Secondary Logo

Journal Logo

Research Report

Effects of Intensive Locomotor Treadmill Training on Young Children with Cerebral Palsy

Mattern-Baxter, Katrin PT, PCS, DPT; Bellamy, Sandra PT, DPT, PCS; Mansoor, Jim K. PhD

Author Information
doi: 10.1097/PEP.0b013e3181bf53d9
  • Free


Children with cerebral palsy (CP) show different motor patterns than children with typical development. Their movements are defined by excessive muscle cocontraction, altered joint kinematics, and decreased postural reactions, resulting in difficulty with ambulation for 90% of these children.1 Ambulation with or without assistance is important for various reasons. It has been shown that children who are ambulatory are more accomplished in their daily activities and peer interactions compared with children who use a wheelchair.2 The additional benefits of walking include increased bone mineral density, cardiopulmonary endurance, and obesity prevention.3,4 Furthermore, the majority of parents of young children with CP state ambulation as an explicit goal for their child.5 Young children with CP achieve their motor milestones at an accelerated rate compared with older children with CP.6 In light of this, it is crucial to provide intensive physical-therapy intervention for children with CP during the earlier years of childhood. In the past 10 years, locomotor training on a treadmill (LTT) has been used in the treatment of children with CP in an attempt to maximize walking independence, gait speed, and walking endurance. Locomotor training is designed to provide task-specific training with multiple repetitions of the walking task. Active participation of the child and explicit feedback in the form of verbal and tactile reinforcement have been shown to aid in the achievement of new motor skills.7–9

Locomotor training has been studied in both ambulatory and nonambulatory children with CP, with most studies having been conducted with children ages 5 to 18 years.10–15 Improvements in walking speed and endurance were shown in a clinically controlled study of 14 school-age children ages 5.5 to 14.7 years with moderate to severe impairments (GMFCS levels III and IV).14 In a study of 6 children ages 6 to 14 years with mild impairments (GMFCS level I), significant improvements were seen in the Energy Expenditure Index and gait speed.12 When examining the effects of a 3-month LTT intervention on gross motor function in 10 school-age children from 6 to 18 years (GMFCS levels not specified), statistically significant results in Gross Motor Function Measure (GMFM) Dimensions D (standing) and E (walking, running, and jumping) were found.11 In another group of 6 children ages 2.3 to 9.7 years with a mean age of 6.8 years and with the GMFCS levels ranging from I to V, improvement in gross motor function and endurance were found after an LTT program that was offered 3 to 4 times per week for 4 weeks.16 In a study of 8 children ages 3.5 to 6.3 years with a mean age of 4.5 years and GMFM levels II and III, who received LTT 2 to 3 times per week for 36 weeks, significant improvements in gross motor function and stride length were found.17

Despite these encouraging results, there is still a relative paucity of literature on LTT related to younger children with CP. In a case study, a 5-month-old infant with grade III intraventricular hemorrhage and at high risk for developmental disabilities was offered LTT for 23 weeks. This infant, who did not have a diagnosis of CP, made great improvement in gross motor function, including increased symmetry of gait, and eventually achieved independent ambulation.18 In an extended study of 4 infants with CP who were nonambulatory ages 1.7 to 2.3 years (mean age 2.0 years), LTT was provided 3 to 4 times per week for 4 months as an adjunct to traditional physical therapy. All children showed improved gross motor function, and 2 children attained independent walking.19 Both studies, however, were carried out for several months and maturation effects could have played an important role. The author of a recent systematic review concluded that there was a lack of evidence regarding the effects of intensive, short-term LTT on young children with CP younger than the age of 4 years.20

Accordingly, we examined the effects of intensive LTT in children younger than 4 years with different types and severity of CP with respect to gross motor function, in particular, standing and walking function. We used a short-term, intensive intervention period of 4 weeks to limit gross motor changes because of maturation. In particular, we were interested in determining whether this short-term, intensive program would (1) increase the walking speed and the endurance in children who were ambulatory with or without supporting devices; (2) decrease the amount of assistance needed for ambulation; (3) improve gross motor skills related to standing and walking; and (4) decrease the amount of caregiver assistance for the children.


