Objective: To examine sleep and wake disorders following traumatic brain injury (TBI) and their impact on recovery of cognition, communication and mood. Research Design: This three-manuscript thesis comprises an introduction to sleep in the context of human function and development. It is followed by a systematic review of the literature pertaining to sleep and wake disorders following TBI, and then explores the relationship between sleep and arousal disturbance and functional recovery of cognitive-communication through a single case study, pre–post intervention. Finally, a larger study longitudinally explores the impact of treatment to optimize sleep and wakefulness on recovery of cognition, communication and mood through objective and subjective measures, pre-post intervention. The thesis concludes with a chapter that addresses the implications of findings for rehabilitation from the perspective of the International Classification of Functioning, Disability and Health (ICF), and a presentation of future research directions for the field Methods: The first manuscript involved a systematic review and rating of the quality of evidence. The second manuscript involved the evaluation of sleep and wakefulness by objective measures, and longitudinally by self-report through the Daily Cognitive-Communication and Sleep Profile (DCCASP, © Wiseman-Hakes 2008, see Appendix S). Cognitive-communication abilities were also measured by the DCCASP. The third manuscript utilized a single case series and cohort design to evaluate sleep and wakefulness, and to examine cognition, communication and mood at baseline and following optimization of sleep and wakefulness. Results: For Manuscript One, 43 articles were reviewed for levels and quality of evidence across 5 domains: epidemiology, pathophysiology, neuropsychological implications, intervention and paediatrics. In Manuscript Two, we showed that there was a statistically and functionally significant relationship between perceived quality of sleep and language processing, attention and memory, seen across the phases of the intervention. In Manuscript Three, we showed that there were statistically and functionally significant improvements across several domains of cognition, communication and mood in response to treatment. Conclusions: Sleep and wake disorders after TBI are pervasive, and can negatively impact rehabilitation and recovery. There is a need for systematic evaluation and intervention for these disorders in all persons with TBI.

This thesis investigates head injuries (HIs) occurring in sports and recreation (SR) that presented to emergency departments (EDs) in Edmonton, Alberta, from April 1st, 1997 to March 31st, 2008. There were 4,950 SR HIs identified over the 11 years (5.13 HIs per 1,000 ED injury visits). Individuals less than 18 years old were 3.4 times more likely, and males were 1.25 times more likely to present with a head injury from SR (p < 0.0001). A history of one or two previous SR HIs increased the odds of subsequent HI by 2.62 and 5.94 times, respectively, while children aged 7 13 and 14 17 were more than four times more likely to sustain a HIs from SR (p < 0.001). The effects of multiple HIs occurring in children participating in SR activities needs to be addressed, due to the chronic neurobehavioral effects of HIs. / Epidemiology

Traumatic Brain Injury (TBI) mainly affects young persons in traffic accidents and the elderly in fall accidents. Improvements in the clinical management have significantly improved the outcome following TBI but survivors still suffer from depression, memory problems, personality changes, epilepsy and fatigue. The initial injury starts a series of events that give rise to a secondary injury process and despite several clinical trials there is no drug available for clinical use that targets secondary brain injury mechanisms. Some recovery of function is seen during the first months following injury but is usually limited and there are no drugs that stimulate the recovery of lost function. Some of the recovery is attributed to plasticity, the brains ability to adapt to new circumstances, and enhancing plasticity via increased axonal growth has the potential to partly restore lost function. In this thesis mice were subjected to the controlled cortical impact model of TBI and functional outcome was evaluated using Morris water maze, the cylinder test and the rotarod. Brain tissue loss was measured in all Papers but the additional histological analyses differ among the Papers. Attempts to increase axonal growth were made by interfering with Nogo receptor function in Paper I and by conditional knockout of ephA4 in Paper II. Contrary to the hypothesis cognition was impaired in Paper I but otherwise no effects of treatment were detected in Paper I and II. Much is still unknown about plasticity and despite the discouraging results of Papers I and II this treatment approach is still worth further exploration. It is firmly established that TBI results in an inflammatory response and some aspects of it may damage brain tissue. In Papers III and IV the inflammatory response was attenuated using an IL-1β directed antibody which resulted in reduced tissue loss and edema while improving cognitive function. The results from Papers III and IV are encouraging and the possibility to find a treatment based on IL-1β inhibition appears promising.

