Background: Osteoarthritis (OA) is one of the most common musculoskeletal conditions (MSCs) worldwide, most commonly affecting those over 60 years of age. OA is associated with pain and disability, and also has negative social and emotional effects on the individual. End stage management for OA of the knee, when the individual is no longer receiving adequate relief from conservative measures, is total knee arthroplasty (TKA). Long waiting lists are a challenge globally and in South Africa (SA) where patients often have to wait years to receive their surgery. Evidence suggests that the longer an individual is in pain, the more likely they are to stay in pain and thus long waiting lists may result in poor post-operative outcomes. It has been found that 10-34% of patients undergoing TKA do not gain adequate relief from surgery. Literature suggests that self-management interventions using exercise and education are effective in reducing pain and improving function in patients who are on a waiting list, in a South African context. The current literature on selfmanagement programmes does not include post-operative follow-up and thus the current study was warranted to determine the effects of a six-week pre-operative physiotherapist-led exercise and education intervention on post-operative recovery in patients with OA undergoing total knee replacement. Methods: A single-blind randomized controlled trial was conducted at Tygerberg Hospital (TBH) in the Western Cape, SA. Pain was the primary outcome measure, with disability (WOMAC), healthrelated quality of life (EQ-5D-5L), fear of movement (TSK) and function (ALF) as secondary outcomes. Participants were randomly placed into the experimental or control group. The experimental group took part in a six-week, group programme, which they attended at TBH. Classes were two hours long and consisted of education, exercise and relaxation training. The control group’s management remained unchanged and they received their usual care. Measurements were taken at recruitment, post-intervention, pre-operatively, post-operatively, and at six weeks and three-months postoperatively by a blinded research assistant. Analysis was by intention to treat. Due to the small sample size, non-parametric analysis was conducted and results are presented as median and range throughout. Effect sizes were calculated with a 95% confidence interval. Results: There were no significant changes between groups on the primary outcome measure of pain. The only outcome revealing any significant change between groups was the TSK, with a large effect size of 1.39 (0.41 – 2.26 95% CI), indicating that the experimental group performed better than the control group at three months post-operative follow-up. Conclusion: Participation in the intervention provided no significant benefit to the experimental group over the control group and did not change post-operative outcomes. There were challenges with regard to participation and completion of the measurement tools used, raising questions regarding issues of pain and fear, self-efficacy, social learning, social cohesion and behavior change. The suitability of this intervention for use in a pre-operative population for post-operative benefits, needs to be reviewed; it would appear that what is suitable for those on a waiting list, with no clarity on when surgery will occur, may not necessarily apply to those who are actually due to undergo surgery. Future research should consider modifying the intervention, including information appropriate to pre and post-operative needs of the population. The roles of self-efficacy, social cohesion and social learning in behavior change should be considered and, levels of self-efficacy and fear of movement beliefs assessed prior to randomization, in order to adequately establish the effect of the intervention on these factors.

