- Email: [email protected]
Upper-limb function in Australian children with traumatic brain injury: A controlled, prospective study
642
Upper-Limb Function in Australian Children With Traumatic Brain Injury: A Controlled, Prospective Study Margaret A. Wallen, MA, Suzanne Mackay, BAppSc, Sharon M. Duff, BAppSc, Lynn C. McCartney, BAppSc, Stephen J. O’Flaherty, FRACP ABSTRACT. Wallen MA, Mackay S, Duff SM, McCartney LC, O’Flaherty SJ. Upper-limb function in Australian children with traumatic brain injury: a controlled, prospective study. Arch Phys Med Rehabil 2001;82:642-9. Objective: To describe upper-limb function in children with mild and severe traumatic brain injury (TBI), by using both quantitative and qualitative measures. Design: Controlled, prospective cohort study with assessment points initially, at 6 months, and at 2 years after TBI. Setting: A tertiary pediatric trauma center in Australia. Patients: Fifty-one children, ranging in age up to 14 years, who were consecutive admissions with TBI. On the basis of initial and persisting abnormal coma score and persistence of posttraumatic amnesia, they were assigned to either a mild (n ⫽ 26) or a severely injured (n ⫽ 25) group. Thirty children admitted with non-TBI trauma were recruited as a control group. Main Outcome Measures: Quantitative measures included Bruininks-Oseretsky Test of Motor Proficiency and Peabody Developmental Motor Scales. Qualitative measures included Brunnstrom Recovery Stages (adapted), categoric scales of muscle tone, grasp used when handwriting, quality of writing product, bilateral activity, and splint use. Results: There was little difference between the groups on the standardized assessments for subjects who could complete the tests. Qualitative measures showed the severe TBI group to have more difficulties with gross arm control, hand control, and hand function. Conclusion: Children with severe TBI experience more and persisting difficulties with upper-limb function. It is essential to include both quantitative and qualitative measures in this type of research. Key Words: Arm; Brain injuries; Disabled children; Rehabilitation. © 2001 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
From the Occupational Therapy (Wallen, Mackay) and Rehabilitation Departments (Duff, McCartney, O’Flaherty), The New Children’s Hospital, Westmead, Australia. Accepted in revised form July 26, 2000. Supported by the Motor Accidents Authority of New South Wales (grant no. RE92/28) and Australian Commonwealth Department of Human Services and Health (grant no. 93/02972). Presented in part at OT Australia–NSW 11th NSW Conference, October 15–16, 1998, in Jamberoo, NSW, Australia. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Margaret Wallen, MA, Senior Occupational Therapist/Research Coordinator, The New Children’s Hospital, PO Box 3515, Parramatta NSW 2124, Australia, e-mail: [email protected]. 0003-9993/01/8205-6154$35.00/0 doi:10.1053/apmr.2001.22620
Arch Phys Med Rehabil Vol 82, May 2001
NFORMATION ABOUT REHABILITATION outcomes Iaboutafter traumatic brain injury (TBI) guides decision making clinical and resource management. Several studies 1-12
have examined the outcomes in children with TBI, but none have studied the rehabilitation outcomes of Australian children across the age ranges. The pediatric rehabilitation service at Westmead Hospital, Sydney, Australia (now amalgamated with The Children’s Hospital at Westmead, Australia) completed a prospective, controlled, multidisciplinary outcomes study of children with TBI. Upper-limb function was one of the areas studied by the occupational therapists in the research team. Upper-limb function is important in a child’s ability to participate in daily activities and to integrate into home, school, and community life.13 It is, therefore, an integral part of occupational therapy involving children with TBI. An overview of the results of upper-limb function has been reported elsewhere14 as part of the overall study’s method and results. This article discusses in greater detail the results and implications of the upper-limb assessments. UPPER-LIMB FUNCTION Rehabilitation staff members must understand the course, nature, and extent of recovery of upper-limb function in children with TBI. This is important if resources are to be appropriately directed and motor difficulties are adequately managed so to facilitate optimal recovery, quality of movement, and functional use. Studies that have motor outcomes as a variable being investigated report that physical disability is common in severely injured children, but is less common in children with mild TBI.2,4,15,16 Few studies, however, specify upper-limb function as a separate variable, and even fewer use objective tools to measure that function. Rutter et al16 used parent interviews to determine the incidence of physical disability, but did not report specific upper-limb involvement. Eiben et al10 reported fine motor difficulties in their subjects, but included self-care items such as dressing and feeding in the fine motor category without describing performance in those areas. Ewing-Cobbs et al2 used motor scales from 2 psychometric assessments, the Bayley Scales of Infant Development and the McCarthy Scales of Children’s Abilities, to evaluate children after brain injury. Various components of the Weschler Intelligence Scale for Children–Revised and the Woodcock-Johnson Scale of Independent Behaviour have also been used.4 Although these are objective evaluations, they do not provide specific details of upper-limb performance. One study that specified a test of upper-limb performance used the Bruininks-Oseretsky Test of Motor Proficiency.1 The study found no difference between children with brain injury and control children on the fine motor composite score. Children with brain injury, however, scored significantly lower (p ⫽ .001) on the upper-limb speed and dexterity subtest. This finding is consistent with the literature, suggesting that children with TBI are significantly slower on timed tests of fine motor performance.
643
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen
Given the reported incidence of motor dysfunction in severely brain-injured children, and the impact of diminished speed of motor activity on performance in functional tasks, it appears that motor difficulties are a significant issue for children with brain injury and are likely to impact on their ability to participate in daily activities. Most studies that have investigated upper-limb function have focused on severely injured children.3,7,8,10,17-20 There is debate, however, about the longterm outcome of children with less-severe brain injury. Clinically, it is important to know whether children who sustain a mild TBI are likely to have long-term difficulties. This knowledge would allow appropriate resource allocation, including the provision of appropriate follow-up and intervention. Further research would, therefore, be useful to clarify the incidence, impact, and nature of upper-limb disability. In summary, there is a need to identify the impact of both mild and severe TBI on upper-limb abilities in children. This study was designed to describe and compare the upper-limb function of groups of Australian children with non-TBI trauma, mild TBI, and severe TBI at 6 months and 2 years postinjury. Both quantitative and qualitative measures were used to ensure a comprehensive assessment, within the context of a prospective, multidisciplinary outcome study. METHOD This study is a component of a large multidisciplinary prospective, longitudinal, cohort follow-up study (the detailed methodology was published elsewhere14). A summary of the methods and details of the assessment of upper-limb function are presented here. Subjects and Sampling Twenty-six children with severe TBI, 25 children with mild TBI, and 30 control children were recruited for the study. Subjects included all children and adolescents, up to 14 years of age, who were admitted consecutively to Westmead Hospital after a TBI during July 1992 through December 1994. The mean age of the subjects recruited initially was: control ⫾ standard deviation (SD), 7.5 ⫾ 4 years (range, 10mo–14yr); mild TBI, 5.5 ⫾ 4.7 years (range, 1–14yr); and severe TBI, 7.6 ⫾ 4.8 years (range, 1mo–13yr). See table 1 for other demographic information. There were a number of dropouts by the 6-month (25%) and 2-year (40%) assessment points. We did not specifically record their reasons to discontinue, but mostly families chose not to continue participating or could not be contacted. The dropouts did not differ from the rest of the group in respect to age, gender, parental occupational prestige, poverty status, marital status, or maternal education.14 Subjects had to have experienced head trauma, a loss of consciousness that was observed by a reliable witness, and/or a lowered coma score within 24 hours of the TBI. Coma score refers either to the Adelaide Paediatric Coma Score, which was used with subjects younger than 5 years old, or to the Glasgow Coma Score (GCS), which was used with the older subjects. Exclusionary criteria included a history of preexisting neurologic conditions, previous neurologic insult or loss of consciousness, developmental, intellectual, or learning disability, nonaccidental injury, presence of psychiatric illness, epilepsy, or progressive disease. Eligible subjects were then assigned to “severe” or “mild” TBI groups on the basis of: (1) initial coma score, (2) abnormal coma score over the first week postinjury, and (3) persistence of posttraumatic amnesia (PTA) in subjects older than 7 years of age. The Westmead Post-Traumatic Amnesia Scale was used to measure PTA.21-23 The control group consisted of 30 children admitted consecutively to Westmead Hospital during the study period with
Table 1: Demographic Information Control TBI
Gender Boys Girls Initial GCS % (n) 0–8 9–12 13–15 Time ICP was raised 0h ⬍12h 12–24h ⬎24h Days in ICU % (n) 0 1–3 ⬎3 Mean days Range Duration of PTA % (n)* 0d 1–3d 4–7d ⬎7d Mean days Range
22 8
Mild TBI
19 6
Severe TBI
22 4 100 (16)
6 (1) 94 (15)
38 (6) 63 (10)
100 (16)
100 (16)
94 (16)
73 (11) 26 (4)
6 (1) 0.5 0–8 100 (16)
0 0
0.5 0–3
38 13 13 38
(6) (2) (2) (6)
26 (4) 74 (12) 12.6 1–65
81 (13) 19 (3)
0.2 0–1
13 (1) 88 (7) 85.6 7–180
Abbreviations: ICP, intercranial pressure; ICU, intensive care unit. * PTA data not available for all severely injured children.
