Introduction: A review of the literature shows that standard clinical balance measures do not adequately predict fall risk in community-dwelling older individuals. There is significant evidence demonstrating the interactions of fear, anxiety, and confidence with the control of standing posture. Little is known however about the nature of this relationship under more challenging balance conditions, particularly in the elderly. The primary purpose of this work was to evaluate the relationship between fear of falling, clinical balance measures and fall-risk. Methods: Three studies were conducted evaluating the effects of postural threat (manipulated by support surface elevation) and/or cognitive loading (working memory secondary task) on clinical balance performance and task-specific psychological measures. Predictive and construct validity as well as test-retest reliability was evaluated for measures used to assess fear of falling and related psychological constructs . Results: Postural threat resulted in reduced balance confidence and perceived stability as well as increased state anxiety and fear of falling. These changes were significantly correlated to decrements in performance of clinical balance tasks. Neither standard clinical scales of balance and mobility nor generalized psychological measures, alone or in combination, could predict falls in community-dwelling elderly. However, combined scores on selected challenging clinical balance tasks could significantly predict falls. Furthermore, improved predictive precision resulted from having these tasks performed under combined postural threat and cognitive loading. Finally, the inclusion of task-specific psychological measures resulted in further improvements to predictive precision. Psychological measures demonstrated fair to excellent test-retest reliability in both healthy young and independent-living older individuals. Conclusions: Clinical balance tasks performed under more challenging conditions likely better reflect everyday experiences in which a fall is likely to occur. Incorporating easy-to-administer task-specific psychological evaluations and self-reported health estimates with clinical balance assessments might improve the likelihood of correctly identifying community-dwelling individuals at risk for falls. Improved estimates of fall-risk may lead to a reduction in the number of falls experienced in this population, thereby reducing the significant burden of fall-related hospitalizations, treatments and rehabilitation on the individual, families and health care system.

Introduction: A review of the literature shows that standard clinical balance measures do not adequately predict fall risk in community-dwelling older individuals. There is significant evidence demonstrating the interactions of fear, anxiety, and confidence with the control of standing posture. Little is known however about the nature of this relationship under more challenging balance conditions, particularly in the elderly. The primary purpose of this work was to evaluate the relationship between fear of falling, clinical balance measures and fall-risk. Methods: Three studies were conducted evaluating the effects of postural threat (manipulated by support surface elevation) and/or cognitive loading (working memory secondary task) on clinical balance performance and task-specific psychological measures. Predictive and construct validity as well as test-retest reliability was evaluated for measures used to assess fear of falling and related psychological constructs . Results: Postural threat resulted in reduced balance confidence and perceived stability as well as increased state anxiety and fear of falling. These changes were significantly correlated to decrements in performance of clinical balance tasks. Neither standard clinical scales of balance and mobility nor generalized psychological measures, alone or in combination, could predict falls in community-dwelling elderly. However, combined scores on selected challenging clinical balance tasks could significantly predict falls. Furthermore, improved predictive precision resulted from having these tasks performed under combined postural threat and cognitive loading. Finally, the inclusion of task-specific psychological measures resulted in further improvements to predictive precision. Psychological measures demonstrated fair to excellent test-retest reliability in both healthy young and independent-living older individuals. Conclusions: Clinical balance tasks performed under more challenging conditions likely better reflect everyday experiences in which a fall is likely to occur. Incorporating easy-to-administer task-specific psychological evaluations and self-reported health estimates with clinical balance assessments might improve the likelihood of correctly identifying community-dwelling individuals at risk for falls. Improved estimates of fall-risk may lead to a reduction in the number of falls experienced in this population, thereby reducing the significant burden of fall-related hospitalizations, treatments and rehabilitation on the individual, families and health care system.

Background: The incidence of falls and severity of fall related injuries steadily increase with age. As well as physical injury, falls can lead to adverse psychological and social consequences, which can further reduce older adults’ quality of life. The most commonly reported cause of falls in older persons is tripping over an obstacle, which may reflect the difficulty many older adults have in making an appropriate stepping response. In order to reduce the number of falls experienced by older adults, individuals at high risk of falling need to be identified so that targeted interventions can be implemented. Aims: This series of studies aimed to develop a new Multiple-Lunge test to distinguish between Fallers and Non-fallers in independent older adults, aged 65 years and over. The studies sought to determine the test-retest reliability of the Multiple-Lunge test; as well its validity to predict faller status in a sample of community-dwelling older adults. Methods: One hundred and thirty community-dwelling older adults, aged 65 – 93 years (mean age 77 ± 7 years) with (n = 40) and without (n = 90) a history of falls were administered the Multiple-Lunge test. For the purpose of this study, a Faller was classified as an older adult with a history of one fall, or a Multiple-faller if there was a history of two or more falls in the previous 12 months. The Multiple-Lunge test required the individual to lunge forward to a step length determined as 60% of their leg length, and return to start position, for a total of five repetitions. Two trials were performed after a familiarisation trial. The number of correct steps and the total time for the five steps were recorded for each trial. The highest number of correct steps and the lowest total time of the two trials were used for data analysis. Test-retest reliability of the Multiple-Lunge test was established across two testing occasions from a sub-sample of the validity study (n = 14, mean age 79 ± 6 years). A cross-sectional design was used to establish the sensitivity and specificity of the Multiple-Lunge test to predict faller status based on retrospective self-reported fall history. Initial analysis using the number of correct steps and total time, was followed by a linear regression analysis to determine the validity of the test to predict falls. Results: The Multiple-Lunge test was found to be reliable across trials (ICC = 0.79 for number of correct steps; ICC = 0.86 for total time). The change in the mean for steps was small and similar across four trials (-0.43 steps, -0.36 steps, -0.50 steps). The change in the mean for time showed a gradual decrease in time scores across trials (-0.69 seconds, -0.73 seconds, -0.93 seconds). Sensitivity and specificity values were calculated as 73% and 63% for predicting Multiple-fallers using the measure of all five steps done correctly. Linear regression analysis did not indicate that the Multiple-Lunge test could be used to predict faller status for either of the step predictor variables (0/5 steps or 5/5 steps). However, a very high sensitivity value (98%) was observed for the variable of both steps and time in predicting Fallers. Also a very high specificity value (99%) was recorded for the variable of time to predict Multiple-fallers. Conclusions: The Multiple-Lunge test is an easily administered test for independent older adults. Due to the challenging nature of the test it may be well suited to detect subtle differences in abilities of higher functioning older adults. The test mimics the movements needed in response to a trip, the most common cause of falls in older adults. This test is a reliable and reasonably valid measure of falls risk. A practitioner can be confident in 7 out of 10 cases that an older adult who cannot complete all five steps of the Multiple-Lunge test is at high risk of falls. The results of this thesis suggest that there is potential for the Multiple-Lunge test to be used in clinical practice and fall prevention research. However, additional research on how to further increase its validity and/or to determine the most appropriate populations with which to administer this test appears warranted.

