This research studies delegation in outpatient physiotherapy and attempts to solve its deficiency by designing a theoretical model of constructive delegation (CD model). The CD model is functional and uses a systematic and rational approach to plan the level of delegation by using task and cost-benefit analysis and it supports delegation dynamically by organising training, working partnerships, communications and the working environment. An initial survey of tasks carried out by physiotherapists and assistants at ten sites found inconsistent approaches to delegation, with 80% of physiotherapists expressing concerns. This mirrored similar experiences in the literature where there was evidence of some delegation of technical clinical tasks, but also resistance within the profession. Generally in the literature delegation was defined and eluded to, but was not constructed in order to provide a planned system. The CD model was offered as a tool to, by construction, implement delegation safely and without loss of quality to ensure appropriate skills for appropriate tasks and to analyse current practice and implicitly suggest improvements.

No description available.

No / In October 2015 I started an optional Student Supervised Physiotherapy clinic for neurologically impaired patients. The primary aim of the clinic is to provide an experiential learning environment for Levels 5 and 6 students. The secondary aim of the clinic is to provide an assessment, review and treatment to local service users experiencing neurological problems and who are seeking further/alternate physiotherapy within a learning environment. The clinic is the first dedicated student supervised neurological physiotherapy clinic in the UK that we are aware of.

The purpose of this study was to evaluate the methods currently available to measure the functional outcomes of physiotherapy treatment for low back problems. As a preliminary step, all extant questionnaires were located and evaluated against practical criteria to determine their likely utility in clinical practice. This process identified a large number of questionnaires, however, only six back-specific questionnaires fulfilled the practical criteria for clinical application. Four of these questionnaires were selected for further evaluation along with a generic health status assessment instrument, the SF-36 Health Survey. Current recommendations suggest that a low-back specific and a generic questionnaire are required for comprehensive assessment of the impact of low back problems. The four back-specific questionnaires selected were the Oswestry Disability Questionnaire, the Quebec Back Pain Disability Scale, the Roland-Morris Disability Questionnaire and the Waddell Disability Index. An evaluation of the literature on the clinimetric properties of these questionnaires revealed that little information was available for the Quebec and Waddell questionnaires and no information was available for any of the questionnaires for a clinical population of people with low back pain seeking physiotherapy treatment in an Australian setting. The primary aim of the research was identify which, if any, of the questionnaires should be recommended for measuring outcomes of physiotherapy treatment for low back pain. Consecutive ambulatory (non-admitted) patients presenting for physiotherapy treatment at three public hospital physiotherapy outpatient departments, three community health services, and four private practices were invited to enter the study. Patients were included if they were seeking treatment for a low back problem, were aged 18 or over, and could read and write English. Subjects completed the questionnaires on two occasions six-weeks apart. One hundred and forty subjects returned the first set of questionnaires, and 106 the second set. The mean age of the sample at pre-test was 51 (SD 17) and ranged from 18 to 89 years. Sixty-six percent were female, 41% were employed and 12% were receiving compensation for their back problem. Duration of the back complaint was more than six weeks for 56% of subjects, and 60% reported five or more previous episodes or continuous pain. Referred pain in the buttock thigh or leg was reported by 70% of subjects. The first aim was to compare the questionnaires for acceptability and comprehensibility. Data quality was high for all the questionnaires (less than 5% missing data). As expected, subjects found the more complex SF-36 Health Survey more difficult to complete than three of the low-back questionnaires. However, less than 10% of subjects found any of the questionnaires more than a little difficult to complete. The next aim was to explore the internal structure and inter-relationships of the low-back questionnaires and the three physical scales of the SF-36 Health Survey (Physical Functioning, Role-Physical and Bodily Pain). Analysis of item-item correlation, item-total correlation and Cronbach�s alpha confirmed that all scales were internally consistent. Factor analysis