Psychosocial factors represent complex and enduring challenges for people with a traumatic brain injury (TBI), their families and health and rehabilitation systems, particularly in relation to social isolation, change/loss of role and identity issues. Traditional rehabilitation approaches to psychosocial issues target cognitive rehabilitation and psychological adjustment of the individual with a TBI, which reflect the medical model of health. However, the International Classification of Functioning, Disability and Health (ICF) ushers a new era in TBI research and practice, which advances the conceptualisation of disability, to include the impact of environmental factors. The ICF identifies the Internet as one such environmental factor, which has the potential to facilitate or impede social participation of people with a disability. Empirical studies suggest the potential of the Internet to empower people with disabilities, via opportunities for social connection; social support; social role participation and identity experimentation. However, research has focused on people with sensory, physical, and mental health issues, with little known of the potential of the Internet for people with a TBI. This area of deficit warrants investigation, in view of the potential of the Internet to address many dimensions of psychosocial concern following TBI. This study explores the role of the Internet in the lives of people with a TBI, with a focus on the potential of the Internet to influence psychosocial recovery. Psychosocial recovery is conceptualised as the reconstruction of a positive identity, as found in subjective accounts of people with a TBI (Lewington, 1996), and strongly supported in the mental health consumer literature. Using a social constructionist approach, this study explores how people with a TBI use the Internet and how they make meaning of their Internet experience. This approach acknowledges the socially situated nature of Internet use and validates the subjective accounts of people with a TBI, whose perspectives are underrepresented in the rehabilitation literature. This study also trials the method of email-facilitated qualitative interviewing, to address face-to-face interviewing barriers, relating to cognitive-linguistic impairments (Lloyd, Gatherer, & Kalsy, 2006; Paterson & Scott-Findley, 2002). Thus, the Internet is the focus of the enquiry and the mediator of the method. The findings highlight the positive potential of the Internet to facilitate social participation for people with a TBI. Participants reported that features such as asynchronicity, reduced cues and anonymity made the Internet an accessible and usable technology, for social connection; social support; social role participation and identity reconstruction. A major theme emerging from the data was control of self, which this study conceptually linked to identity reconstruction and psychosocial recovery. Findings allayed concerns regarding the negative potential of the Internet to increase social isolation of vulnerable people, as participants regarded face-to-face relationships as more fulfilling than online friendships, supporting classic communication theories (Rice, 1987; Rice & Love, 1987), which propose that the Internet is a less personal medium than face-to-face communication, due to the feature of reduced cues. The findings of the method indicated that email facilitated qualitative interviewing addressed face-to-face interviewing barriers related to cognitive linguistic impairments; mobility factors; chronic health issues and environmental stimuli. Most participants indicated that asynchronicity, reduced cues, and anonymity facilitated control of communication, cognition and identity, thus enabling interview participation. The method had also advantages for the researcher, including time for reflection and the ability to yield richer data than in face-to-face contexts. However, the method was resource intensive, requiring information technology proficiency, familiarly with the impact of cognitive-linguistic impairments in online contexts and counselling experience. In addition, ethical guidelines required the involvement of a support person for the emotional protection of participants. This study makes two contributions to knowledge. One contribution relates to the potential of the Internet as an ICF environmental factor to address long-term psychosocial concerns, in addition to positively influencing psychosocial recovery from TBI, as reported by participants. This study contributes to a new era of research, which considers the impact of environmental factors on the experience of TBI, as framed by the ICF. The second contribution relates to the method of email facilitated qualitative interviewing, which advances knowledge of interviewing barriers for people with a TBI and addresses calls for innovative methods with this population. The findings of the method bring into question long held assumptions about the capacity of people with a TBI to participate in research and have implications for research design in qualitative and quantitative methodologies.

