Abstract
Purpose:
To evaluate changes in gross motor function in children with cerebral palsy (CP), following multiple intensive rehabilitation periods.
Materials and methods:
In this retrospective longitudinal study, we assessed 46 children with CP, aged 2–12 years, Gross Motor Function Classification System (GMFCS) levels I–V. Participants underwent 2–12 intensive rehabilitation periods, incorporating goal-directed group activities. Long-term change in gross motor function was analyzed using a linear mixed model based on reference percentiles for the Gross Motor Function Measure-66 (GMFM-66). Comparisons were made between GMFCS levels and age groups.
Results:
GMFM-66 scores increased during each intensive rehabilitation period, with significant improvements in period 1–4. The number of participants decreased as the number of periods increased. GMFM-66 percentiles increased by 10.9 (p < 0.001) per period, with an additive gain of 2.4 (p = 0.124) GMFM-66 percentiles per added intensive rehabilitation period. A decrease in percentiles was observed between periods. Gross motor function enhanced across all GMFCS levels, particularly levels I and V. Children under 6 years showed greater improvements than older children.
Conclusions:
Intensive rehabilitation may be associated with improved gross motor function in children with CP, irrespective of GMFCS levels. Repeated periods appear especially beneficial in early childhood, supporting their strategic use in long-term rehabilitation.
Introduction
Cerebral palsy (CP) is the most common childhood disability () and is traditionally defined as “a group of permanent disorders of the development of movement and posture, causing activity limitation, that are attributed to non-progressive disturbances that occurred in the developing fetal or infant brain” (). More recently, CP has been conceptualized as an early-onset, lifelong neurodevelopmental condition characterized by limitations in activity due to impaired development of movement and posture, resulting from non-degenerative brain dysplasia or injury in early life, and frequently accompanied by additional impairments across multiple domains (). CP is associated with a wide range of activity and participation limitations (, ), with severity of impaired gross motor function classified by the five-level Gross Motor Function Classification System (GMFCS) (). Motor impairments are often accompanied by secondary musculoskeletal conditions and disturbances in sensory-, cognitive-, communicative- and behavioral functions (), which consequently may require a multidisciplinary approach throughout life. As activity limitations in gross motor function are a core symptom in CP (), identifying strategies to improve gross motor skills is important for enhancing the quality of life (, ). Improving gross motor function may positively influence a child's ability to navigate their environment, engage in play, and interact in educational and community activities (, ).
The International Classification of Functioning, Disability, and Health (ICF) (), along with system theories on motor learning (, ), consider gross motor progress a result of interactions between multiple factors, including child characteristics, environmental factors and interventions (). In recent years, the CP treatment evidence base has expanded rapidly, providing information about more effective interventions (). High-intensity, task-oriented training that incorporates real-life, goal-directed activities and motor learning principles (e.g., intensity, specificity, salience) has shown positive effects on gross motor function in children with CP (, –). Thus suggesting that training must be sufficiently intense to stimulate experience-dependent neural plasticity ().
Motor function serves as the most common outcome parameter in research involving individuals with CP. This is frequently evaluated using the Gross Motor Function Measure-66 (GMFM-66) due to its robust psychometric properties (, ). Particular attention is directed towards intensive treatments, indicating that high-intensity training is effective in improving short-term gross motor progress in children with CP (–). In Norway, intensive rehabilitation programs function as a supplemental approach to local habilitation services, with the primary objective of amplifying activity levels and participation in daily life (, ).
Generally, children classified as GMFCS levels I–II exhibit more pronounced improvements in gross motor functions after intensive rehabilitation compared to those classified in levels III–V (, ). Similarly, the most significant gross motor progress primarily occurs early in life and in younger children (, , ). However, it should be noted that children with CP typically show an increase in gross motor function up to the age of 6–7 years under standard care (, ). Long-term changes in gross motor function have only been explored in a limited number of studies (, , ), which have reported cumulative effects on gross motor function following multiple periods of intensive rehabilitation. Nevertheless, the intensity and longevity of the intensive periods vary significantly in the studies, resulting in insufficient evidence to the role the amount and intensity of intensive rehabilitation plays in changes in gross motor function beyond natural progression.
Longitudinal monitoring of gross motor function allows for the identification of developmental trajectories and facilitating early intervention strategies (). Assessing the long-term implications intensive rehabilitation has on changes in gross motor function may provide important insights for healthcare professionals in optimizing treatment plans for children with CP. Therefore, the primary aim of this longitudinal study was to evaluate which changes in gross motor function could be observed in children with CP after repeated periods of intensive rehabilitation, assessed using the GMFM-66. The secondary aim was to investigate whether changes in gross motor function differ between GMFCS levels, and if children exhibit different patterns of change dependent on age groups.
Materials and methods
Design
This study uses a retrospective longitudinal research design, incorporating repeated data from the Children's Physiotherapy Center (CPC) in Bergen, Norway, over 6 years, from 2017 to 2023. The sample of children was obtained from the data materials of CPC's project “Intensity matters” (). Participants were recruited from different regions of Norway. Inclusion criteria of the original sample were children diagnosed with significantly reduced motor function due to congenital or early acquired brain damage, from the age of 6 months to 16 years. Exclusion criteria were parents with poor understanding of Norwegian or English.
Participants
Forty-six children aged between 2 and 12 years, diagnosed with CP across all GMFCS severity levels, were included in this study. The selection aimed to include children with CP across all severity levels, and to use gross motor reference percentiles, developed for children with CP, as the primary dependent variable for evaluating longitudinal changes.
