Abstract
This study compared the efficacy of two intervention schedules of bilateral intensive training (intensive and distributed) in children with unilateral cerebral palsy.
Upper limb dysfunction is a common and disabling consequence of children with cerebral palsy (CP). Several promising therapeutic models have been reported for children with unilateral CP (UCP), including constraint-induced movement therapy (CIMT); hand–arm intensive bimanual therapy, also known as HABIT; goal‐directed therapy; hybrid therapy; and virtual reality (Fandim et al., 2021; Hoare et al., 2019; Novak et al., 2020; Ouyang et al., 2020). With most daily activities requiring cooperative use of both hands, bimanual coordination impairments can interfere with a child’s participation beyond the unilateral impairments (Cohen-Holzer et al., 2017; Souto et al., 2022). Thus, it is essential to encourage the simultaneous use of both hands to enhance functional performance and, thus, lead to the increasing development of various bilateral intensive training (BIT) programs (Brandão et al., 2018; Gardas et al., 2023; Reid et al., 2015).
BIT is based on the strategies of CIMT, aiming to address the bimanual coordination while preserving the benefits of intensive training and repetitive practices (Fedrizzi et al., 2013; Gordon et al., 2011; Klepper et al., 2017). This approach offers intensive training with graded functional activities tailored to the child’s specific impairments and interests, with a direct focus on bimanual hand use (Charles & Gordon, 2006). Instead of merely encouraging the use of the more affected hand, BIT strives for children to use it similarly to the way a typically developing child uses their nondominant hand, particularly emphasizing two-hand coordination (Charles & Gordon, 2006; Gordon et al., 2007). Hence, BIT is proven to be an effective intervention to facilitate both the motor function of affected hands and bilateral hand coordination in children with UCP (Alahmari et al., 2020; Demont et al., 2022; Fedrizzi et al., 2013; Gordon, 2011; Hung et al., 2011; Klepper et al., 2017; Sakzewski et al., 2014).
Investigating the effectiveness of different delivering ways of BIT is crucial for clients to choose the optimal and feasible intervention option. Previous studies have investigated the treatment effectiveness of BIT across different dosages (Gordon, 2011; Ouyang et al., 2020), but few have compared the effectiveness of various dosing schedules. The two common ways of delivering dosage are through a training program with long daily session duration, high frequency, and a short program length (identified as an intensive schedule) and through a program with short daily session duration, low frequency, and a longer program length (identified as a distributed schedule; Wang et al., 2023). The intensive schedule accelerates the completion of treatment in a short period, enabling participants to see results from the intervention at an early stage and often takes the form of camp-based programs. In contrast, the distributed schedule follows a more traditional clinical approach, offering training activities in natural environments that can be integrated into a child or family’s weekly routine.
The influence of dosing schedules has been extensively studied in CIMT (DeLuca et al., 2012; Wang et al., 2023; Wu et al., 2020). However, for the BIT, only Brandão et al. (2018) have investigated different dosing arrangements, demonstrating similar effectiveness between a 90-hr intervention delivered over 3-wk and two 45-hr interventions with a 6-mo resting period between them. This study provided valued insights into how to arrange the training period while maintaining the same daily session duration and frequency. However, the influence of different dosing schedules, such as delivering training activities intensively over a short program length versus distributing them over a longer program length, remains unknown and requires further investigation.
The purpose of this study was to compare the treatment effectiveness on motor skills of different dosing schedules of BIT with the same total number of training hours. The total number of training hours was 36 for each schedule, which was beyond the minimum recommended dose of 30 hr (Jackman et al., 2020). Under the equivalent training hours, the intensive BIT was delivered within 1 wk, and the distributed BIT was delivered over 8 wk. The treatment effects with the same number of intervention hours (36 hr) and the treatment effects over the same period (8 wk after treatment initiation) were investigated. In addition, previous studies have shown that intensive neurorehabilitative protocols, with various levels of family involvement and burden, can affect parental stress status (Lin et al., 2011; Wang et al., 2023). Therefore, understanding parental stress levels across dosing schedules is crucial for identifying the most effective protocols.
Method
Participants
The study was a secondary analysis research using data extracted from two registered clinical trials (ClinicalTrials.gov IDs NCT02801279 and NCT04516876). The distributed BIT group in this study was derived from the first trial, which compared the effectiveness of the Kinect-based BIT and conventional BIT for children with CP. The intensive BIT group in this study was drawn from the second trial, which investigated the feasibility, efficacy, and acceptability of a camp-based BIT for children with CP. All participants were recruited from the CP associations, medical centers, and special education systems. The inclusion criteria included congenital UCP; ages 6 to 12 yr; active extension at the wrist and metacarpophalangeal joint of the more affected hand ≥10 degrees; no excessive muscle tone before treatment; the ability to follow instructions according to medical documents, parental reports, and clinical observation; and no injections of botulinum toxin type A or operations on the hand within 6 mo.