non– brain-injury trauma. The same exclusionary criteria were applied to the control group. Twenty-six were admitted for nonskull fractures and the remaining 4 had soft tissue injuries or burns to less than 10% of the body. The reason for using a control group that had experienced trauma was to allow for the effects of hospitalization and the problems associated with trauma because these factors may influence recovery and outcome. Wesson et al9 found that 50% of children hospitalized with severe non– brain-related injuries had disability persisting for at least 6 months postinjury. We also wanted a comparison group that may have had predisposing behaviors that put those group members at risk of sustaining injury. Finally, we wanted to differentiate changes on testing resulting from development, recovery, and the effect of repeated testing. Procedure The Western Sydney Area Health Service Ethics Committee granted its approval for the study. Once the eligible subjects and controls were identified, informed consent was obtained from the subjects’ parents or caregivers. All subjects completed a full post-PTA assessment (quantitative and qualitative measures) once they were medically stable. In mildly and severely injured children, this assessment was completed within a week after they emerged from PTA or had achieved a coma score normal for their age. A screening assessment of baseline, qualitatively measured upper-limb function was completed before this point on mildly and severely injured children who were stable but still in PTA. A second stage of assessments was completed 6 months after the post-PTA assessment. A final assessment was completed 2 years postinjury. Control subjects were assessed initially, at 6 months, and 2 years postinjury. Arch Phys Med Rehabil Vol 82, May 2001
644
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen
Multiple data collectors, including occupational therapists or final-year occupational therapy students, were trained in the study’s administration procedures, however, interrater reliability was not completed between data collectors. Tools Both quantitative and qualitative aspects of upper-limb performance were evaluated. The quantitative evaluation consisted of standardized norm-referenced tests. Haley et al13 recommended the tests to obtain a precise estimation of recovery rate and comparison with an age-appropriate reference. Qualitative assessment of upper-limb performance was included to describe movement pattern difficulties and to ensure that data were available for all subjects, including those for whom standardized assessments were inappropriate (eg, subjects with severe motor problems or with difficulty comprehending or following test instructions). Quantitative tools. The Peabody Developmental Motor Scales–Fine Motor24 (PDMS-FM) and the Bruininks-Oseretsky Test of Motor Proficiency–Fine Motor Composite25 (BOTMPFMC) were used as the standardized norm-referenced tests of upper-limb function. The PDMS was used with subjects who, at the initial evaluation, were younger than 4 years, 11 months of age. The BOTMP was used with subjects 5 years of age and older. This age cutoff meant that subjects could be reassessed at the 2-year follow-up evaluation by using the same measure that was used at the post-PTA assessment. The 4 categories of the PDMS-FM (grasping, hand use, eye-hand coordination, manual dexterity) were completed. Reliability has been stable over time and between raters, particularly for the total score.26,27 Content- and criterion-related validity have been established as being reasonable.26,27 The 3 subtests of the BOTMP-FMC (response speed, visual motor control, upper-limb speed and dexterity) were used. The test manual reports evidence of validity, particularly the correlation of scores with age, internal consistency of the subtests,
and the ability to discriminate between children with and without disabilities. Test-retest and interrater reliability are also reported as satisfactory. Qualitative tools. The Brunnstrom Recovery Stages28 were adapted for use with subjects aged 16 months and older. The adaptation allowed a description of qualitative gross arm control and hand control, including tremors and ataxia. Categoric scales were developed to gather information on muscle tone at all assessment points, on grasp used when handwriting, the quality of writing product, bilateral activity, and splint usage at the 2-year assessment. Data Analysis Data were analyzed by using SPSS.a Most variables had insufficient numbers for statistical analysis so descriptive summaries were used to examine the data. One-way analysis of variance was used to compare groups for the continuous data of the standardized upper-limb function assessments. A few children in the severe TBI group were unable to complete the standardized assessment because of significant physical disability. To include these children in the results for standardized assessments, the percentage of children who scored below average in the assessments was calculated. This percentage included children unable to be assessed on standardized tools. RESULTS Screening Assessment Information was only gathered at this point from subjects with mild TBI (n ⫽ 4) and with severe TBI (n ⫽ 16) who were in PTA. All mild TBI subjects presented with normal muscle tone. Most severely injured subjects presented with abnormal tone or motor control (table 2). The older subjects were assessed on the modified Brunnstrom scales. No mild TBI subjects had difficulties with arm or
Table 2: Percentage of Control, Mild, and Severe Subjects in Each Muscle Tone Rating at Screening, Post-PTA, 6 Months, and 2 Years, for Left and Right Upper Limbs Screening Muscle Tone
Normal L R Abnormal Spasticity L R Hypotonia L R Fluctuating L R Ataxia L R
n (total patients in group) L R
Mild TBI (%)
Severe TBI % (n)
100 100
27 (4) 19 (3)
Post-PTA
6 Months
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
100 100
50 (13) 54 (14)
2 Years
Control (%)
Mild TBI % (n)
Severe TBI % (n)
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
95 (18) 95 (18)
82 (18) 64 (14)
100 100
100 100
73 (11) 60 (9)
20 (3) 25 (4)
15 (4) 12 (3)
5 (1) 18 (4)
27 (4) 44 (7)
15 (4) 15 (4)
5 (1) 9 (2)
20 (3) 40 (6)
7 (1) 6 (1) 20 (3) 6 (1)
3 4
15 16
Abbreviations: L, left; R, right.
Arch Phys Med Rehabil Vol 82, May 2001
19 (5) 19 (5)
28 28
23 24
26 26
20 20
5 (1) 5 (1)
9 (2) 9 (2)
19 19
22 22
7 (1)
17 17
16 16
15 15
645
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen Table 3: Percentage of Control, Mild, and Severe TBI Subjects in Each Category of the Modified Brunnstrom Scales for Left and Right Gross Arm Control at Screening, Post-PTA, 6 Months, and 2 Years Screening Modified Brunnstrom Scales
Gross arm control No abnormalities L R Poor quality L R Ataxia/no coordination L R Independence from synergies L R Movement deviating from synergies L R Voluntary synergies L R Synergies developing L R No movement L R
n (total in group) L R
Mild TBI (%)
Severe TBI % (n)
100 100
33 (4) 8 (1)
Post-PTA
6 Months
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
100 100
57 (13) 57 (13)
2 Years
Control (%)
Mild TBI (%)
Severe TBI % (n)
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
100 100
79 (16) 74 (13)
100 100
100 100
69 (11) 56 (9)
17 (2) 15 (2)
22 (5) 13 (3)
5 (1) 5 (1)
6 (1) 6 (1)
17 (2) 8 (1)
13 (3) 9 (2)
11 (2) 16 (3)
6 (1) 13 (1)
6 (1) 4 (1)
9 (2)
5 (1)
6 (1)
6 (1) 13 (2)
17 (2) 23 (3)
9 (2)
17 (2) 46 (6) 3 4
12 13
5 (1)
6 (1) 6 (1)
9 (2) 26 26
22 23
hand control. Most of the subjects with severe TBI had some abnormality resulting from their brain injuries (tables 3, 4). Post-PTA Thirty control subjects, 24 with mild TBI, and 26 with severe TBI were assessed post-PTA. Full data sets were not available because not all subjects completed all the assessments. All control subjects and those with mild TBI presented with normal muscle tone. About half of the severely injured subjects continued to experience difficulties with muscle tone (table 2) and difficulties on the modified Brunnstrom scales (tables 3, 4). Quantitative assessment: PDMS. The control and mild TBI groups obtained mean scores within the normal range. The severe TBI group mean was slightly more than 1 SD below the mean of the normative sample (table 5). There was no significant difference between groups. All group means were within the normal range on the subtests of the PDMS except for the severe TBI group, which scored just below the normal range on eye-hand coordination. There was no significant difference between groups on any of the subtests of the PDMS. Twenty-five percent of the control group and 31% and 54% of the mild and severe TBI subjects, respectively, had total scores below the average or were unable to complete the assessment (table 5). Most of these difficulties appeared to be experienced in the eye-hand coordination subtest. Two subjects
23 23
19 19
19 19
19 19
17 17
16 16
16 16
