Increasing attention has been paid to expected functional outcomes for children with disabilities in a variety of environments. The Individuals with Disabilities Education Act (IDEA), Amendments of 2004 Part B (20 U.S.C. §1401) identifies an expectation of educational performance in least restrictive environments. Recent studies have confirmed the need to consider environmental factors when making decisions about functional walking of children with cerebral palsy,1 pedestrians older than 65 years,2 and older adults with and without mobility disabilities.3 Physical therapists (PTs) serving children in school settings need to consider environmental and task variables when making decisions about individual student accommodations or planning interventions.
A review of the literature suggests that walking speed has traditionally been measured in a gait lab with specialized equipment and with one child tested at a time. When characteristics of gait are used as normative reference values for child development or as an outcome following physical therapy, a variety of parameters have been identified including stride and step length, foot angle, base of support, cadence, and velocity.4–9 Any significant change in body structure such as an increase in range of motion, body function such as increased strength, or a reduction in an activity limitation was typically considered a success regardless of whether a change occurred in day-to-day walking. Even when walking speed is measured in day-to-day environments, walking was typically performed in a standardized manner on a predetermined path instead of a path used daily by the subjects.10–12 Ecological assessments that evaluate children in natural environments enhance the validity of the performance being measured.13,14
The students whom PTs serve in schools may have cerebral palsy, myelomeningocele, orthopedic impairments, arthrogryposis, or other conditions resulting in limited walking ability. To prevent impairments of body structure and function and limitations of activity from restricting participation in the general education environment, students with disabilities need to demonstrate day-to-day performance similar to their peers. Individualized Education Program (IEP) team members discuss a student's needs, goals, and the amount and type of accommodations needed to allow the student with a mobility limitation to participate in the general education environment. This discussion typically includes how the student will move in and around the school grounds. In elementary schools, teachers typically monitor the hallways and in a general sense set the speed at which students walk in the halls. The IEP team might consider the following possibilities: (1) The student can move at the same speed as peers; therefore, no accommodation is needed. (2) The student will not be able to independently move at a speed “good enough” to keep up with his or her peers in the halls; therefore, another person will have to assist the student to move in and around the school, or the student will move independently but too slowly to keep up with his peers. (3) The student may need an IEP goal related to mobility and physical therapy intervention to improve his or her speed. It might be hypothesized that physical therapy intervention could improve the student's speed or peers could slow down enough to match the student's speed. This slower speed would be identified as the “good enough” speed for participating with peers. The IEP team's decision should include input from the PT regarding the student's potential to increase his or her mobility speed, from the parents regarding whether this is a priority need for their child, and from the general education classroom teacher regarding how slow the class could move in the hallway that would still be acceptable. A peer-based school standard for mobility in hallways would help families, school personnel, and therapists identify realistic student-specific IEP goals.
Few data have been collected on actual walking speeds in elementary schools. Knutson et al10 identified distances walked in 30 seconds for students, aged six to 13 years, along a preselected path in a gymnasium. Two hundred twenty-seven children without disabilities were asked to pretend that they were line leaders. This design had compromised validity since neither the task nor the environment could be considered natural.
The purpose of this study was to collect direct observational data of walking speeds in elementary school hallways. The typical method of mobility in elementary schools is walking in-line in the hallway as a class behind a chosen line leader. Often, but not always, the teacher accompanies the class. Since IDEA encourages the use of task accommodations to permit students with disabilities to participate with their peers, a second set of data was collected. The purpose of the second part of the study was to collect data on what a teacher's slowest but “good enough” walking speed would be if he or she had to slow down his or her class's line to accommodate a student with a mobility limitation.
The actual walking speeds measured in the hallway used during day-to-day transitions between classes will be different for different grades. In the second part of the study, it was hypothesized that teachers have a mental standard for a range of walking speeds that would be acceptable for their class and it was assumed that teachers will be able to demonstrate their slowest “good enough” speed. This “good enough” speed could be the standard for acceptable accommodation in the least restrictive environment.
Elementary school class-lines were chosen based on convenience and permission from the school principal or other administrative personnel. The 43 elementary schools where data collection took place had hallways that were at least 50 ft long, were level, and without stairs or closed doors. The speed of the class-line was operationally defined as the speed of the first student in line. All elementary schools were in Iowa and all students were expected to walk in line in the hallways. Descriptive data for the participating elementary schools are found in Table 1. The line leaders were classified by grade level and not by age. All measurements were taken in mid-winter so that prior experience walking with the teacher and class was similar for all grades. Because walking speed is independent of gender of the child,9 no attempt was made to identify or control for gender. The elementary school teachers who determined “good enough” times were identified by their presence in the school where the class-line measurements were taken and were typically the teachers of the timed classes. They were a sample of convenience. Verbal agreement to participate was obtained after the study was explained to them.