Intervention
The multidisciplinary intensive rehabilitation program at the CPC was carried out by a team of physiotherapists, occupational therapists, sports educators, and pediatricians. The program consisted of two three-week periods over a year, incorporating group training sessions Monday through Thursday, each lasting 5 h (). Training groups, composed of 3–6 children and their parents, were tailored to align participants in terms of age, cognitive and functional ability. Before the intervention, individualized goals and plans were developed in collaboration with each child, their parent or guardian, and the therapist. Activities were designed to promote mastery, function, enjoyment of movement, development, and independence, with a focus on gross motor skills and goal-directed activities relevant to daily life, such as riding a bike, walking outdoors and climbing stairs (). Tasks were a combination of structured group activities and individually tailored, goal-directed tasks.
The parents were active in the training and received close follow-up from the therapists (). The parents learned through “hands on” training how they could continue training, stimulation, and facilitation at home. Contact was established between the CPC and the health service in the child's municipality to exchange knowledge about the participants and their goals. Between intensive rehabilitation periods, children received follow-up from local physiotherapists, focusing on set goals. However, the amount or content of local follow-up is not specified.
Children were initially offered two intensive rehabilitation periods. Eligibility for subsequent periods was determined by regional health authorities and, upon new referral, granted in sets of two. Recruitment occurred continuously from 2017 to 2023, resulting in variation in the total number of periods completed. Those enrolling earlier in the study could complete more intensive rehabilitation periods than those enrolled later. Consequently, the number of periods per child depended on enrollment timing and the study's data collection period. Occasionally, periods were postponed due to illness, medical treatment, or family circumstances, leading to long intervals of 2–3 years before resuming participation; however, such instances were rare.
Outcome measures
The GMFM-66 was used to monitor changes in gross motor function in children with CP over time, due to its strong psychometric properties (, , –). The GMFM-66 assesses performance across five dimensions of gross motor activities, with a four-point scoring scale based on the extent of skill mastery rather than the quality of performance (). The scores are calculated on a 0–100 interval scale, with children expected to start near 0 as newborns and acquire skills rapidly in early years, slowing as they reach their potential (). The minimum clinically important difference (MCID) in change score has been studied, with a score of 1.58 indicating general improvement and a score of 3.71 indicating a large improvement ().
In 2002, Rosenbaum and colleagues () published five motor development curves, corresponding to the GMFCS levels. These curves illustrated changes in GMFM-66 scores across severity levels, indicating both development pace and an anticipated ceiling for functional capabilities (). The curves typically show rapid progression to peak GMFM-66 scores at younger ages, exhibiting a leveling of scores in levels I and II, or a decline followed by leveling for level III–V ().
To enable more nuanced evaluations, Hanna et al. (, ) developed reference percentiles, providing an appropriate normative interpretation of GMFM-66 scores by age within each GMFCS level. These percentiles account for natural maturation and development, as they are based on a longitudinal sample of children with CP, reflecting typical age-related changes in motor function (). Despite large variability (), children with CP are generally expected to follow their GMFM-66 percentile over time (), thus a positive change in the GMFM-66 percentile may indicate better motor development beyond the expected curve. Hanna et al. () reported mean percentile changes of 3.0, −0.8, 3.3, 2.5, and 3.6 for GMFCS levels I–V, respectively. Furthermore, a second assessment indicated about a 50% chance of being within ±10.5, 10.5, 8.4, 8.0, and 8.9 percentiles for each GMFCS level. In this study, GMFM-66 percentiles were used as the primary outcome measure, due to their ability to account for age-related expectations within each GMFCS level.
Method of data collection
Children were assessed using the GMFM-66 on the first and last day of the program, and assessments were conducted at the CPC. The assessors were experienced physiotherapists familiar with the GMFM-66 measure. In most cases the same assessor was involved in both administering the pre- and post GMFM-66 testing and contributing to the group intervention as part of a multidisciplinary team.
Statistical analysis
Statistical analyses were performed in IBM SPSS Statistics (Version 29) and Microsoft Excel. Descriptive statistics are presented as count (%) for categorical variables, and as mean (SD, min-max) for continuous variables. The Wilcoxon Signed Rank Test was applied to analyze changes throughout each intensive period.
The association between repeated periods of intensive rehabilitation and changes in gross motor function was analyzed utilizing a linear mixed model, with GMFM-66 percentiles as the primary dependent variable. The covariates were post-test (vs. pre-test), and the number of intensive rehabilitation periods. We included child as a random effect, with random intercept and random slope for the number of intensive rehabilitation periods. GMFM-66 scores were converted to GMFM-66 percentiles and subsequently to z-scores. A z-score equals the number of standard deviations from the mean. To better interpret the clinical implications of the results, outcomes using both percentiles and z-scores as dependent variables were reported. Participants with missing data at one or more time points are included in the linear mixed models with their available data. Results from linear mixed model analyses are unbiased if data are missing at random, while a complete case analysis including only participants with complete data would be unbiased only if data are missing completely at random, and also have less power ().
A visual inspection of box plots revealed similar patterns in both models, whether using z-scores or GMFM-66 percentiles as the dependent variable. Consequently, due to its clinical interpretability, we present results for the GMFM-66 percentiles as the primary dependent variable.
To provide a more accurate representation of the intervention's impact on gross motor function according to the level of impairment and age progress, each GMFCS level and children under and over 6 years (at baseline) were analyzed separately. We conducted additional analyses that incorporated GMFCS or age group and their interactions with posttest and intensive rehabilitation periods. The selection of an age division under and over 6 years is based on the understanding that the most significant gross motor progress primarily occurs in early childhood and in younger children (). Similarly, children typically show a plateauing in their GMFM-scores at that age ().
To provide some protection against false positive findings due to multiple tests, we regard two-sided p-values under 0.01 to represent statistical significance.