Mild Traumatic Brain injury (mTBI) is one of the most common forms of acquired neurological damage. However, the term 'mild' TBI is misleading because the physical, cognitive and emotional impairments that can follow from mTBI can be significant. In order to provide objective, prognostic measures for diagnosing the severity of mTBI and identifying individuals who may be at risk for poor outcomes a battery of neuropsychological measures for detecting cognitive impairment was evaluated. The Rapid Screen of Concussion (RSC) is a collection of tests assessing verbal recall, orientation, processing speed and speed of language comprehension. Previous studies have demonstrated that the RSC has acceptable reliability, validity and sensitivity to cognitive impairment that arises during the acute stages of injury. However, no studies have investigated the predictive validity of this instrument. Moreover it is unclear what additional patient or post injury variables could assist in identifying those individuals who may be at risk of poor neuropsychological outcomes following mTBI. These were among the main issues that were addressed across the five empirical studies in this thesis. A pragmatic, prospective, longitudinal and cross-sectional study of the sequelae of mTBI in patients presenting to the Department of Emergency Medicine of the Royal Brisbane and Women's hospital was the basis of this project. The first empirical chapter (chapter 2), examined the psychometric properties of two measures of verbal learning and memory and investigated their potential for discriminating between mTBI and orthopaedic controls. The performance of 93 patients with mTBI and 68 participants with orthopaedic injuries was analysed to identify the number of individuals who performed at ceiling on the Hopkins Verbal Learning test (HVLT-R) versus a 5-word test of immediate and delayed recall. While both of these verbal recall measures were effective in separating the mTBI and orthopaedic groups, overall, the HVLT-R was shown to be a more suitable measure for screening for deficits in verbal learning and memory after mTBI. Given the superiority of the HVLT-R as a measure of verbal learning and memory, chapter 3 aimed to examine whether inclusion of this test could improve the sensitivity of the RSC in mTBI compared to orthopaedic and uninjured control samples. Results were generally within the direction predicted. Significant differences were found between groups on the majority of cognitive indices assessed. Both the orthopaedic and mTBI group performed more poorly than the uninjured group on all measures except the Hopkins delayed recognition. Additional performance decrements shown by the mTBI group compared to the orthopaedic group illustrate that factors beyond the general effects of trauma influence performance and may be related to cognitive impairment specific to sustaining mTBI. Overall it was concluded that the revised RSC is a sensitive instrument deserving investigation in assessing the more long term cognitive effects following mTBI. Chapter 4 applied this sensitive battery for investigation of group and individual recovery of neuropsychological test performance and post-concussive symptom reporting up to 3-months after mTBI. A sample of 30 mTBI participants and 30 uninjured controls were serially assessed on cognitive measures and symptom report scales immediately after injury and after 1-week, 1-month and 3-months. Symptom reporting on the Rivermead post-concussive inventory separated the mTBI and control groups after 1-week but diagnostic accuracy was no greater than chance at 1 and 3-months. In contrast the mTBI group performed more poorly than controls on measures on neuropsychological measures acutely, at 1-week and 1-month, with group differences still evident after 3-months. Nonethless, a trend of progressive recovery over time was seen in the mTBI group. In chapter 5, criteria utilising the concepts of reliable and statistically significant change were applied to the data. Overall, 73% of mTBI patients were impaired on one or more tests acutely. Significant recovery was demonstrated by 20% of mTBI participants by 3-months; however recovery remained incomplete for half of the mTBI participants by 3-months. These results highlighted the importance of an individual approach to the assessment of mTBI and support the notion that a proportion of mTBI cases may have protracted difficulties. Chapter 6 extended these findings by showing that the RSC has prognostic ability. It was found that acute neuropsychological performance on the RSC was a significant predictor of performance on an extended battery at 3-months. The final chapter provides a general discussion and synthesis of the findings. In summary, the present dissertation demonstrated that inclusion of a sensitive measure of verbal recall led to improved diagnostic validity of the RSC. Neuropsychological measures rather than symptom reporting were sensitive in detecting cognitive impairment at 3-months. Analysis of individuals showed that up to 50% of the group had failed to – demonstrate reliable recovery – that is, make improvements over and above practice effects after 3-months. Finally, acute neuropsychological performance was predictive of long term performance. Overall, the present thesis has identified a short battery of tests that is suitable for assessment of mTBI within 24 hours and may assist in identifying individuals at risk of poor cognitive outcomes after mTBI.