Study Participants

Approval for the study was obtained from the Institutional Review Board of University of the Pacific, California. Written informed consent was obtained from the parents, and additional verbal assent was obtained from those children who were old enough to understand the purpose of the study. A convenience sample of 6 children with CP was recruited from parent support groups, from referrals from pediatric physical therapists in the community, and from Early Intervention Programs in the Northern California region. The inclusion criteria included (1) a diagnosis of cerebral palsy, (2) age of 1 to 5 years, (3) weight less than 40 kg, (4) the ability to bear weight on legs with or without support, and (5) parental ability to provide transportation. The exclusion criteria included (1) a medical contraindication for standing or walking, (2) history of untreated cardiac problems, (3) history of uncontrolled seizures, (4) history of orthopedic surgery, and (5) use of medication to control spasticity, including a baclofen pump or Botox injections in the past 6 months. Ten children were screened in their homes for participation in the study, with 6 children meeting the inclusion criteria. All children who were selected participated in the study, and 5 children attended all scheduled sessions. Only 1 child missed 2 sessions because of respiratory illness.

The age of the children ranged from 2.5 to 3.9 years with a mean age of 3.1 years, with 3 boys and 3 girls. Among them, 1 was GMFCS level I, 2 were GMFCS level II, 1 was GMFCS level III, and 2 were GMFCS level IV; 3 children had spastic diplegia, 1 had spastic quadriplegia, 1 had dystonic quadriplegia, and 1 had hypotonic CP. Of them, 3 children were able to ambulate with supervision and supporting devices at the start of the study, with 1 child beginning to take independent steps. The other 3 children used gait trainers with maximal assist (see Table 1 for subject characteristics).

Characteristics of Participants at Entry into Study

Study Design and Outcome Measures

The intervention consisted of 12 treadmill training sessions that were offered 3 times per week for a total of 4 weeks, with 1 or 2 days of rest between sessions. The training sessions consisted of 2 sets of treadmill walking with a small break between sets. The children were encouraged to walk for as long as possible and as fast as possible during each set. All children were assessed at baseline within 10 days of the start of the intervention period. The post-assessment was conducted within 7 days after completion of the intervention period, and a 1-month follow-up was conducted 1 month after the post-assessment. All dimensions of the Gross Motor Function Measure (GMFM) and 3 domains of the Pediatric Evaluation of Disability Inventory (PEDI) (Mobility-Functional Skills, Mobility-Caregiver Assistance, and Self-Help-Caregiver Assistance) were used as tests of gross motor function. Additionally, performance on the timed 10-m walk test, the 6-minute walk test, a treadmill walk test, and standing balance on 2 feet was measured. The walking tests were conducted at a local facility in Davis, California, by the same Pediatric Certified Specialist (PCS) to ensure the same conditions for each child. All children used their customary lower extremity orthotics during the walking sessions. Four children wore bilateral lower extremity orthotics and shoes, and 2 children walked only in shoes (Table 1).

Tests of Gross Motor Function.

GMFM: This a standardized clinical instrument, which is designed to evaluate changes in gross motor function in children with CP.6 The GMFM has been shown to have high validity and reliability.21 The GMFM-66 item version was used in this study. All the GMFM dimensions (Dimension A: lying and rolling; Dimension B: sitting; Dimension C: crawling and kneeling; Dimension D: standing; Dimension E: walking, running, and jumping) were measured in this study. The children were videotaped in their homes during all 3 assessments by a PCS with 22 years of pediatric experience. The videotapes were analyzed, and the children were assessed for GMFM levels by a different blinded PCS with 9 years of pediatric experience.

PEDI: This is an instrument that provides a clinical assessment of a child’s current functional performance or status.22 The PEDI is designed to evaluate 3 domains: self-care, mobility, and social function. These domains are evaluated through parent interviews, direct observations, and testing of the functional abilities of the children. The PEDI includes caregiver assistance scales for each domain. The Mobility-Functional Skills domain and the Caregiver Assistance domains for Mobility and Self-Help were used in this study. PEDI administration was done in the children’s homes by the same PCS who videotaped the children for the GMFM.23

Walking Tests.

Timed 10-Meter Walk Test: This is a valid and reliable measure to assess walking speed in children with CP.24 It can be used to assess self-selected walking speed or maximum walking speed. The children walked as fast as possible without running and were timed for 10 m. The test was performed 2 consecutive times with a short break between, and the faster time achieved was reported. The children subsequently rested until resting heart rate (HR) returned to pretest values, and then they proceeded to the 6-minute walk test.

Six-Minute Walk Test: This test is a reliable and valid measure to assess walking endurance in children with CP.25 The children were encouraged to walk at a self-selected walking speed but were discouraged from running and were allowed to vary their pace or rest as needed. The total walking distance in meters was measured with a tape measure.