confirmed item homogeneity (unidimensionality) of all questionnaires except the Roland-Morris Disability Questionnaire. The questionnaires were significantly intercorrelated, but correlations exceeded .70 only for the Oswestry, Quebec and Waddell questionnaires. The next aim of the research was to compare test-retest reliability of the questionnaires. The Oswestry, Quebec and SF-36 Physical Functioning scale had sufficient reliability and scale width for clinical application. Despite previous reports of high reliability, the Roland-Morris scale was significantly less reliable than several of the other questionnaires. This indicates the importance of establishing the measurement properties of a test in the population or setting in which it will be used. The Waddell Disability Index, and the SF-36 Role-Physical and Bodily Pain scales had insufficient scale width to be useful in clinical practice. More than 15% of respondents had an initial score on these scales that would not allow change to be detected with 90% confidence. The next aim of the research was to compare the responsiveness of the questionnaires. None of the questionnaires was consistently identified as more or less responsive than the others although two methods (effect size and Liang�s standardized response mean) suggested the SF-36 Bodily Pain scale was more responsive than some other questionnaires. A secondary aim of this section was to evaluate the validity of the many available responsiveness indices and a novel �reliable change� method. A �known groups� strategy was used to determine whether the responsiveness index could discriminate between the low-back relevant questionnaires and the SF-36 General Health scale, the scores of which did not change across the retest period. With the exception of the novel �reliable change� method the responsiveness indices were all found to be valid indicators of responsiveness. Guyatt�s Responsiveness Index, effect size and Liang�s standardized response mean discriminated at 95% confidence between the reference scale and all the low-back questionnaires. The standardized response mean, t-test, correlation and ROC methods discriminated between the reference scale and five or six of the seven other questionnaires. Guyatt�s index was recommended as the best of the criterion-based methods, and the effect size the best of the distribution-based methods. The three questionnaires identified as having sufficient reliability and scale width, the Oswestry, Quebec and SF-36 Physical Functioning scale, were next analysed for data fit to a Rasch model. All three questionnaires had good data fit and item function was not affected by time, age, gender or whether or not subjects reported avoiding bending. The final aim of this research was to identify by Rasch analysis items to supplement the SF-36 Physical Functioning scale. The new scale, named the Low-Back SF-36 Physical Functioning18, showed comparable reliability and responsiveness to the SF-36 Physical Functioning scale. Further research is required to establish the measurement properties of the Low-Back SF-36 Physical Functioning18 scale in an independent sample. However, it has the potential to improve the clinical measurement of function by providing clinicians with a single measurement tool for comprehensive assessment of patients with low back pain.

Although traditional prevention and management strategies for hamstring injuries have focussed on optimising muscle strength, flexibility and endurance, incidence and/or recurrence rates of these injuries remains high. A theoretical framework was developed considering additional factors that increase the stabilising demand of the hamstrings. These factors included loss of related stability at the knee and lumbopelvic regions and extrinsic factors during functional and sporting activities. The aims of this research were to determine whether electromyographic (EMG) derived hamstrings, quadriceps and gluteal muscle activation patterns as well as isokinetic torque generation patterns could differentiate athletes who had incurred a hamstring injury from uninjured control athletes. It was hypothesised that the EMG activity of the injured participants would be decreased compared to uninjured control participants during maximal activities, but increased during weight bearing activities. The research included the identification of laboratory-based tasks relevant to the function of the hamstring muscles; test-retest reliability of EMG variables recorded during these tasks; and a comparative cross-sectional study of hamstring-injured (hamstring group, HG) and control athletes (control group, CG). Electromyographic activation patterns were determined during assessment of concentric and eccentric isokinetic strength of the thigh muscles, during transition from double- to single-leg stance, and forward lunging. Isokinetic and EMG onset and amplitude variables were compared both within- and between-groups. Despite no significant differences for