Mild Traumatic Brain injury (mTBI) is one of the most common forms of acquired neurological damage. However, the term 'mild' TBI is misleading because the physical, cognitive and emotional impairments that can follow from mTBI can be significant. In order to provide objective, prognostic measures for diagnosing the severity of mTBI and identifying individuals who may be at risk for poor outcomes a battery of neuropsychological measures for detecting cognitive impairment was evaluated. The Rapid Screen of Concussion (RSC) is a collection of tests assessing verbal recall, orientation, processing speed and speed of language comprehension. Previous studies have demonstrated that the RSC has acceptable reliability, validity and sensitivity to cognitive impairment that arises during the acute stages of injury. However, no studies have investigated the predictive validity of this instrument. Moreover it is unclear what additional patient or post injury variables could assist in identifying those individuals who may be at risk of poor neuropsychological outcomes following mTBI. These were among the main issues that were addressed across the five empirical studies in this thesis. A pragmatic, prospective, longitudinal and cross-sectional study of the sequelae of mTBI in patients presenting to the Department of Emergency Medicine of the Royal Brisbane and Women's hospital was the basis of this project. The first empirical chapter (chapter 2), examined the psychometric properties of two measures of verbal learning and memory and investigated their potential for discriminating between mTBI and orthopaedic controls. The performance of 93 patients with mTBI and 68 participants with orthopaedic injuries was analysed to identify the number of individuals who performed at ceiling on the Hopkins Verbal Learning test (HVLT-R) versus a 5-word test of immediate and delayed recall. While both of these verbal recall measures were effective in separating the mTBI and orthopaedic groups, overall, the HVLT-R was shown to be a more suitable measure for screening for deficits in verbal learning and memory after mTBI. Given the superiority of the HVLT-R as a measure of verbal learning and memory, chapter 3 aimed to examine whether inclusion of this test could improve the sensitivity of the RSC in mTBI compared to orthopaedic and uninjured control samples. Results were generally within the direction predicted. Significant differences were found between groups on the majority of cognitive indices assessed. Both the orthopaedic and mTBI group performed more poorly than the uninjured group on all measures except the Hopkins delayed recognition. Additional performance decrements shown by the mTBI group compared to the orthopaedic group illustrate that factors beyond the general effects of trauma influence performance and may be related to cognitive impairment specific to sustaining mTBI. Overall it was concluded that the revised RSC is a sensitive instrument deserving investigation in assessing the more long term cognitive effects following mTBI. Chapter 4 applied this sensitive battery for investigation of group and individual recovery of neuropsychological test performance and post-concussive symptom reporting up to 3-months after mTBI. A sample of 30 mTBI participants and 30 uninjured controls were serially assessed on cognitive measures and symptom report scales immediately after injury and after 1-week, 1-month and 3-months. Symptom reporting on the Rivermead post-concussive inventory separated the mTBI and control groups after 1-week but diagnostic accuracy was no greater than chance at 1 and 3-months. In contrast the mTBI group performed more poorly than controls on measures on neuropsychological measures acutely, at 1-week and 1-month, with group differences still evident after 3-months. Nonethless, a trend of progressive recovery over time was seen in the mTBI group. In chapter 5, criteria utilising the concepts of reliable and statistically significant change were applied to the data. Overall, 73% of mTBI patients were impaired on one or more tests acutely. Significant recovery was demonstrated by 20% of mTBI participants by 3-months; however recovery remained incomplete for half of the mTBI participants by 3-months. These results highlighted the importance of an individual approach to the assessment of mTBI and support the notion that a proportion of mTBI cases may have protracted difficulties. Chapter 6 extended these findings by showing that the RSC has prognostic ability. It was found that acute neuropsychological performance on the RSC was a significant predictor of performance on an extended battery at 3-months. The final chapter provides a general discussion and synthesis of the findings. In summary, the present dissertation demonstrated that inclusion of a sensitive measure of verbal recall led to improved diagnostic validity of the RSC. Neuropsychological measures rather than symptom reporting were sensitive in detecting cognitive impairment at 3-months. Analysis of individuals showed that up to 50% of the group had failed to – demonstrate reliable recovery – that is, make improvements over and above practice effects after 3-months. Finally, acute neuropsychological performance was predictive of long term performance. Overall, the present thesis has identified a short battery of tests that is suitable for assessment of mTBI within 24 hours and may assist in identifying individuals at risk of poor cognitive outcomes after mTBI.