with severe TBI and significant motor difficulties were unable to complete the assessment. Inability to test these children, and, therefore, to include their scores on the PDMS, skewed the results for the severe TBI group to being less impaired than it actually was. Quantitative assessment: BOTMP. Older subjects were assessed by using the BOTMP. All group means were within the normal range, though the severe TBI group scored significantly lower than the mild TBI group (table 5). All group means were within the normal range for all the subtests, though the severely injured group scored significantly lower than the mild TBI group on the visuomotor and upper-limb speed and dexterity subtests. Despite means within the normal range, 57% of severely injured subjects had scores below average or were unable to complete the assessment (table 5), with most of the difficulties occurring in the response speed and upper-limb speed and dexterity subtests. Six Months Twenty control subjects, 19 with mild TBI, and 22 with severe TBI were assessed, though complete data sets were not available for all subjects. Qualitative assessment. All control subjects, 95% (n ⫽ 18) of subjects with mild TBI, and most of the subjects with severe TBI had normal tone (table 2). All control subjects had normal Arch Phys Med Rehabil Vol 82, May 2001
646
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen Table 4: Percentage of Control, Mild, and Severe TBI Subjects on Each Category of the Modified Brunnstrom Scales for Left and Right Hand Control at Screening, Post-PTA, 6 Months, and 2 Years Screening
Modified Brunnstrom Scales
Hand control No abnormalities L R Poor quality L R Ataxia/no coordination L R Palmar prehension L R Thumb release L R Mass grasp and wrist extension L R Mass finger flexion L R No movement L R
n (total in group) L R
Mild TBI (%)
Severe TBI % (n)
100 100
25 (3) 8 (1)
Post-PTA
6 Months
Control (%)
Mild TBI % (n)
Severe TBI % (n)
100 100
96 (21) 96 (22)
67 (14) 50 (11)
17 (2) 15 (2)
Control (%)
Severe TBI % (n)
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
95 (18) 95 (18)
79 (15) 63 (12)
100 100
100 100
69 (11) 56 (9)
5 (1) 5 (1)
11 (2) 16 (3)
6 (1) 13 (2)
11 (2) 11 (2)
6 (1) 6 (1)
14 (3) 18 (4)
17 (2) 8 (1)
4 (1) 4 (1)
2 Years
Mild TBI % (n)
10 (2) 14 (3)
6 (1) 5 (1) 5 (1)
6 (1)
8 (1)
17 (2) 23 (3)
5 (1)
5 (1)
6 (1) 6 (1)
10 (2) 5 (1)
5 (1)
6 (1)
17 (2) 46 (6) 3 4
12 13
6 (1) 6 (1) 25 24
22 23
movement on the Brunnstrom scales, as did all but 1 subject with mild TBI. Several of the subjects with severe TBI continued to have difficulties when assessed with the Brunnstrom scales (tables 3, 4). Quantitative assessment: PDMS. The control group mean for the total score was just below the average range. Although both the mild and severe TBI groups had scores within the normal range, the severely injured group had a significantly lower mean than the group with mild TBI (table 5). All group means for subtests were within the average range except for the control group for eye-hand coordination. There were no differences between groups on any of the subtests. Forty percent of the control group and 8% of the subjects with mild TBI had scores in the below-average range. Forty percent of the severe TBI subjects had scores in the belowaverage range, including 1 child who was unable to complete the assessment and was thus classified as below average (table 5). Quantitative assessment: BOTMP. All group means were within the average range, though the severe TBI group mean was significantly lower than both the control and mild TBI groups. The severe TBI group mean was also significantly lower than the mild TBI and control groups for both the visuomotor and upper-limb speed and dexterity subtests, though the upper-limb speed and dexterity subtest mean was the only one that was in the below-average range. Seven percent of the control group and 17% of the subjects with Arch Phys Med Rehabil Vol 82, May 2001
21 22
19 19
19 19
19 19
17 17
16 16
16 16
severe TBI were below the average range or were unable to complete the assessment (table 5). None of the subjects in the mild TBI group were below the average range or unable to complete the assessment. Two Years Seventeen control subjects, 16 with mild TBI, and 16 with severe TBI were assessed at this point. Full data sets were not available for all subjects. Qualitative assessment. Only subjects with severe TBI experienced difficulties with muscle tone (table 2), with 3 subjects requiring wrist/hand splints (2 subjects with single wrist/hand splints, the other with bilateral wrist/hand splints). No control subjects or subjects with mild TBI experienced difficulties on the Brunnstrom arm or hand scales. Many of the severe TBI group continued to have difficulty on both these scales (tables 3, 4). Other final qualitative outcome variables (hand dominance, written communication, handwriting grasp, bilateral activity) are also reported as follows. Most subjects had normal hand dominance (control, 94%; mild TBI, 94%; severe TBI, 73%). One child in the control group had not yet established a dominance, which was considered abnormal for age. One subject with mild TBI had changed dominance, with 4 severe TBI subjects having abnormal hand dominance (ie, changing dominance as a result of TBI or a lack of established dominance, which was considered inappropriate for age).
647
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen Table 5: Scores Achieved on the PDMS and BOTMP by Control, Mild, and Severe TBI Subjects at Post-PTA, 6 Months, and 2 Years Post-PTA
PDMS total score n* Mean ⫾ SD Range p‡ Percentage below average (n)* BOTMP-FMC scores n Mean ⫾ SD Range p‡ Percentage below average
Control
Mild TBI
4 42 ⫾ 9.2 29–50 0.2 (NS) 25
13 46 ⫾ 9.9 27–61
16 57 ⫾ 9.1 41–76
9 65 ⫾ 6.6† 49–70 .03 0
0
31
6 Months Severe TBI
9 39 ⫾ 7.6 27–51 55 (11) 10 48 ⫾ 20.7† 20–79 57
2 Years
Control
Mild TBI
Severe TBI
Control
Mild TBI
Severe TBI
5 39 ⫾ 10.8 27–51
12 50 ⫾ 9.5† 27–60 .05 8
9 41 ⫾ 8.5† 31–56
5 42 ⫾ 4.8 37–48 0.8 (NS) 40
9 44 ⫾ 6.7 31–55
4 45 ⫾ 6.5 40–54
11
33 (6)
6 59 ⫾ 9.8† 44–73
12 48 ⫾ 12† 20–67
10 56 ⫾ 10.3 40–70 .08 (NS) 0
5 63 ⫾ 6.7 52–69
5 46 ⫾ 15 36–72
0
40
40 14 58 ⫾ 8.4† 37–70 .03 7
0
40 (10)
17
NOTE. Scores for PDMS and BOTMP are expressed as T scores with a mean ⫾ SD of 50 ⫾ 10. Scores of ⱕ 39 are considered below average. Abbreviation: NS, nonsignificant. * n is the number of children who completed the assessment in each group. (n) is the revised n where those children unable to be assessed by using the standardized assessments because of severity of disability have been included in a calculation of percentage in each group who were below average (ie, more than 1 SD below the mean). Only children in the severe TBI group were unable to complete the standardized assessment, thus, these data only appear in the severe TBI column. † Significant difference between these means. At the 6-month assessment on the BOTMP, the control and mild groups both differed significantly from the severe group. ‡ One-way ANOVA used for the 3 groups.
All of the control subjects, those with mild TBI and 73% of the subjects with severe TBI, used handwriting as their primary source of written communication. All of the subjects with mild TBI had good handwriting, but 7% (n ⫽ 1) of the control subjects wrote poorly. Twenty-seven percent of subjects with severe TBI had poor handwriting and another 27% used a combination of keyboard and handwriting to record written work. Most of the control subjects and subjects with mild TBI (94% and 93%, respectively) used a normal or variant of the normal dynamic tripod grasp when handwriting, in comparison with 64% of subjects with severe TBI. The remainder of the severe subjects had a delayed (7%) or abnormal (29%) grasp. All the control subjects and mild TBI subjects had functional bilateral upper-limb activity, compared with only 69% of the severely head-injured subjects. Thirteen percent of severe TBI subjects had no bilateral activity and 19% had limited bilateral activity. Quantitative assessment: PDMS. All group means were in the average range for the total score and subtests with no significant group differences. Forty percent of the control group and 11% and 33% of the mild and severe TBI subjects, respectively, had scores in the impaired range of the PDMS. This included 2 children in the severe TBI group who were unable to complete the assessment because of significant physical disability (table 5). Quantitative assessment: BOTMP. All groups had means within the normal range on the total score, response speed, and visuomotor subtests. The severe TBI subjects had just below average mean on the upper-limb speed and dexterity subtest. There was no significant difference between groups for the total score or subtests, though the severe TBI group difference on upper-limb speed and dexterity approached significance (p ⫽ .06). All the scores for control and mild TBI subjects were within the normal range, but 40% of severe TBI subjects had scores in the impaired range (table 5).