This was a descriptive observational study. The study design was reviewed and approved by the Iowa Department of Education, Bureau of Children, Family, and Community Services Research Committee. Review guidelines are consistent the Family Educational Rights and Privacy Act (FERPA). FERPA standards are used by educational agencies receiving federal money. The law exempts organizations conducting studies for educational agencies for the purpose of developing, validating, or administering predictive tests from obtaining individual parental permission if the study is conducted in a manner that will not permit personal identification of the students and their parents and if the information is destroyed when it is no longer needed for the purposes of the project.
A pilot study was conducted in Iowa elementary schools the year prior to this study. Procedures for measuring walking speeds in school hallways were identified and refined. In the pilot study, individual students were asked to pretend to be a line leader even though other students were not present. While this procedure was feasible, it was decided that a more valid method of measuring walking speeds would be a direct measurement of class-lines as children moved from one location to another as a part of the regular school day.
A letter was then sent to all Iowa PTs providing physical therapy to students with an IEP goal. The therapists were invited to participate and were offered a monetary incentive to be used for future staff development. Forty-one PTs from across the state of Iowa volunteered to participate and 30 PTs actually submitted data at the end of the year. To determine the interrater reliability of measuring walking speed along a 50-ft path using a stopwatch, two PTs were selected and they timed the same class-lines. In a convenient elementary school, meeting the same criteria as the schools in this study, each therapist measured a 50-ft path on opposite sides of the hall and timed the same line leader walking along her path. There were 20 timed observations. Thirteen of the 20 timings were scored identical, and seven timings differed by only one second. An intraclass correlation method 2 (covariance matrix) analysis resulted in a high interrater reliability (intraclass correlation coefficient = 0.99).
Test Protocol for Line-leader Walking Speeds
The therapists identified the approximate times each elementary school class moved in line in the hallway. Using a tape measure, the therapist marked a distance of 50 ft in the hallway. Paths did not include stairs or closed doors. If there was not a 50-ft path, actual distances used were recorded. Using a standard school stopwatch, the therapist measured the time that the first student in line, the line leader, took to walk along the 50-ft path. Times were recorded to the nearest second. Information on the use of prompts or the absence of prompts by the teacher during the timing was specifically requested on the protocol chart. Therapists were encouraged to make additional comments on environmental conditions and other observed variables such as where the class was going, the position of the teacher, and hallway disturbances. Each recorder included the date, grade, building, and teacher's name for each timing. The data collection forms contained space to record other comments. Each rater was asked to collect data on two classes in each grade, kindergarten through sixth, that used a line to walk from place to place.
Test Protocol for Elementary School Teacher “Good Enough” Times
This part of the project aimed to identify the slowest speed that the teacher would find acceptable if one of his or her students had a mobility problem. When the line-leader data were being collected, teachers may have noticed the therapist in the hallway but did not know ahead of time that the line was being timed and did not know the purpose of the study. The same or similar 50-ft path used for gathering the children's data was used to collect data from the teachers. The project was explained to the teacher and verbal permission was obtained to participate in the study. The test protocol instructions for the teachers were the following: “We are trying to identify what would be the slowest acceptable walking (moving) speed a student with a mobility limitation would need to achieve in order to walk in line with your class.” It was explained that this hypothetical student might be wearing braces, using an assistive device such as a walker, or using a wheelchair. The therapist demonstrated what a slow walking speed might look like, and the teacher was given the opportunity to practice if he or she wanted to. When the teacher was comfortable with his or her “good enough” standard, a stopwatch was used to record the time that it took the teacher to walk 50 ft at his or her selected, “good enough” speed. Times were recorded to the nearest second. The measurement was repeated. Therapists were asked to collect data from two different teachers in each elementary grade. The PT recorded the teacher's name, grade, building, environmental conditions, the two timings, and any additional comments.
Line-leader Walking Speeds.
A one-way analysis of variance (ANOVA) was used to identify differences among the group means of the times to walk 50 ft. Multiple comparison tests (Scheffé) were used to evaluate more specific differences in walking distances across the grades, based on the presence of equal variances across grade levels.