Treadmill Walk: The third method of measuring walking ability was done by measuring the total distance and the speed walked on the treadmill during the 3 testing periods. A pediatric weight-support harness system with a hydraulic lifting mechanism (LiteGait Walkable 100) was placed over a treadmill with adjustable speed (GaitKeeper 18WST). The subjects were fitted into the harness and lowered onto the treadmill and were encouraged to take as much weight as possible on their legs without buckling. The percentage of weight support was calculated by weighing the children while in the harness and calculating the percentage in relation to their full body weight. All the children held on to an adjustable handle bar during the treadmill training. The children received assistance with hand placement if they could not maintain their grasp during walking. The children were given 1 initial training session on the treadmill to determine optimal walking speed and harness support. The optimal walking speed was determined to be the speed at which the children were able to take continuous steps without dragging their feet for more than 5 seconds. The treadmill distance and the walking speed were measured at the first intervention training session and the last intervention training session and at the 1-month follow-up.


Children who were able to stand independently were asked to stand on both feet as long as possible, without stepping or external support. Two consecutive trials were given, and the longer of the 2 trials was recorded in seconds.

Intervention Protocol.

The training sessions were scheduled 3 days per week for 4 weeks, with 1 or 2 rest days after each training day. The intervention lasted for 1 hour per day. Five children completed all training sessions, and 1 child missed 2 training sessions because of respiratory illness. All children participated in their regularly scheduled physical therapy sessions during the duration of the study. None of the children received additional treadmill training during the study period. All children were allowed to engage in their normal everyday activities, including walking.

The starting treadmill speed was determined during the initial training session and was increased as quickly as possible throughout the sessions. The speed was increased when the children could move their feet independently, with verbal cues or minimal manual cues at the pelvis without dragging their feet for more than 5 seconds. The children walked on the treadmill for as many minutes as possible to the point of fatigue. During all treadmill walking, the children were monitored with a HR monitor (Polar Monitor Model 515), and the HR was recorded at 1-minute intervals. The maximum allowable HR was 80% of the age-predicted maximum HR. None of the children reached the 80% maximum HR. The HR was also continuously monitored after the treadmill walk until it returned to resting levels, at which point a second treadmill walk was initiated. The children were not allowed to walk for more than a total of 40 minutes per training session. The amount of weight support was decreased as quickly as possible over the duration of the study determined by their ability to take continuous steps without dragging their feet for more than 5 seconds. By week 2 of the intervention period, 5 of 6 children did not require weight support anymore. The parents were present during all the sessions.

The same physical therapist provided facilitation at the pelvis when the child stumbled or stopped stepping. The facilitation was not intended to correct the gait pattern of the child. The children were allowed to make mistakes in step length and height but were encouraged with verbal and tactile cues to take symmetrical steps. The children were encouraged to look up during walking and were motivated with singing, favorite toys, or stuffed animals.

Data Analysis

Nonparametric statistics were used in all analyses because of the lack of a normal population distribution. For the primary analysis of all data, a Friedman analysis of variance (pre-intervention versus post-intervention versus 1-month follow-up) was performed. Post hoc analysis for pairwise comparisons was performed using the minimum significant difference.26 An alpha level of ≤0.05 was used in all analyses. For tests that the subjects were unable to complete (eg, the 10-m and 6-minute walk tests), they were assigned the worst ranks for data analysis purposes. The data are presented as means plus or minus standard deviations, unless otherwise indicated. Statistical analyses were performed using SPSS version 16.0 for the MAC.


Tests of Gross Motor Function


GMFM Dimensions A and B showed no statistically significant changes between pre-intervention and post-intervention. Statistically significant changes were found in the GMFM Dimensions C (p = 0.05), D (p = 0.007), and E (p = 0.01) in the primary analysis. The post hoc analysis revealed a significant difference between pre-intervention and post-intervention and between pre-intervention and 1-month follow-up, for both Dimensions D and E, indicating that the children improved their functional standing and walking abilities (Fig. 1).

Fig. 1.:
Individual GMFM Scores for Dimensions D and E. Subject 2 had scores of 0 for pre-intervention Dimension D and for pre-intervention, post-intervention, and 1-month follow-up for Dimension E. Means were statistically different between pre-intervention and post-intervention for Dimensions D and E.


