Share this article on:

Item-Specific Functional Recovery in Children and Youth with Acquired Brain Injury

Tokcan, Gul MS, PT; Haley, Stephen M. PhD, PT; Gill-Body, Kathleen M. DPT, MS, NCS; Dumas, Helene M. MS, PT, PCS

Pediatric Physical Therapy: April 2003 - Volume 15 - Issue 1 - p 16-22
doi: 10.1097/01.PEP.0000050765.71924.6C
Research Report

Purpose The purpose of this study was to examine changes in item-specific functional activities and caregiver support in children and youths with acquired brain injury (ABI) in an inpatient setting using the Pediatric Evaluation of Disability Inventory (PEDI).

Methods The PEDI was administered by therapy staff at admission and discharge to a consecutive sample of 94 children and youths with brain injury (62 males and 32 females; age range = one to 19 years of age) admitted to the inpatient service at Franciscan Children's Hospital and Rehabilitation Center during a two-year, 10-month period. Individual PEDI item scores were abstracted from the medical records.

Results Using the Wilcoxon matched-pairs signed rank test (two-tailed), significant differences between admission and discharge scores were found for most mobility, self-care, and social activity item groups and for all domains of caregiver support. Patterns of activity-specific improvements were contrasted between children with traumatic ABI and those with nontraumatic ABI.

Conclusions Recovery in children with ABI can be detected during inpatient rehabilitation across many areas of functioning as represented by item content of the PEDI. Changes at the item level may serve as a guide for defining specific physical therapy goals for inpatient episodes. Item groups in which changes are not detected provide information about sections of the PEDI that are less relevant for hospital-based recovery of children with ABI.

The authors found that items of the PEDI detect recovery in children with acquired brain injury. As a result, these items may help therapists establish inpatient physical therapy goals for children with acquired brain injury.

Rehabilitation Hospital of Rhode Island (G.T.), Providence, RI; Department of Rehabilitation Sciences and Center for Rehabilitation Effectiveness, Sargent College of Health and Rehabilitation Sciences, Boston University (S.M.H.), Boston, Mass; Graduate Programs in Physical Therapy, Massachusetts General Hospital Institute of Health Professions (S.M.H., K.M.G.-B.), and The Research Center for Children with Special Health Care Needs, Franciscan Children's Hospital and Rehabilitation Center (H.M.D.), Boston, Mass

Address correspondence to: Stephen M. Haley, Center for Rehabilitation Effectiveness, Sargent College of Health and Rehabilitation Sciences, 635 Commonwealth Avenue, Boston University, Boston, MA 02215-1605. Email:

Back to Top | Article Outline


Limitations in functional activities and need for caregiver support beyond that expected for age are common long-term sequelae associated with acquired brain injury (ABI) in children and youths. 1,2 Etiologies of ABI include traumatic (the largest group seen in inpatient pediatric rehabilitation programs) and nontraumatic, such as stroke, seizure disorder, infection, anoxia, or brain tumor. Although deficits in cognitive and communicative function are thought to contribute proportionally more to long-term disability than loss of physical functioning, 3–5 all areas of functioning may be affected. Therefore, a comprehensive functional assessment to guide physical therapy management is crucial. 6

Pediatric physical therapists have increasingly focused on evaluating and documenting functional outcomes to provide evidence of meaningful results of services to the child and family. 7,8 Using a standardized functional measure, such as the Pediatric Evaluation of Disability Inventory (PEDI), 9 to document functional changes for various groups of children enables those designing rehabilitation programs to identify patterns of expected recovery and anticipated outcomes. Dudgeon et al 10 used this strategy to describe functional profiles of change in two groups of children undergoing selective dorsal rhizotomy (SDR). Using the PEDI at six-month and 12-month follow-up periods after SDR, the authors reported greater gains in children with diplegia than in children with quadriplegia. Furthermore, the patterns of change were also different; children with diplegia made the greatest changes in body movements and dynamic balance, whereas improvements in children with quadriplegia were generally restricted to transfer skills.

