Introduction

Cerebral palsy (CP), defined broadly as “a non-progressive motor impairment syndrome caused by a problem in the developing brain,” is the most common pediatric physical disability, with an incidence of 2.11 per 1,000 live births [1]. Children with hemiparesis or substantially greater deficit in one upper extremity than the other have been reported to account for a considerable 15.3~36% of the CP population [2–5]. Such children tend not to use the affected hand, which leads to a decrease in strength and motor control and further interferes with the development of bilateral coordination in multiple domains (e.g., play, self-care, and daily activities) [6]. The tendency of a child to underuse or disuse the affected upper limb is called developmental disregard [7–9].

Developmental disregard has been referred to as a process wherein a child learns not to use the affected hand during the development of motor skills [2, 9, 10]. It is observed in 50%-90% in children with hemiplegic cerebral palsy [8, 11]. Children with hemiplegia have received positive reinforcement from successful experiences of using the unaffected hand and negative reinforcement from unsuccessful experiences of using the affected hand. The long-term interaction between positive and negative reinforcements causes them to develop adaptive strategies of preferring only to use the unaffected hand in daily activities, regardless of the preserved capacity of the affected hand [12]. This lack of spontaneous use of the affected hand in bimanual tasks is known as developmental disregard [8, 13, 14], and the discrepancy between the ability to use (forced-use capacity) and actual use (real-world performance) of the affected hand is identified as the severity of the developmental disregard.

Few assessments have been developed to investigate the phenomenon of developmental disregard in children with CP. The Pediatric Motor Activity Log-Revised (PMAL-R) [15], revised from the Motor Activity Log used in adult stroke patients [16, 17], is a questionnaire-based instrument for assessing how much and how well a child actually uses his or her more-affected arm in daily life. While the PMAL-R gathers meaningful data on the spontaneous use of the affected upper limb in daily life, it also includes many unilateral tasks (e.g., turning on or off a light, picking up small objects, and pointing at a target). Since it is common to use the dominant hand to perform unilateral tasks, even in the normal population, evaluating the use of the affected hand with bimanual activities could more directly capture the real upper limb performance [11, 14, 18]. Furthermore, the PMAL-R, a questionnaire-based test, does not measure the motor capacity of a child’s affected upper limb. Thus, the discrepancy between capacity and performance cannot be obtained using this instrument.

The revised Video-Observation Aarts and Aarts module: Determine Developmental Disregard (the VOAA-DDD-R) [8, 11, 18], an observation-based instrument, has also been developed to measure upper limb capacity, performance, and developmental disregard in children with CP. The VOAA-DDD-R includes two tasks: a bead-stringing task and a muffin-decorating task. The bead-stringing task, which requires children to use both upper limbs together to string beads, can be used to evaluate the capacity of the affected upper limb. The muffin-decorating task, which is designed to encourage (stimulating, but not requiring) children to use both upper limbs together, can be used to evaluate spontaneous use of the affected upper limb. Developmental disregard is calculated by the discrepancy between the capacity (the bead task) and the performance (the muffin task) [8].

Although the VOAA-DDD-R is able to measure the capacity, performance, and developmental disregard of affected limbs in children with CP, a major concern is that it lacks daily living tasks that would reflect the children’s actual upper limb functions in daily routines. In addition, since task properties could influence a child’s motor performance [19, 20], the use of two different tasks to estimate the discrepancy between capacity and performance might lead to construct errors. In order to reduce the influence of task properties, having participants perform the same task in both the spontaneous use and the forced use conditions might be a more feasible and reasonable way to determine the phenomenon of developmental disregard.

In summary, the current promising assessments for measuring developmental disregard can be classified as subjective and objective tools. PMAL-R, the subjective questionnaire, obtains data on parental perceptions of how much and how well a child actually uses his or her more-affected arm by daily tasks. However, it includes many unilateral tasks that may not capture the real upper limb performance of the affected hand [11, 14, 18]. The VOAA-DDD-R, the objective observational-based assessment, examines the capacity, performance, and developmental disregard of the affected upper limb in children with CP. However, it does not include enough activities of daily living to represent the use of the affected upper limb in daily routines.

Thus, the purposes of the current study were to develop a reliable and valid assessment, named the Observatory Test of Capacity, Performance, and Developmental Disregard (OTCPDD), for measuring the performance of affected upper limbs in the daily routines in children with CP. To be specific, the OTCPDD retained the important features of the PMAL-R [15] in terms of measuring the amount and quality of the use of the affected hand functions in daily activities, but we have modified all items into bimanual activities. In addition, the OTCPDD used the same scoring structure as the VOAA-DDD-R in terms of capacity, performance, and developmental disregard [8] and employed an observation-based manner, but we have expanded the number of items from 2 to 18 to sufficiently represent the use of the affected upper limb in daily routines.