Background Neuropathic pain affects 7 – 10% of people and responds poorly to pharmacotherapy. Numbers needed to treat for first line drugs range from 4 – 8. Therefore, there is an obvious need for improved understanding of the mechanisms of neuropathic pain to inform improved treatments. Mechanistic research on neuropathic pain frequently uses a human surrogate model of secondary hyperalgesia that is a common feature of neuropathic pain. The value of experimentally inducing secondary hyperalgesia is that one can then test the influence of different pharmacological and nonpharmacological interventions. This may shed some light on the physiological mechanisms within the spinal cord, which possibly also translates to the effects of the interventions on other pathways that are involved in processing signals that may be related to pain. Additionally, pain is known to be influenced by the threat value of the situation. Many South Africans live under constant threat: less than a third of South Africans feel safe walking alone at night. This constant threat may be perpetuating the pain problem in South Arica. However, the mechanism by which threat achieves this influence on pain is unclear. This project is focused on one possible mechanistic hypothesis: that threat influences pain by affecting central physiological changes within the dorsal horn of the spinal cord. These central changes often present clinically as secondary hyperalgesia. A thorough understanding of these mechanisms will inform improved treatment strategies. Methods Phase one: systematic review and meta-analysis The aim of this systematic review was to identify, describe, and compare methods that have been used to manipulate experimentally induced secondary hyperalgesia in healthy humans. A systematic search strategy (conducted on 01 October 2019) was supplemented by reference list checks and direct contact with identified laboratories to maximise the identification of data reporting the experimental manipulation of experimentally induced secondary hyperalgesia in humans. Studies were only included if they were published and in-press or accepted records for which the title, abstract, and full-text versions were available in English. Duplicate screening, risk of bias assessment, and data extraction procedures were used. Risk of bias was appraised for the following domains: selection, performance, detection, attrition, measurement, reporting, and other sources of bias. Data were extracted using a standardised data extraction form. This form was piloted and refined beforehand. Authors were asked to provide data were necessary. Data were pooled by method of v manipulation and outcome (intensity of secondary hyperalgesia, area of secondary hyperalgesia, or both). Phase two: experimental paradigm An experimental study was developed and conducted to investigate the effect of a stimulus threat on secondary hyperalgesia. The aim of this study was to investigate the effect of a manipulation of the threat value of a stimulus on experimentally induced secondary hyperalgesia in healthy human volunteers. All participants underwent a sham skin examination (the threat stimulus) at both the experimental and control sites. Through this sham skin examination, participants were informed that their skin integrity was fragile at the experimental site and robust at the control site. Secondary hyperalgesia was induced with high-frequency electrical stimulation at both the experimental and control site. Sensory testing was conducted at the experimental and control site, providing a withinsubject comparison of the intensity and area of secondary hyperalgesia at each site. It was hypothesised that greater threat will be associated with (hypothesis 1) greater intensity and (hypothesis 2) greater surface area of induced secondary hyperalgesia. Results Phase one: systematic review Twenty-one studies with non-pharmacological manipulations were included. Nine (out of 21) studies assessed intensity of secondary hyperalgesia after manipulation. Nicotine deprivation and negative expectations about the induction increased the intensity of secondary hyperalgesia. Three studies using attentional modulation and cognitive loading reported conflicting results with two studies having no effect and the other reporting a decrease in the intensity of secondary hyperalgesia. Emotional disclosure decreased the intensity of secondary hyperalgesia at four days and at one month after the manipulation. Hot/cold application, and verbal suggestion had no effect on the intensity of secondary hyperalgesia. Seventeen (out of 21) studies assessed area of secondary hyperalgesia after manipulation. Nicotine deprivation and sleep deprivation increased the area of secondary hyperalgesia. Hyperbaric oxygen therapy, cognitive behavioural therapy, emotional disclosure, spinal manipulation, transcranial direct current stimulation, and placebo analgesia decreased the area of secondary hyperalgesia. Interestingly, the effects of emotional disclosure and hyperbaric oxygen therapy were evident one month after manipulation. Acupuncture had no significant effect on the area of secondary hyperalgesia. Four studies assessed the effect of hot/cold application. Three studies reported no effect and one study reported an increase in the area of secondary hyperalgesia after cold application.

Includes abstract. / Includes bibliographical references (leaves 235-263). / Deficient truncal postural control is said to be the main factor in the development of motor disorders, however in cerebral palsy (CP) the role of the trunk especially in lower limb function is not clearly understood. This research project aimed to investigate the role of the abdominal muscles and determine its contribution to the motor dysfunction seen in stance and during gait in children with spastic-type CP (STCP).