Ethical considerations
This study utilizes previously collected data, where the ethical approval was obtained from the Regional Committee for Medical and Health Research Ethics (2018/1781/REK vest) and by the Norwegian Social Science Data Service (project no. 898475). No protocol has been registered in ClinicalTrials, as the study is based on established practice. Parents or legal guardians signed informed consent letters before enrollment. If parents did not consent to participation, it did not impact their child's rehabilitation services.
Results
Demographical data and intervention periods
Demographic statistics are shown in Table 1. The age at the start of the intervention ranged from 24 to 131 months, with a mean of 61.8 months. All GMFCS levels were represented, but levels I, III and IV were most common. The majority had spastic bilateral CP. There were slightly more boys participating.
Table 1
| Variable | n | % |
|---|---|---|
| Age at baselinea, in months; mean (SD), min-max. | 61.8 (30.5), 24–131 | |
| Sex | ||
| Female | 21 | 45.7 |
| Male | 25 | 54.3 |
| GMFCS level | ||
| I | 10 | 21.7 |
| II | 5 | 10.9 |
| III | 10 | 21.7 |
| IV | 14 | 30.4 |
| V | 7 | 15.2 |
| CP subtypes | ||
| Spastic unilateral | 10 | 21.7 |
| Spastic bilateral | 29 | 63.0 |
| Ataxic | 2 | 4.3 |
| Dyskinetic | 4 | 8.7 |
| Unclassified | 1 | 2.2 |
Demographics of participants (N = 46).
Categorical variables are presented as count (%), and continuous variables as mean (SD, min-max). GMFCS, gross motor function classification system.
First intervention period.
All participating children received at least one period with intensive rehabilitation, with GMFM-66 assessments pre- and post-intervention. While the program was initially planned to consist of two three-week periods over the course of a year, the intervals between these periods varied. Intervals between intensive rehabilitation periods ranged from 3 to 38 months, with a median (IQR) of 7.0 (5.0–8.0) months and a mean (SD) of 7.4 (4.4) months. The total distribution of intervals between intensive rehabilitation periods was calculated among participants who completed two or more periods, allowing for comparison of intervals between sessions. The number of participating children decreased as the number of intensive rehabilitation periods increased (Figure 1). Children with GMFCS level II comprised the largest group that underwent intensive rehabilitation for five or more periods.
Figure 1
Long term change in gross motor function
Analysis of pre and post GMFM-66 scores, using the Wilcoxon signed rank test, demonstrated an increase in post-scores during all the intensive rehabilitation periods. The improvements were statistically significant for periods 1–4 (Table 2). For the majority of the intensive periods, changes exceeded the estimated MCID score (), indicating a clinical meaningful progress. For the remaining periods, the number of participants decreased and the changes from pre- to post-intervention were not statistically significant (p ≥ 0.01).
Table 2
| Period | n | Pre-test median (min, max) | n | Post-test median (min, max) | Change score | p |
|---|---|---|---|---|---|---|
| 1 | 46 | 47.2 (14.8, 96.0) | 44 | 50.1 (21.3, 100) | 2.9b | <.001* |
| 2 | 38 | 48.5 (8.7, 81.9) | 38 | 51.0 (10.3, 100) | 2.5b | <.001* |
| 3 | 27 | 56.9 (19.7, 100) | 27 | 60.6 (26.0, 100) | 3.7b | <.001* |
| 4 | 18 | 60.8 (24.0, 87.9) | 17 | 64.8 (26.6, 90.6) | 4.0b | <.001* |
| 5 | 10 | 64.4 (24.0, 82.4) | 9 | 67.8 (40.7, 88.0) | 3.4b | .012 |
| 6 | 5 | 68.5 (43.6, 70.8) | 5 | 69.2 (43.4, 81.9) | 0.7 | .080 |
| 7 | 5 | 69.2 (42.2, 76.0) | 5 | 69.6 (45.3, 81.9) | 0.4 | .043 |
| 8 | 5 | 68.0 (44.1, 79.9) | 5 | 69.6 (49.6, 82.9) | 1.6b | .109 |
| 9 | 3 | 69.2 (53.8, 80.9) | 3 | 72.1 (55.1, 88.0) | 2.9b | .109 |
| 10 | 3 | 74.2 (52.1, 76.8) | 3 | 77.5 (53.8, 82.23) | 3.3b | .109 |
| 11 | 2 | 80.6 (76.0, 85.2) | 2 | 82.6 (79.9, 85.2) | 2.0b | .317 |
| 12 | 2 | 81.4 (80.9, 81.9) | 2 | 85.8 (81.9, 89.7) | 4.4b | .180 |
| 13 | 1a | 52.1 | 1a | 54.7 | 2.6b |
Comparison of median (min-max) GMFM-66 raw scores pre- and post-intervention for each intensive rehabilitation period.
The p-value cannot be computed because there is only one observation.
Change in GMFM-66 scores above the minimum clinically important difference (MCID).
p < .001, statistically significant.
To investigate long-term changes in GMFM-66 percentiles after several periods of intensive rehabilitation, a linear mixed model was used (Table 3). Each period of intensive rehabilitation resulted in an estimated mean change of 10.9 (95% CI: 8.5–13.3. p < 0.001) in GMFM-66 percentiles, surpassing the expected average change in percentile score (). An additive gain of multiple periods showed a mean increase of 2.4 GMFM-66 percentiles per added intensive rehabilitation period. However, this change was not statistically significant (p = 0.124).
Table 3
| Independent variables | Estimate | 95% CI | p |
|---|---|---|---|
| Intercept | 53.0 | 45.4 to 60.7 | <.001 |
| Post test | 10.9 | 8.5 to 13.3 | <.001 |
| Intensive rehabilitation periods | 2.4 | −0.7 to 5.6 | .124 |
Estimated mean change (95% CI) in GMFM-66 percentilesa according to each intensive rehabilitation period.