Interventions
In this study, the core principles of BIT were providing repetitive, intensive, and structured practice of functional tasks grounded in motor learning theory, with an emphasis on using both hands in daily activities. We also incorporated various types of target movements, such as grasping; releasing; carrying; stabilizing; and in-hand manipulation into intervention activities, identified through a comprehensive literature review (Charles & Gordon, 2006). Additionally, whole-task and part-task practice were tailored to the children’s individual abilities. The amount of training for the intensive and distributed BIT groups was designed to total 36 hr. The intensive BIT and distributed BIT were different in the delivered dosing schedules. The intensive BIT has 36 hr of training in 1 week. During the intervention, participants received a theme camp–based treatment in their summer vacation. They engaged in therapeutic BIT activities at the same time, which were graded individually by occupational therapists. The distributed BIT, however, has 36 hr of training with 2.25 hr/day, twice a week, for 8 weeks. Participants in both groups were allowed to continue their usual rehabilitation care during the study period. Occupational therapists implemented the intensive BIT at the university’s occupational therapy department, whereas the distributed BIT was implemented in participants’ homes or schools. These therapists had completed a 2-day training program on the BIT protocols. The interventions were documented according to the Template for Intervention Description and Replication guidelines and are detailed in Tables A.1 and A.2 in the Supplemental Material (available online with this article at https://research.aota.org/ajot).
Outcome Measures
We examined motor outcomes using three assessments for unilateral performance and two for bilateral performance. Unilateral motor performance was assessed with the Box and Block Test (BBT; Mathiowetz et al., 1985), the Melbourne Assessment–2 (MA2; Randall et al., 2014), and the Pediatric Motor Activity Log–Revised (PMAL–R; Uswatte et al., 2012). Bilateral performances were assessed with the Bruininks–Oseretsky Test of Motor Proficiency, Second Edition (BOT–2; Bruininks, 2005), and the ABILHAND–Kids test (Arnould et al., 2004).
BBT
The BBT assesses coordination of the upper limbs by transferring blocks across the partition with one hand in 60 s (Mathiowetz et al., 1985). To verify the improvement of the more affected hands, we counted the number of blocks transferred with the more affected hand in this study. The BBT demonstrates high test–retest reliability (r = .98) and good construct validity in children with CP (Liang et al., 2021).
MA2
The MA2 contains 14 items to assess the unilateral upper limb function (Randall et al., 2014). It evaluates motor abilities through four dimensions, including range of motion, fluency, accuracy, and dexterity. The MA2 demonstrates strong psychometric properties, including high test–retest reliability (rs = .92–.98), good concurrent validity, and high responsiveness (Wang et al., 2017).
PMAL-R
The PMAL–R is a questionnaire designed to measure children’s spontaneous use of their more affected hands (Uswatte et al., 2012). There are 22 items, each rated from 0 (the child did not use the weaker arm for the activity) to 5 (the child’s weaker arm did the activity normally) points in increments of 0.5 points, that assess the frequency of usage (how often) and the quality of movement (how well) in children’s daily activities. The PMAL–R, developed for children with CP, demonstrates high internal consistency (Cronbach’s α = .93), strong test–retest reliability (r = .89), and fair concurrent and predictive validity (rs = .31–.48; Lin et al., 2012; Uswatte et al., 2012).
BOT–2
The BOT–2 assesses both fine and gross motor skills in children (Bruininks, 2005). It has been reported to exhibit excellent test–retest reliability (r = .99), good internal consistency (Cronbach’s α = .92), and strong construct validity (Deitz et al., 2007; Wuang & Su, 2009). Specifically, we conducted Subtest 3 of the BOT–2, which measures manual dexterity through five activities involving manipulation with objects with upper limbs.
ABILHAND–Kids
The ABILHAND–Kids is a parent-reported scale to evaluate children’s bilateral hand performance in daily life (Arnould et al., 2004). It contains 21 daily activities that require the use of both hands, such as opening jars, taking off a T-shirt, zipping, and other bilateral tasks. Caregivers are required to rate each item according to how difficult it is for the children to perform a task. The ABILHAND–Kids demonstrates good construct validity, strong internal consistency reliability (r = .94), and excellent test–retest reliability (r = .91; Arnould et al., 2004). The outcome of ABILHAND–Kids is converted from an ordinal score into an interval logit score for subsequent analysis.