Teacher “Good Enough” Times.
This analysis was run using the mean of the two times recorded for the teacher to walk 50 ft at a “good enough” speed, both when there were two recorded or just the first if that was all that was obtained. A one-way ANOVA was used to identify any significant differences among teachers of different grade levels. Tamhane multiple comparisons15 were used because of the unequal variances among grades.
Line-leader Walking Speeds
Line-leader walking time was measured for 376 elementary classes in the selected schools. All except 17 timings were taken using the recommended 50-ft distance. The 17 instances, when the distance walked was different from 50 ft, were converted to seconds per 50 ft for subsequent data analysis. Data from six line leaders were not included because the data collection deviated from the protocol. The low number of sixth grade line leaders is due to the frequent use of a middle school model in Iowa with sixth grade in a different building. Table 2 presents the mean, standard deviation, median, and range of the students' waking speeds. The seconds recorded for each line leader to walk 50 ft was used to determine speed expressed in feet per second.
Significant differences were found among grades in mean seconds to walk 50 ft (F = 10.92, df = 6, p < 0.001). There was a monotonic decrease (consistent decrease of speeds) in mean seconds required to walk 50 ft, from 13.5 seconds in kindergarten to 10.6 seconds for sixth grade. Multiple comparison tests (Scheffé) (Table 3) showed that mean times were significantly higher in first grade than in fifth and sixth grades, significantly higher in second grade than in fifth grade, and significantly higher in kindergarten than in fourth to sixth grades.
Variables Reported During Line-leader Timings
Therapists recorded information on teacher prompts for 319 of the 370 lines (Table 4). Of these, 30% of the teachers prompted their class-line at least once. The percentage of teacher prompts by grade decreased from 31% for kindergarten lines to 0% for sixth grade. Therapists were also asked to comment on the position of the teacher during the timing. Two hundred seventy of the 370 recordings (73%) included information about the position of the teacher during the recording of line-leader speed (Table 4). Of these, the teacher was described as being at the front of the line 130 times (48%). Teacher-in-front positions were described as “in front of the line,” “in front and off to the side,” “leading the line,” “leading but walking backward,” “beside the leader,” “teacher set the pace,” or “holding line leader's hand.” As with teacher prompts, the percentage of times that the teacher led the line decreased by grade; 31% for kindergarten lines and 0.5% for sixth grade. Forty-six percent of the class-lines had some type of teacher control (prompting or leading).
Only a few therapists commented on the general hall environment or described the class-line. The descriptions of the conditions identified fell into three categories: noise level, number of other people in the hall, and degree of precision of line spacing.
Elementary School Teacher “Good Enough” Times
Twenty-seven PTs submitted data on 290 teachers participating in this part of the study (Table 5). A one-way ANOVA revealed significant differences among grades (F = 3.90, df = 6, p < 0.001). Tamhane multiple comparisons, used because of unequal variances among grades, showed that times were significantly higher in third grade than in fifth and sixth grades. Teacher “good enough” times were significantly different from the actual line-leader times at every grade level (p < 0.001) (Fig. 1). The difference of the mean teacher “good enough” time from the mean line-leader time at each grade ranged from 5.8 to 12.2 seconds.
When a PT recorded two “good enough” speeds for the same teacher, the reliability between the two timings (n = 234) was very high. The intraclass correlation coefficient (method 2, covariance matrix) was 0.99.
Some therapists commented on the teacher's response to being asked to walk slowly. During the recording of the “good enough times,” 24 teachers reported that they have or have had a student with a mobility problem in their class.
Typical examples of teachers' comments were as follows:
- “If a student makes an effort, I will accept any speed.”
- “If a student fell, we would just stop and wait for her.”
- “If it was that child's time to be line leader, he or she could go as slowly as needed.”
- “They [other students] want to run all the time. If we go slowly, they'll run right over the child with a disability.”
- “This is hard. It depends on where the class is going and whether we're early or late.”
- “I will tolerate any speed as long as the student isn't falling.”
- “The child [previously in my class with a walking limitation] usually walked at the end because she was more comfortable walking behind the others. Then when she fell, she didn't trip anyone.”
- “If they took too long, they'd need to be in a wheelchair and then they could be pushed.”
- “I haven't had a student with a disability related to mobility, so I don't have experience with this.”
- “We would accommodate if the student were in a wheelchair. The kids would have to accept that we have to leave earlier.”