Post-test: change according to each intensive rehabilitation period, Intensive rehabilitation periods: the additive gain of multiple periods.
Dependent variable.
An inspection of pre- and post-test GMFM-66 percentiles (Figure 2) showed a positive trend, with the largest improvements during the first five periods. A decrease in percentiles was observed after each rehabilitation period, but subsequent periods typically started, on average, at a higher level. A decrease in GMFM-66 percentiles during periods 5 and 6 may partly be due to the remaining participants after period 4 having lower scores (Figure 2).
Figure 2
Comparison of changes in gross motor function across GMFCS levels
Comparisons revealed that intensive rehabilitation improved gross motor development across all GMFCS levels (p values ≤ 0.001). The greatest improvements were seen in children classified within GMFCS level I and V, with percentile changes of 18.7 (95% CI: 9.2–28.2) and 16.4 (95% CI: 7.7–25.1) respectively (Table 4).
Table 4
| GMFCS | Independent variables | Estimate | 95% CI | p |
|---|---|---|---|---|
| GMFCS I (n = 10) | Intercept | 50.8 | 32.4 to 69.2 | <.001 |
| Post test | 18.7 | 9.2 to 28.2 | <.001 | |
| Intensive rehabilitation periods | 1.1 | −8.5 to 10.7 | .801 | |
| GMFCS II (n = 5) | Intercept | 69.4 | 44.4 to 94.3 | .001 |
| Post test | 6.7 | 3.5 to 9.9 | <.001 | |
| Intensive rehabilitation periods | 2.2 | −4.8 to 9.3 | .373 | |
| GMFCS III (n = 10) | Intercept | 49.7 | 25.0 to 74.3 | .001 |
| Post test | 9.4 | 5.8 to 13.0 | <.001 | |
| Intensive rehabilitation periods | 2.9 | −2.3 to 8.2 | .225 | |
| GMFCS IV (n = 14) | Intercept | 51.3 | 37.3 to 65.2 | <.001 |
| Post test | 9.9 | 5.5 to 14.4 | <.001 | |
| Intensive rehabilitation periods | 4.4 | −3.2 to 12.0 | .216 | |
| GMFCS V (n = 7) | Intercept | 49.6 | 31.9 to 67.3 | <.001 |
| Post test | 16.4 | 7.7 to 25.1 | .001 | |
| Intensive rehabilitation periods | −13.2 | −63.9 to 37.4 | .411 |
Estimated mean change (95% CI) in GMFM-66 reference percentilesa according to each intensive rehabilitation period across GMFCS levels.
Post-test: change according to each intensive rehabilitation period, Intensive rehabilitation periods: the additive gain per period. GMFCS, gross motor function classification system. Estimates with wide CI should be interpreted with caution.
Dependent variable.
The box plots, shown in Figure 3, suggest an increase in GMFM-66 percentiles after each intensive rehabilitation period. Most observations were in GMFCS levels II and IV, with a noticeable decrease in percentiles after intensive rehabilitation, particularly in GMFCS level II and somewhat in level IV.
Figure 3
To further illustrate these findings, individual GMFM-66 scores for children in GMFCS level II were plotted against the reference percentiles from Hanna et al. (), providing a visual representation of each child's motor function trajectory relative to expected development (Figure 4). GMFCS level II was chosen, as this subgroup included the largest number of participants who completed the most intensive rehabilitation periods, allowing for a clear visualization of longitudinal changes in gross motor function. Most children showed improvements in GMFM-66 scores over time following intensive rehabilitation periods, with gains generally aligning with or exceeding expected trajectories (). Similarly, as seen in Figure 3, there is an observed decrease in GMFM-66 scores after each intensive rehabilitation period.
Figure 4
Comparison of changes in gross motor function across age
An analysis was conducted to assess the effect of age on intensive rehabilitation effectiveness, separately for children below and above 6 years of age. Out of 46 participants, 31 were under 6 years at baseline. Both groups showed a significant increase in the GMFM-66 percentile (p value <0.001), with a mean change of 11.7 (95% CI: 8.6–14.9) for children under 6 years and 9.3 (95% CI: 5.8–12.9) for those over 6 years (Table 5). Additional analyses including GMFCS level or age group and their interactions with posttest and intensive rehabilitation periods revealed no significant effects.
Table 5
| Age at baseline | Independent variables | Estimate | 95% CI | p |
|---|---|---|---|---|
| Under 6 years | Intercept | 51.7 | 42.0 to 61.4 | <.001 |
| Post test | 11.7 | 8.6 to 14.9 | <.001 | |
| Intensive rehabilitation periods | 2.5 | −2.4 to 7.3 | .301 | |
| Over 6 years | Intercept | 56.6 | 42.7 to 70.5 | <.001 |
| Post test | 9.3 | 5.8 to 12.9 | <.001 | |
| Intensive rehabilitation periods | 0.8 | −1.6 to 3.2 | .423 |
Estimated mean change (95% CI) in GMFM-66 reference percentilesa based on a linear mixed model, separately for the age groups.
Post-test: change according to each intensive rehabilitation period, Intensive rehabilitation periods: the additive gain per period.
Dependent variable.
Discussion
In our study, we aimed to evaluate changes in gross motor function in children with CP, aged 2–12 years, after several periods of intensive rehabilitation. These changes were assessed using the GMFM-66.
This study demonstrated that intensive rehabilitation may be associated with improvements in gross motor function in children, aged 2–12 years, with CP, across all GMFCS levels. An improvement in the average GMFM-66 scores following each period of intensive rehabilitation were noted. Children exceeded their anticipated GMFM-66 percentiles scores, with an additive gain over multiple periods. Significant changes were primarily observed during the initial periods, while a decrease in percentiles was evident between intervals of intensive rehabilitation. However, initial points for subsequent periods started, on average, at a higher level. Notably, improvements were most substantial in children classified as GMFCS levels I and V. Furthermore, the results were statistically significant for both age groups.