Disability is a complex construct, and our understanding of it has evolved over the years from a purely medical description to encapsulating the experience of those with disability in the context in which they live. The International Classification of Functioning, Disability and Health (ICF) provides a framework to explore the concept in a biopsychosocial framework taking into account the interaction of a person with a health condition with their environment. The central purpose of this thesis was to explore disability within an under resourced context in order to provide data to service planners to improve the health and well-being of those affected. The exploration of disability involved a cross-sectional survey using instruments based on the ICF framework including the Washington Group Short Set of Questions on Disability, the WHODAS-2, the WHOQOL-BREF and the EQ-5D. The objectives were to establish the prevalence of disability and the description of the impairments, functional limitations and participation restrictions of those identified with disabilities. A total of 950 households were visited in Oudtshoorn (a semi-rural town) and Nyanga (a peri-urban area) and information was gathered on 7336 individuals with a mean age of 30.5 years. The majority of the participants were women. Both areas presented with estimates higher than those from the national census (5.0-6,7%). The urban area of Nyanga presented with a disability prevalence of 13.1% and the semi-rural area of Oudtshoorn with a prevalence of 6.8%. Overall the disability rate was 9.7%. The types of impairment and functional limitations were similar in the two areas, but more severe disability was reported in the semi-rural area, that also had significantly more elderly people. Non-communicable diseases were identified as the major cause of disability in both areas, followed by communicable diseases in Oudtshoorn and unintentional trauma in Nyanga. However, a person was twice as likely to be disabled due to non-communicable disease (Odds Ratio 2.2) when living in Oudtshoorn, and three times more likely to be disabled due to intentional trauma when living in Nyanga (Odds Ratio 0.3). Non-communicable diseases were responsible for the largest number of healthy life years lost. Those living in Nyanga had a higher burden of disability due to their lower quality of life scores as measured by the EQ-5D. Respondents in Nyanga consistently scored higher (worse) on all domains of the WHODAS-2 compared to respondents in Oudtshoorn. Living in Nyanga was associated with a 10% increase in domain scores. However, the pattern of scoring was similar and both areas reported worst functioning for the domains of Getting Around and Life Activities, which are associated with physical mobility. Respondents in Oudtshoorn reported better QoL and HRQoL than those in Nyanga. Functional level predicted the QoL scores, with Nyanga reporting worse functioning. Being employed and married was associated with a higher (better) EQ-5D VAS score, while mobility problems, pain or discomfort and anxiety or depression decreased the score. Transport was the most commonly identified barrier in both areas. Major barriers for those living in Oudtshoorn were Surroundings and Help in the home, whereas Help in the home and Prejudice and discrimination were viewed as the major barriers for respondents in Nyanga. The elderly were the ones most likely to not receive the rehabilitation services that they needed. The conclusions that can be drawn from this research are that context influences the experience of disability, and that disability prevalence alone is an insufficient basis for service planning. Those who experience greater deprivation also have a worse experience of disability. It is therefore essential for South Africa policy makers to view disability through a socio-political lens to ensure the equalisation of opportunities for people with disabilities. Improved quality of life for those living in under-resourced communities should be a priority. Service providers should have a broad range of skills to enable them to address not only the rehabilitation needs of people with disabilities, but also their social needs.

Includes bibliographical references. / This thesis investigated the effects on critically ill, mechanically ventilated paediatric patients of two related, frequently performed physiotherapy procedures: nonbronchoscopic bronchoalveolar lavage (NB-BAL) and endotracheal (ET) suctioning. General aims: To investigate un- or poody-documented complications of paediatric NBBAL and ET suctioning, and to test a method for each procedure of reducing the incidence and/or severity of these complications.