peak torque, the HG injured limb generated lower average eccentric flexor torque towards the outer range of motion in comparison to the HG uninjured limb (P = 0.034) and the CG bilateral average (P = 0.025). Furthermore, the EMG root mean square (RMS) decrease from the start to the end range of the eccentric flexor contraction was greater for the HG injured limb hamstrings than the CG bilateral average. During the transition from double- to single-leg stance, the EMG onsets of the HG injured limb (biceps femoris [BF] P < 0.001, medial hamstrings [MH] P = 0.001), and the HG uninjured limb (BF P = 0.023, MH P = 0.011) were earlier in comparison to the CG bilateral average. The transition normalised EMG RMS was significantly higher for the HG injured side BF (P = 0.032), MH (P = 0.039) and vastus lateralis (VL, P = 0.037) in comparison to the CG bilateral average. During the forward lunge, no significant differences were observed within- and between-groups for the normalised EMG amplitude prior to and following initial foot contact. These results suggest that during maximal isokinetic eccentric flexor contractions, the average torque and EMG activity is decreased towards the lengthened position of the hamstring-injured limb. This may be due to structural changes or neurophysiological inhibitory mechanisms. During the static weight bearing task an earlier onset of the HG hamstring muscles was evident in comparison to controls. The hamstrings and the VL of the injured limbs were activated at greater normalised amplitude. The increased muscle activation in the hamstring-injured limbs during the support phase may indicate a greater demand towards stability of the kinetic chain or changes in proprioceptive function. Future research should consider the mechanisms and clinical implications underlying a loss of eccentric flexor torque towards the outer range of contraction, and investigate why increased activation of thigh muscles occurs during the static weight bearing task in hamstring-injured athletes.

During pregnancy the female body must accommodate the enlarging gravid uterus and increased mass. Therefore the maternal musculoskeletal system is required to adapt in both morphology and functional workload. After childbirth there is a rapid change in both mass and dimensions, requiring further adaptations. The objectives of the study were to investigate seated and standing upper body posture, the kinematics of seated and standing trunk motion, and the three dimensional kinematics and kinetics during rising to stand from a chair, as pregnancy progressed and in the early post-birth period. Nine maternal subjects (aged 28 to 40 years) were tested at less than 16 weeks, 24 weeks, 30 weeks, 38 weeks gestation and at 8 weeks postbirth. The subjects, fitted with 37 retroreflective markers, were filmed during upright sitting, quiet standing, and four trials each of maximum seated and standing trunk forward flexion, side to side flexion and during maximum seated axial rotation. Three trials each of constrained and free rising to stand from a height adjustable stool and with each foot placed on a forceplate were also recorded. An eight-camera motion analysis system was used to record movements of the body segments and synchronised force plate variables in three dimensions. Motion of the ankle, knee and hip joints, pelvic, thoracic and head segments and the thoracolumbar and cervicothoracic spines and shoulder joints were investigated. Twelve nulliparous subjects (aged 21 to 35 years) were used as controls to provide standard descriptive data and to investigate the consistency of the selected biomechanical variables with repeated testing. A repeated measures ANOVA was used to investigate the possibility of linear and quadratic trends showing systematic changes within the maternal group, over the four test sessions during pregnancy for each variable. Two tailed Student t-tests were used to compare the maternal postbirth variable results with the control group. There was no significant effect of pregnancy on the upper body posture during upright sitting and quiet standing. Postbirth, the pelvic segment had a smaller anterior orientation and the thoracolumbar spine was less extended, indicating a flatter spinal curve. The maternal subjects were similar to the control subjects in early pregnancy and postbirth for trunk segment motions during seated and standing forward flexion and side to side flexion and seated axial rotation. Strategies, such as increasing the width of the base of support and reducing obstruction to movements from other body parts, were used in late pregnancy in attempts to minimise the effects of increased trunk mass and circumference. For seated and standing side to side flexion, the strategies were successful and no significant decreases in range of motion were seen. For seated and standing forward flexion and seated axial rotation, motion of the thoracic segment and the thoracolumbar spine were significantly reduced, although