Results viewed in the context of previous research
This study corroborates previous research, asserting that intensive training can enhance gross motor function in children with CP, irrespective of GMFCS levels (, , , , , ). Our findings underscore the potential benefits of repeated periods of intensive rehabilitation for long-term functional improvements. Observed sustained improvements in gross motor function, going beyond natural progression, align with similar studies (, ). Sorsdahl () found that at least two intensive periods were required to obtain improvements in gross motor function. Størvold's research () further supports that, showing an additive gain on GMFM-66 percentiles with multiple intensive periods, a finding that resonates with our results. However, the lack of statistical significance for the additive gain in our study may hint at potential confounding variables (). This non-significant result may reflect limited statistical power due to smaller sample sizes in later periods, or a plateau effect in motor gains over time. Thus, the additive effect should be interpreted with caution.
Notably, initial periods of intensive rehabilitation may be associated with greater improvements in gross motor function than subsequent periods, an observation underpinned by prior research (). Contrary to Sorsdahl's study (), which reported minimal regression following two periods, we observed a regression between all intensive periods, which seemed to increase with the number of periods. The noted regression might reflect the adaptation period, where children assimilate newly acquired motor skills into everyday life. Although the program was designed to include two three-week periods over the course of a year, the actual intervals between periods varied. This variation may have impacted the consistency of the intervention and should be considered when interpreting the results.
In this study, significant improvements in GMFM-66 percentiles were particularly noted among children with GMFCS levels I and V. Prior studies have shown that children within GMFCS levels I and II, generally, exhibit more pronounced improvements in gross motor functions post-intensive rehabilitation compared to those classified in levels III–V (, , , ). Studies suggest that few children classified at GMFCS level V exceed their predicted percentile scores, as these individuals typically face a decline in gross motor function over time (, ). Nevertheless, given the small sample size in the current study's subgroup, these findings should be interpreted with caution. Individual variability, potential selection bias, or attrition of participants with decreased functionality, may have influenced the results. Additionally, the wide confidence intervals observed, particularly in GMFCS level V, further indicate uncertainty in these estimates and limit the precision with which conclusions can be drawn.
Our study found that children under the age of six have a more pronounced response to the intervention, which aligns with previous research (, ). The professional rationale for early intervention is based on theories of motor learning and neuroplasticity (, ). Children are generally expected to progress faster to their peak GMFM-score at younger ages, often plateauing around age 6–7 (), demonstrating a stabilization of scores in GMFCS level I and II, or decline followed by a plateau in levels III–V (). However, it's important to note that in our study significant gross motor progress was observed in both age-groups, with minimal variation. The less substantial improvements in older children may reflect their proximity to peak motor development. On the other hand, younger children might already be performing at superior levels for their age group, thereby limiting the room for further advancement. The improvement seen in the older group could also be attributed to their familiarity with organized activities, experience with functional training, and comfort in group settings (), thus reflecting the interaction between individual maturation, task-related demands, and the broader environmental context.
Factors contributing to enhanced motor function
Several factors are known to enhance motor function, encompassing child characteristics, alongside environmental factors and psychosocial factors, and characteristics with interventions (, ). The intensive rehabilitation program in this study incorporate principles know to support neuroplasticity (). It emphasized the repeated, intensive practice of meaningful, functional activities to enhance motor, communication, social, and/or intellectual development (). This may have contributed to the positive development in gross motor function observed in our study.
The repeated periods of intervention offered multiple learning opportunities, allowing for skills to be practiced, learned, and reinforced over time. This repeated exposure is beneficial for the consolidation and generalization of learned motor skills, with theories based on motor learning and neuroplasticity (). The social aspect of a group setting can stimulate motivation and encourage active participation (), in alignment with the activity and participation domains of the ICF (). In addition, the intervention is designed to be child-centered, with goals set by both the child and their parents, further bolstering motivation and personal investment in the program.
Strengths and limitations
This study addresses a clinically relevant question regarding the long-term development of gross motor function in children with CP. Our longitudinal study design, along with the use of repeated measurements over time, is a key strength of this study. Another strength is that the intervention is carried out at the same center with a well-established intensive rehabilitation program. Although conducted in a group setting, the program was individualized to each child's unique needs and abilities, ensuring targeted and personalized intervention strategies (). Active parental involvement, along with targeted communication to the local physiotherapist, were seen as strengths in the program.
Another significant strength of the study lies in its utilization of the GMFM-66. The psychometric properties have been judged excellent (, , ), and the sensitivity of the GMFM-66 to detect changes in gross motor function has been well documented. We have therefore all the reason to believe that the changes observed in our study truly reflect increase in gross motor function. The assessors were familiar with the GMFM-66 assessment tool before initiation, but were, however, not blinded. Their dual role as both assessors and contributors to the intervention could introduce bias, including potential confirmation bias due to expectations about children's progress. Although the assessors were experienced and followed standardized procedures, the lack of assessor blinding remains a limitation that may affect the internal validity of the findings.
The retrospective design and lack of a control group are important limitations, both of which reduce internal validity and limit causal inference. These limitations should therefore be considered when interpreting the results. As a result, the findings suggest an association, and definitive conclusions about causality cannot be drawn. However, the use of percentiles allows for comparisons with typical developmental trajectories, which helps to mitigate the limitation of not having a control group, as it provides a recognized benchmark for progress evaluation. Although a prospective controlled design would provide stronger internal validity and allow more robust causal inferences, it was not feasible in this context because the intervention was delivered as part of an established clinical service offered to families across Norway. Implementing randomization or withholding access to the program would have posed ethical and logistical challenges.