- “This is a good thing to think about.”
Typical examples of therapists' comments were as follows:
- “A few teachers were very slow and stated that they or a relative have a disability.”
- “Some teachers were uncomfortable doing this [walking slow].”
- “Teachers took this very seriously and tried to do their best at the ‘good enough' speed.”
- “Some teachers' ‘good enough' speed reflected past experience with a student who walked slowly.”
Direct measurement of walking speeds during a functional, environmentally referenced task provides PTs, IEP team members, and families a standard that can be used to establish the performance needed for mobility at school. Our original hypothesis that actual walking speeds would be different for different grades was confirmed. These results support the existence of a standard for walking in elementary school when class-lines are used. Multiple environmental variables were identified by the PTs. Examples of environmental variables found in the school hallways include very quiet, very noisy, lots of students and teachers in hall in addition to the line, no one else was in the hallway, halls were clean with no obstacles or distractions, halls had pieces of winter clothing (hats, boots, gloves, snow pants) randomly placed along the sides, many art projects on walls and some hanging from the ceiling, custodian waxing the side of the hallway, another class was in hallway dressing for recess, distractions in side hallways, groups of teachers talking in hall, and three lines were walking in the hall at the same time.
Teachers were described as being at the front of the line or prompting the line 46% of the time. One could assume that this would demonstrate a high degree of teacher influence on the speed of these lines. As might be expected, the percentage of teacher control over the speed of the class-lines decreased with increasing grade. This investigator also hypothesized that teachers have a range of walking speeds that would be acceptable for their class. Intra-teacher “good enough” times were reproducible. Though teachers had never practiced this skill prior to data collection, they seemed to have a standard in mind when performing a “good enough” walking speed. However, “good enough” times among teachers were more variable than we expected. Past experiences and other environmental factors including attitudes toward individuals with disabilities can affect participation. A positive attitude would encourage participation, whereas a negative attitude would create unnecessary barriers. In this study, comments suggested positive attitudes toward individuals with disabilities and toward school accommodations for students with mobility limitations.
The ability to generalize our study is limited by not obtaining a randomized sample and a potentially homogeneous school culture. However, data collection was not difficult and school norms for hallway walking could be established for any elementary school in any location. The model of data collection in this study could be replicated without considerable time or money.
Since we could not find other measures of walking speeds for students in everyday activities, we compared our results to speeds for children walking along an indoor track and speeds measured in gait labs. Knutson et al10 measured walking speeds around the perimeter of a gymnasium. We chose gait lab measurements of children by Sutherland et al16 as our second comparison. All mean walking speeds were converted to feet per second and are presented in Table 6.
The only overlapping data from all three studies was for six and seven year olds. Walking speeds in the gait lab were the slowest, whereas speeds used when walking along an indoor path were considerably higher than ours. The walking speeds of our fourth (4.3 ft/s) and fifth (4.6 ft/s) grade lines are similar to the 4.4 ft/s used as a normal adult comparison speed.12 In this study, the speed of the line leader in the natural school environment appeared to be affected by multiple variables including the teacher's position relative to the line and verbal prompts. Future studies should examine these environmental variables and their effect on walking speed in more depth.
Teacher expectations for “good enough” walking were significantly less than the actual line-leader means at every grade level. This acceptable accommodation will permit many students with slower walking speeds to be a member of their class-line when moving in their school hallways. For example, Baratonek et al17 measured walking velocity for four children (nine to 11 years old, fourth to sixth grades) with low and mid-lumbar myelomeningocele who used ankle-foot orthoses (AFOs) but no other assistive device. Using a self-selected velocity, none could walk the entire 102-m (335-ft) indoor path. Three children walked from 90 to 100 m with an average speed of 3.6 ft/s. In our study, fourth grade line leaders averaged 4.3 ft/s. Fourth grade teachers' “good enough” times averaged 2.3 ft/s. One might predict that these three students with myelomeningocele could keep up with their peers only if the teacher slowed the line to accommodate them. The fourth student with myelomeningocele could walk only 64 m (210 ft) with a velocity of 2.4 ft/s. This student might be able to keep up with a slowed-down line but might feel fatigued by the time they reached any destination 200 ft or more from their starting point. Students who cannot reach a “good enough” walking speed for a functional distance should use an alternate method of mobility, such as a manual or power wheelchair, so that they can continue to move in the halls with their peers. When an alternative method cannot match even the “good enough” time, accommodations including starting or leaving earlier than one's peers should be considered. The IEP team, including the student, may also decide that an adult or peer needs to assist the student so that participation in general education activities is not diminished. Participation may be more important than independence.