The psychometric properties of the OTCPDD, including the internal consistency, the intrarater and interrater reliabilities, and the convergent and discriminate validities, were investigated. In addition, since the subjective questionnaire might reflect participants’ (or their caregivers’) real perceptions of their motor performance in daily life and might provide important information for evidence-based rehabilitation [21, 22], we also developed an OTCPDD item-matched self-report questionnaire, named the Questionnaire of Developmental Disregard (QDD), to provide supplemental subjective information. The results of the OTCPDD and the QDD were also compared.

Methods

Results

Participants

In all, 48 participants, 24 children with hemiplegic CP and 24 matched typically developing children, were recruited for the study. Their characteristics are shown in Table 1. No significant differences were found between these two groups (p>0.05).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic data of the children with cerebral palsy (CP) and the typically developing children (TDC).

https://doi.org/10.1371/journal.pone.0151798.t001

Reliability

The internal consistency (the Cronbach’s alpha) of the OTCPDD was higher than 0.9 on all of the OTCPDD scores, indicating excellent internal consistency in each measure (Table 2). The intrarater and interrater reliabilities were also excellent, with ICCs ranging from 0.902 to 0.995 (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Reliabilities of the Observatory Test of Capacity, Performance, and Developmental Disregard (OTCPDD).

https://doi.org/10.1371/journal.pone.0151798.t002

Discussion

The results of this study indicate that the OTCPDD is a reliable and valid assessment for evaluating the capacity, performance, and developmental disregard in children with CP. The unique feature of the OTCPDD is that it detects the general affected hand functions by assessing the performance of bimanual daily-relevant tasks (including learning, self-care, and play activities) in children with CP. Developmental disregard is defined as a tendency to underuse or disuse the affected upper limb in daily routines, which is not parallel to preserved capacity in daily life [7, 8, 13, 14]. Thus, the OTCPDD, combining the strengths of the PMAL-R [15] (using daily activities) and the VOAA-DDD-R [8] (using observation-based bimanual activities), was developed to provide comprehensive investigation of developmental disregard.

The high Cronbach’s alpha (alpha>0.9) indicated that the 18 items reflected coherent abilities. The external consistency of the scoring system was supported by the excellent intra-rater and inter-rater reliabilities of all of the OTCPDD scores (ICCs>0.9) found in our study. The excellent external consistency indicates that the OTCPDD is able to consistently reflect subjects’ performance regardless of external factors, such as the scoring performance of the same rater across a period of time or the scoring performances of different raters. These results were contributed by our preliminary work on the OTCPDD scoring system. In that work, two raters together reviewed 15 videos of the OTCPDD and scored the motor components and movement qualities for all of the 18 tasks thoroughly. After reaching consensus on the scoring criteria, a new rater, who was blind to the study purpose, was trained for the investigation of the inter-rater reliability. The high inter-rater reliability (between the OTCPDD developer and a new rater) not only confirmed the stability of the OTCPDD scoring but also supported the feasibility of rater training.

The moderate significant correlation (r = 0.446–0.487, p<0.05) between the OTCPDD and the PMAL-R supports the consistency of these two tests. This moderate correlation coefficient meets our expectation that the OTCPDD is able to assess how much a child actually uses his or her more-affected arm in daily activities when the testing items (unilateral or bilateral) and methods (subjective or objective) are different. Further studies investigating the convergent validities between the OTCPDD and other assessments should be conducted to verify this finding.

As for the discriminant validity, a significant difference (p<0.05) was found between typically developing children and children with CP in the scores of the OTCPDD AOU and OTCPDD QOM. This finding supports the ability of the OTCPDD to discriminate between children in these two groups. In addition, we further analyzed the discriminant ability of each of the ten motor components, and our findings echoed those of Aart et al.’s [15, 16] studies, which found that even fundamental motor functions (e.g., grasp, hold, and release) were sensitive enough to discriminate the motor performances of typically developing children from those of children with CP. However, advanced motor functions (e.g., adjust, manipulate, and pinch) might be necessary for documentation of changes or improvements in motor functions over time. Further studies will be necessary to examine that possibility.