A common practice in outcome studies is to report summary change scores for groups of children within a program without examining the underlying pattern of change at the item level. 11 For example, the PEDI has demonstrated good sensitivity to global change in children over short- 12–14 and long-term recovery periods 15,16 after brain injuries. Yet reliance on aggregate scores (percentage of change score or an average change between two assessment points) is not optimal for making decisions about whether anticipated goals for a particular clinical group are being met or if key functional items are being achieved before discharge home. We have recommended that clinicians do not rely solely on the amount of change to understand if an intervention was advantageous but additionally on the pattern of functional activities that have changed. 17 The PEDI may be particularly useful for establishing individual goals and tracking progress in key areas of functioning for individual children. 18

Return of basic functional skills has been demonstrated to be slower in children and youths with nontraumatic brain injuries than with traumatic brain injuries. 19,20 We have consistently noted more rapid and complete short-term motor recovery 14,21 in children with traumatic injuries. However, we have not investigated the specific patterns of recovery of functional items between the two groups. The purpose of this study was to examine changes in item-specific functional activities and caregiver support in children and youths with ABI (traumatic and nontraumatic) in an inpatient setting. A secondary purpose was to evaluate the sensitivity of the different content areas of the PEDI in children with ABI.

Back to Top | Article Outline



The sample consisted of 94 children and youths with brain injury (62 males and 32 females) with an age range from one to 19 years (mean = 9.3, SD = 5.3) admitted to the inpatient service at Franciscan Children's Hospital and Rehabilitation Center between October 1994 and July 1997. Children with less than three-day hospital stays or children who were admitted for evaluation and observation only were excluded from this study (about 10% of cases). Diagnostic classifications of eligible cases included 53 children with traumatic brain injury (56.4%) and 41 children with brain injuries of nontraumatic origin. Included in this latter category were children with primary diagnoses of seizures (12.8%), brain tumors (12.8%), hemorrhage (9.5%), anoxia (6.4%), and cerebral infarct (2.1%). Average length of stay was 60 days (SD = 54.3). The traumatic and nontraumatic groups were not different in demographic characteristics, functional status on admission, or length of hospital stay (Table 1). The Human Subject Committees at Franciscan Children's Hospital and Rehabilitation Center and Spaulding Rehabilitation Hospital, Boston, Mass, approved this study.



Back to Top | Article Outline

Inpatient Rehabilitation Program

Children and youths admitted to Franciscan Children's Hospital and Rehabilitation Center's (FCH) hospital-based program are provided a structured environment with regulated stimulation, a consistent approach to behavior management, and enhancement of mobility, self-care, communication, and cognitive skills. Children and youths admitted to the program are medically stable and demonstrate a prognosis for improvement in the areas of self-care, mobility, safety, communication, cognition, and behavior. They typically receive a minimum of three hours per day of therapy, including individual physical therapy twice per day for 45 to 60 minutes per session and a weekly group therapy session emphasizing functional movement activities.

In addition to physical therapists, the core treatment team at FCH includes a pediatric physiatrist, rehabilitation nurse, occupational therapist, speech-language pathologist, therapeutic recreation specialist, special education teacher, dietitian, school liaison, social worker, and case manager. The primary physical therapist works closely with the patient, family or guardian, and treatment team to develop and implement a comprehensive therapy program based on the identified individual needs of the patient. Throughout the child's admission, safety and family training are addressed. Home visits and community reintegration activities are incorporated into the child's plan of care when appropriate. Functional activity training is typically a significant focus of the plan of care for each child and youth.

Back to Top | Article Outline


The PEDI 14 is a comprehensive, standardized, clinical assessment that samples functional activities in children within three content domains: self-care, mobility, and social function. The PEDI was designed for children with a variety of disabling conditions and for use in different settings, including inpatient rehabilitation. The PEDI has been standardized on a normative sample of nondisabled children between the ages of six months and 7.5 years, but it is also used for older children if functional development is delayed and abilities fall below those expected of a 7.5-year-old child with no disabilities. A normative sample of 412 infants and children without disabilities was used to develop standardized scores, and initial clinical validation work was conducted with a sample of children with mild traumatic brain injury in an acute care rehabilitation setting. Reliability studies have demonstrated good interobserver reliability. 22,23 Concurrent validity has been established with other pediatric functional measures. 18,24 Construct validity studies have demonstrated that the PEDI can effectively discriminate between children with and without disabilities and among severity levels of children with similar diagnoses (eg, children with osteogenesis imperfecta). 25

The major purposes of the PEDI are to provide a clear description of functional status and to evaluate change after rehabilitation interventions. Functional performance is measured in two ways: capability and performance of daily functional activities. Capability refers to the child's performance of tasks in a standardized or ideal situation. It provides information concerning the child's best performance. The PEDI Functional Skills Scale measures capability by assessing functional skills for which the child has demonstrated mastery and competence. Performance of daily functional activities refers to the measurement of functional behaviors as they actually occur in the environment. The Functional Skills Scale consists of 197 items under three domains: 1) self-care, 2) mobility, and 3) social function. The self-care domain includes eating, grooming, dressing, bathing, and toileting tasks. The mobility domain includes floor mobility, ambulation, transfers, and locomotion tasks in different environments. The social function domain is concerned with comprehension, communication, problem solving, time orientation, self-protection, playing with objects, and social interaction with peers and adults in the home environment and community. Each of the 73 items in the self-care domain, 59 items in the mobility domain, and 65 items in the social function domain is scored as “unable/limited in capability to perform in most situations” or “capable of performing the item in most situations.”