Background: Physiotherapists have been identified as key role players in health promotion (HP) as well as in the prevention and treatment of non-communicable diseases (NCDs). As key role players engaged in physical activity (PA) HP who society observe as role models, it is valuable to investigate the health behaviours of physiotherapists and physiotherapy students alike. There is a paucity of evidence on the levels of PA of physiotherapists and physiotherapy students, making it difficult to predict their effectiveness as role models and the effectiveness of HP strategies they use. Aim: The aim of this study is to describe the perceived benefits and barriers to exercise and their association with levels of PA in physiotherapy students (first to fourth year) attending university in the Western Cape province of South Africa. Method: This study follows a quantitative, cross-sectional, survey design. Two hundred and ninety-six participants (median age = 22 years) were recruited from the three universities in the Western Cape (University of Cape Town, University of the Western Cape and University of Stellenbosch). Participants completed a demographic questionnaire (DQ), 43 item exercise benefits and barriers scale (EBBS) and the international physical activity questionnaire short form (IPAQ-short). Results: Female students accounted for 83% of the sample. From the 296 participants, 58% lived off campus and 65% were involved in sporting activities six hours per week. The median score on the EBBS was 136 (54-167) for all years. Responses with the highest agreement for perceived benefits were: (1) Exercise increases my level of physical fitness, (2) Exercise improves functioning of my cardiovascular system and (3) exercise improves the way my body looks. Alternatively, the top three responses for perceived barriers were: (1) exercise tires me, (2) I am fatigued by exercise and (3) Exercise is hard work for me. The IPAQ scores revealed that the majority of students had low PA. Only 37.5% students engaged in high PA. Scores on the EBBS categorised by level of physical activity (low, moderate and high) showed that students with high PA had significantly higher scores on the EBBS than those with moderate and low PA. There were no significant differences between the EBBS scores of students with high, moderate and low PA leves in the first year of study [H (2, N=41) =3.01 p=0.22]. However, students in the second (n=61), third (n=111) and fourth years of study with high PA had significantly better scores on the EBBS than those with low and moderate PA. Conclusion: Undergraduate physiotherapy students in the WC across all three universities do not engage in adequate PA. In this group of students, benefits associated with high PA related to physical performance and barriers associated with low levels of PA related to physical exertion. Clinical relevance: Physiotherapists who do not practice what they preach are not effective role models and may not be effective in obtaining behavior change through PA HP. Global mortality by NCDs is on the rise and physiotherapists need to respond by taking a leadership role. Physiotherapy students should be encouraged to participate in PA as part of their undergraduate training. Methods of promoting PA in this population should take into account the perceived barriers and benefits of this group of students.

Background Physical fitness is an important indicator of health in childhood and a strong predictor of wellbeing in later life. Adiposity, decreased aerobic fitness and low levels of physical activity (PA) are associated with cardiovascular disease risk in children (Andersen et al., 2004). Estimates of PA and fitness in young children at the population level is still lacking for many African countries such as Ghana. One major challenge to conducting research on PA and fitness in children in these settings is the lack of accurate and reliable measures of these constructs. Several valid and reliable field-based physical fitness tests require test kits that are too expensive for people working in low socio-economic settings and lack norms for this population (Smits-Engelsman et al., 2020a). The Alpha-fit is one of the most used tests in Europe (Cvejić et al., 2013; Kolimechkov et al., 2019; Ruiz et al., 2011) and is considered to be a valid, reliable, feasible and safe health-related fitness test battery (Assessing Levels of Physical Activity (ALPHA), 2009; Ruiz et al., 2011). However, the Alpha-fit does not assess movement skills (e.g. throwing and catching, hopping, and balance), and has no norms for children living in Sub Saharan Africa. In recognition of these challenges, the Performance and Fitness (PERF-FIT) test battery was developed to provide a set of valid, reliable fieldbased motor performance items for the assessment of health-related physical fitness in low resource communities (Smits-Engelsman, 2018). However, the validity of the PERF-FIT has not been examined in children within the Ghanaian context. The aims of this study were to determine: (1) the physical fitness levels of school children aged 6 to 12 years in a lowincome urban area in Ghana using the Alpha-fit and (2) the construct validity of the PERFFIT among school children aged 6 to 12 years in a low-income urban area in Ghana by comparing it to tests of the Alpha-fit which measure similar constructs. Method A cross-sectional study design was adopted. Three public elementary schools located in La, a suburb of the La Dade Kotopon Municipal Assembly (LADMA), Accra, Ghana which met the criteria of low socio-economic status, (on the basis of geographical location) was selected using convenience sampling. The total number of pupils aged 6 - 12 years in all three schools were 376. A total of 186 children (aged 6-12 years) took part in the study. Children who provided assent after their parents had consented to participate in the study were recruited. Prior to testing, demographic information, health status and PA levels of children provided by a researcher-developed parent questionnaire were documented. Children with physical or medical conditions [determined with the Children‟s Physical Activity Readiness Questionnaire] that would have made testing unsafe were excluded. The two-item Physical Activity Questionnaire (Prochaska et al., 2001) was used to measure participants‟ PA and to identify those meeting the World Health Organization (WHO) PA guidelines for children. Testing was done on a playground/designated classroom at the participants‟ school during Physical Education periods. Weight and height measurements were taken using an electronic weighing scale and a wall-mounted tape measure. Cardiorespiratory endurance, power, agility and motor skills performance were assessed using the PERF-FIT and Alpha-fit assessment batteries. Only the 20-meter shuttle run test (20mSRT), long jump, Body Mass Index (BMI), waist circumference and 4 x10 meter shuttle run (4 x10m SR) which are items 1, 3, 4, 5 and 7 respectively of Alpha-fit were assessed. The tests were carried out in accordance with the instruction manuals. In describing the fitness levels of the participants, the Alpha-fit norms were used (Kolimechkov et al., 2019), and scores for each component were categorized into very poor (below 3rd percentile), below mean (between 3rd and 25th percentile), normal range (between 25th to 75th percentile), above mean (between 75th and 97th percentile) and very good (above 97th percentile). For the PERF-FIT, scores were collated as raw scores. Descriptive statistics: mean, median, standard deviation, percentages, and frequencies were used to summarize descriptive data. To compare differences of scores on the PERF-FIT and Alpha-fit between those who meet the WHO recommended guidelines and those who did not, the Mann-Whitney U test was used. To test the construct validity, hypotheses were formulated and tested for the expected magnitude of the relation between scores on the PERF-FIT and Alpha-fit intended to measure partly comparable constructs. The level of significance used was 0.05 Results One hundred and eighty-six children (96 boys and 90 girls) from low-income homes participated in this study. The median age of participants was 10.00 years. Only 12% of fathers and 3% of mothers had tertiary education whereas 40% of fathers were Artisans and self-employed, 68% of mothers were traders. Participants who were found to have met the WHO recommendation for moderate-tovigorous physical activity (MVPA) were 65.6% (n = 120),only 15.6% (n = 29) participate in sports outside school. Majority of them, 86.6% (n= 161) walk to school and 28.5% (n = 53) were underweight. Categorizing the Alpha-fit items scores, for item 1(20mSRT), 72 % (n = 134) were below the mean, 1.3 % (n = 3) performed very poorly, 25.3 % (n = 47) had normal scores, only 0.5 % (n=1) scored above mean. For item 3 (long jump), 2.2 % (n = 4) scored very good, 23.7% (n = 44) were above the mean, 59.7% (n = 111) scored normal, 12.9% (n = 24) scored below the mean and 1.1% (n = 2) performed very poorly. For item 7 (4 x10m SR), 22.6% (n = 42) scored very good, 33.2% (n = 50), scored above the mean, 30.1 % (n = 56) scored normal range, 11.8% (n = 22) scored below the mean while 2.7% (n = 5) scored very poor. No significant difference on the PERF-FIT ladder run, ladder step, side jump, long jump and overhand throw (p = 0.26,0.16,0.54,0.90 and 0.99 respectively) and Alpha-fit longjump,4 x 10m SR, and 20mSR level (p = 0.99,0.77 and 0.10 respectively) between the group that met the WHO recommendation for PA and those who did not were found. However, there was a statistically significant difference between the two groups on BMI (p =0.05, with those not meeting the recommendation having a higher BMI). Significant