Another limitation pertains to the lack of formal documentation on reasons for attrition, particularly evident after the fourth and fifth intervention periods. All children were initially offered two intensive rehabilitation periods, with eligibility for additional periods determined by regional health authorities. Therefore, a child may complete the two assigned periods without requiring additional ones, reflecting completion rather than dropout. Unfortunately, no data were available as to why health authorities did not assign children to further periods. Other factors may also have contributed to children not continuing participation, including family choice, logistical challenges, and medical considerations (e.g., illness, fatigue, competing priorities). A sociocultural perspective may explain why some decline further periods of intensive rehabilitation, as the intervention requires significant time (), capacity and adaptation from both child and family. These factors may contribute to retention challenges, which should be considered when interpreting the reduced sample size in later periods and the potential for attrition bias.
While our study demonstrated improvements in GMFM-66 scores following each intensive period, the interpretation of long-term gross motor progress is more complex due to potential confounding factors. The lack of data on prior or intervening medical or surgical interventions, including selective dorsal rhizotomy (SDR), botulinum toxin and intrathecal baclofen, as well as information on socioeconomic factors () such as parental education and the presence of additional comorbid conditions, represents an important limitation and introduces potential confounding bias. Medical and surgical interventions may enhance motor function, subsequently leading to increased GMFM-66 scores (, ). Consequently, the observed improvement may not be entirely attributed to our intervention. Additionally, intellectual disability and reduced lower limb range of motion are associated with negative influences on long-term gross motor progress (), but such information was unfortunately not available in our dataset. In addition to the structured intervention at the CPC, participants may have received varying levels of follow-up from local physiotherapists between intensive rehabilitation periods, contributing to the uncertainty of attributing observed motor function changes. Data on the amount and content of local follow-up were not available. Despite briefing local therapists on pre-established goals, the potential impact of these confounders, and our inability to control them, remains a study limitation.
Clinical implications and future research
This study's findings highlight the significance of applying intensive treatment interventions when emphasizing gross motor progress. Future research should prioritize the implementation of standardized follow-up protocols and thorough documentation of local interventions to enhance the reliability of outcomes. Future research should aim to include prospective controlled designs, ideally with matched comparison groups, as the inclusion of a control group is essential for more robust comparisons and stronger causal inference. To corroborate our findings, additional long-term studies based on GMFM-66 percentiles, are recommended. The need for larger sample sizes is crucial to ensure precise effect size estimates, bolster statistical power, and mitigate selection bias risks, thus enhancing the generalizability of results, especially within smaller GMFCS level subgroups.
Conclusion
This study indicates that intensive rehabilitation may be associated with improvements in gross motor function in children with CP, aged 2–12 years, spanning all GMFCS levels. Notably, improvements appeared most substantial in children classified as GMFCS levels I and V; however, the wide confidence intervals in level V, reflecting greater variability in this subgroup, warrant cautious interpretation. The results were statistically significant for all age groups, although children under 6 years at baseline showed a higher average increase. In conclusion, these findings support the potential advantages of intensive rehabilitation in enhancing gross motor outcomes in children with CP and emphasize the value of implementing repeated periods of intensive rehabilitation. This offers important insight for pediatric physiotherapists and other health care professionals working with children with CP, in optimizing treatment plans.
Statements
Data availability statement
The raw data supporting the conclusions of this article are available upon reasonable request.
Ethics statement
This study utilizes previously collected data, where the ethical approval was obtained from the Regional Committee for Medical and Health Research Ethics (2018/1781/REK vest) and by the Norwegian Social Science Data Service (project no. 898475). The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants' legal guardians/next of kin. Written informed consent was obtained from the minor(s)' legal guardian/next of kin for the publication of any potentially identifiable images or data included in this article.
Author contributions
VO: Writing – review & editing, Conceptualization, Formal analysis, Writing – original draft, Visualization, Project administration, Data curation, Methodology, Validation. CS: Data curation, Formal analysis, Methodology, Project administration, Supervision, Investigation, Writing – review & editing, Validation, Writing – original draft, Conceptualization. KT: Data curation, Writing – original draft, Conceptualization, Project administration, Formal analysis, Methodology, Validation, Investigation, Writing – review & editing. SL: Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draft, Writing – review & editing. GS: Writing – original draft, Writing – review & editing, Formal analysis, Methodology, Conceptualization. RS: Formal analysis, Writing – original draft, Writing – review & editing, Conceptualization, Supervision, Methodology.
Funding
The author(s) declare that financial support was received for the research and/or publication of this article. This article was developed as part of a Master's degree in Health Sciences, physiotherapy for children and adolescents, at OsloMet, with the university providing support and resources for the project. The Directorate for Health and Social Affairs in Norway have financially supported the intervention. In addition, The GC Rieber Foundation has financially supported the CPC to plan, develop and execute the program. The funding sources were not involved in the study design, data collection, or interpretation of results.
Acknowledgments
We thank all children, parents, or legal guardians who participated in the study. We would also like to acknowledge the physiotherapists that collected data for the study. We used Microsoft Copilot and Sikt-AI, version GPT-4 Turbo, provided by OsloMet as an AI-tool for writing assistance during autumn 2023 and spring 2024. The tool was applied to support phrasing, grammar, and clarity. All content generated or edited with the assistance of this tool has been reviewed and verified by the authors for accuracy, references, and integrity. No AI tool is listed as author.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declare that Generative AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1.
McIntyreSGoldsmithSWebbAEhlingerVHollungSJMcConnellKet alGlobal prevalence of cerebral palsy: a systematic analysis. Dev Med Child Neurol. (2022) 64:1494–506. 10.1111/dmcn.15346
2.