movement of the pelvis was less affected. In early pregnancy and postbirth the kinematics and kinetics of the lower limbs and upper body segment kinematics during constrained and free rising were generally similar to the control subjects. As pregnancy progressed there were increases in mass and dimensions of body segments. The effect of increased mass was seen in increased ground reaction forces and sagittal plane lower limb joint external moments. An increased base of support width was found in association with an increased lateral ground reaction force and ankle inversion moment from each foot, which would move the body centre of mass medially. There was little change in the three dimensional kinematics of the thoracolumbar and cervicothoracic spine, although the contribution of the upper body segments differed for each rise condition. There were also few significant changes in the displacement of the ankle, knee and hip, and the angular velocity of ankle and knee joints. The maternal subjects were thus able to flex the upper body forward, raise the body and maintain stability as pregnancy progressed, regardless of whether the rise to stand was performed in a natural manner or under constrained conditions. The overall results show that, contrary to expectations as pregnancy progressed, maternal subjects minimised propulsion rather than increasing it to overcome the increased mass and possibly limited trunk flexion. A fear of postural instability may have made the subjects more cautious and as they were able to adequately flex the trunk forward, propulsion was minimised in favour of maintaining upright terminal balance.

The occupation of sheep-shearing is classified as heavy to very-heavy physical work requiring a high level of energy expenditure while performing a repetitive task in a predominantly sustained trunk flexion posture. The task is further complicated by unpredictable animal behaviour. Shearing is considered to carry a high risk of injury particularly to the lumbar spine and epidemiological studies have demonstrated that the prevalence of low back pain (LBP) in shearers is high relative to other occupations. The emergence of a commercial shearing-aid (Warrie Back-Aid TM) that is purported to decrease loading on the spine, reduce injury rate and alleviate symptoms of spinal origin has been welcomed by many sectors of the wool-harvesting industry. However, the precise biomechanical influence of the Warrie Back-Aid TM (WBA) on the sheep-shearing task has not been quantified. The purpose of this study was to investigate the effect of the WBA on three-dimensional kinematics and electromyography of the trunk for 12 experienced shearers during the occupational task of sheep-shearing. Sheep-shearing is a highly patterned process comprised of specific interlinked phases. Nine distinct phases of the shearing-task where the harness was worn during the removal of the sheep�s fleece were investigated. The study used an opto-electronic motion analysis system and rigid body dynamic modelling. The shearer�s trunk and pelvis were considered as a series of three coupled rigid segments; pelvis, lumbar, and thorax with four passive retro-reflective markers defining each segment. The kinematic variables of angular displacement, velocity and acceleration at the thoraco-lumbar and lumbo-pelvic joint centres were calculated for each shearer while shearing with and without the WBA. An eight channel sEMG telemetry system was used to simultaneous record activity in four pairs of trunk muscles. Temporal analysis of the sEMG signal gave information on the duration and relative intensity of trunk muscle activity. The results demonstrated that the task of sheep-shearing required endurance-based muscle activity and the adoption of quasi-static posturing combined with complex asymmetrical trunk motion for extensive periods of the task time. There was considerable variability in the trunk motion of individual shearers despite the pattern-taught and repeatable nature of the shearing task. The introduction of the WBA had no effect on the time taken to shear or trunk kinematics however it resulted in reductions in muscular activity of the trunk extensors. When the complete shearing task was analysed there was a statistically significant reduction in mean intensity of muscle activity for the left multifidus (p = 0.010), right multifidus (p = 0.001), right iliocostalis (p = 0.004) and right longissimus (p = 0.002) when the WBA was used. A reduction in muscular activity of the trunk extensors during the sheep-shearing task may result in a decrease in spinal loading, energy expenditure and muscular fatigue. The clinical recommendation based on the biomechanical results of this and other studies is that the WBA should be incorporated into the practice of sheep-shearing to assist in the reduction of inherent risks associated with the shearing task. Prospective studies into the effect of the harness on LBP are required to endorse this recommendation.