correlations were observed between the PERF-FIT (power and agility) items and the Alpha-fit 4 x10m SR and Long jump. The Alpha-fit 4 x10m SR showed low to moderate correlation with the PERF-FIT ladder run, ladder step and long jump (rs= 0.26, 0.20, and 0.35 respectively) while the Alpha- fit long jump showed moderate to high correlations with the PERF-FIT side jump, long jump and overhand throw (rs= 0.36, 0.84 and 0.62 respectively). Conclusion Participants in the study demonstrated good muscular strength and agility measured by the long jump and 4 x10m SR as a result of habitual PA. However, the low participation in organized sports observed by this study could have accounted for the poor aerobic fitness demonstrated by low scores on the 20mSRT. The majority of the children were physically active and met global (WHO) recommendations for PA as reported by their parents. They mostly employed active transportation to get in and out of school (walking). Nevertheless, the BMI of the group of children not meeting the WHO recommendations was significantly higher than the group that met the recommendations. Scores on both Alpha-fit and PERF-FIT were however not significantly different between the two groups. The relation between PERF-FIT and Alpha-fit was mainly (80%) of the hypothesised magnitude. Majority the PERF-FIT (power and agility) items showed moderate to high correlation with the Alpha-fit 4 x10m SR and long jump which shows that the PERF-FIT measures a partly comparable construct to these tests. Identifying deficits in physical fitness in children living in low socioeconomic settings is recommended as a critical step toward the development of a large-scale effective prevention and/or intervention for children with low levels of physical fitness. Participants understood all the PERF-FIT instructions and carried out tests accordingly, testing was time-efficient since participants could be tested in small groups. Assessors found it easy to score participants. Our study findings show that the PERF-FIT performed well in this setting and may be used to assess fitness levels of children in this context.

Background: South Africa has the highest prevalence rate (8.3%) of childhood obesity in sub-Saharan Africa. Obesity is a complex condition to control as it has environmental, as well as genetic factors that influence its prevalence. Childhood obesity, a rising problem worldwide and within South Africa, has been negatively linked with both physical fitness and physical activity. PA and school-based nutrition intervention programmes have been shown to have positive effects on diet and PA behaviours in children. However, there is minimal literature reporting on the effectiveness of school-based interventions in a South African setting. The aim of this study was to determine the prevalence of obesity, levels of physical fitness and physical activity in adolescents attending school in an urban setting. Together with the literature, an intervention programme was designed addressing key outcome measures identified from the sample. Methodology: The study was conducted in two phases. Ten schools were randomly selected from the school education district in KwaZulu-Natal Department of Education (KZNDoE) for phase one. A total of 400 eligible students were identified by the respective school teachers and parental consent was obtained, with assent obtained from the participants. A sample of 278 (girls n = 150, boys n = 128) participants with a mean age of 15yrs 1 month (CI 95% 12,1 – 17.1 and SD 1.14), was measured for height, weight, hip circumference, waist circumference and physical fitness (standing broad jump, sit-ups in 30s, hand grip strength, sit-and-reach and 20m shuttle run). BMI and WHR were calculated using anthropometric measurements. Physical activity (PA) was self-reported using the Physical Activity Questionnaire for Adolescents (PAQ-A) and the International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE) demographic form were completed by the participants' guardians/parents. Phase two involved the design of a PA intervention programme using evidence-based outcome measures from previous studies and problem areas identified in this study. Results: The mean BMI z-score was 0.30 (CI 95% 0.12 to 0.49) with boys displaying a mean of 0.19 (CI 95% -0.89 to 0.46) and girls 0.39 (CI 95% 0.15 to 0.65). There was a 16.2% prevalence of overweight (1 < BMI-z score ≤ 2), 13.3% prevalence of obesity (2 < BMI-z score ≤ 3) and 2.9% prevalence of morbid obesity (3 < BMI-z score ≤ 4). Mean WHR for the sample was 0.82 (CI 95% 0.81 – 0.83). Low levels of PA were reported by 45% (girls n = 83, boys n = 42) of the sample and 53.6% (girls n = 66, boys n = 83) reported moderate levels of PA. Girls in the sample had a higher mean sit-and-reach (t = 4.68, p < 0.05) and hand grip strength (t = 5.49, p < 0.05) than the normative