonek A, Tedroff K, Orefelt C, Haglund-Akerlind Y, Guiterrez-Farewik EM. Botulinum toxin A does not improve the results of cast treatment for idiopathic toe-walking
: a randomized controlled trial. J Bone Joint Surg Am. 2013;95:400–407.
7. Williams CM, Michalitsis J, Murphy A, Rawicki B, Haines TP. Do external stimuli impact the gait of children with idiopathic toe walking
? A study protocol for a within-subject randomized control trial. BMJ Open. 2013;3:e0002389.
8. Pendharkar G, Percival P, Morgan D, Lai D. Automated method to distinguish toe walking
strides from normal strides in the gait of idiopathic toe walking
children from heel accelerometry data. Gait Posture. 2012;35:478–482.
9. Fox A, Deskin S, Pettigrew G, Paton R. Serial casting in the treatment of idiopathic toe-walkers and review of the literature. Acta Orthop Belg. 2006;72:722–730.
10. Eastwood DM, Menelaus MB, Dickens DR, Broughton NS, Cole WG. Idiopathic toe-walking
: does treatment alter the natural history? J Pediatr Orthop B. 2000;9:47–49.
11. Rathinam C, Bateman A, Peirson J, Skinner J. Observational gait assessment tools in paediatrics—a systematic review. Gait Posture. 2014;40:279–285.
12. Toro B, Nester C, Farren P. A review of observational gait assessment in clinical practice. Physiother Theory Pract. 2003;19:137–149.
13. Donnor A, Eliasziw M. Sample size requirements for reliability studies. Stat Med. 1987;6:441–448.
14. Williams CM, Tinley P, Curtin M. The toe walking
tool: a novel method for assessing idiopathic toe walking
children. Gait Posture. 2010;32:508–511.
15. Christensen C, Haddad A, Maus E. The validation of an accelerometer used to measure step count in children with idiopathic toe walking
. Pediatr Phys Ther. 2017;29:153–157.
16. Duncan JS, Schofield G, Duncan EK, Hinckson EA. Effects of age, walking
speed, and body composition on pedometer accuracy in children. Res Q Exerc Sport. 2007;78:420–428.
17. Wondra VC, Pitetti KH, Beets MW. Gait parameters in children with motor disabilities using an electronic walkway system: assessment of reliability. Pediatr Phys Ther. 2007;19:326–331.
18. Bilney B, Morris M, Webster K. Concurrent related validity of the GAITRite walkway system for quantification of the spatial and temporal parameters of gait. Gait Posture. 2003;17:68–74.
19. Westberry DE, Davids JR, Davis RB, de Morais Filho MC. Idiopathic toe walking
: a kinematic and kinetic profile. J Pediatr Orthop. 2008;28:353–358.
Keywords:Copyright © 2017 Wolters Kluwer Health, Inc. and Section on Pediatrics of the American Physical Therapy Association. All rights reserved.
ankle/physiopathology; equinus gait; idiopathic toe walking; outcome measure; pediatrics; toe walking; walking