The OTCPDD is the first assessment to estimate the real discrepancy between capacity and performance on the same tasks. We asked a child to perform the tasks under natural conditions to measure real-world performance, followed by a second trial, in which the child was guided to use both hands to measure forced-use capacity. This approach could reduce internal errors and decrease the influence of task properties [19]. The discrepancy between the use of the affected hand in real-world performance and the forced-use capacity is represented by the scores of the DD-AOU and DD-QOM. The DD-AOU score indicates the quantity of the preserved capacities that children with CP disregard. For example, if a child obtains a DD-AOU score of 10 points, it indicates that this child with CP disregards 10 motor components in real-world performance. This child has the capacity for the 10 hidden motor components but does not transfer that capacity to real-world performance. In addition, the score of the DD-QOM indicates the quality of movement for the above preserved capacities. These scores could help researchers and therapists further to understand the discrepancy between motor capacity and motor performance of children and to evaluate the potential for improvement of the affected upper limbs. The results may also guide clinicians to provide adequate rehabilitation programs.

Interestingly, we also found that the scores of the DD-AOU were negatively associated with those of the P-QOM (r = -0.855, p<0.05), indicating that the poorer the quality of movement of the children, the more preserved capacities they may disregard. This finding may be explained by the suggestion of Houwink et al. [7] that children with poorer movement quality may experience over-loading of cognitive efforts while performing a task. Children with hemiplegic CP have relatively few experiences using their affected hand that make them prefer not to use their affected upper limb when the demand of task is high. Otherwise, they have to put their efforts both on their affected hand as well as the demanding of task at the same time. In this case, the automatizing of motor capacity into motor performance was limited. These unique scores could further help researchers and clinicians to understand children’s movement quality and motor capacity and present the potential capability for stimulation of the use of the affected limb in their daily routines. The results may also provide clinical proof that intensive training of the affected limb (such as Constraint-Induced Movement Treatment) can help to improve the movement quality and thus reduce the phenomenon of developmental disregard [24, 25].

This study also investigated the feasibility of the matched questionnaire (DDQ). The significant high correlations between the DDQ and the PMAL-R (r = 0.716–0.813, p<0.05) support the promising subjective outcomes of the DDQ in measuring affected hand functions in daily routines. In addition, the moderate to high relationships between the subjective and the objective outcomes of the same item design, the OTCPDD and the DDQ, (r = 0.508–0.513, p<0.05) fit our expectations and are in line with findings based on typically developing children [22]. This study also adds further support to the idea that parents’ perceptions of their children’s motor skill performance are correlated moderately-to-largely with their children’s real performance, not only in typically developing children but also in children with CP. Furthermore, according to our observations, we found that all the inconsistencies occurred in those parents who had children with moderate movement quality. Such children tend to have fundamental functional skills but perform tasks more slowly or clumsily. Thus, their parents tended to provide relatively lower scores on the DDQ, even when the child performed well in observed tests (the OTCPDD). The results might be explained by the high expectations of the parents and their magnification of the children’s motor deficits in reference to typically developing children. Further studies should investigate the potential factors (e.g., parental education, expectations, and stress level) that lead to this inconsistency to verify this hypothesis.

A few limitations of this study warrant consideration. First, the participants were a convenience sample, and only children with mild-to-moderate CP were included, which might limit the generalizability of the findings. A larger sample of children with CP with differing levels of motor impairment is necessary to verify the results of the current study. Second, the OTCPDD was developed mainly to measure a child’s motor functions based on daily relevant activities with minimum cognitive demands. However, data on cognitive function (e.g., IQ) were not collected in this study. Since cognition may be an important factor in children’s development, future studies should investigate the influence of cognitive levels on children’s motor performance. In addition, the OTCPDD was developed for documenting the treatment effects of developmental disregard. Further investigation of its clinimetric properties (including the responsiveness and the clinically important difference) is necessary to determine whether it can sufficiently detect changes and whether those changes are clinically relevant.

Conclusions

In conclusion, the OTCPDD is a reliable and valid observation-based assessment instrument that uses daily bimanual tasks for measuring capacity, performance, and developmental disregard in children with CP. The OTCPDD has several unique features. First, it detects the capacity, performance, and developmental disregard through observation of performing daily bimanual tasks to represent the general affected hand functions in daily routines. Second, the OTCPDD measures a child’s capacity and performance with the same task in both spontaneous use and forced use conditions to reduce the influence of task properties. The OTCPDD can also quantify the severity of developmental disregard and thereby help researchers and clinicians to understand children’s affected hand functions comprehensively and could be used as a potential outcome indicator for treatment effectiveness. Third, the subjective item-matched questionnaire, the DDQ, can provide supplemental information to reflect the parents’ real perceptions of their children’s motor performance in daily life. It is suggested that further studies recruit larger samples of children with CP, examine children at different levels of severity, and examine the clinimetric properties of the OTCPDD.

Supporting Information