Identifying realistic functional goals is an important component of decision making regarding the need for physical therapy intervention for children with mobility limitations. Using a functional standard decreases the potential for overuse of therapy due to unrealistic expectations. When the standard for discontinuing therapy is a normal walking parameter such as stride or step length, foot angle, base of support, or cadence instead of an ecological standard, the decision to discontinue physical therapy is potentially never made because the standard is never met.
The median seconds used by class-lines to walk 50 ft in the hallway decreases with increasing grade from kindergarten to sixth grade. Teachers were willing to slow down the speed of the line within a “good enough” range of speeds. The best standard of comparison for an individual student with a mobility limitation is always the actual walking speed for his or her peer-led line. But with the information from this study, families, students, and therapists can project future needs more realistically and make informed, evidence-based decisions about future goals, expectations, and the need for physical therapy intervention. Additional information about walking speeds for groups of individuals with specific disabilities would provide a broader range of standards. For example, if there were a standard for walking speeds on level, smooth surfaces for children with cerebral palsy or spastic diplegia level III,18 then more specific mobility expectations could be made. Realistic goals for physical therapy intervention could be established with greater confidence. If the projected walking speed for a student with a mobility limitations was too slow for function in an elementary school environment, then more time and effort could be placed on identifying appropriate accommodations to eliminate participation restrictions.
We thank all the Iowa Area Education Agency PTs who assisted with the protocols and collected the data for this study. Our appreciation goes to Dr. Mack Shelley, Iowa State University, for assistance with data analysis. This project could not have been completed without the cooperation of the teachers, students, and building principals of the participating schools. We thank them all.
1.Palisano RJ, Tieman BL, Walter SD, et al. Effect of environmental setting on mobility methods of children with cerebral palsy. Dev Med Child Neurol
2.Knoblauch RL, Pietrucha MT, Nitzburg M. Field studies of pedestrian walking speeds and start-up time. Pedestrian bicycle research. Transportation Record 1538
3.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
4.Adams MA, Chandler LS, Schuhmann K. Gait changes in children with cerebral palsy following a neurodevelopmental treatment course. Pediatr Phys Ther
5.Dogra I, Sene G. Analysis of gait in healthy children in different age groups. Dev Med Child Neurol
. 2001;43(Suppl 89):37.
6.Foley CD, Quanbury AO, Steinke T. Kinematics of normal child
locomotion—a statistical study based on TV data. J Biomech
7.Rose SA, Ounpuu S, DeLuca PA. Strategies for the assessment of pediatric gait in the clinical setting. Phys Ther
8.Statham L, Murray MP. Early walking patterns of normal children. Clin Orthop
9.Wheelwright EF, Minns RA, Law HT, et al. Temporal and spatial parameters of gait in children. I: Normal control data. Dev Med Child Neurol
10.Knutson LM, Schimmel PA, Ruff A. Standard task measurements for mobility: thirty-second walk test. Pediatr Phys Ther
11.Lepage C, Noreau L, Bernard P. Association between characteristics of locomotion and accomplishment of life habits in children with cerebral palsy. Phys Ther
12.Waters RL, Lunsford VE, Perry J, et al. Energy-speed relationship of walking: standard tables. J Orthop Res
13.Karnish K, Bruder MB, Rainforth B. A comparison of physical therapy in two school based treatment contexts. Phys Occup Ther Pediatr
14.Fetters L, Tucker CA. Developmental concepts for therapeutic intervention. In: Damiano D, ed. Topics in Physical Therapy: Pediatrics, Lesson 10, pp 1–18
. Alexandria, VA: APTA, 2001.
15.Hochberg Y, Tambane AC. Multiple Comparison Procedures.
New York: Wiley; 1987.
16.Sutherland DH, Olshen RA, Biden EN, et al. The Development of Mature Walking
. London: MacKeith Press; 1988.
17.Bartonek A, Ericksson M, Saraste H. Heart rate and walking velocity during independent walking in children with low and midlumbar myelomeningocele. Pediatr Phys Ther
18.Palisano R, Rosenbaum R, Walter S, et al. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol
Keywords:© 2005 Lippincott Williams & Wilkins, Inc.
reference standards; child; walking/physiology; time factors; comparative study; physical therapy/methods