Mild Traumatic Brain injury (mTBI) is one of the most common forms of acquired neurological damage. However, the term 'mild' TBI is misleading because the physical, cognitive and emotional impairments that can follow from mTBI can be significant. In order to provide objective, prognostic measures for diagnosing the severity of mTBI and identifying individuals who may be at risk for poor outcomes a battery of neuropsychological measures for detecting cognitive impairment was evaluated. The Rapid Screen of Concussion (RSC) is a collection of tests assessing verbal recall, orientation, processing speed and speed of language comprehension. Previous studies have demonstrated that the RSC has acceptable reliability, validity and sensitivity to cognitive impairment that arises during the acute stages of injury. However, no studies have investigated the predictive validity of this instrument. Moreover it is unclear what additional patient or post injury variables could assist in identifying those individuals who may be at risk of poor neuropsychological outcomes following mTBI. These were among the main issues that were addressed across the five empirical studies in this thesis. A pragmatic, prospective, longitudinal and cross-sectional study of the sequelae of mTBI in patients presenting to the Department of Emergency Medicine of the Royal Brisbane and Women's hospital was the basis of this project. The first empirical chapter (chapter 2), examined the psychometric properties of two measures of verbal learning and memory and investigated their potential for discriminating between mTBI and orthopaedic controls. The performance of 93 patients with mTBI and 68 participants with orthopaedic injuries was analysed to identify the number of individuals who performed at ceiling on the Hopkins Verbal Learning test (HVLT-R) versus a 5-word test of immediate and delayed recall. While both of these verbal recall measures were effective in separating the mTBI and orthopaedic groups, overall, the HVLT-R was shown to be a more suitable measure for screening for deficits in verbal learning and memory after mTBI. Given the superiority of the HVLT-R as a measure of verbal learning and memory, chapter 3 aimed to examine whether inclusion of this test could improve the sensitivity of the RSC in mTBI compared to orthopaedic and uninjured control samples. Results were generally within the direction predicted. Significant differences were found between groups on the majority of cognitive indices assessed. Both the orthopaedic and mTBI group performed more poorly than the uninjured group on all measures except the Hopkins delayed recognition. Additional performance decrements shown by the mTBI group compared to the orthopaedic group illustrate that factors beyond the general effects of trauma influence performance and may be related to cognitive impairment specific to sustaining mTBI. Overall it was concluded that the revised RSC is a sensitive instrument deserving investigation in assessing the more long term cognitive effects following mTBI. Chapter 4 applied this sensitive battery for investigation of group and individual recovery of neuropsychological test performance and post-concussive symptom reporting up to 3-months after mTBI. A sample of 30 mTBI participants and 30 uninjured controls were serially assessed on cognitive measures and symptom report scales immediately after injury and after 1-week, 1-month and 3-months. Symptom reporting on the Rivermead post-concussive inventory separated the mTBI and control groups after 1-week but diagnostic accuracy was no greater than chance at 1 and 3-months. In contrast the mTBI group performed more poorly than controls on measures on neuropsychological measures acutely, at 1-week and 1-month, with group differences still evident after 3-months. Nonethless, a trend of progressive recovery over time was seen in the mTBI group. In chapter 5, criteria utilising the concepts of reliable and statistically significant change were applied to the data. Overall, 73% of mTBI patients were impaired on one or more tests acutely. Significant recovery was demonstrated by 20% of mTBI participants by 3-months; however recovery remained incomplete for half of the mTBI participants by 3-months. These results highlighted the importance of an individual approach to the assessment of mTBI and support the notion that a proportion of mTBI cases may have protracted difficulties. Chapter 6 extended these findings by showing that the RSC has prognostic ability. It was found that acute neuropsychological performance on the RSC was a significant predictor of performance on an extended battery at 3-months. The final chapter provides a general discussion and synthesis of the findings. In summary, the present dissertation demonstrated that inclusion of a sensitive measure of verbal recall led to improved diagnostic validity of the RSC. Neuropsychological measures rather than symptom reporting were sensitive in detecting cognitive impairment at 3-months. Analysis of individuals showed that up to 50% of the group had failed to – demonstrate reliable recovery – that is, make improvements over and above practice effects after 3-months. Finally, acute neuropsychological performance was predictive of long term performance. Overall, the present thesis has identified a short battery of tests that is suitable for assessment of mTBI within 24 hours and may assist in identifying individuals at risk of poor cognitive outcomes after mTBI.

Mild Traumatic Brain injury (mTBI) is one of the most common forms of acquired neurological damage. However, the term 'mild' TBI is misleading because the physical, cognitive and emotional impairments that can follow from mTBI can be significant. In order to provide objective, prognostic measures for diagnosing the severity of mTBI and identifying individuals who may be at risk for poor outcomes a battery of neuropsychological measures for detecting cognitive impairment was evaluated. The Rapid Screen of Concussion (RSC) is a collection of tests assessing verbal recall, orientation, processing speed and speed of language comprehension. Previous studies have demonstrated that the RSC has acceptable reliability, validity and sensitivity to cognitive impairment that arises during the acute stages of injury. However, no studies have investigated the predictive validity of this instrument. Moreover it is unclear what additional