The analysis of the Functional Skills Mobility Scale, the Caregiver Assistance Mobility Scale, and the Caregiver Assistance Self-Help Scale of the PEDI revealed statistically significant changes in the Functional Skills Mobility Scale (p = 0.022) and the Caregiver Assistance Mobility Scale (p = 0.018) but not the Caregiver Assistance Self-Help Scale. The post hoc analysis of the Functional Skills Mobility Scale showed a significant difference between pre-intervention and 1-month follow-up but not between pre-intervention and post-intervention. The post hoc analysis of the Caregiver Assistance Mobility Scale revealed significant differences between pre-intervention and post-intervention and between pre-intervention and 1-month follow-up. This indicates that the children increased their independence in the functional mobility tasks and relied less on their caregivers for assistance (Fig. 2).

Fig. 2.:
Individual PEDI Standard Scores for Functional Skills Mobility Scale, Caregiver Assistance Mobility Scale, and Caregiver Assistance Self-Help Scale. Means were statistically different for the Functional Skills Mobility Scale between pre-intervention and 1-month follow-up and for the Caregiver Assistance Mobility Scale between pre-intervention and post-intervention, and between pre-intervention and 1-month follow-up.

Walking Tests

Timed 10-Meter Walk Test.

Three children were able to complete the 10-m walk test at the pre-intervention assessment; 4 children were able to complete the test at post-intervention assessment, whereas all 6 children were able to complete this test at the 1-month follow-up. Of the children who learned to propel their gait trainers during the study period, the steering wheels had to be locked during the test because of their inconsistent ability to walk in a straight line. Statistically significant changes were found in the primary analysis (p = 0.011). Although all the children improved their walking speed between pre-intervention and post-intervention, the post hoc test showed a significant difference only between pre-intervention and 1-month follow-up (Fig. 3A).

Fig. 3.:
Individual scores for the 10-m walk test (A) and 6-minute walk test (B). Subjects 1, 2, and 3 were not able to complete the 10-m walk test and the 6-minute walk test for some of the conditions. Means were statistically different for the 10-m walk test between pre-intervention and 1-month follow-up and for the 6-minute walk test between pre-intervention and 1-month follow-up.

Six-Minute Walk Test.

Three children were able to ambulate for 6 minutes during the pre-intervention testing. By the 1-month follow-up, all the children were able to take steps in their gait trainers independently and were thus able to participate in the 6-minute walk test. The primary analysis showed statistically significant differences (p = 0.029), with nonsignificant improvements occurring between pre-intervention and post-intervention but significant improvements occurring between pre-intervention and 1-month follow-up (Fig. 3B).

Treadmill Walk.

Statistically significant differences were found in the primary analysis for the total distance walked (p = 0.009) and walking speed (p = 0.002). The post hoc analysis was significant for both variables between pre-intervention and post-intervention and between pre-intervention and 1-month follow-up. Some children required partial weight support during this activity. The mean walking speed was 0.08 m/sec during the first training session with a range of 0.04 to 0.13 m/sec. The mean walking speed during the last training session was 0.25 m/sec with a range of 0.13 to 0.45 m/sec. The mean distance walked during the first training session was 46.3 m with a range of 4.0 to 79.0 m, and the mean distance walked at the last training session was 151.8 m with a range of 36.0 to 434.0 m (Table 2).

Progression of Treadmill Ambulation

In addition to these results, 1 child who was ambulatory with a walker at the beginning of the study achieved independent walking during the intervention period. Three children who used gait trainers with maximal assist at the beginning of the study improved to intermittent assistance by the end of the study. This intermittent support was for steering of the gait trainer and for occasional help when getting stuck, but all 3 children were able to self-propel their gait trainers, which they had previously been unable to do.


Because of their young age and the functional level of the children, standing balance on 1 foot could not be attained by any of the children. Therefore, standing balance on 2 feet was used as a measure of balance, which could only be attained by 2 of the children. Both the children approximately doubled their standing balance time from pre-intervention to post-intervention and continued to show gains at the 1-month follow-up.


The findings of this study add to the body of knowledge that functional standing and walking skills can be improved by intensive treadmill training in children with CP. In this study, young children of ages 2.5 to 3.9 years with various types of CP and different functional levels were able to make significant improvements in their walking ability, as measured by walking distance and gait speed. They also showed improvement in functional gross motor skills related to standing and walking. These changes, in general, were greater in children with higher GMFCS levels compared to children with lower GMFCS levels at study onset. Additionally, no adverse effects such as excessive fatigue or harness discomfort from the intensive LTT training program used in this study were observed.