RosenbaumPPanethNLevitonAGoldsteinMBaxMDamianoDet alA report: the definition and classification of cerebral palsy April 2006. Dev Med Child Neurol Suppl. (2007) 109:8–14.
3.
DanBRosenbaumPCarrLGoughMCoughlanJNwekeN. Proposed updated description of cerebral palsy. Dev Med Child Neurol. (2025) 67:700–9. 10.1111/dmcn.16274
4.
ØstensjøSCarlbergEBVøllestadNK. Motor impairments in young children with cerebral palsy: relationship to gross motor function and everyday activities. Dev Med Child Neurol. (2004) 46:580–9. 10.1017/s0012162204000994
5.
ChiarelloLAPalisanoRJOrlinMNChangH-JBegnocheDMAnM. Understanding participation of preschool-age children with cerebral palsy. J Early Interv. (2012) 34:19–3. 10.1177/1053815112443988
6.
PalisanoRRosenbaumPWalterSRussellDWoodEGaluppiB. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. (1997) 39:214–23. 10.1111/j.1469-8749.1997.tb07414.x
7.
RosenbaumPLWalterSDHannaSEPalisanoRJRussellDJRainaPet alPrognosis for gross motor function in cerebral palsy: creation of motor development curves. J Am Med Assoc. (2002) 288:1357–63. 10.1001/jama.288.11.1357
8.
LeeBH. Relationship between gross motor function and the function, activity and participation components of the international classification of functioning in children with spastic cerebral palsy. J Phys Ther Sci. (2017) 29:1732–6. 10.1589/jpts.29.1732
9.
ChiarelloLAPalisanoRJBartlettDJMcCoySW. A multivariate model of determinants of change in gross-motor abilities and engagement in self-care and play of young children with cerebral palsy. Phys Occup Ther Pediatr. (2011) 31:150–68. 10.3109/01942638.2010.525601
10.
JackmanMSakzewskiLMorganCBoydRNBrennanSELangdonKet alInterventions to improve physical function for children and young people with cerebral palsy: international clinical practice guideline. Dev Med Child Neurol. (2022) 64:536–49. 10.1111/dmcn.15055
11.
World Health Organization. International Classification of Functioning, Disability and Health (ICF). Geneva: WHO (2001).
12.
Shumway-CookAWoollacottMH. Motor Control: Translating Research into Clinical Practice. 5th ed.Philadelphia, PA: Wolters Kluwer (2017).
13.
PalisanoRJOrlinMNSchreiberJ. Campbell’s Physical Therapy for Children. 5th ed.St. Louis, Miss: Elsevier (2017).
14.
StørvoldGVJahnsenRBEvensenKAIRomildUKBratbergGH. Factors associated with enhanced gross motor progress in children with cerebral palsy: a register-based study. Phys Occup Ther Pediatr. (2018) 38:548–61. 10.1080/01942638.2018.1462288
15.
NovakIMorganCFaheyMFinch-EdmondsonMGaleaCHinesAet alState of the evidence traffic lights 2019: systematic review of interventions for preventing and treating children with cerebral palsy. Curr Neurol Neurosci Rep. (2020) 20:3. 10.1007/s11910-020-1022-z
16.
FaccioliSPaglianoEFerrariAMaghiniCSianiMFSgherriGet alEvidence-based management and motor rehabilitation of cerebral palsy children and adolescents: a systematic review. Front Neurol. (2023) 14:1171224. 10.3389/fneur.2023.1171224
17.
ZaiWXuNWuWWangYWangR. Effect of task-oriented training on gross motor function, balance and activities of daily living in children with cerebral palsy: a systematic review and meta-analysis. Medicine. (2022) 101:e31565. 10.1097/md.0000000000031565
18.
DemontAGeddaMLagerCde LattreCGaryYKeroulleEet alEvidence-based, implementable motor rehabilitation guidelines for individuals with cerebral palsy. Neurology. (2022) 99:283–97. 10.1212/wnl.0000000000200936
19.
WangHYYangYH. Evaluating the responsiveness of 2 versions of the gross motor function measure for children with cerebral palsy. Arch Phys Med Rehabil. (2006) 87:51–6. 10.1016/j.apmr.2005.08.117
20.
OeffingerDBagleyARogersSGortonGKryscioRAbelMet alOutcome tools used for ambulatory children with cerebral palsy: responsiveness and minimum clinically important differences. Dev Med Child Neurol. (2008) 50:918–25. 10.1111/j.1469-8749.2008.03150.x
21.
KachmarOMysulaIKushnirAVoloshynTMatiushenkoOHasiukMet alChanges in motor functions in children with cerebral palsy after the course of intensive neurophysiological rehabilitation: a single-blind study. Int Neurol J. (2019) 5:5–11. 10.22141/2224-0713.5.107.2019.17670
22.
KushnirAKachmarO. Intensive neurophysiological rehabilitation system for children with cerebral palsy: a quasi-randomized controlled trial. BMC Neurol. (2023) 23:157. 10.1186/s12883-023-03216-4
23.
Janssen-PottenYJMRoksLRoijenRVermeulenRJRameckersEEA. Effectiveness of functional intensive therapy on mobility and self-care activities in children and adolescents with cerebral palsy - a prospective clinical study. Disabil Rehabil. (2022) 45:3529–38. 10.1080/09638288.2022.2130445
24.
Norsk kvalitets- og oppfølgingsregister for cerebral parese. Cerebral parese (NorCP). (2024). Available online at:https://metodebok.no/cp (Accessed July 01, 2024).
25.
AndersenGLHollungSJKlevbergGLJahnsenRStadskleivKKløveN. Norsk kvalitets- og oppfølgingsregister for cerebral parese (NorCP). Årsrapport for 2021 med plan for forbedringstiltak. (2022).