DISCUSSION Both qualitative measures and standardized assessments were used to evaluate the effects of TBI on upper-limb function in subjects with mild or severe TBI. The results from using the qualitative tools showed that the severe TBI group had more upper-limb function difficulties at each data collection point and that these were maintained at the 2-year follow-up. For instance, 25% to 50% of severe TBI subjects had abnormal muscle tone, arm and hand control dysfunction, had poor handwriting (or required a keyboard), had an abnormal or delayed handwriting grasp, and had difficulty with bilateral activity. The subjects with mild TBI and the control subjects had no or fewer difficulties with most of these variables. This persistence of physical disability in the severe TBI group is consistent with other researchers’ findings.2,4,15,16 In contrast with the qualitative assessments, the severe TBI group had results statistically equivalent to the mild TBI and control groups, except on the PDMS at 6 months and the BOTMP at post-PTA and 6 months. At 2 years, the severe TBI group mean appeared lower, but not to a statistically significant degree. In addition, the means for the severe TBI group were within the normal range, except on the PDMS at post-PTA. Chaplin et al1 reported no difference between control and brain-injured subjects on the fine motor composite score of the BOTMP. One reason for the discrepancy between the qualitative and standardized measures in our findings is that some individuals in the severe TBI group were unable to complete the standardized assessments because of the severity of their disability. One effect of this is that subject numbers for statistical comparison were low. Subjects who were able to complete the standardized assessments had less-affected fine motor ability. We categorized all subjects as either functioning normally, or as having difficulty completing, or being unable to complete, the standardized assessments. We found that the severe TBI group did have more fine motor difficulties (table 5, “% below average”). Arch Phys Med Rehabil Vol 82, May 2001
648
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen
These results highlight the importance of including qualitative measures to reflect more accurately the impact of brain injury. There did appear to be a discrepancy between the severe TBI and the other 2 groups in the visuomotor subtest of the BOTMP at post-PTA and 6 months, and in the upper-limb speed and dexterity subtest at all points. Chaplin1 similarly found the upper-limb speed and dexterity subtest to discriminate between the subjects with brain injury and the control subjects. This is consistent with literature that reports that timed fine motor tests pose difficulties for children with brain injury. These severe TBI children, therefore, may not be able to perform tasks as well as their peers, in day-to-day situations when time conditions apply (eg, in the classroom, dressing for school). The upper-limb function of the children with mild TBI at both the 6-month and 2-year data collection points was unimpaired on all the variables measured. This is consistent with the conclusions of others4,15-17 and suggests that we can reassure families early after a mild TBI injury that long-term upper-limb function difficulties are unlikely. In addition, this knowledge helps clinicians justify allocation of resources to children with severe TBI who have greater need. Our results also indicate that most children with severe TBI had reached their maximum level of qualitative upper-limb function by 6 months postinjury. Although the data were not sufficiently robust to rule out continuing improvement after 6 months, the 6-month mark is useful as a point at which more definitive statements may be made about a severely brain injured child’s outcome, particularly about resource requirements for the future. Some limitations of the study are presented elsewhere.14 A major consideration is that children who needed therapeutic intervention received it. This study, therefore, incorporates any influences resulting from those interventions. There has been no attempt to separate the effects of rehabilitation intervention from the natural outcome in this group of children. Other limitations pertinent to this part of our total study include some of the measuring tools, the number of data collectors, and sample sizes for many variables. The use of standardized fine motor tools precluded inclusion of the children with significant disabilities. Not only did this decrease the sample size for these variables, but it also biased the results so to make it appear that the subjects with severe TBI were not substantially different in these areas from the other 2 groups. Furthermore, the standardized assessments were developed in the United States and some are quite old (eg, BOTMP, 1978). It is uncertain, therefore, how relevant these assessments are to Australian children, though the control group should alleviate this weakness. The data collection period spanned 4.5 years and staff changes meant that many occupational therapists collected the data. Interrater reliability was not established and, though the evaluations were all part of standard clinical practice, it is unclear how this may have influenced the results. Finally, many of the conclusions are drawn from qualitative assessment of descriptive statistics. Small sample sizes for many variables, caused partly by large attrition rates, precluded meaningful statistical analyses. Such conclusions should be viewed with caution. Future studies should greatly increase the sample sizes, use qualitative evaluations, ensure rater reliability in administering and scoring assessments, especially if multiple data collectors are necessary, and should use more current standardized assessments, particularly assessments that have an established relevance to Australian children. Arch Phys Med Rehabil Vol 82, May 2001
CONCLUSIONS This part of a multidisciplinary prospective and controlled study describes the upper-limb function of children with mild or severe brain injury. The study’s purpose was to develop accurate information with which to plan future interventions. The information is also needed to set clinical priorities and to allocate resources rationally. The qualitative evaluations clearly indicated that many severely brain-injured children have persistent difficulties with upper-limb function. Subjects with more severe disabilities were unable to complete standardized assessments, therefore, the relative equivalence of control, mild, and severe TBI groups on these measures may be misleading. Qualitative measures that ensure that all children are assessed are essential in this type of outcome research. This study’s main conclusion is that children with mild TBI present with upper-limb function within expected limits at 6 months and 2 years after trauma. These children probably do not require continued follow-up as part of routine clinical practice for upper-limb function, though intervention in other areas may still be warranted. Children with severe TBI, however, will have continuing disability that is likely to interfere with their daily activities. These children, therefore, require ongoing follow-up and intervention to help maximize their independence in all their activities. Acknowledgment: The authors thank Dr. Karen Byth for her assistance with data analysis. References 1. Chaplin D, Deitz J, Jaffe K. Motor performance in children after traumatic brain injury. Arch Phys Med Rehabil 1993;74:161-4. 2. Ewing-Cobbs L, Miner M, Fletcher J, Levin H. Intellectual, motor, and language sequelae following closed head injury in infants and preschoolers. J Pediatr Psychol 1989;14:531-47. 3. Heiden J, Small R, Caton W, Weiss M, Kurze T. Severe head injury. Clinical assessment and outcome. Phys Ther 1983;63: 1946-51. 4. Fay G, Jaffe K, Polissar N, Liao S, Rivara M, Martin K. Outcome of pediatric traumatic brain injury at three years: a cohort study. Arch Phys Med Rehabil 1994;75:733-41. 5. Jaffe K, Fay G, Polissar N, Martin K, Shurtleff H, Rivara M, et al. Severity of pediatric traumatic brain injury and neurobehavioral recovery at one year: a cohort study. Arch Phys Med Rehabil 1993;74:587-95. 6. Coster W, Haley S, Baryza M. Functional performance of young children after traumatic brain injury: a 6-month follow-up study. Am J Occup Ther 1994;48:211-8. 7. Brink J, Imbus C, Woo-Sam J. Physical recovery after severe closed head trauma in children and adolescents. J Pediatr 1980; 97:721-7. 8. Brink J, Garrett A, Hale W, Woo-Sam J, Nickel V. Recovery of motor and intellectual function in children sustaining severe head injuries. Dev Med Child Neurol 1970;12:565-71. 9. Wesson D, Williams J, Spence L, Filler R, Armstrong P, Pearl R. Functional outcome in pediatric trauma. J Trauma 1989;29:58992. 10. Eiben C, Anderson T, Lockman L, Matthews D, Dryja R, Martin J, et al. Functional outcome of closed head injury in children and young adults. Arch Phys Med Rehabil 1984;65:168-70. 11. Berger M, Pitts L, Lovely M, Edwards M, Bartkowski H. Outcome from severe head injury in children and adolescents. J Neurosurg 1985;62:194-9. 12. Thompson N, Francis D, Stuebing K, Fletcher J, Ewing-Cobbs L, Miner M, et al. Motor, visual-spatial, and somatosensory skills after closed head injury in children and adolescents: a study of change. Neuropsychology 1994;8:333-42. 13. Haley S, Baryza M, Lewin J, Cioffi M. Sensorimotor dysfunction in children with brain injury: development of a data base for evaluative research. Phys Occup Ther Pediatr 1991;11:1-26.
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen
14. O’Flaherty SO, Chivers A, Hannan TJ, Kendrick LM, McCartney LC, Wallen MA, et al. The Westmead pediatric TBI multidisciplinary outcome study: use of functional outcomes data to determine resource prioritization. Arch Phys Med Rehabil 2000;81:723-9. 15. Jaffe K, Fay G, Polissar N, Martin K, Shurtleff H, Rivara M, et al. Severity of pediatric traumatic brain injury and early neurobehavioral outcome: a cohort study. Arch Phys Med Rehabil 1992;73: 540-7. 16. Rutter M, Chadwick O, Shaffer D, Brown G. A prospective study of children with head injuries: I. Design and methods. Psychol Med 1980;10:633-45. 17. Massagli T, Jaffe K. Pediatric traumatic brain injury: prognosis and rehabilitation. Pediatr Ann 1994;23:29-36. 18. Costeff H, Groswasser Y, Landman Y, Brenner T. Survivors of severe traumatic brain injury in childhood. I. Late residual disability. Scand J Rehabil Med Suppl 1985;12:10-5. 19. Mayer T, Walker M, Shasha I, Matlak M, Johnson D. Effect of multiple trauma on outcome of pediatric patients with neurologic injuries. Childs Brain 1981;8:189-97. 20. Koskiniemi M, Kyykka T, Nybo T, Jarho L. Long-term outcome after severe brain injury in preschoolers is worse than expected. Arch Pediatr Adolesc Med 1995;149:249-54. 21. Marosszeky NEV, Ryan L, Shores EA, Batchelor J, Marosszeky JE. The PTA protocol: guidelines for using the Westmead PostTraumatic Amnesia (PTA) Scale. Sydney (Aust): Wild & Woolley; 1998.
649
22. Marosszeky NEV, Batchelor J, Shores EA, Marosszeky JE, KleinBoonschate M, Fahery PP. The performance of hospitalized, non head-injured children on the Westmead PTA Scale. Clin Neuropsychol 1993;7:86-95. 23. Shores EA, Marosszeky JE, Sandanam J, Batchelor J. Preliminary validation of a clinical scale for measuring the duration of posttraumatic amnesia. Med J Aust 1986;144:569-72. 24. Folio MR, Fewell RR. Peabody Developmental Motor Scales and Activity Cards. Allen (TX): DLM Teaching Resources; 1983. 25. Bruininks R. Bruininks-Oseretsky Test of Motor Proficiency. Circle Pines (MN): American Guidance Service; 1978. 26. Hinderer KA, Richardson PK, Atwater SW. Clinical implications of the Peabody Developmental Motor Scales: a constructive review. Phys Occup Ther Pediatr 1989;9:81-102. 27. Stokes NA, Deitz J, Crowe TK. The Peabody Developmental Fine Motor Scale: an interrater reliability study. Am J Occup Ther 1990;44:334-40. 28. Scott AD. The Brunnstrom approach: movement therapy. In: Trombly CA, editor. Occupational therapy for physical dysfunction. 2nd ed. Baltimore: Williams & Wilkins; 1983. p 97-105. Supplier a. SPSS, version 9.0; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.