The occupation of sheep-shearing is classified as heavy to very-heavy physical work requiring a high level of energy expenditure while performing a repetitive task in a predominantly sustained trunk flexion posture. The task is further complicated by unpredictable animal behaviour. Shearing is considered to carry a high risk of injury particularly to the lumbar spine and epidemiological studies have demonstrated that the prevalence of low back pain (LBP) in shearers is high relative to other occupations. The emergence of a commercial shearing-aid (Warrie Back-Aid TM) that is purported to decrease loading on the spine, reduce injury rate and alleviate symptoms of spinal origin has been welcomed by many sectors of the wool-harvesting industry. However, the precise biomechanical influence of the Warrie Back-Aid TM (WBA) on the sheep-shearing task has not been quantified. The purpose of this study was to investigate the effect of the WBA on three-dimensional kinematics and electromyography of the trunk for 12 experienced shearers during the occupational task of sheep-shearing. Sheep-shearing is a highly patterned process comprised of specific interlinked phases. Nine distinct phases of the shearing-task where the harness was worn during the removal of the sheep�s fleece were investigated. The study used an opto-electronic motion analysis system and rigid body dynamic modelling. The shearer�s trunk and pelvis were considered as a series of three coupled rigid segments; pelvis, lumbar, and thorax with four passive retro-reflective markers defining each segment. The kinematic variables of angular displacement, velocity and acceleration at the thoraco-lumbar and lumbo-pelvic joint centres were calculated for each shearer while shearing with and without the WBA. An eight channel sEMG telemetry system was used to simultaneous record activity in four pairs of trunk muscles. Temporal analysis of the sEMG signal gave information on the duration and relative intensity of trunk muscle activity. The results demonstrated that the task of sheep-shearing required endurance-based muscle activity and the adoption of quasi-static posturing combined with complex asymmetrical trunk motion for extensive periods of the task time. There was considerable variability in the trunk motion of individual shearers despite the pattern-taught and repeatable nature of the shearing task. The introduction of the WBA had no effect on the time taken to shear or trunk kinematics however it resulted in reductions in muscular activity of the trunk extensors. When the complete shearing task was analysed there was a statistically significant reduction in mean intensity of muscle activity for the left multifidus (p = 0.010), right multifidus (p = 0.001), right iliocostalis (p = 0.004) and right longissimus (p = 0.002) when the WBA was used. A reduction in muscular activity of the trunk extensors during the sheep-shearing task may result in a decrease in spinal loading, energy expenditure and muscular fatigue. The clinical recommendation based on the biomechanical results of this and other studies is that the WBA should be incorporated into the practice of sheep-shearing to assist in the reduction of inherent risks associated with the shearing task. Prospective studies into the effect of the harness on LBP are required to endorse this recommendation.