The Caregiver Assistance Scale also measures performance because it identifies the level of caregiver assistance needed to accomplish major functional activities. The Caregiver Assistance Scale consists of 20 items in three domains: (1) self-care, (2) mobility, and (3) social function. In the Caregiver Assistance Scale, a six-point scale is used to assess the amount of caregiver assistance required for the performance of the complex functional skills in each of the eight items in the self-care domain, the seven items in the mobility domain, and the five items in the social function domain. A rating of 5 indicates independence with no caregiver assistance required, whereas a rating of 0 indicates that a large amount of caregiver assistance is required. Operational definitions for all activities and scales were used as provided in the PEDI Development, Standardization, and Administration Manual. 9

Changes in overall PEDI summary scores in cohorts similar to this group have been reported elsewhere. 12,13,21 For this report, the focus is on specific areas of functional change that make up these global summary scores. For each domain in the Functional Skills Scale, task-specific item-group scores were determined by adding together scores on several individual items that assess similar constructs. For example, in the self-care domain, the four items that assess eating different food consistencies constitute an item group labeled “food textures”. Four to five items were grouped in this manner for each task-specific item group within each PEDI Functional Skills content domain. Rating scale changes for each item in the Caregiver Assistance section were analyzed individually.

Back to Top | Article Outline


The PEDI is routinely administered to children with brain injury at Franciscan Children's Hospital and Rehabilitation Center at admission to and discharge from the hospital. Therapists attend in-service training sessions and complete the case studies in the PEDI manual in preparation for administering the PEDI in routine clinical practice. The self-care domains were administered by an occupational therapist; the mobility domains, by a physical therapist; and the social function domains, by a speech pathologist. Admission, discharge, and individual PEDI item scores for each domain in the Functional Skills and Caregiver Assistance Scales were collected from the medical records of all the subjects. Variables of age, gender, length of hospital stay, and diagnosis were also collected from the medical records.

Back to Top | Article Outline

Data Analysis

Wilcoxon matched-pairs tests (α < 0.01) were conducted for the total sample and for the two ABI subgroups (traumatic and nontraumatic) to determine the change within the PEDI item-group scores in the self-care, mobility, and social function domains of the Functional Skills Scale and individual items in self-care, mobility, and social function domains of the Caregiver Assistance Scale. We used an α level of 0.01 to reduce type I error rate given the multiple tests conducted for item groups and caregiver assistance items.

Back to Top | Article Outline


Changes in the task-specific item groups for each domain in the Functional Skills and Caregiver Assistance Scales for the total group and two subgroups are displayed in Tables 2 through 4. Most children either improved during the rehabilitation hospitalization or showed no change (noted as a tie). A small number of subjects regressed (negative results). In the combined sample, significant changes were found in all of the Functional Skills item groups and all of the Caregiver Assistance individual items.







Children in the traumatic group demonstrated significant changes in all but two Functional Skill item groups (household chores and self-protection tasks) and in all the Caregiver Assistance items. In contrast, fewer Functional Skill item groups and Caregiver Assistance items showed significant changes in the nontraumatic group. No changes in the nontraumatic group were noted in 11 Functional Skills item groups (food textures, hair brushing, nose care, fasteners, toileting tasks, bowel and bladder management, toilet transfers, home-based chores, self-protection, and community functions) and in four Caregiver Assistance items (grooming, bladder and bowel management, and peer play).

Back to Top | Article Outline


In general, the overall ABI sample showed significant improvements in the performance of functional skills and a decrease in the amount of caregiver assistance needed at the end of inpatient rehabilitation. In the analysis of the total sample, all of the Functional Skill item groups and Caregiver Assistance items improved significantly from admission to discharge. This was driven largely by the strong improvements in the traumatic group. As expected, children with traumatic injuries demonstrated earlier functional improvements by the time of rehabilitation hospital discharge than those in the nontraumatic group.

For the subgroup of children with traumatic injuries, item groups of household chores and self-protection did not show a significant improvement. Even though the PEDI manual provides examples of how both of these items can be scored in a hospital environment, these items are often perceived as community-based items only and thus are thought to be more difficult to accurately assess outside the home. There was also a smaller sample of responses for these item groups than for other social function items.