patient or post injury variables could assist in identifying those individuals who may be at risk of poor neuropsychological outcomes following mTBI. These were among the main issues that were addressed across the five empirical studies in this thesis. A pragmatic, prospective, longitudinal and cross-sectional study of the sequelae of mTBI in patients presenting to the Department of Emergency Medicine of the Royal Brisbane and Women's hospital was the basis of this project. The first empirical chapter (chapter 2), examined the psychometric properties of two measures of verbal learning and memory and investigated their potential for discriminating between mTBI and orthopaedic controls. The performance of 93 patients with mTBI and 68 participants with orthopaedic injuries was analysed to identify the number of individuals who performed at ceiling on the Hopkins Verbal Learning test (HVLT-R) versus a 5-word test of immediate and delayed recall. While both of these verbal recall measures were effective in separating the mTBI and orthopaedic groups, overall, the HVLT-R was shown to be a more suitable measure for screening for deficits in verbal learning and memory after mTBI. Given the superiority of the HVLT-R as a measure of verbal learning and memory, chapter 3 aimed to examine whether inclusion of this test could improve the sensitivity of the RSC in mTBI compared to orthopaedic and uninjured control samples. Results were generally within the direction predicted. Significant differences were found between groups on the majority of cognitive indices assessed. Both the orthopaedic and mTBI group performed more poorly than the uninjured group on all measures except the Hopkins delayed recognition. Additional performance decrements shown by the mTBI group compared to the orthopaedic group illustrate that factors beyond the general effects of trauma influence performance and may be related to cognitive impairment specific to sustaining mTBI. Overall it was concluded that the revised RSC is a sensitive instrument deserving investigation in assessing the more long term cognitive effects following mTBI. Chapter 4 applied this sensitive battery for investigation of group and individual recovery of neuropsychological test performance and post-concussive symptom reporting up to 3-months after mTBI. A sample of 30 mTBI participants and 30 uninjured controls were serially assessed on cognitive measures and symptom report scales immediately after injury and after 1-week, 1-month and 3-months. Symptom reporting on the Rivermead post-concussive inventory separated the mTBI and control groups after 1-week but diagnostic accuracy was no greater than chance at 1 and 3-months. In contrast the mTBI group performed more poorly than controls on measures on neuropsychological measures acutely, at 1-week and 1-month, with group differences still evident after 3-months. Nonethless, a trend of progressive recovery over time was seen in the mTBI group. In chapter 5, criteria utilising the concepts of reliable and statistically significant change were applied to the data. Overall, 73% of mTBI patients were impaired on one or more tests acutely. Significant recovery was demonstrated by 20% of mTBI participants by 3-months; however recovery remained incomplete for half of the mTBI participants by 3-months. These results highlighted the importance of an individual approach to the assessment of mTBI and support the notion that a proportion of mTBI cases may have protracted difficulties. Chapter 6 extended these findings by showing that the RSC has prognostic ability. It was found that acute neuropsychological performance on the RSC was a significant predictor of performance on an extended battery at 3-months. The final chapter provides a general discussion and synthesis of the findings. In summary, the present dissertation demonstrated that inclusion of a sensitive measure of verbal recall led to improved diagnostic validity of the RSC. Neuropsychological measures rather than symptom reporting were sensitive in detecting cognitive impairment at 3-months. Analysis of individuals showed that up to 50% of the group had failed to – demonstrate reliable recovery – that is, make improvements over and above practice effects after 3-months. Finally, acute neuropsychological performance was predictive of long term performance. Overall, the present thesis has identified a short battery of tests that is suitable for assessment of mTBI within 24 hours and may assist in identifying individuals at risk of poor cognitive outcomes after mTBI.

The overall aim of this thesis is to contribute to the development and implementation of evidence-based physical activity promotion strategies for people with acquired brain injury (ABI). A randomized controlled trial (RCT) will furnish the highest level of evidence regarding the efficacy of a given intervention, but the immaturity of this area of research means that, currently, an RCT is not an appropriate research design. Therefore the purpose of the program of research presented in this thesis was to strategically contribute to the evidence base required to justify the conduct of a well designed RCT of an intervention promoting physical activity for people with ABI. By increasing participation in free-living, moderate intensity walking, people with ABI may reduce the oxygen cost of walking, thereby improving everyday functioning. The first study (presented in Chapter 3) evaluated a novel over-ground walking protocol in which a third party assists participants with ABI to reproduce previously self-selected comfortable and brisk walking speeds. Such a protocol could be used in conjunction with a portable indirect calorimetry to evaluate changes in oxygen cost of self-selected walking speeds over time. Thirteen people with ABI