Five of the 6 children were able to improve in their functional standing skills as measured by Dimension D and were able to maintain those skills at the 1-month follow-up. The child who did not show any changes in Dimension D was ambulatory with a walker at the onset of the study but had a diagnosis of chronic lung disease causing him to miss 2 training sessions because of respiratory illness. This child did, however, make improvements in his functional walking skills as measured by Dimension E of the GMFM. An additional 4 children showed improvement in Dimension E. One child with spastic CP did not improve in this dimension. This child, who was nonambulatory GMFCS level IV, and dependent in all gravity-dependent positions, had to use partial weight support on the treadmill throughout the entire study, and had strong lower extremity spasticity. Schindl et al11 found that after a 3-month intensive treadmill-training program a decrease in assistance was required for ambulation in 3 of 6 children ages 6 to 18 years with spastic tetraparesis who were previously nonambulatory. This suggests that improvements in gross motor skills related to ambulation might necessitate a prolonged intensive treadmill training program for children with lower GMFCS levels and higher levels of spasticity.

In an effort to assess the children’s functioning in societal roles according to the International Classification of Functioning, Disability, and Health Model of the World Health Organization,27 we used the PEDI Caregiver Assistance Scale.22 Caregiver assistance is an important factor that contributes to a child’s ability to participate in society at this young age. Parents and guardians play a central role in the child’s ability to fulfill social roles at this age and are often the sole providers of support during outings in the community.28 This was particularly true for the 3 children in this study who were nonambulatory and who completely relied on their caregivers’ ability to provide mobility because none of the children had access to independent power mobility devices. Although these 3 children used gait trainers, they required maximum assistance for ambulation and were usually in adaptive strollers or carried by their parents during community outings. It might be debatable whether the significant results in the caregiver portion of the Mobility Scale represented increased participation in societal roles in these children, but it clearly reflected a decreased burden on caregivers. Indeed, the PEDI is a reliable and valid assessment tool that reflects caregivers’ perceptions of the performance of their child and is sensitive to change over time.29

The primary emphasis of this study was to enable young children with CP to take independent steps on the treadmill with as little support or facilitation as possible. This invariably led to more mistakes during LTT, and a chance for the children to self-correct before outside correction was provided. Although the children received ongoing explicit verbal feedback, tactile feedback was kept to a minimum, and the children were not corrected regarding step height or step length. This is in contrast to other studies in which facilitation was provided at the hips, knees, and feet by 1, 2, and, in some instances, 3 therapists.11,14,15,17,19 Furthermore, the amount of weight support was decreased, and the treadmill speed was increased as quickly as possible throughout the intervention period. Three of 4 children who required weight support at study onset no longer required weight support at the end of the intervention period. All children increased their treadmill speed by at least 100%, and 1 child increased her speed by 500% by the end of the 4-week intervention period. Increases in treadmill speed have been previously reported in school-age children with CP12,16 and in studies with adults after stroke30 but were not as large as in this sample of young children with CP. Begnoche and Pittetti16 reported increases in treadmill speed of 50% to 80% over 4 weeks in a study of 5 children with a mean age of 6.7 years. Provost et al12 progressed the treadmill speed by 61% to 65% in 6 children with a mean age of 10.5 years. However, the starting speeds in our study were lower than those in other studies because of the young age and the inexperience of our subjects in walking on a treadmill. It was observed initially that the children with lower GMFCS levels tended to sit in the harness during treadmill training and seemed unsure how to move their legs—a trend that was reversed with the training at progressively higher treadmill speeds and decreased weight support. This finding is consistent with studies of school-age children when weight support was decreased as quickly as possible to allow maximal active involvement of the children.12,16 This improvement in stepping ability with decreased weight support might have been triggered by an increase in proprioceptive input from the increased joint pressure throughout the kinetic chain.31 Similar increases in stepping ability with decreased weight support over time have been observed in pediatric patients with Down syndrome when the children received additional proprioceptive input by use of ankle weights during treadmill walking.32–34 Additionally, improved stepping ability at faster treadmill speeds may have been due to a greater stretch of the hip flexors, leading to an increased activation of central pattern generators mediated at the spinal level.35