Parenting Stress Index–Short Form
To confirm the stress level experienced by the caregivers of participants, we used the Parenting Stress Index–Short Form (PSI–SF; Yeh et al., 2001). This 5-point questionnaire consists of 36 items, with ratings ranging from 1 (strongly disagree) to 5 (strongly agree) points. A higher score on the PSI–SF indicates a lower level of stress. The PSI–SF demonstrates good reliability and solid construct validity (Yeh et al., 2001).
Measurement Time Points
Outcome assessments were conducted at pretreatment and posttreatment, with an additional follow-up evaluation at Week 8 for the intensive BIT group. Specifically, participants in the distributed BIT group underwent assessments at the beginning of the intervention (Week 0) and the end (Week 8). Participants in the intensive BIT group received assessments at the beginning of the intervention (Week 0), the end of the intervention (Week 1), and 8 wk after the beginning of the intervention (Week 8). We evaluated the posttreatment measurements at Week 1 of intensive BIT (W1int) and Week 8 of distributed BIT (W8dis) to clarify the immediate treatment effects of the two groups. Additionally, we assessed the outcomes from Week 8 of intensive BIT (W8int) and those from W8dis to minimize the impact of differences in children’s motor development caused by the different lengths of the intensive and distributed interventions.
Statistical Analysis
We conducted the statistical analyses using IBM SPSS Statistics (Version 22) with a significance level set at α < 0.05. Demographic variables between groups were analyzed using χ2 tests for categorical variables, such as gender and affected side. We assessed the normality of distribution for each group using the Kolmogorov-Smirnov test (p > .05). We analyzed continuous demographic variables, such as age, using either the independent t test or the Mann-Whitney U test, depending on the data distribution. For outcome measures that did not follow a normal distribution, a log transformation was applied before analysis, specifically for the accuracy subtest in the MA2. We conducted the outcomes comparison using analysis of covariance with pretreatment as the covariate, which could be used to assess the changes from pretreatment to posttreatment between groups. We performed two comparisons between the intensive and distributed groups. First, to purely compare the immediate treatment effects with the same dose delivered by different dosing schedules, the posttreatment outcomes of the two groups were compared (W1int vs. W8dis). Second, to assess the treatment effects and eliminate the influence of the children’s development, we conducted a comparison between the two groups at Week 8 after the beginning of intervention (W8int vs. W8dis). Moreover, the percentage of participants with improvements exceeding the minimal clinically important difference (MCID) was estimated. The MCID values for the four subdomains of the MA2 (range of motion, fluency, accuracy, and dexterity) were 2.85, 1.63, 1.97, and 1.84, respectively (Wang et al., 2017). The MCID values were 3.52 for the How Often subscale of the PMAL–R, 3.3 for the How Well subscale of the PMAL–R, and 0.925 for the BOT–2 (Lin et al., 2012; Wang et al., 2017). Currently, there is no established MCID for the ABILHAND–Kids.
Results
A total of 35 children were included in this study, comprising 15 children in the intensive group and 20 children in the distributed group. Table 1 presents the demographic characteristics and baseline clinical characteristics of participants; there were no significant differences between groups.
Demographics and Baseline Clinical Characteristics of Participants
Note. BIT = bilateral intensive training; MACS = Manual Ability Classification System.
Table 2 presents the results of participants’ unilateral and bilateral motor performances and the percentage of participants whose improvements exceeded the MCID, as well as the stress levels of their caregivers. Immediate treatment effects (W1int vs. W8dis) were compared posttreatment: There were no significant differences between groups (ps = .175–.817). However, in the comparison of the intensive BIT and distributed BIT groups at 8 wk (W8int vs. W8dis), the intensive BIT group increased significantly compared with the distributed BIT group on the frequency and quality subtests of the PMAL–R group (ps = .005–.029).
Outcome Measures at Pretreatment, Posttreatment, and the 8-wk After Treatment Begins
Note. ANCOVA = analysis of covariance; BBT = Box and Block Test; BIT = bilateral intensive training; BOT–2 = Bruininks–Oseretsky Test of Motor Proficiency, Second Edition; MA2 = Melbourne Assessment–2; MCID = minimal clinically important difference; ROM = range of motion; PMAL–R = Pediatric Motor Activity Log–Revised; PSI–SF = Parenting Stress Index–Short Form; W1int = Week 1 evaluation of intensive BIT; W8dis = Week 8 evaluation of distributed BIT; W8int = Week 8 evaluation of intensive BIT.
*p < .05.