and related gait pattern impairment (age 31 + 8 yrs) completed two familiarization and two testing sessions. The first testing session used a self-paced (SP) protocol in which participants walked for 6 minutes at two self-selected speeds – comfortable and brisk paces. The second, conducted one week later, used an externally-paced (EP) protocol in which participants were instructed to walk at the pace indicated by the test administrator, who verbally and visually guided the participant to walk at their previous self-selected comfortable and brisk paces. In each testing session participants wore a portable indirect calorimeter (Cosmed K4b2). Measures obtained were oxygen cost of walking (l.min-1) and distance walked (m). Analysis demonstrated that the EP protocol reproduced distances walked at comfortable and brisk speeds with a high degree of accuracy and that agreement between measures of oxygen cost obtained during the SP and EP protocols were acceptable. Steady-state oxygen uptake is characteristic of a constant workload and was elicited during both EP walking trials, indicating that walking speeds were constant throughout the walk trials. It was concluded that the EP protocol described is a valid means of assisting people with ABI to reproduce overground walking speeds and that the protocol would be useful for evaluating changes in the oxygen cost of those walking speeds that may occur over the course of an intervention. The purpose of the second study (presented in Chapter 4) was to evaluate the validity of a uniaxial accelerometer (MTI Actigraph) for measuring physical activity in people with acquired brain injury (ABI) using portable indirect calorimetry (Cosmed K4b2) as a criterion measure. Fourteen people with ABI and related gait pattern impairment (age 32 + 8 years), wore an MTI Actigraph that measured activity (counts.min-1) and a Cosmed K4b2 that measured oxygen consumption (ml.kg-1.min-1) during four activities – quiet sitting (QS), comfortable paced (CP), brisk paced (BP) and fast paced (FP) walking. MET levels were predicted from Actigraph counts using a published equation and compared with Cosmed measures. Predicted METs for each of the 56 activity bouts (14 participants x four bouts) were classified (light, moderate, vigorous or very vigorous intensity) and compared with Cosmed-based classifications. Results indicated that Actigraph counts provide a valid index of activity across the intensities investigated in this study. For light to moderate activity, Actigraph-based estimates of METs are acceptable for group-level analysis and are a valid means of classifying activity intensity. The Actigraph significantly underestimated higher intensity activity although, in practice, this limitation will have minimal impact on activity measurement of most community-dwelling people with ABI as higher intensity activity is likely to be rare in this group. In the third study (presented in Chapter 5), 18 community-dwelling adults with acquired brain injury and a related gait pattern impairment (32.3 + 7.5 yrs) participated in an 8-week intervention promoting lifestyle physical activity. The aims were threefold: to evaluate the physical and psychosocial effects of the intervention; to assess whether the intervention increased the physical activity of participants; and to qualitatively evaluate the perceived effectiveness and acceptability of the intervention. Data were collected at six time points over 28 weeks: three pre-intervention, one each at 12 weeks prior (T1), 11 weeks prior (T2) and immediately pre-intervention (T3); and three post-intervention at immediately after (T4), four weeks after (T5) and eight weeks after (T6) the intervention. Physical outcomes measured were oxygen cost of comfortable and brisk-paced walking and maximum distance walked in three minutes. Psychosocial outcomes measured were SF-36, Depression Anxiety and Stress Scales (DASS), Rosenberg Self-Esteem Scale (RSES), Satisfaction with Life Scale (SWLS) and the Barriers to Health Activities for Disabled Persons (BHADP). Physical activity was sampled 6 days at T3, T4 and T6 and two measures were obtained – Actigraph counts per day and total minutes of activity greater than or equal to moderate intensity. Semi-structured interviews were used to evaluate the perceived effectiveness and acceptability of the intervention. The intervention comprised weekly, home-based, interactive problem-solving sessions designed to identify and overcome barriers to activity and to promote walking, together with facilitation of a community based leisure activity of the participant’s choice. Results indicated that the intervention improved important aspects of physical and psychosocial health for community-dwelling people with ABI. Compared with mean baseline measures, improvements in oxygen cost of brisk walking and self-esteem occurred that were both clinically and statistically significant at T6 (p < 0.01). Significant changes in two subscales of the SF-36 and the SWLS also occurred, although they were not sustained at T6. Measures of physical activity increased but not to an extent that was statistically significant. Qualitative data were principally positive, with 100% of participants and their significant others indicating they would recommend the program to another person with ABI. Given the particularly low levels of physical activity in the ABI population, and the correspondingly large individual and community benefits of that will be accrued if their physically active behavior can be increased, the promising results from this program of research indicate that there is a strong justification for allocating the resources necessary to conduct a sufficiently powered, randomized controlled trial of a lifestyle physical activity intervention for people with ABI.