Background: The incidence of falls and severity of fall related injuries steadily increase with age. As well as physical injury, falls can lead to adverse psychological and social consequences, which can further reduce older adults’ quality of life. The most commonly reported cause of falls in older persons is tripping over an obstacle, which may reflect the difficulty many older adults have in making an appropriate stepping response. In order to reduce the number of falls experienced by older adults, individuals at high risk of falling need to be identified so that targeted interventions can be implemented. Aims: This series of studies aimed to develop a new Multiple-Lunge test to distinguish between Fallers and Non-fallers in independent older adults, aged 65 years and over. The studies sought to determine the test-retest reliability of the Multiple-Lunge test; as well its validity to predict faller status in a sample of community-dwelling older adults. Methods: One hundred and thirty community-dwelling older adults, aged 65 – 93 years (mean age 77 ± 7 years) with (n = 40) and without (n = 90) a history of falls were administered the Multiple-Lunge test. For the purpose of this study, a Faller was classified as an older adult with a history of one fall, or a Multiple-faller if there was a history of two or more falls in the previous 12 months. The Multiple-Lunge test required the individual to lunge forward to a step length determined as 60% of their leg length, and return to start position, for a total of five repetitions. Two trials were performed after a familiarisation trial. The number of correct steps and the total time for the five steps were recorded for each trial. The highest number of correct steps and the lowest total time of the two trials were used for data analysis. Test-retest reliability of the Multiple-Lunge test was established across two testing occasions from a sub-sample of the validity study (n = 14, mean age 79 ± 6 years). A cross-sectional design was used to establish the sensitivity and specificity of the Multiple-Lunge test to predict faller status based on retrospective self-reported fall history. Initial analysis using the number of correct steps and total time, was followed by a linear regression analysis to determine the validity of the test to predict falls. Results: The Multiple-Lunge test was found to be reliable across trials (ICC = 0.79 for number of correct steps; ICC = 0.86 for total time). The change in the mean for steps was small and similar across four trials (-0.43 steps, -0.36 steps, -0.50 steps). The change in the mean for time showed a gradual decrease in time scores across trials (-0.69 seconds, -0.73 seconds, -0.93 seconds). Sensitivity and specificity values were calculated as 73% and 63% for predicting Multiple-fallers using the measure of all five steps done correctly. Linear regression analysis did not indicate that the Multiple-Lunge test could be used to predict faller status for either of the step predictor variables (0/5 steps or 5/5 steps). However, a very high sensitivity value (98%) was observed for the variable of both steps and time in predicting Fallers. Also a very high specificity value (99%) was recorded for the variable of time to predict Multiple-fallers. Conclusions: The Multiple-Lunge test is an easily administered test for independent older adults. Due to the challenging nature of the test it may be well suited to detect subtle differences in abilities of higher functioning older adults. The test mimics the movements needed in response to a trip, the most common cause of falls in older adults. This test is a reliable and reasonably valid measure of falls risk. A practitioner can be confident in 7 out of 10 cases that an older adult who cannot complete all five steps of the Multiple-Lunge test is at high risk of falls. The results of this thesis suggest that there is potential for the Multiple-Lunge test to be used in clinical practice and fall prevention research. However, additional research on how to further increase its validity and/or to determine the most appropriate populations with which to administer this test appears warranted.

No description available.

Falling in indoor home setting can be dangerous for elderly population (in USA and globally), causing hospitalization, long term reduced mobility, disability or even death. Prevention of fall by monitoring different human activities or identifying the aftermath of fall has greater significance for elderly population. This is possible due to the availability and emergence of miniaturized sensors with advanced electronics and data analytics tools. This thesis aims at developing machine learning models to classify fall activities and non-fall activities. In this thesis, two types of neural networks with different parameters were tested for their capability in dealing with such tasks. A publicly available dataset was used to conduct the experiments. The two types of neural network models, convolution and recurrent neural network, were developed and evaluated. Convolution neural network achieved an accuracy of over 95% for classifying fall and non-fall activities. Recurrent neural network provided an accuracy of over 97% accuracy in predicting fall, non-fall and a third category activity (defined in this study as “pre/postcondition”). Both neural network models show high potential for being used in fall prevention and management activity. Moreover, two theoretical designs of fall detection systems were proposed in this thesis based on the developed convolution and recurrent neural networks.