We saw significant decreases in the 10-m walk test times and significant increases in the 6-minute walk test distance in our young subjects. In contrast to other studies in which the timed 10-m walk test was used to assess comfortable walking speed,12,14,16 we used it to examine maximum walking speed over 10 m. Although other investigators have observed improvements in self-selected walking speed over 10 m in children of ages 5 to 18 years,12,14,16 our study is the first to show improvements in maximal walking speed over 10 m in children younger than 4 years of age with CP as a consequence of locomotor training. Three children with GMFCS levels I and II were able to complete the timed 10-m walk test and the 6-minute walk test as part of the pretest. By the posttest, 1 additional child at GMFCS level III was able to complete the 10-m walk test by independently moving his gait trainer. At the 1-month follow-up, 2 additional children at GMFCS level IV were able to walk 10 m with their gait trainers. However, the 3 children who used gait trainers for locomotion continued to need assistance for steering and were able to move their gait trainers only on level, smooth ground. Although these gains were only functional in an optimal environment without barriers, they were reported as major improvements by these young children’s parents. Similar results have been found after a 4-month LTT intervention protocol in 4 toddlers ages 1.7 to 2.3 years.19 However, these toddlers were younger than the subjects in our study, and the intervention period spanned a 4-month period, indicating that maturation may have been responsible for some of the positive changes. Although some degree of maturation cannot be ruled out in the 2-month period between pre-intervention assessment and 1-month follow-up, the greatest gains in our study were made immediately after the 4-week intervention period, probably indicating that the children improved because of intervention rather than maturation. Moreover, we cannot discount the role of continued practice after the intervention period, which may have played an important role in the improvements seen at the 1-month follow-up.

Our self-selected walking speed was calculated from the 6-minute walk test, and all our subjects showed improvements attributable to the LTT. These improvements in self-selected walking speed were similar to those found by Provost et al,12 who reported significant improvements in self-selected walking speed in children who were ambulatory ages 5 to 18 years after 6 weeks of LTT. However, in 2 other studies on school-age children,14,16 improvements in self-selected walking speed did not reach statistical significance. The 3 children in our study who were ambulatory with supporting devices at study onset made relatively larger gains than did those who were nonambulatory. Their ages were 2.5, 3.1, and 3.9 years. Children who develop typically in this age range achieve self-selected mean walking speeds of 0.86 to 0.99 m/sec, whereas mean self-selected walking speed increases to 1.12 and 1.16 m/sec by kindergarten age and school age, respectively.36,37 The children in our study who were ambulatory showed self-selected walking speeds of 0.14, 0.55, and 0.59 m/sec by the 1-month follow-up, which is still considerably slower than those of peers with typical development. However, their maximum walking speeds by the 1-month follow-up were 0.24, 0.86, and 1.34 m/sec, indicating that, with effort, these children could temporarily keep up with their peers.38

There were significant improvements in the distance walked during the 6-minute walk test between preassessment and 1-month follow-up in our study. The lack of significance between preassessment and postassessment (immediately after the intervention period) might indicate that a period longer than 4 weeks is necessary to make significant physiological changes in endurance in young children with CP. These results are similar to those of 2 other studies that did not find significant improvements in the 6- or 10-minute walk test after a 6-week LTT intervention that was offered 2 times per week to children with CP aged 5 to 14 years.12,14 These findings indicate that prolonged, more intensive LTT programs might be necessary to make physiological changes in endurance.

An additional interesting finding was the significant correlation (r = 0.98; p = 0.004) between self- selected walking speed in the 6-minute walk test and walking speed on the treadmill. We determined treadmill walking speed based on the children’s ability to step without dragging their feet for more than 5 seconds. This indicates that the criterion for selecting treadmill speed in this study was a good reflection of the child’s self-selected over-ground walking speed. This finding suggests that a reliable, clinical estimate of a child’s gait speed could be made by observing the ability to advance the legs on the treadmill without dragging the feet for more than 5 seconds. This might prove a useful approach for practitioners when selecting treadmill walking speed for children with CP (Table 3).

Comparison of Walking Speed Across Tests (m/sec)


A limitation of this study is the lack of a control group for this convenience sample of children with CP. Because our study occurred over a 2-month period, history and maturation effects cannot be ruled out; however, none of the children engaged in other intensive interventions during the study, except for their regularly scheduled physical therapy sessions. Additional limitations are the small sample size and the lack of homogeneity in our group, with children having different types of CP and GMFCS levels I through IV. The results of this study are preliminary for this small sample of young children, limiting external validity. Although a blinded assessor scored the GMFM from videotapes of the children at preassessment, postassessment, and 1-month follow-up, the walking and balance tests were conducted by the same nonblinded assessor. Although these tests were measured objectively by timing the children and measuring the walking distance, the children might have performed better because they were more comfortable with the assessor, the facility, or both. The parents of the children in this sample were motivated to drive to a facility 3 times per week and were able to take time out of their week to participate in the intervention. This type of intensive training might be limited to children whose parents are able to provide time and transportation. Outside the research setting, it may be difficult to obtain funding for this type of intensive treadmill training, creating a socioeconomic advantage to those families who can afford it.