Posttreatment of intensive BIT versus posttreatment of distributed BIT.
Follow-up at 8 wk for intensive BIT versus posttreatment of distributed BIT.
Log transformation was applied.
The results of bilateral hand performance, assessed with the BOT–2 and ABILHAND–Kids, are shown in Table 2. There were no significant differences between intensive BIT and distributed BIT groups at posttreatment (W1int vs. W8dis; ps = .28–.44) and 8 wk after the treatment began (W8int vs. W8dis; ps = .56–.90). In addition, the parents of the children in both groups were assessed with the PSI–SF at the pretreatment and posttreatment time points. For the intensive BIT group, a follow-up assessment at Week 8 was also conducted to monitor the parents’ stress status during study period. The results showed no significant increases in stress for either intervention.
Discussion
In this study, we investigated the efficacy of different dosing schedules for BIT in children with unilateral CP. Overall, there were comparable improvements between the two dosing schedules at both the immediate posttreatment and 8-wk follow-up periods. However, the intensive BIT showed a significant benefit in using the more affected hand in daily activities compared with distributed BIT, which was measured by the PMAL–R, at 8 wk after the initiation of intervention.
For the unilateral motor outcomes, the results demonstrated no significant difference between groups immediately after the intervention or at the 8-wk comparison time point, which indicated that changes of motor efficiency (as measured by the BBT) and motor quality (measured with the MA2) were similar for both the intensive and distributed programs. Although the parent-rated amount and quality use of the more affected hands in daily activities (measured with the PMLA–R) showed similar improvement after the 36-hr intervention, significant differences emerged between groups at the 8-wk comparison time point. The frequency and quality scores from the PMAL–R in the intensive BIT group demonstrated a larger improvement at Week 8 compared with the distributed BIT group. These results may indicate that once children in the intensive BIT group achieve 36 training hours in 1 week, they continued to show additional improvement in the frequency and quality use of the more affected hand at 8 wk. This finding supports the idea that transferring enhanced motor function into daily activities is a continuous process that demands sustained practice; therefore, it requires extended duration to improve consistently over time (Fedrizzi et al., 2013; Gordon, 2011). In addition, these findings also align with previous CIMT studies, which suggest that once a child completed the intervention, they may naturally continue practicing daily activities by themselves, thereby gaining more improvement in the use of the more affected hand (Wang et al., 2023). However, investigations of additional 8-wk follow-ups in the distributed program for integrating improvements into daily activities (i.e., W8int vs. Week 16 in distributed BIT) are warranted in future studies.
For the bilateral motor outcomes, study results presented similar improvements between the intensive and distributed BIT groups in bilateral coordination (as measured by the BOT–2) and parent-rated daily bilateral performance (as measured with the ABILHAND–Kids). There were no significant differences in bilateral hand use immediately after the intervention or at the 8-wk comparison time point, which indicates that the same number of training hours could lead to similar improvements on motor outcomes (Liang et al., 2023), regardless of the intensive or distributed dosing schedule. It is interesting to note that when comparing the differences between the two dosing schedules at 8 wk after the initiation of treatment, the effect of transferring enhanced motor function into daily activities only presented in the unilateral hand use, not in the bilateral daily activities. This result could be explained by the fact that bilateral hand use is more complex than unilateral performance. Children and their parents might not have been sufficiently sensitive to notice the role of the more affected hand in different bimanual tasks, or they may have lacked the motor knowledge to enhance bilateral coordination in natural contexts (Ferre et al., 2017; Liang et al., 2023). Caregivers face more challenges and difficulties in training bilateral hand performance compared with implementing unilateral practice (Gordon, 2011). Therefore, the continuous improvement of daily bilateral activities potentially requires more therapist assistance.
The intervention environment can play a significant role in the effectiveness of both intensive and distributed BIT. Previous studies have indicated that children who receive rehabilitation in home settings (e.g., CIMT) tend to achieve better motor functional outcomes compared with those in clinical settings (Rostami et al., 2012). Furthermore, group-based intervention might reduce the intensity of using the more affected hand because of the nature of the activities (e.g., waiting and taking turns; Sakzewski et al., 2015), which could potentially diminish improvement. However, in the present study, both distributed BIT, conducted individually in a natural environment, and intensive BIT, delivered in a group-based day camp setting, demonstrated similar immediate outcomes. The results suggest that the environmental differences in this study may have minimal impact on treatment effectiveness. One possible explanation is the consistent 1:1 therapist-to-child ratio maintained across both groups, which ensured program quality regardless of the different settings. Future studies could explore other therapist-to-child ratios that are more representative of actual clinical settings to more accurately capture the potential influences of context.