Arch Phys Med Rehabil Vol 82, May 2001
Upper-Limb Function in Australian Children With Traumatic Brain Injury: A Controlled, Prospective Study Margaret A. Wallen, MA, Suzanne Mackay, BAppSc, Sharon M. Duff, BAppSc, Lynn C. McCartney, BAppSc, Stephen J. O’Flaherty, FRACP ABSTRACT. Wallen MA, Mackay S, Duff SM, McCartney LC, O’Flaherty SJ. Upper-limb function in Australian children with traumatic brain injury: a controlled, prospective study. Arch Phys Med Rehabil 2001;82:642-9. Objective: To describe upper-limb function in children with mild and severe traumatic brain injury (TBI), by using both quantitative and qualitative measures. Design: Controlled, prospective cohort study with assessment points initially, at 6 months, and at 2 years after TBI. Setting: A tertiary pediatric trauma center in Australia. Patients: Fifty-one children, ranging in age up to 14 years, who were consecutive admissions with TBI. On the basis of initial and persisting abnormal coma score and persistence of posttraumatic amnesia, they were assigned to either a mild (n ⫽ 26) or a severely injured (n ⫽ 25) group. Thirty children admitted with non-TBI trauma were recruited as a control group. Main Outcome Measures: Quantitative measures included Bruininks-Oseretsky Test of Motor Proficiency and Peabody Developmental Motor Scales. Qualitative measures included Brunnstrom Recovery Stages (adapted), categoric scales of muscle tone, grasp used when handwriting, quality of writing product, bilateral activity, and splint use. Results: There was little difference between the groups on the standardized assessments for subjects who could complete the tests. Qualitative measures showed the severe TBI group to have more difficulties with gross arm control, hand control, and hand function. Conclusion: Children with severe TBI experience more and persisting difficulties with upper-limb function. It is essential to include both quantitative and qualitative measures in this type of research. Key Words: Arm; Brain injuries; Disabled children; Rehabilitation. © 2001 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
From the Occupational Therapy (Wallen, Mackay) and Rehabilitation Departments (Duff, McCartney, O’Flaherty), The New Children’s Hospital, Westmead, Australia. Accepted in revised form July 26, 2000. Supported by the Motor Accidents Authority of New South Wales (grant no. RE92/28) and Australian Commonwealth Department of Human Services and Health (grant no. 93/02972). Presented in part at OT Australia–NSW 11th NSW Conference, October 15–16, 1998, in Jamberoo, NSW, Australia. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Margaret Wallen, MA, Senior Occupational Therapist/Research Coordinator, The New Children’s Hospital, PO Box 3515, Parramatta NSW 2124, Australia, e-mail: [email protected]. 0003-9993/01/8205-6154$35.00/0 doi:10.1053/apmr.2001.22620
Arch Phys Med Rehabil Vol 82, May 2001
NFORMATION ABOUT REHABILITATION outcomes Iaboutafter traumatic brain injury (TBI) guides decision making clinical and resource management. Several studies 1-12
have examined the outcomes in children with TBI, but none have studied the rehabilitation outcomes of Australian children across the age ranges. The pediatric rehabilitation service at Westmead Hospital, Sydney, Australia (now amalgamated with The Children’s Hospital at Westmead, Australia) completed a prospective, controlled, multidisciplinary outcomes study of children with TBI. Upper-limb function was one of the areas studied by the occupational therapists in the research team. Upper-limb function is important in a child’s ability to participate in daily activities and to integrate into home, school, and community life.13 It is, therefore, an integral part of occupational therapy involving children with TBI. An overview of the results of upper-limb function has been reported elsewhere14 as part of the overall study’s method and results. This article discusses in greater detail the results and implications of the upper-limb assessments. UPPER-LIMB FUNCTION Rehabilitation staff members must understand the course, nature, and extent of recovery of upper-limb function in children with TBI. This is important if resources are to be appropriately directed and motor difficulties are adequately managed so to facilitate optimal recovery, quality of movement, and functional use. Studies that have motor outcomes as a variable being investigated report that physical disability is common in severely injured children, but is less common in children with mild TBI.2,4,15,16 Few studies, however, specify upper-limb function as a separate variable, and even fewer use objective tools to measure that function. Rutter et al16 used parent interviews to determine the incidence of physical disability, but did not report specific upper-limb involvement. Eiben et al10 reported fine motor difficulties in their subjects, but included self-care items such as dressing and feeding in the fine motor category without describing performance in those areas. Ewing-Cobbs et al2 used motor scales from 2 psychometric assessments, the Bayley Scales of Infant Development and the McCarthy Scales of Children’s Abilities, to evaluate children after brain injury. Various components of the Weschler Intelligence Scale for Children–Revised and the Woodcock-Johnson Scale of Independent Behaviour have also been used.4 Although these are objective evaluations, they do not provide specific details of upper-limb performance. One study that specified a test of upper-limb performance used the Bruininks-Oseretsky Test of Motor Proficiency.1 The study found no difference between children with brain injury and control children on the fine motor composite score. Children with brain injury, however, scored significantly lower (p ⫽ .001) on the upper-limb speed and dexterity subtest. This finding is consistent with the literature, suggesting that children with TBI are significantly slower on timed tests of fine motor performance.
643
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen
Given the reported incidence of motor dysfunction in severely brain-injured children, and the impact of diminished speed of motor activity on performance in functional tasks, it appears that motor difficulties are a significant issue for children with brain injury and are likely to impact on their ability to participate in daily activities. Most studies that have investigated upper-limb function have focused on severely injured children.3,7,8,10,17-20 There is debate, however, about the longterm outcome of children with less-severe brain injury. Clinically, it is important to know whether children who sustain a mild TBI are likely to have long-term difficulties. This knowledge would allow appropriate resource allocation, including the provision of appropriate follow-up and intervention. Further research would, therefore, be useful to clarify the incidence, impact, and nature of upper-limb disability. In summary, there is a need to identify the impact of both mild and severe TBI on upper-limb abilities in children. This study was designed to describe and compare the upper-limb function of groups of Australian children with non-TBI trauma, mild TBI, and severe TBI at 6 months and 2 years postinjury. Both quantitative and qualitative measures were used to ensure a comprehensive assessment, within the context of a prospective, multidisciplinary outcome study. METHOD This study is a component of a large multidisciplinary prospective, longitudinal, cohort follow-up study (the detailed methodology was published elsewhere14). A summary of the methods and details of the assessment of upper-limb function are presented here. Subjects and Sampling Twenty-six children with severe TBI, 25 children with mild TBI, and 30 control children were recruited for the study. Subjects included all children and adolescents, up to 14 years of age, who were admitted consecutively to Westmead Hospital after a TBI during July 1992 through December 1994. The mean age of the subjects recruited initially was: control ⫾ standard deviation (SD), 7.5 ⫾ 4 years (range, 10mo–14yr); mild TBI, 5.5 ⫾ 4.7 years (range, 1–14yr); and severe TBI, 7.6 ⫾ 4.8 years (range, 1mo–13yr). See table 1 for other demographic information. There were a number of dropouts by the 6-month (25%) and 2-year (40%) assessment points. We did not specifically record their reasons to discontinue, but mostly families chose not to continue participating or could not be contacted. The dropouts did not differ from the rest of the group in respect to age, gender, parental occupational prestige, poverty status, marital status, or maternal education.14 Subjects had to have experienced head trauma, a loss of consciousness that was observed by a reliable witness, and/or a lowered coma score within 24 hours of the TBI. Coma score refers either to the Adelaide Paediatric Coma Score, which was used with subjects younger than 5 years old, or to the Glasgow Coma Score (GCS), which was used with the older subjects. Exclusionary criteria included a history of preexisting neurologic conditions, previous neurologic insult or loss of consciousness, developmental, intellectual, or learning disability, nonaccidental injury, presence of psychiatric illness, epilepsy, or progressive disease. Eligible subjects were then assigned to “severe” or “mild” TBI groups on the basis of: (1) initial coma score, (2) abnormal coma score over the first week postinjury, and (3) persistence of posttraumatic amnesia (PTA) in subjects older than 7 years of age. The Westmead Post-Traumatic Amnesia Scale was used to measure PTA.21-23 The control group consisted of 30 children admitted consecutively to Westmead Hospital during the study period with
Table 1: Demographic Information Control TBI
Gender Boys Girls Initial GCS % (n) 0–8 9–12 13–15 Time ICP was raised 0h ⬍12h 12–24h ⬎24h Days in ICU % (n) 0 1–3 ⬎3 Mean days Range Duration of PTA % (n)* 0d 1–3d 4–7d ⬎7d Mean days Range
22 8
Mild TBI
19 6
Severe TBI
22 4 100 (16)
6 (1) 94 (15)
38 (6) 63 (10)
100 (16)
100 (16)
94 (16)
73 (11) 26 (4)
6 (1) 0.5 0–8 100 (16)
0 0
0.5 0–3
38 13 13 38
(6) (2) (2) (6)
26 (4) 74 (12) 12.6 1–65
81 (13) 19 (3)
0.2 0–1
13 (1) 88 (7) 85.6 7–180
Abbreviations: ICP, intercranial pressure; ICU, intensive care unit. * PTA data not available for all severely injured children.