The results of this study provide preliminary evidence that short-term intensive treadmill training improves measures of gross motor function, maximum and self-selected walking speed, and walking distance in a small sample of young children with CP 2.5 to 3.9 years of age. These changes were maintained after a 1-month follow-up period, potentially indicating long-term changes. Although the results of this study are encouraging, this type of intensive intervention at a facility requires a large time commitment for parents and children and may be difficult to fund outside the research setting. Other researchers have found home-based intervention on small portable treadmills feasible and beneficial in children with Down syndrome.33 Although one study exists to date that showed the feasibility of LTT in an infant at high risk for developmental disabilities,18 future research should examine whether home-based treadmill training is feasible for young children with CP and whether it can lead to a more convenient and cost-effective delivery model for intensive intervention.


The authors thank Physical Edge, a private physical therapy practice in Davis, Calif, for providing their space and support during the intervention period. They thank the parent groups, physical therapists, and other early intervention providers in the community who referred children to this study. They also thank the student volunteers Jennifer Reynolds, Erin Turner, Marina Nguyen, and Jessica Johnson from the University of California, Davis, Calif, and Seth Mailloux and Kyleigh Short from Sacramento State University, Sacramento, Calif. Part of the equipment used in this study was obtained with a Scholarly/Artistic Activity Grant from the University of the Pacific, Stockton, Calif. The treadmill was provided on loan from Mobility Research, Tempe, Ariz. Most importantly, the authors thank the children and their parents who participated in this study.