The purpose of this study was to evaluate the methods currently available to measure the functional outcomes of physiotherapy treatment for low back problems. As a preliminary step, all extant questionnaires were located and evaluated against practical criteria to determine their likely utility in clinical practice. This process identified a large number of questionnaires, however, only six back-specific questionnaires fulfilled the practical criteria for clinical application. Four of these questionnaires were selected for further evaluation along with a generic health status assessment instrument, the SF-36 Health Survey. Current recommendations suggest that a low-back specific and a generic questionnaire are required for comprehensive assessment of the impact of low back problems. The four back-specific questionnaires selected were the Oswestry Disability Questionnaire, the Quebec Back Pain Disability Scale, the Roland-Morris Disability Questionnaire and the Waddell Disability Index. An evaluation of the literature on the clinimetric properties of these questionnaires revealed that little information was available for the Quebec and Waddell questionnaires and no information was available for any of the questionnaires for a clinical population of people with low back pain seeking physiotherapy treatment in an Australian setting. The primary aim of the research was identify which, if any, of the questionnaires should be recommended for measuring outcomes of physiotherapy treatment for low back pain. Consecutive ambulatory (non-admitted) patients presenting for physiotherapy treatment at three public hospital physiotherapy outpatient departments, three community health services, and four private practices were invited to enter the study. Patients were included if they were seeking treatment for a low back problem, were aged 18 or over, and could read and write English. Subjects completed the questionnaires on two occasions six-weeks apart. One hundred and forty subjects returned the first set of questionnaires, and 106 the second set. The mean age of the sample at pre-test was 51 (SD 17) and ranged from 18 to 89 years. Sixty-six percent were female, 41% were employed and 12% were receiving compensation for their back problem. Duration of the back complaint was more than six weeks for 56% of subjects, and 60% reported five or more previous episodes or continuous pain. Referred pain in the buttock thigh or leg was reported by 70% of subjects. The first aim was to compare the questionnaires for acceptability and comprehensibility. Data quality was high for all the questionnaires (less than 5% missing data). As expected, subjects found the more complex SF-36 Health Survey more difficult to complete than three of the low-back questionnaires. However, less than 10% of subjects found any of the questionnaires more than a little difficult to complete. The next aim was to explore the internal structure and inter-relationships of the low-back questionnaires and the three physical scales of the SF-36 Health Survey (Physical Functioning, Role-Physical and Bodily Pain). Analysis of item-item correlation, item-total correlation and Cronbach�s alpha confirmed that all scales were internally consistent. Factor analysis confirmed item homogeneity (unidimensionality) of all questionnaires except the Roland-Morris Disability Questionnaire. The questionnaires were significantly intercorrelated, but correlations exceeded .70 only for the Oswestry, Quebec and Waddell questionnaires. The next aim of the research was to compare test-retest reliability of the questionnaires. The Oswestry, Quebec and SF-36 Physical Functioning scale had sufficient reliability and scale width for clinical application. Despite previous reports of high reliability, the Roland-Morris scale was significantly less reliable than several of the other questionnaires. This indicates the importance of establishing the measurement properties of a test in the population or setting in which it will be used. The Waddell Disability Index, and the SF-36 Role-Physical and Bodily Pain scales had insufficient scale width to be useful in clinical practice. More than 15% of respondents had an initial score on these scales that would not allow change to be detected with 90% confidence. The next aim of the research was to compare the responsiveness of the questionnaires. None of the questionnaires was consistently identified as more or less responsive than the others although two methods (effect size and Liang�s standardized response mean) suggested the SF-36 Bodily Pain scale was more responsive than some other questionnaires. A secondary aim of this section was to evaluate the validity of the many available responsiveness indices and a novel �reliable change� method. A �known groups� strategy was used to determine whether the responsiveness index could discriminate between the low-back relevant questionnaires and the SF-36 General Health scale, the scores of which did not change across the retest period. With the exception of the novel �reliable change� method the responsiveness indices were all found to be valid indicators of responsiveness. Guyatt�s Responsiveness Index, effect size and Liang�s standardized response mean discriminated at 95% confidence between the reference scale and all the low-back questionnaires. The standardized response mean, t-test, correlation and ROC methods discriminated between the reference scale and five or six of the seven other questionnaires. Guyatt�s index was recommended as the best of the criterion-based methods, and the effect size the best of the distribution-based methods. The three questionnaires identified as having sufficient reliability and scale width, the Oswestry, Quebec and SF-36 Physical Functioning scale, were next analysed for data fit to a Rasch model. All three questionnaires had good data fit and item function was not affected by time, age, gender or whether or not subjects reported avoiding bending. The final aim of this research was to identify by Rasch analysis items to supplement the SF-36 Physical Functioning scale. The new scale, named the Low-Back SF-36 Physical Functioning18, showed comparable reliability and responsiveness to the SF-36 Physical Functioning scale. Further research is required to establish the measurement properties of the Low-Back SF-36 Physical Functioning18 scale in an independent sample. However, it has the potential to improve the clinical measurement of function by providing clinicians with a single measurement tool for comprehensive assessment of patients with low back pain.