values. Boys in the sample had higher mean (t = 13.2, p < 0.05) sit-and-reach than normative values. Multiple regression models were applied and BMI was found to be an inverse predictor for sit-ups ß = -0.07 (p < 0.05, CI 95% -1.0 to -0.03) and broad jump ß = -0.01 (p < 0.05, CI 95% -0.01 to -0.005). BMI was also a direct predictor for hand grip strength ß = -0.07 (p < 0.05, CI 95% 0.05 to 0.09). Discussion: The 16.2% prevalence of obesity was higher than the reported 8.5% in another South African study on adolescent scholars and the national reported level of 5.5%. The waist-hip ratio (WHR) was lower than the national reference level of 0.84 and higher compared to values obtained from adolescents in Europe (comparisons were made to age and gender specific normative values obtained in Europe, as there are no national reference physical fitness values for adolescents). Girls displayed lower values in three of the five physical tests (standing broad jump, sit-ups and 20m shuttle run), while boys displayed lower values for four (standing broad jump, hand grip strength, sit-ups and 20m shuttle run) of the five physical fitness tests. PA is indirectly correlated with BMI, and with BMI being a predictor for physical fitness, the promotion of PA amongst adolescents is imperative and urgent to curb the growing prevalence of obesity. Using evidence from previous research an intervention programme was designed to address the problems areas identified. These were high prevalence of obesity and overweight, low PA levels, poor cardiovascular function , low muscular power and poor muscular endurance across the sample. Boys also exhibited low levels of muscular strength than the normal population mean. PA programmes must be 60 minutes of duration implemented at least three times per week over 12 weeks minimum in order to have a reduction in BMI. Programmes should incorporate aerobic and anaerobic training as concurrent training has shown best results for decreasing BMI. The aerobic component of the programme must include a high intensity interval training (HIIT) as this has positive benefits on cardiovascular markers. The anaerobic component must include free weights as this showed better improvements than using resisted exercise devices in body composition, increases in upper and lower limb muscle strength in boys and improved lower limb muscle power especially in girls. Programmes should be implemented in schools with trained instructors and incorporate the family as a combination of these showed favourable compliance and overall better outcome measures. PA programmes that have a circuit formation and that incorporate a fun element has also shown to have better outcome measures and compliance. Conclusion: Physical activity levels are on the decline and obesity prevalence is on a rapid upward trajectory amongst South African adolescents. These trends could be ominous for the next workforce generation as these have been linked to numerous non-communicable diseases where the already constrained health system is placed under added pressure. PA programmes should be incorporated into schools' curricula as this may be a viable way to implement successful interventions to address obesity, physical fitness and physical activity.

Includes abstract. Includes bibliographical references.

Includes bibliographical references (p.181-200). / Background: Balance and postural control are an integral part of gross motor function in activities of daily living. Studies have shown that children with hemiplegic cerebral palsy have poor directional specificity as well as problems with the temporal and spatial modulation of appropriate muscle action in response to balance perturbations. Children with hemiplegia have also been shown to develop direction-specific postural control at a slower pace than typically developing children. Apart from their postural muscle coordination problems, these children have difficulties with sensory integration which contributes to increased reaction time. Research on balance training in children with cerebral palsy has demonstrated that improved balance translates into more effective gross motor function. It appears that postural control mechanisms are still modifiable for children with cerebral palsy even in elementary to middle school ages. Physiotherapy treatment for children with cerebral palsy should therefore involve balance training as a focus of intervention. Literature on balance control and virtual reality rehabilitation justifies investigating the use of a commercially ~vailable gaming system, such as the Nintendo Wii Fit, as a rehabilitation tool to improve balance control and therefore gross motor function in children with cerebral palsy. Objective: To determine the effect of an intervention with the Nintendo Wii Fit on the balance control and gross motor function of children with spastic hemiplegic cerebral palsy.