non– brain-injury trauma. The same exclusionary criteria were applied to the control group. Twenty-six were admitted for nonskull fractures and the remaining 4 had soft tissue injuries or burns to less than 10% of the body. The reason for using a control group that had experienced trauma was to allow for the effects of hospitalization and the problems associated with trauma because these factors may influence recovery and outcome. Wesson et al9 found that 50% of children hospitalized with severe non– brain-related injuries had disability persisting for at least 6 months postinjury. We also wanted a comparison group that may have had predisposing behaviors that put those group members at risk of sustaining injury. Finally, we wanted to differentiate changes on testing resulting from development, recovery, and the effect of repeated testing. Procedure The Western Sydney Area Health Service Ethics Committee granted its approval for the study. Once the eligible subjects and controls were identified, informed consent was obtained from the subjects’ parents or caregivers. All subjects completed a full post-PTA assessment (quantitative and qualitative measures) once they were medically stable. In mildly and severely injured children, this assessment was completed within a week after they emerged from PTA or had achieved a coma score normal for their age. A screening assessment of baseline, qualitatively measured upper-limb function was completed before this point on mildly and severely injured children who were stable but still in PTA. A second stage of assessments was completed 6 months after the post-PTA assessment. A final assessment was completed 2 years postinjury. Control subjects were assessed initially, at 6 months, and 2 years postinjury. Arch Phys Med Rehabil Vol 82, May 2001
644
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen
Multiple data collectors, including occupational therapists or final-year occupational therapy students, were trained in the study’s administration procedures, however, interrater reliability was not completed between data collectors. Tools Both quantitative and qualitative aspects of upper-limb performance were evaluated. The quantitative evaluation consisted of standardized norm-referenced tests. Haley et al13 recommended the tests to obtain a precise estimation of recovery rate and comparison with an age-appropriate reference. Qualitative assessment of upper-limb performance was included to describe movement pattern difficulties and to ensure that data were available for all subjects, including those for whom standardized assessments were inappropriate (eg, subjects with severe motor problems or with difficulty comprehending or following test instructions). Quantitative tools. The Peabody Developmental Motor Scales–Fine Motor24 (PDMS-FM) and the Bruininks-Oseretsky Test of Motor Proficiency–Fine Motor Composite25 (BOTMPFMC) were used as the standardized norm-referenced tests of upper-limb function. The PDMS was used with subjects who, at the initial evaluation, were younger than 4 years, 11 months of age. The BOTMP was used with subjects 5 years of age and older. This age cutoff meant that subjects could be reassessed at the 2-year follow-up evaluation by using the same measure that was used at the post-PTA assessment. The 4 categories of the PDMS-FM (grasping, hand use, eye-hand coordination, manual dexterity) were completed. Reliability has been stable over time and between raters, particularly for the total score.26,27 Content- and criterion-related validity have been established as being reasonable.26,27 The 3 subtests of the BOTMP-FMC (response speed, visual motor control, upper-limb speed and dexterity) were used. The test manual reports evidence of validity, particularly the correlation of scores with age, internal consistency of the subtests,
and the ability to discriminate between children with and without disabilities. Test-retest and interrater reliability are also reported as satisfactory. Qualitative tools. The Brunnstrom Recovery Stages28 were adapted for use with subjects aged 16 months and older. The adaptation allowed a description of qualitative gross arm control and hand control, including tremors and ataxia. Categoric scales were developed to gather information on muscle tone at all assessment points, on grasp used when handwriting, the quality of writing product, bilateral activity, and splint usage at the 2-year assessment. Data Analysis Data were analyzed by using SPSS.a Most variables had insufficient numbers for statistical analysis so descriptive summaries were used to examine the data. One-way analysis of variance was used to compare groups for the continuous data of the standardized upper-limb function assessments. A few children in the severe TBI group were unable to complete the standardized assessment because of significant physical disability. To include these children in the results for standardized assessments, the percentage of children who scored below average in the assessments was calculated. This percentage included children unable to be assessed on standardized tools. RESULTS Screening Assessment Information was only gathered at this point from subjects with mild TBI (n ⫽ 4) and with severe TBI (n ⫽ 16) who were in PTA. All mild TBI subjects presented with normal muscle tone. Most severely injured subjects presented with abnormal tone or motor control (table 2). The older subjects were assessed on the modified Brunnstrom scales. No mild TBI subjects had difficulties with arm or
Table 2: Percentage of Control, Mild, and Severe Subjects in Each Muscle Tone Rating at Screening, Post-PTA, 6 Months, and 2 Years, for Left and Right Upper Limbs Screening Muscle Tone
Normal L R Abnormal Spasticity L R Hypotonia L R Fluctuating L R Ataxia L R
n (total patients in group) L R
Mild TBI (%)
Severe TBI % (n)
100 100
27 (4) 19 (3)
Post-PTA
6 Months
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
100 100
50 (13) 54 (14)
2 Years
Control (%)
Mild TBI % (n)
Severe TBI % (n)
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
95 (18) 95 (18)
82 (18) 64 (14)
100 100
100 100
73 (11) 60 (9)
20 (3) 25 (4)
15 (4) 12 (3)
5 (1) 18 (4)
27 (4) 44 (7)
15 (4) 15 (4)
5 (1) 9 (2)
20 (3) 40 (6)
7 (1) 6 (1) 20 (3) 6 (1)
3 4
15 16
Abbreviations: L, left; R, right.
Arch Phys Med Rehabil Vol 82, May 2001
19 (5) 19 (5)
28 28
23 24
26 26
20 20
5 (1) 5 (1)
9 (2) 9 (2)
19 19
22 22
7 (1)
17 17
16 16
15 15
645
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen Table 3: Percentage of Control, Mild, and Severe TBI Subjects in Each Category of the Modified Brunnstrom Scales for Left and Right Gross Arm Control at Screening, Post-PTA, 6 Months, and 2 Years Screening Modified Brunnstrom Scales
Gross arm control No abnormalities L R Poor quality L R Ataxia/no coordination L R Independence from synergies L R Movement deviating from synergies L R Voluntary synergies L R Synergies developing L R No movement L R
n (total in group) L R
Mild TBI (%)
Severe TBI % (n)
100 100
33 (4) 8 (1)
Post-PTA
6 Months
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
100 100
57 (13) 57 (13)
2 Years
Control (%)
Mild TBI (%)
Severe TBI % (n)
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
100 100
79 (16) 74 (13)
100 100
100 100
69 (11) 56 (9)
17 (2) 15 (2)
22 (5) 13 (3)
5 (1) 5 (1)
6 (1) 6 (1)
17 (2) 8 (1)
13 (3) 9 (2)
11 (2) 16 (3)
6 (1) 13 (1)
6 (1) 4 (1)
9 (2)
5 (1)
6 (1)
6 (1) 13 (2)
17 (2) 23 (3)
9 (2)
17 (2) 46 (6) 3 4
12 13
5 (1)
6 (1) 6 (1)
9 (2) 26 26
22 23
hand control. Most of the subjects with severe TBI had some abnormality resulting from their brain injuries (tables 3, 4). Post-PTA Thirty control subjects, 24 with mild TBI, and 26 with severe TBI were assessed post-PTA. Full data sets were not available because not all subjects completed all the assessments. All control subjects and those with mild TBI presented with normal muscle tone. About half of the severely injured subjects continued to experience difficulties with muscle tone (table 2) and difficulties on the modified Brunnstrom scales (tables 3, 4). Quantitative assessment: PDMS. The control and mild TBI groups obtained mean scores within the normal range. The severe TBI group mean was slightly more than 1 SD below the mean of the normative sample (table 5). There was no significant difference between groups. All group means were within the normal range on the subtests of the PDMS except for the severe TBI group, which scored just below the normal range on eye-hand coordination. There was no significant difference between groups on any of the subtests of the PDMS. Twenty-five percent of the control group and 31% and 54% of the mild and severe TBI subjects, respectively, had total scores below the average or were unable to complete the assessment (table 5). Most of these difficulties appeared to be experienced in the eye-hand coordination subtest. Two subjects
23 23
19 19
19 19
19 19
17 17
16 16
16 16