26.
StarkCDuranIMartakisKSpiessKSemlerOSchoenauE. Effect of long-term repeated interval rehabilitation on the gross motor function measure in children with cerebral palsy. Neuropediatrics. (2020) 51:407–16. 10.1055/s-0040-1715489
27.
SorsdahlABMoe-NilssenRKaaleHKRieberJStrandLI. Change in basic motor abilities, quality of movement and everyday activities following intensive, goal-directed, activity-focused physiotherapy in a group setting for children with cerebral palsy. BMC Pediatr. (2010) 10:26. 10.1186/1471-2431-10-26
28.
HannaSEBartlettDJRivardLMRussellDJ. Reference curves for the gross motor function measure: percentiles for clinical description and tracking over time among children with cerebral palsy. Phys Ther. (2008) 88:596–607. 10.2522/ptj.20070314
29.
SorsdahlABMoe-NilssenRLarsenEMLundalSHRieberJSkarsteinEet alLong-term change of gross motor function in children with cerebral palsy; an observational study of repeated periods of intensive physiotherapy in a group setting. Eur J Physiother. (2020) 22:148–54. 10.1080/21679169.2018.1564363
30.
StørvoldGVJahnsenRBEvensenKAIBratbergGH. Is more frequent physical therapy associated with increased gross motor improvement in children with cerebral palsy? A national prospective cohort study. Disabil Rehabil. (2020) 42:1430–8. 10.1080/09638288.2018.1528635
31.
Barnas Fysioterapisenter. Intensiv tverrfaglig habilitering Barnas Fysioterapisenter (2023). Available online at:https://www.bfsnett.no (Accessed January 10, 2024).
32.
Arnevik AustrheimKSkagenCRieberJMelfald TvetenK. Practice, play, repeat - individualized outcomes after the “intensity matters!”-program for children with disabilities - a descriptive multicase study. Disabil Rehabil. (2024) 46:2847–52. 10.1080/09638288.2023.2233905
33.
RussellDJAveryLMRosenbaumPLRainaPSWalterSDPalisanoRJ. Improved scaling of the gross motor function measure for children with cerebral palsy: evidence of reliability and validity. Phys Ther. (2000) 80:873–85. 10.1093/ptj/80.9.873
34.
HannaSERosenbaumPLBartlettDJPalisanoRJWalterSDAveryLet alStability and decline in gross motor function among children and youth with cerebral palsy aged 2 to 21 years. Dev Med Child Neurol. (2009) 51:295–302. 10.1111/j.1469-8749.2008.03196.x
35.
RussellDJAveryLMWalterSDHannaSEBartlettDJRosenbaumPLet alDevelopment and validation of item sets to improve efficiency of administration of the 66-item gross motor function measure in children with cerebral palsy. Dev Med Child Neurol. (2010) 52:e48–54. 10.1111/j.1469-8749.2009.03481.x
36.
RussellDRosenbaumPLWrightM. GMFM (GMFM-66 & GMFM-88) User’s Manual. 2nd ed.London: Mac Keith Press (2013).
37.
HannaSEBartlettDJRivardLMRussellDJ. Tabulated reference percentiles for the 66-item gross motor function measure for use with children having cerebral palsy. (2008). Available online at:https://www.canchild.ca (Accessed Feburary 11, 2024).
38.
O'KellyMRatitchB. Clinical Trials With Missing Data: A Guide for Practitioners. Chichester: Wiley (2014).
39.
AranedaRDricotLEbner-KarestinosDParadisJGordonAMFrielKMet alBrain activation changes following motor training in children with unilateral cerebral palsy: an fMRI study. Ann Phys Rehabil Med. (2021) 64:101502. 10.1016/j.rehab.2021.101502
40.
BleyenheuftYDricotLEbner-KarestinosDParadisJSaussezGRendersAet alMotor skill training may restore impaired corticospinal tract fibers in children with cerebral palsy. Neurorehabil Neural Repair. (2020) 34:533–46. 10.1177/1545968320918841
41.
BrodalP. The Central Nervous System: Structure and Function. New York: Oxford University Press (2010).
42.
NyquistAJahnsenRBMoserTUllenhagA. The coolest I know – a qualitative study exploring the participation experiences of children with disabilities in an adapted physical activities program. Disabil Rehabil. (2020) 42:2501–9. 10.1080/09638288.2019.1573937
43.
HannaSE. Longitudinal studies of development are good – but what are they good for?Dev Med Child Neurol. (2016) 58:997–8. 10.1111/dmcn.13189
44.
Assis-MadeiraEACarvalhoSGBlascovi-AssisSM. Functional performance of children with cerebral palsy from high and low socioeconomic status. Rev Paul Pediatr. (2013) 31:51–7. 10.1590/s0103-05822013000100009
Summary
Keywords
cerebral palsy, GMFM-66, intensive rehabilitation, gross motor function, group training, longitudinal study
Citation
Opheim V, Skagen C, Tveten KM, Lydersen S, Størvold GV and Sæther R (2025) Changes in gross motor function in children with cerebral palsy following repeated intensive rehabilitation periods: a longitudinal study. Front. Pediatr. 13:1636955. doi: 10.3389/fped.2025.1636955
Received
28 May 2025
Accepted
31 October 2025
Published
18 November 2025
Volume
13 - 2025
Edited by
Anna Maria Lavezzi, University of Milan, Italy
Reviewed by
Qamar Mahmood, National Institute of Rehabilitation Medicine (NIRM), Pakistan
Kanishka Baduni, University of Georgia, United States
Updates
Copyright
© 2025 Opheim, Skagen, Tveten, Lydersen, Størvold and Sæther.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Rannei Sæther rannei.sather@ntnu.no
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.