The overall aim of this thesis is to contribute to the development and implementation of evidence-based physical activity promotion strategies for people with acquired brain injury (ABI). A randomized controlled trial (RCT) will furnish the highest level of evidence regarding the efficacy of a given intervention, but the immaturity of this area of research means that, currently, an RCT is not an appropriate research design. Therefore the purpose of the program of research presented in this thesis was to strategically contribute to the evidence base required to justify the conduct of a well designed RCT of an intervention promoting physical activity for people with ABI. By increasing participation in free-living, moderate intensity walking, people with ABI may reduce the oxygen cost of walking, thereby improving everyday functioning. The first study (presented in Chapter 3) evaluated a novel over-ground walking protocol in which a third party assists participants with ABI to reproduce previously self-selected comfortable and brisk walking speeds. Such a protocol could be used in conjunction with a portable indirect calorimetry to evaluate changes in oxygen cost of self-selected walking speeds over time. Thirteen people with ABI and related gait pattern impairment (age 31 + 8 yrs) completed two familiarization and two testing sessions. The first testing session used a self-paced (SP) protocol in which participants walked for 6 minutes at two self-selected speeds – comfortable and brisk paces. The second, conducted one week later, used an externally-paced (EP) protocol in which participants were instructed to walk at the pace indicated by the test administrator, who verbally and visually guided the participant to walk at their previous self-selected comfortable and brisk paces. In each testing session participants wore a portable indirect calorimeter (Cosmed K4b2). Measures obtained were oxygen cost of walking (l.min-1) and distance walked (m). Analysis demonstrated that the EP protocol reproduced distances walked at comfortable and brisk speeds with a high degree of accuracy and that agreement between measures of oxygen cost obtained during the SP and EP protocols were acceptable. Steady-state oxygen uptake is characteristic of a constant workload and was elicited during both EP walking trials, indicating that walking speeds were constant throughout the walk trials. It was concluded that the EP protocol described is a valid means of assisting people with ABI to reproduce overground walking speeds and that the protocol would be useful for evaluating changes in the oxygen cost of those walking speeds that may occur over the course of an intervention. The purpose of the second study (presented in Chapter 4) was to evaluate the validity of a uniaxial accelerometer (MTI Actigraph) for measuring physical activity in people with acquired brain injury (ABI) using portable indirect calorimetry (Cosmed K4b2) as a criterion measure. Fourteen people with ABI and related gait pattern impairment (age 32 + 8 years), wore an MTI Actigraph that measured activity (counts.min-1) and a Cosmed K4b2 that measured oxygen consumption (ml.kg-1.min-1) during four activities – quiet sitting (QS), comfortable paced (CP), brisk paced (BP) and fast paced (FP) walking. MET levels were predicted from Actigraph counts using a published equation and compared with Cosmed measures. Predicted METs for each of the 56 activity bouts (14 participants x four bouts) were classified (light, moderate, vigorous or very vigorous intensity) and compared with Cosmed-based classifications. Results indicated that Actigraph counts provide a valid index of activity across the intensities investigated in this study. For light to moderate activity, Actigraph-based estimates of METs are acceptable for group-level analysis and are a valid means of classifying activity intensity. The Actigraph significantly underestimated higher intensity activity although, in practice, this limitation will have minimal impact on activity measurement of most community-dwelling people with ABI as higher intensity activity is likely to be rare in this group. In the third study (presented in Chapter 5), 18 community-dwelling adults with acquired brain injury and a related gait pattern impairment (32.3 + 7.5 yrs) participated in an 8-week intervention promoting lifestyle physical activity. The aims were threefold: to evaluate the physical and psychosocial effects of the intervention; to assess whether the intervention increased the physical activity of participants; and to qualitatively evaluate the perceived effectiveness and acceptability of the intervention. Data were collected at six time points over 28 weeks: three pre-intervention, one each at 12 weeks prior (T1), 11 weeks prior (T2) and immediately pre-intervention (T3); and three post-intervention at immediately after (T4), four weeks after (T5) and eight weeks after (T6) the intervention. Physical outcomes measured were oxygen cost of comfortable and brisk-paced walking and maximum distance walked in three minutes. Psychosocial outcomes measured were SF-36, Depression Anxiety and Stress Scales (DASS), Rosenberg Self-Esteem Scale (RSES), Satisfaction with Life Scale (SWLS) and the Barriers to Health Activities for Disabled Persons (BHADP). Physical activity was sampled 6 days at T3, T4 and T6 and two measures were obtained – Actigraph counts per day and total minutes of activity greater than or equal to moderate intensity. Semi-structured interviews were used to evaluate the perceived effectiveness and acceptability of the intervention. The intervention comprised weekly, home-based, interactive problem-solving sessions designed to identify and overcome barriers to activity and to promote walking, together with facilitation of a community based leisure activity of the participant’s choice. Results indicated that the intervention improved important aspects of physical and psychosocial health for community-dwelling people with ABI. Compared with mean baseline measures, improvements in oxygen cost of brisk walking and self-esteem occurred that were both clinically and statistically significant at T6 (p < 0.01). Significant changes in two subscales of the SF-36 and the SWLS also occurred, although they were not sustained at T6. Measures of physical activity increased but not to an extent that was statistically significant. Qualitative data were principally positive, with 100% of participants and their significant others indicating they would recommend the program to another person with ABI. Given the particularly low levels of physical activity in the ABI population, and the correspondingly large individual and community benefits of that will be accrued if their physically active behavior can be increased, the promising results from this program of research indicate that there is a strong justification for allocating the resources necessary to conduct a sufficiently powered, randomized controlled trial of a lifestyle physical activity intervention for people with ABI.