During pregnancy the female body must accommodate the enlarging gravid uterus and increased mass. Therefore the maternal musculoskeletal system is required to adapt in both morphology and functional workload. After childbirth there is a rapid change in both mass and dimensions, requiring further adaptations. The objectives of the study were to investigate seated and standing upper body posture, the kinematics of seated and standing trunk motion, and the three dimensional kinematics and kinetics during rising to stand from a chair, as pregnancy progressed and in the early post-birth period. Nine maternal subjects (aged 28 to 40 years) were tested at less than 16 weeks, 24 weeks, 30 weeks, 38 weeks gestation and at 8 weeks postbirth. The subjects, fitted with 37 retroreflective markers, were filmed during upright sitting, quiet standing, and four trials each of maximum seated and standing trunk forward flexion, side to side flexion and during maximum seated axial rotation. Three trials each of constrained and free rising to stand from a height adjustable stool and with each foot placed on a forceplate were also recorded. An eight-camera motion analysis system was used to record movements of the body segments and synchronised force plate variables in three dimensions. Motion of the ankle, knee and hip joints, pelvic, thoracic and head segments and the thoracolumbar and cervicothoracic spines and shoulder joints were investigated. Twelve nulliparous subjects (aged 21 to 35 years) were used as controls to provide standard descriptive data and to investigate the consistency of the selected biomechanical variables with repeated testing. A repeated measures ANOVA was used to investigate the possibility of linear and quadratic trends showing systematic changes within the maternal group, over the four test sessions during pregnancy for each variable. Two tailed Student t-tests were used to compare the maternal postbirth variable results with the control group. There was no significant effect of pregnancy on the upper body posture during upright sitting and quiet standing. Postbirth, the pelvic segment had a smaller anterior orientation and the thoracolumbar spine was less extended, indicating a flatter spinal curve. The maternal subjects were similar to the control subjects in early pregnancy and postbirth for trunk segment motions during seated and standing forward flexion and side to side flexion and seated axial rotation. Strategies, such as increasing the width of the base of support and reducing obstruction to movements from other body parts, were used in late pregnancy in attempts to minimise the effects of increased trunk mass and circumference. For seated and standing side to side flexion, the strategies were successful and no significant decreases in range of motion were seen. For seated and standing forward flexion and seated axial rotation, motion of the thoracic segment and the thoracolumbar spine were significantly reduced, although movement of the pelvis was less affected. In early pregnancy and postbirth the kinematics and kinetics of the lower limbs and upper body segment kinematics during constrained and free rising were generally similar to the control subjects. As pregnancy progressed there were increases in mass and dimensions of body segments. The effect of increased mass was seen in increased ground reaction forces and sagittal plane lower limb joint external moments. An increased base of support width was found in association with an increased lateral ground reaction force and ankle inversion moment from each foot, which would move the body centre of mass medially. There was little change in the three dimensional kinematics of the thoracolumbar and cervicothoracic spine, although the contribution of the upper body segments differed for each rise condition. There were also few significant changes in the displacement of the ankle, knee and hip, and the angular velocity of ankle and knee joints. The maternal subjects were thus able to flex the upper body forward, raise the body and maintain stability as pregnancy progressed, regardless of whether the rise to stand was performed in a natural manner or under constrained conditions. The overall results show that, contrary to expectations as pregnancy progressed, maternal subjects minimised propulsion rather than increasing it to overcome the increased mass and possibly limited trunk flexion. A fear of postural instability may have made the subjects more cautious and as they were able to adequately flex the trunk forward, propulsion was minimised in favour of maintaining upright terminal balance.