Four general functional areas did not change beyond chance levels in children with nontraumatic injuries across Functional Skill item groups and the Caregiver Assistance items: 1) toileting tasks, 2) eating (pertaining to oral-motor functioning), 3) advanced self-care skills requiring good fine motor skills (eg, hair brushing, fasteners), and 4) independence in peer play. It is interesting to note that these functional areas include very basic skills (toileting and eating) and higher-level fine motor and social skills. This may suggest that some children in the nontraumatic group (eg, those with anoxia and severe stroke) are not able to recover these basic skills during hospital rehabilitation episodes, at least not to the extent of their peers with traumatic injury. Similarly, achieving changes in more challenging fine motor and social activities may be less likely in the nontraumatic injury group than in the children with traumatic brain injury.

An important contributor to the changes noted here is the level of associated pre-existing congenital problems that were present in the nontraumatic group. Although we cannot report premorbid data, anecdotally we noted that many of the children in the nontraumatic group who had seizures, strokes (eg, Moyamoya syndrome), and other neurological conditions also had multiple medical and congenital anomalies that may have limited functional progress during inpatient rehabilitation. In contrast, for most, if not all, of the children in the traumatic group, this hospitalization was in response to an accidental injury unrelated to any significant prior developmental or cognitive delay.

For the most part, differences in functional change between groups reflect a true disparity in recovery during inpatient stay. No substantial ceiling or floor effects at admission were noted to interfere with the interpretation of these item-specific change analyses. As noted in Table 1, initial status across all functional areas is strikingly similar. For example, in the item group “bladder management,” approximately 50% of the children in both groups were totally dependent in bladder management at admission. At discharge, 41.2% of the nontraumatic group continued to be dependent, whereas the percentage in the traumatic group who were totally dependent at discharge fell to 24.5%. Cognitive and behavioral limitations noted in the nontraumatic group likely contributed to the lack of progress in the toileting area. We did not note any indications of premorbid sphincter problems in the nontraumatic group. Furthermore, differences in sensitivity cannot be attributed to group differences in hospital length of stay, age at admission, and or initial levels of functioning (Table 1).

A potential limitation in the interpretation of the results is the lack of formal reliability checks in use of the PEDI during the recruitment period of the study. As noted in the data collection procedures, in-service training programs were conducted during the data collection period to address specific questions about the administration of the PEDI and to reach a consensus on how to score in specific instances when the PEDI scoring rules were ambiguous. Yet the magnitude and consistency of changes noted in most of the item groups in this study are not likely to be due to measurement error from inappropriate administration and scoring of the PEDI. However, clinical programs using the PEDI on a consistent basis should collect data to document the extent of measurement error introduced by multiple raters.

Analyses of change at the item level can provide information to guide program planning for individual children. For example, 60% of all the children made some improvements in the capability to perform chair transfers. Sixty-three percent of children in the traumatic injury group made some improvements in transfers; however, only 43% of the children with nontraumatic injuries improved in transfers. These data can be useful for setting appropriate and realistic goals for hospital discharge.

Back to Top | Article Outline


In this study, our primary purpose was to determine changes in item-specific functional activities and caregiver support in children and youths with ABI in an inpatient setting. Children with nontraumatic injuries had more difficulty recovering basic toileting and eating skills, higher-level personal care activities requiring fine motor skills, and social skills such as play. In general, most items from the PEDI seem to be responsive in detecting changes in functional skills and caregiver assistance of children with ABI in an inpatient setting. The small number of items in the PEDI that were not responsive to change in this sample may reflect functional activities that are not routinely addressed in intervention programs during hospitalization and that may be better represented in community-based assessments. These results can guide the development of treatment goals focusing on the functional skills that are likely to improve in children and youths with acquired brain injury during inpatient stays.

Back to Top | Article Outline


This article was adapted from a thesis submitted by G. Tokcan to the Faculty of the Graduate Programs in Physical Therapy, MGH Institute of Health Professions, Boston, Mass, in partial fulfillment of the requirements for the degree of Master of Science in Physical Therapy.