The assessment of parental stress levels using the PSI–SF at pretreatment, at posttreatment, and during the 8-wk follow-up time point indicated that there were no significant increases in stress for parents in either intervention group. This finding suggests that the interventions were well tolerated and did not impose additional stress on the parents, ensuring that caregivers’ stress remained stable regardless of the dosing schedules. For the intensive schedule, the camp-based design delivered intervention-embedded activities in novel themes to improve the children’s motivation and engagement. Consequently, parental stress was not mentioned as a concern. On the contrary, in the distributed schedule, previous studies have demonstrated that for children with CP and their caregivers, home-based training for 2 hr/day is more preferable and comfortable than training for 3 to 4 hr/day (Eliasson et al., 2005; Gelkop et al., 2015; Wu et al., 2020). This study arranged as intervention delivery schedules of 2.25 hr/day, twice a week for 8 wk, to fit the ecological features of local families’ routines, which did not increase parental stress. Because parental stress levels influence parents’ well-being and children’s development, delivering intervention protocols that do not increase parents’ burden is important. The stability of stress levels over time further underscores the feasibility and acceptability of the interventions, supporting their potential for broader application and long-term use across various settings.
Several limitations of this study need to be considered. First, the bimanual motor outcomes as measured by the ABILHAND–Kids focused on the perceived of difficulty level in performing bilateral tasks, without subscores for the frequency and quality as in the PMAL–R. This might present different aspects of unilateral and bilateral hand use and requires cautious interpretation of the study findings. Second, to maintain the quality of treatment, we used a 1:1 therapist-to-participant ratio in both settings. However, this design might challenge the cost-effectiveness of the intervention. Third, we did not investigate motor control strategies through kinematic measurements, which are reported to be reliable and sensitive. In future studies, researchers may consider the inclusion of kinematic analyses to broaden the understanding of motor outcomes. Fourth, the Assisting Hand Assessment (AHA) is considered the gold standard for evaluating the effectiveness of bimanual performance. Future studies should consider comparing these outcomes with the AHA for further validation. Finally, this study lacks long-term follow-up data to fully assess the sustained impact of the interventions. Because treatment effects may be maintained over time, future research should incorporate follow-up assessments at 6 mo or 1 yr after intervention completion. This would provide a more comprehensive understanding of the long-term effects and the sustainability of outcomes for both intensive and distributed BIT dosing schedules.
Implications for Occupational Therapy Practice
The study results have the following clinical implications for occupational therapy practice: Both intensive and distributed BIT showed similar motor improvements immediately after the 36-hr intervention, indicating that, with the same total training hours, a distributed schedule (36 hr over 8 wk) is as effective as an intensive one (36 hr in 1 wk). This allows clinicians and clients greater flexibility in choosing a delivery method that aligns with family routines and preferences. Intensive BIT led to greater improvement in daily use of the more affected hand at the 8-week follow-up, suggesting that once the habit of using the more affected hand is established in children with UCP, it may result in further enhancements in daily activities. Enhanced motor function transferred to daily activities was observed only in unilateral hand use, not in bilateral tasks. This may be due to the complexity of bilateral coordination; caregivers may lack the sensitivity or motor knowledge needed to support bilateral hand performance effectively in natural contexts, highlighting the need for therapist guidance.
Conclusion
This study demonstrated that both intensive and distributed dosing schedules for BIT in children with unilateral CP resulted in comparable improvements in motor outcomes immediately posttreatment. Notably, the intensive BIT schedule showed significant benefits in the use of the more affected hand in daily activities at 8 wk, suggesting that establishing the ability and habit of using the more affected hand earlier can continuously enhance its use in daily activities. For bilateral motor outcomes, the similar improvements between the two dosing schedules indicate that, rather than the distribution, total training hours are crucial for motor improvement. Parental stress levels remained stable, highlighting the acceptability of both intervention schedules. A strictly randomized controlled trial with a larger sample size is recommended for future studies to validate our findings.
Supplemental Material
Supplementary material for Effectiveness of Intensive and Distributed Bilateral Intensive Training for Children With Unilateral Cerebral Palsy
Supplementary material, sj-pdf-1-aot-10.5014_ajot.2025.051039.pdf for Effectiveness of Intensive and Distributed Bilateral Intensive Training for Children With Unilateral Cerebral Palsy by Kai-Jie Liang, Zhi-Chi Weng, Hao-Ling Chen and Tien-Ni Wang in The American Journal of Occupational Therapy
Footnotes
Acknowledgments
We thank the children and their families for participating in this study.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