1. Leonard CT, Hirschfeld H, Forssberg H. The development of independent walking in children with cerebral palsy. Dev Med Child Neurol. 1991;33:567–577.
2. Lepage C, Noreau L, Bernard P. Association between characteristics of locomotion and accomplishment of life habits in children with cerebral palsy. Phys Ther. 1998;78:458–469.
3. Wilmshurst S, Ward K, Adams JE, et al. Mobility status and bone density in cerebral palsy. Arch Dis Child. 1996;75:164–165.
4. Chien F, DeMuth S, Knutson L, et al. The use of the 600 yard walk-run test to assess walking endurance and speed in children with cerebral palsy. Pediatr Phys Ther. 2006;18:86–87.
5. Hutton JL, Pharoah P. Effects of cognitive, motor, and sensory disabilities on survival in cerebral palsy. Arch Dis Child. 2002;86:84–90.
6. Rosenbaum P, Walter S, Hanna S, et al. Prognosis for gross motor function in cerebral palsy: creation of motor development curves. JAMA. 2002;288:1357–1363.
7. Schmidt RA, Young DE, Swinnen S, et al. Summary knowledge of results for skill acquisition: support for the guidance hypothesis. J Exp Psychol Learn Mem Cogn. 1989;15:352–359.
8. Newell KM. Motor skill acquisition. Annu Rev Psychol. 1991;42:213–237.
9. Hadders-Algra M. Early brain damage and the development of motor behavior in children: clues for therapeutic intervention? Neural Plast. 2001;8:31–49.
10. Unnithan V, Kenne E, Logan L, et al. The effect of partial body weight support on the oxygen cost of walking in children and adolescents with spastic cerebral palsy. Pediatr Exerc Sci. 2006;17:11–21.
11. Schindl MR, Forstner C, Kern H, et al. Treadmill training with partial body weight support in nonambulatory patients with cerebral palsy. Arch Phys Med Rehabil. 2000;81:301–306.
12. Provost B, Dieruf K, Burtner P, Phillips J, et al. Endurance and gait in children with cerebral palsy after intensive body weightsupported treadmill training. Pediatr Phys Ther. 2007;19:2–10.
13. McNevin NH, Coraci L, Schafer J. Gait in adolescent cerebral palsy: the effect of partial unweighting. Arch Phys Med Rehabil. 2000;81:525–528.
14. Dodd KJ, Foley S. Partial body-weight-supported treadmill training can improve walking in children with cerebral palsy: a clinical controlled trial. Dev Med Child Neurol. 2007;49:101–105.
15. Day J, Fox EJ, Lowe J, et al. Locomotor training with partial body weight support on a treadmill in a nonambulatory child with spastic tetraplegic cerebral palsy: a case report. Pediatr Phys Ther. 2004;16:106–113.
16. Begnoche D, Pitetti K. Effects of traditional treatment and partial body weight treadmill training on the motor skills of children with spastic cerebral palsy: a pilot study. Pediatr Phys Ther. 2007;19:11–19.
17. Cherng R, Liu C, Lau T, Hong R. Effect of treadmill training with body weight support on gait and gross motor function in children with spastic cerebral palsy. Phys Med Rehabil. 2007;86:548–555.
18. Bodkin AW, Baxter RS, Dobkin BH. Treadmill training for an infant born preterm with a grade III intraventricular hemorrhage. Phys Ther. 2003;83:1107–1118.
19. Richards CL, Malouin F, Dumas F, et al. Early and intensive treadmill locomotor training for young children with cerebral palsy: a feasibility study. Pediatr Phys Ther. 1997;9:158–165.
20. Mattern-Baxter K. Effects of partial body weight supported treadmill training on children with cerebral palsy. Pediatr Phys Ther. 2009;21:12–22.
21. Russell DJ, Avery LM, Rosenbaum PL, et al. Improved scaling of the gross motor function measure for children with cerebral palsy: evidence of reliability and validity. Phys Ther. 2000;80:873–885.
22. Feldman A, Haley S, Coryell J. Concurrent and construct validity of the pediatric evaluation of disability inventory. Phys Ther. 1990;70:602–610.
23. 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.
24. Boyd R, Fantone S, Rodda J, et al. High- or low-technology measurements of energy expenditure in clinical gait analysis? Dev Med Child Neurol. 1999;41:676–682.
25. Li AM, Yin J, Yu CCW, et al. The six minute walk test in healthy children: reliability and validity. Eur Respir J. 2005;25:1057–1060.
26. Portney L, Watkins M. Foundations of Clinical Research. 3rd ed. Upper Saddle River, NJ: Prentice Hall; 2009.
27. Rosenbaum P, Stewart D. The world health organization international classification of functioning, disability, and health: a model to guide clinical thinking, practice and research in the field of cerebral palsy. Semin Pediatr Neurol. 2004;11:5–10.
28. Østensjø S, Brogren E, Carlberg E, et al. Everyday functioning in young children with cerebral palsy: functional skills, caregiver assistance, and modifications of the environment. Develop Med Child Neurol. 2003;45:603–612.
29. Nichols D, Case-Smith J. Reliability and validity of the pediatric evaluation of disability inventory. Pediatr Phys Ther. 1996;8:15–24.
30. Sullivan K, Brown D, Klasses T, et al. Effects of taskspecific locomotor and strength training in adults who were ambulatory after stroke: results of the STEPS randomized clinical trial. Phys Ther. 2007:1580–1600.
31. Lam T, Pearson K. The role of proprioceptive feedback in the regulation and adaptation of locomotor activity. Adv Exp Med Biol. 2002;508:343–355.
32. Ulrich B, Ulrich D, Collier D, et al. Developmental shifts in the ability of infants with Down syndrome to produce treadmill steps. Phys Ther. 1995;75:14–23.
33. Ulrich DA, Lloyd MC, Tiernan C, et al. Effects of intensity of treadmill training on developmental outcomes and stepping in infants with Down syndrome. Phys Ther. 2008;88:114–122.
34. Behrman AL, Nair PM, Bowden MG, et al. Locomotor training restores walking in a nonambulatory child with chronic, severe, incomplete cervical spinal cord injury. Phys Ther. 2008;88:580–590.
35. Dimitrijevic MR, Gerasimenko Y, Pinter MM. Evidence for a spinal central pattern generator in humans. Ann NY Acad Sci. 1998;860:360–376.
36. David KS, Sullivan M. Expectations for walking speeds: standards for students in elementary school. Pediatr Phys Ther. 2005;17:120–127.
37. Waters R, Lunsford BR, Perry J, et al. Energy-speed relationship of walking: standard tables. J Orthop Res. 1988;6:215–222.
38. Sutherland DH, Olshen RA, Biden EN, et al. The Development of Mature Walking. London: MacKeith Press; 1988.

activities of daily living; age factors; cerebral palsy/therapy; child/preschool; exercise therapy; gait; human movement system; physical therapy modalities/instrumentation; treatment outcome; walking

© 2009 Lippincott Williams & Wilkins, Inc.