with severe TBI and significant motor difficulties were unable to complete the assessment. Inability to test these children, and, therefore, to include their scores on the PDMS, skewed the results for the severe TBI group to being less impaired than it actually was. Quantitative assessment: BOTMP. Older subjects were assessed by using the BOTMP. All group means were within the normal range, though the severe TBI group scored significantly lower than the mild TBI group (table 5). All group means were within the normal range for all the subtests, though the severely injured group scored significantly lower than the mild TBI group on the visuomotor and upper-limb speed and dexterity subtests. Despite means within the normal range, 57% of severely injured subjects had scores below average or were unable to complete the assessment (table 5), with most of the difficulties occurring in the response speed and upper-limb speed and dexterity subtests. Six Months Twenty control subjects, 19 with mild TBI, and 22 with severe TBI were assessed, though complete data sets were not available for all subjects. Qualitative assessment. All control subjects, 95% (n ⫽ 18) of subjects with mild TBI, and most of the subjects with severe TBI had normal tone (table 2). All control subjects had normal Arch Phys Med Rehabil Vol 82, May 2001
646
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen Table 4: Percentage of Control, Mild, and Severe TBI Subjects on Each Category of the Modified Brunnstrom Scales for Left and Right Hand Control at Screening, Post-PTA, 6 Months, and 2 Years Screening
Modified Brunnstrom Scales
Hand control No abnormalities L R Poor quality L R Ataxia/no coordination L R Palmar prehension L R Thumb release L R Mass grasp and wrist extension L R Mass finger flexion L R No movement L R
n (total in group) L R
Mild TBI (%)
Severe TBI % (n)
100 100
25 (3) 8 (1)
Post-PTA
6 Months
Control (%)
Mild TBI % (n)
Severe TBI % (n)
100 100
96 (21) 96 (22)
67 (14) 50 (11)
17 (2) 15 (2)
Control (%)
Severe TBI % (n)
Control (%)
Mild TBI (%)
Severe TBI % (n)
100 100
95 (18) 95 (18)
79 (15) 63 (12)
100 100
100 100
69 (11) 56 (9)
5 (1) 5 (1)
11 (2) 16 (3)
6 (1) 13 (2)
11 (2) 11 (2)
6 (1) 6 (1)
14 (3) 18 (4)
17 (2) 8 (1)
4 (1) 4 (1)
2 Years
Mild TBI % (n)
10 (2) 14 (3)
6 (1) 5 (1) 5 (1)
6 (1)
8 (1)
17 (2) 23 (3)
5 (1)
5 (1)
6 (1) 6 (1)
10 (2) 5 (1)
5 (1)
6 (1)
17 (2) 46 (6) 3 4
12 13
6 (1) 6 (1) 25 24
22 23
movement on the Brunnstrom scales, as did all but 1 subject with mild TBI. Several of the subjects with severe TBI continued to have difficulties when assessed with the Brunnstrom scales (tables 3, 4). Quantitative assessment: PDMS. The control group mean for the total score was just below the average range. Although both the mild and severe TBI groups had scores within the normal range, the severely injured group had a significantly lower mean than the group with mild TBI (table 5). All group means for subtests were within the average range except for the control group for eye-hand coordination. There were no differences between groups on any of the subtests. Forty percent of the control group and 8% of the subjects with mild TBI had scores in the below-average range. Forty percent of the severe TBI subjects had scores in the belowaverage range, including 1 child who was unable to complete the assessment and was thus classified as below average (table 5). Quantitative assessment: BOTMP. All group means were within the average range, though the severe TBI group mean was significantly lower than both the control and mild TBI groups. The severe TBI group mean was also significantly lower than the mild TBI and control groups for both the visuomotor and upper-limb speed and dexterity subtests, though the upper-limb speed and dexterity subtest mean was the only one that was in the below-average range. Seven percent of the control group and 17% of the subjects with Arch Phys Med Rehabil Vol 82, May 2001
21 22
19 19
19 19
19 19
17 17
16 16
16 16
severe TBI were below the average range or were unable to complete the assessment (table 5). None of the subjects in the mild TBI group were below the average range or unable to complete the assessment. Two Years Seventeen control subjects, 16 with mild TBI, and 16 with severe TBI were assessed at this point. Full data sets were not available for all subjects. Qualitative assessment. Only subjects with severe TBI experienced difficulties with muscle tone (table 2), with 3 subjects requiring wrist/hand splints (2 subjects with single wrist/hand splints, the other with bilateral wrist/hand splints). No control subjects or subjects with mild TBI experienced difficulties on the Brunnstrom arm or hand scales. Many of the severe TBI group continued to have difficulty on both these scales (tables 3, 4). Other final qualitative outcome variables (hand dominance, written communication, handwriting grasp, bilateral activity) are also reported as follows. Most subjects had normal hand dominance (control, 94%; mild TBI, 94%; severe TBI, 73%). One child in the control group had not yet established a dominance, which was considered abnormal for age. One subject with mild TBI had changed dominance, with 4 severe TBI subjects having abnormal hand dominance (ie, changing dominance as a result of TBI or a lack of established dominance, which was considered inappropriate for age).
647
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen Table 5: Scores Achieved on the PDMS and BOTMP by Control, Mild, and Severe TBI Subjects at Post-PTA, 6 Months, and 2 Years Post-PTA
PDMS total score n* Mean ⫾ SD Range p‡ Percentage below average (n)* BOTMP-FMC scores n Mean ⫾ SD Range p‡ Percentage below average
Control
Mild TBI
4 42 ⫾ 9.2 29–50 0.2 (NS) 25
13 46 ⫾ 9.9 27–61
16 57 ⫾ 9.1 41–76
9 65 ⫾ 6.6† 49–70 .03 0
0
31
6 Months Severe TBI
9 39 ⫾ 7.6 27–51 55 (11) 10 48 ⫾ 20.7† 20–79 57
2 Years
Control
Mild TBI
Severe TBI
Control
Mild TBI
Severe TBI
5 39 ⫾ 10.8 27–51
12 50 ⫾ 9.5† 27–60 .05 8
9 41 ⫾ 8.5† 31–56
5 42 ⫾ 4.8 37–48 0.8 (NS) 40
9 44 ⫾ 6.7 31–55
4 45 ⫾ 6.5 40–54
11
33 (6)
6 59 ⫾ 9.8† 44–73
12 48 ⫾ 12† 20–67
10 56 ⫾ 10.3 40–70 .08 (NS) 0
5 63 ⫾ 6.7 52–69
5 46 ⫾ 15 36–72
0
40
40 14 58 ⫾ 8.4† 37–70 .03 7
0
40 (10)
17
NOTE. Scores for PDMS and BOTMP are expressed as T scores with a mean ⫾ SD of 50 ⫾ 10. Scores of ⱕ 39 are considered below average. Abbreviation: NS, nonsignificant. * n is the number of children who completed the assessment in each group. (n) is the revised n where those children unable to be assessed by using the standardized assessments because of severity of disability have been included in a calculation of percentage in each group who were below average (ie, more than 1 SD below the mean). Only children in the severe TBI group were unable to complete the standardized assessment, thus, these data only appear in the severe TBI column. † Significant difference between these means. At the 6-month assessment on the BOTMP, the control and mild groups both differed significantly from the severe group. ‡ One-way ANOVA used for the 3 groups.
All of the control subjects, those with mild TBI and 73% of the subjects with severe TBI, used handwriting as their primary source of written communication. All of the subjects with mild TBI had good handwriting, but 7% (n ⫽ 1) of the control subjects wrote poorly. Twenty-seven percent of subjects with severe TBI had poor handwriting and another 27% used a combination of keyboard and handwriting to record written work. Most of the control subjects and subjects with mild TBI (94% and 93%, respectively) used a normal or variant of the normal dynamic tripod grasp when handwriting, in comparison with 64% of subjects with severe TBI. The remainder of the severe subjects had a delayed (7%) or abnormal (29%) grasp. All the control subjects and mild TBI subjects had functional bilateral upper-limb activity, compared with only 69% of the severely head-injured subjects. Thirteen percent of severe TBI subjects had no bilateral activity and 19% had limited bilateral activity. Quantitative assessment: PDMS. All group means were in the average range for the total score and subtests with no significant group differences. Forty percent of the control group and 11% and 33% of the mild and severe TBI subjects, respectively, had scores in the impaired range of the PDMS. This included 2 children in the severe TBI group who were unable to complete the assessment because of significant physical disability (table 5). Quantitative assessment: BOTMP. All groups had means within the normal range on the total score, response speed, and visuomotor subtests. The severe TBI subjects had just below average mean on the upper-limb speed and dexterity subtest. There was no significant difference between groups for the total score or subtests, though the severe TBI group difference on upper-limb speed and dexterity approached significance (p ⫽ .06). All the scores for control and mild TBI subjects were within the normal range, but 40% of severe TBI subjects had scores in the impaired range (table 5).
