Factors associated with improvement in the gross motor function outcomes of children with acquired brain injury in a paediatric intermediate care facility in the Western Cape, South Africa: A descriptive study

Master Thesis

2018

Permanent link to this Item
Authors
Supervisors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher

University of Cape Town

License
Series
Abstract
Aim: The study aimed to record the recovery patterns of gross motor function following acquired brain injury (ABI) in 17 participants, aged two to 14 years, receiving physiotherapy intervention at a paediatric intermediate care facility in the Western Cape, South Africa. Objectives of the study were to explore if factors such as age at injury, time since injury, intervention type, injury severity, gender, and the health- related quality of life of the child and caregiver were predictive of improvement of motor function and participation. Methods: A quantitative, longitudinal, prospective cohort design with repeated measures was employed. The children received physiotherapy either intensively or intermittently within eight months post brain insult. The Gross Motor Function Classification System (GMFCS) was used to classify the children according to their functional abilities. The Gross Motor Function Measure 88 (GMFM88) was used to assess changes in the child’s gross motor function. The Paediatric Evaluation of Disability Inventory (PEDI) evaluated functional performance in daily life activities (participation). The EQ-5D-Y proxy version and the Caregiver Strain Index (CSI) evaluated the health-related quality of life of the children and caregivers respectively. Assessments were performed at Baseline, Week 3, 7, 9 and 13 after admission to the study. Analysis: Descriptive statistics were used to describe the demographic data of the participants. Non-parametric analysis was performed to determine the time points at which the greatest improvement occurred because the sample was relatively small to support an adequately powered randomized intermittent trial. The GMFM88 was the primary outcome measure and the changes in score over the 13-week study period were plotted graphically in an attempt to identify patterns of improvement. The change in score from Baseline to Week 3 was also used as the dependent variable in investigating the determinants of short term improvement between the five measurement time points. Scatterplots and Spearman’s rho were used to investigate the relationship between changes in GMFM88 score from Baseline to Week 3 and the age of the child at injury, time from injury and Baseline GMFM88 score. The Kruskal Wallis ANOVA for ordinal data was used to establish whether there were differences in all the outcome measures at the different time points: between Baseline and Week 3, between Week 3 and Week 7 and between Week 7 and Week 9 and, if significant, a post-hoc Sign test was done to see where the differences lay. The effect size was calculated by dividing the z value by the square root of the total number of observations at both time points. Simple regression analysis was used to determine the variance in Week 9 GMFM88% score accounted for by the Baseline score. Results: Seventeen participants met the inclusion criteria and were recruited. There were more boys (n=14) than girls (n=3) admitted with brain injury and the majority of the total sample (n=14) were drawn from families with minimal financial resources. Although no differences were found between the two intervention groups, the greatest improvement in GMFM88 scores was observed within the first three weeks of intervention, regardless of the time since injury. Three patterns of recovery were identified through plotting the GMFM88 scores: Group A, included participants with a high baseline score who showed sustained but small improvement until reaching the ceiling score; Group B, participants who started with lower scores and then improved considerably; and Group C, participants who started with low scores and showed slower and smaller improvement. Most participants (n=13) returned to near normal motor function after Week 9 (over 90% on the GMFM). Four participants did not show the same improvement: The Week 9 GMFM88 score was less than 60% in one participant and less than 30% for the other three. In this study, participants who showed the greatest improvement were those children on GMFCS levels three and four. The Baseline GMFM88 scores were significantly correlated with other time points (rho=0.886 at Week 9 and 0.748 at Week 13), but not with the change in scores. The effect size of the change in GMFM88 scores from one time point to the other were all measured as medium (effect size of 0.5 but less than 1.3) and the largest effect size was seen between Baseline and Week 7 (effect size=.660). Change in score from Baseline to Week 3 was not predicted by gender, cause of injury or method of intervention delivered (intensive or intermittent). The Baseline motor score accounted for 86% of the variance of the Week 9 GMFM88% score and each point in the baseline score increased the Week 9 score by .94. The PEDI Mobility Score demonstrated a similar pattern of improvement to the GMFM88, but the pattern in progression of the PEDI Self-care domain was less clear. The EQ-5D-Y Proxies indicated that at Week 13, at least five children still had problems in one or more of the dimensions, with the greatest number (n=-8) having problems with the Usual Activities and Worried, Sad, or Unhappy domains. The older (above age nine years), higher functioning children made less gains in their functional abilities. The greatest cumulative cause of strain reported by the thirteen respondents, was a change in the former self of the child, followed closely by financial strain and work adjustments. The caregivers of the children in Group C, the most disabled group, experienced double the strain at Week 13 than caregivers of children with mild and moderate brain injury. Although not correlated at Baseline, the CSI was negatively correlated with the proxy EQ-5D-Y reported visual analogue general health scale at Week 13 (n=10, rho=-724, p=.018) .655, p=.021). Conclusions: Although, most participants in the study regained physical functioning, approximately 25% admitted for rehabilitation with ABI might have significant residual motor damage and require on-going rehabilitative support. Although there were no other factors identified which predicted the outcome, these children could be identified based on poor GMFM88 scores on admission to rehabilitation. The EQ-5D-Y indicated that at Week 13, eight of the 12 proxies reported that participants had problems with anxiety and depression. Rehabilitation might therefore need to include a greater emphasis on self-care and mental health of the child, even when motor control has been established. Planning of long-term support should start early to maximise recovery and reduce the stress on caregivers. The study findings cannot be generalised due to the small sample size, but the results should alert therapists at the Facility to the possible long-term outcomes of children admitted with ABI. The study findings can assist with the formulation of patient specific and family centred rehabilitation care plans for children admitted with ABI at the Facility.
Description

Reference:

Collections