Back to Top | Article Outline


1. Kraus J, Fife D, Conroy C. Pediatric brain injuries: the nature, clinical course, and early outcomes in a defined United States population. Pediatrics. 1987; 79: 501–507.
2. Adelson P, Kochaneck P. Head injury in children. J Child Neurol. 1998; 13: 1–15.
3. Berger M, Pitts L, Lovely M, et al. Outcome from severe head injury in children and adolescents. J Neurosurg. 1985; 62: 194–99.
4. Knights R, Ivan L, Vebtureyra E, et al. The effects of head injury on neuropsychological and behavioral functioning. Brain Inj. 1991; 5: 339–551.
5. Jaffe KM, Massagli TL, Martin KM, et al. Pediatric traumatic brain injury: acute and rehabilitation costs. Arch Phys Med Rehabil. 1993; 74: 681–686.
6. Haley SM, Baryza MJ, Webster HC. Pediatric rehabilitation and recovery of children with traumatic injuries. Pediatr Phys Ther. 1992; 24–30.
7. Harris SR. Efficacy of physical therapy in promoting family functioning and functional independence for children with cerebral palsy. Pediatr Phys Ther. 1990; 2: 160–164.
8. Palisano RJ, Campbell SK, Harris SR. Decision making in pediatric physical therapy. In: Campbell SK, Vander Linder D, Palisano RJ, eds. Physical Therapy for Children. Philadelphia, PA: WB Saunders; 2000; 198–224.
9. Haley SM, Coster WJ, Ludlow LH, et al. Pediatric Evaluation of Disability Inventory: Administration Manual. Boston, MA: New England Medical Center; 1992.
10. Dudgeon BJ, Libby AK, McLaughlin JF, et al. Prospective measurement of functional changes after selective dorsal rhizotomy. Arch Phys Med Rehabil. 1994; 75: 46–53.
11. Stucki G, Daltroy L, Katz J, et al. Interpretation of change scores in ordinal clinical scale and health status measures: the whole may not equal the sum of the parts. J Clin Epidemiol. 1996; 49: 711–717.
12. Dumas HM, Haley SM, Fragala MA, et al. Self-care recovery of children with brain injury: descriptive analysis using the Pediatric Evaluation of Disability Inventory (PEDI) functional classification levels. Phys Occup Ther Pediatr. 2001; 21: 7–27.
13. Dumas HM, Haley SM, Bedell GM, et al. Social function changes in children and adolescents with acquired brain injury during inpatient rehabilitation. Pediatr Rehabil. 2001; 4: 177–185.
14. Haley SM, Dumas HM, Ludlow LH. Variation by diagnostic and practice pattern groups in the mobility outcomes of inpatient rehabilitation programs for children and youth. Phys Ther. 2001; 81: 1425–1436.
15. Dumas HM, Haley SM, Rabin JP. Short-term durability and improvement of function in traumatic brain injury: a pilot study using the Pediatric Evaluation of Disability Inventory (PEDI) classification levels. Brain Inj. 2001; 15: 891–902.
16. Coster WJ, Haley S, Baryza MJ. Functional performance of young children after traumatic brain injury: a 6-month follow-up study. Am J Occup Ther. 1994; 48: 211–218.
17. Haley SM, Ludlow LH, Coster WJ. Pediatric Evaluation of Disability Inventory: clinical interpretation of summary scores using Rasch rating scale methodology. Phys Med Rehabil Clin North Am. 1993; 4: 529–540.
18. Ziviani J, Ottenbacher KJ, Shephard K, et al. Concurrent validity of the Functional Independence Measure for Children (WeeFIM) and the Pediatric Evaluation of Disability Inventory with developmental disabilities and acquired brain injury. Phys Occup Ther Pediatr. 2001; 21: 91–101.
19. Heindl U, Laub M. Outcome of persistent vegetative state following hypoxic or traumatic brain injury in children and adolescents. Neuropediatrics. 1996; 27: 94–100.
20. Sazbon L, Zagreba F, Ronen J, et al. Course and outcome of patients in vegetative state of nontraumatic aetiology. J Neurol Neurosurg Psychiatry. 1993; 56: 407–409.
21. Fragala MA, Haley SM, Dumas HM, et al. Classifying mobility recovery in children and youth with brain injury during hospital-based rehabilitation. Brain Inj. 2002; 16: 149–160.
22. Wright FV, Boschen KA. The Pediatric Evaluation of Disability Inventory (PEDI): validation of a new functional assessment outcome instrument. Can J Rehabil. 1993; 7: 41–42.
23. Nichols DS, Case-Smith J. Reliability and validity of the Pediatric Evaluation of Disability Inventory. Pediatr Phys Ther. 1996; 8: 15–24.
24. Nordmark E, Jarnlo G, Hagglund G. Comparison of the Gross Motor Function Measure and the Pediatric Evaluation of Disability Inventory in assessing motor function in children undergoing selective dorsal rhizotomy. Dev Med Child Neurol. 2000; 42: 245–252.
25. Engelbert RHH, Custers JWH, van der Net J, et al. Functional outcomes in Osteogenesis imperfecta: Disability profiles using the PEDI. Pediatr Phys Ther 1997; 9: 18–22.
© 2003 Lippincott Williams & Wilkins, Inc.