DISCUSSION Both qualitative measures and standardized assessments were used to evaluate the effects of TBI on upper-limb function in subjects with mild or severe TBI. The results from using the qualitative tools showed that the severe TBI group had more upper-limb function difficulties at each data collection point and that these were maintained at the 2-year follow-up. For instance, 25% to 50% of severe TBI subjects had abnormal muscle tone, arm and hand control dysfunction, had poor handwriting (or required a keyboard), had an abnormal or delayed handwriting grasp, and had difficulty with bilateral activity. The subjects with mild TBI and the control subjects had no or fewer difficulties with most of these variables. This persistence of physical disability in the severe TBI group is consistent with other researchers’ findings.2,4,15,16 In contrast with the qualitative assessments, the severe TBI group had results statistically equivalent to the mild TBI and control groups, except on the PDMS at 6 months and the BOTMP at post-PTA and 6 months. At 2 years, the severe TBI group mean appeared lower, but not to a statistically significant degree. In addition, the means for the severe TBI group were within the normal range, except on the PDMS at post-PTA. Chaplin et al1 reported no difference between control and brain-injured subjects on the fine motor composite score of the BOTMP. One reason for the discrepancy between the qualitative and standardized measures in our findings is that some individuals in the severe TBI group were unable to complete the standardized assessments because of the severity of their disability. One effect of this is that subject numbers for statistical comparison were low. Subjects who were able to complete the standardized assessments had less-affected fine motor ability. We categorized all subjects as either functioning normally, or as having difficulty completing, or being unable to complete, the standardized assessments. We found that the severe TBI group did have more fine motor difficulties (table 5, “% below average”). Arch Phys Med Rehabil Vol 82, May 2001
648
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen
These results highlight the importance of including qualitative measures to reflect more accurately the impact of brain injury. There did appear to be a discrepancy between the severe TBI and the other 2 groups in the visuomotor subtest of the BOTMP at post-PTA and 6 months, and in the upper-limb speed and dexterity subtest at all points. Chaplin1 similarly found the upper-limb speed and dexterity subtest to discriminate between the subjects with brain injury and the control subjects. This is consistent with literature that reports that timed fine motor tests pose difficulties for children with brain injury. These severe TBI children, therefore, may not be able to perform tasks as well as their peers, in day-to-day situations when time conditions apply (eg, in the classroom, dressing for school). The upper-limb function of the children with mild TBI at both the 6-month and 2-year data collection points was unimpaired on all the variables measured. This is consistent with the conclusions of others4,15-17 and suggests that we can reassure families early after a mild TBI injury that long-term upper-limb function difficulties are unlikely. In addition, this knowledge helps clinicians justify allocation of resources to children with severe TBI who have greater need. Our results also indicate that most children with severe TBI had reached their maximum level of qualitative upper-limb function by 6 months postinjury. Although the data were not sufficiently robust to rule out continuing improvement after 6 months, the 6-month mark is useful as a point at which more definitive statements may be made about a severely brain injured child’s outcome, particularly about resource requirements for the future. Some limitations of the study are presented elsewhere.14 A major consideration is that children who needed therapeutic intervention received it. This study, therefore, incorporates any influences resulting from those interventions. There has been no attempt to separate the effects of rehabilitation intervention from the natural outcome in this group of children. Other limitations pertinent to this part of our total study include some of the measuring tools, the number of data collectors, and sample sizes for many variables. The use of standardized fine motor tools precluded inclusion of the children with significant disabilities. Not only did this decrease the sample size for these variables, but it also biased the results so to make it appear that the subjects with severe TBI were not substantially different in these areas from the other 2 groups. Furthermore, the standardized assessments were developed in the United States and some are quite old (eg, BOTMP, 1978). It is uncertain, therefore, how relevant these assessments are to Australian children, though the control group should alleviate this weakness. The data collection period spanned 4.5 years and staff changes meant that many occupational therapists collected the data. Interrater reliability was not established and, though the evaluations were all part of standard clinical practice, it is unclear how this may have influenced the results. Finally, many of the conclusions are drawn from qualitative assessment of descriptive statistics. Small sample sizes for many variables, caused partly by large attrition rates, precluded meaningful statistical analyses. Such conclusions should be viewed with caution. Future studies should greatly increase the sample sizes, use qualitative evaluations, ensure rater reliability in administering and scoring assessments, especially if multiple data collectors are necessary, and should use more current standardized assessments, particularly assessments that have an established relevance to Australian children. Arch Phys Med Rehabil Vol 82, May 2001
CONCLUSIONS This part of a multidisciplinary prospective and controlled study describes the upper-limb function of children with mild or severe brain injury. The study’s purpose was to develop accurate information with which to plan future interventions. The information is also needed to set clinical priorities and to allocate resources rationally. The qualitative evaluations clearly indicated that many severely brain-injured children have persistent difficulties with upper-limb function. Subjects with more severe disabilities were unable to complete standardized assessments, therefore, the relative equivalence of control, mild, and severe TBI groups on these measures may be misleading. Qualitative measures that ensure that all children are assessed are essential in this type of outcome research. This study’s main conclusion is that children with mild TBI present with upper-limb function within expected limits at 6 months and 2 years after trauma. These children probably do not require continued follow-up as part of routine clinical practice for upper-limb function, though intervention in other areas may still be warranted. Children with severe TBI, however, will have continuing disability that is likely to interfere with their daily activities. These children, therefore, require ongoing follow-up and intervention to help maximize their independence in all their activities. Acknowledgment: The authors thank Dr. Karen Byth for her assistance with data analysis. References 1. Chaplin D, Deitz J, Jaffe K. Motor performance in children after traumatic brain injury. Arch Phys Med Rehabil 1993;74:161-4. 2. Ewing-Cobbs L, Miner M, Fletcher J, Levin H. Intellectual, motor, and language sequelae following closed head injury in infants and preschoolers. J Pediatr Psychol 1989;14:531-47. 3. Heiden J, Small R, Caton W, Weiss M, Kurze T. Severe head injury. Clinical assessment and outcome. Phys Ther 1983;63: 1946-51. 4. Fay G, Jaffe K, Polissar N, Liao S, Rivara M, Martin K. Outcome of pediatric traumatic brain injury at three years: a cohort study. Arch Phys Med Rehabil 1994;75:733-41. 5. Jaffe K, Fay G, Polissar N, Martin K, Shurtleff H, Rivara M, et al. Severity of pediatric traumatic brain injury and neurobehavioral recovery at one year: a cohort study. Arch Phys Med Rehabil 1993;74:587-95. 6. Coster W, Haley S, Baryza M. Functional performance of young children after traumatic brain injury: a 6-month follow-up study. Am J Occup Ther 1994;48:211-8. 7. Brink J, Imbus C, Woo-Sam J. Physical recovery after severe closed head trauma in children and adolescents. J Pediatr 1980; 97:721-7. 8. Brink J, Garrett A, Hale W, Woo-Sam J, Nickel V. Recovery of motor and intellectual function in children sustaining severe head injuries. Dev Med Child Neurol 1970;12:565-71. 9. Wesson D, Williams J, Spence L, Filler R, Armstrong P, Pearl R. Functional outcome in pediatric trauma. J Trauma 1989;29:58992. 10. Eiben C, Anderson T, Lockman L, Matthews D, Dryja R, Martin J, et al. Functional outcome of closed head injury in children and young adults. Arch Phys Med Rehabil 1984;65:168-70. 11. Berger M, Pitts L, Lovely M, Edwards M, Bartkowski H. Outcome from severe head injury in children and adolescents. J Neurosurg 1985;62:194-9. 12. Thompson N, Francis D, Stuebing K, Fletcher J, Ewing-Cobbs L, Miner M, et al. Motor, visual-spatial, and somatosensory skills after closed head injury in children and adolescents: a study of change. Neuropsychology 1994;8:333-42. 13. Haley S, Baryza M, Lewin J, Cioffi M. Sensorimotor dysfunction in children with brain injury: development of a data base for evaluative research. Phys Occup Ther Pediatr 1991;11:1-26.
UPPER-LIMB FUNCTION IN CHILDREN POST-TBI, Wallen
14. O’Flaherty SO, Chivers A, Hannan TJ, Kendrick LM, McCartney LC, Wallen MA, et al. The Westmead pediatric TBI multidisciplinary outcome study: use of functional outcomes data to determine resource prioritization. Arch Phys Med Rehabil 2000;81:723-9. 15. Jaffe K, Fay G, Polissar N, Martin K, Shurtleff H, Rivara M, et al. Severity of pediatric traumatic brain injury and early neurobehavioral outcome: a cohort study. Arch Phys Med Rehabil 1992;73: 540-7. 16. Rutter M, Chadwick O, Shaffer D, Brown G. A prospective study of children with head injuries: I. Design and methods. Psychol Med 1980;10:633-45. 17. Massagli T, Jaffe K. Pediatric traumatic brain injury: prognosis and rehabilitation. Pediatr Ann 1994;23:29-36. 18. Costeff H, Groswasser Y, Landman Y, Brenner T. Survivors of severe traumatic brain injury in childhood. I. Late residual disability. Scand J Rehabil Med Suppl 1985;12:10-5. 19. Mayer T, Walker M, Shasha I, Matlak M, Johnson D. Effect of multiple trauma on outcome of pediatric patients with neurologic injuries. Childs Brain 1981;8:189-97. 20. Koskiniemi M, Kyykka T, Nybo T, Jarho L. Long-term outcome after severe brain injury in preschoolers is worse than expected. Arch Pediatr Adolesc Med 1995;149:249-54. 21. Marosszeky NEV, Ryan L, Shores EA, Batchelor J, Marosszeky JE. The PTA protocol: guidelines for using the Westmead PostTraumatic Amnesia (PTA) Scale. Sydney (Aust): Wild & Woolley; 1998.
649
22. Marosszeky NEV, Batchelor J, Shores EA, Marosszeky JE, KleinBoonschate M, Fahery PP. The performance of hospitalized, non head-injured children on the Westmead PTA Scale. Clin Neuropsychol 1993;7:86-95. 23. Shores EA, Marosszeky JE, Sandanam J, Batchelor J. Preliminary validation of a clinical scale for measuring the duration of posttraumatic amnesia. Med J Aust 1986;144:569-72. 24. Folio MR, Fewell RR. Peabody Developmental Motor Scales and Activity Cards. Allen (TX): DLM Teaching Resources; 1983. 25. Bruininks R. Bruininks-Oseretsky Test of Motor Proficiency. Circle Pines (MN): American Guidance Service; 1978. 26. Hinderer KA, Richardson PK, Atwater SW. Clinical implications of the Peabody Developmental Motor Scales: a constructive review. Phys Occup Ther Pediatr 1989;9:81-102. 27. Stokes NA, Deitz J, Crowe TK. The Peabody Developmental Fine Motor Scale: an interrater reliability study. Am J Occup Ther 1990;44:334-40. 28. Scott AD. The Brunnstrom approach: movement therapy. In: Trombly CA, editor. Occupational therapy for physical dysfunction. 2nd ed. Baltimore: Williams & Wilkins; 1983. p 97-105. Supplier a. SPSS, version 9.0; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.
Arch Phys Med Rehabil Vol 82, May 2001