This study was designed to investigate the relationship between retinol intake, bone mineral density, and falls in 101 elderly women aged 72 to 90 years (78.6 yrs. �� 4.3 yrs.). Bone mineral density (BMD) (g/cm��) of the left hip, anterior-posterior lumbar spine (L3), and lateral spine (L3) was measured using dual-energy x-ray absorptiometry. Dietary intake and physical activity were assessed by validated questionnaires (the 100-item Block Food Frequency Questionnaire and the Physical Activity Scale for the Elderly, respectively). Isometric hip abduction strength of the right and left legs was assessed using a hand-held dynamometer. Fall surveillance was collected using a "postcard" system at three-month intervals over a two-year period. Multiple regression analyses were used to show the predictability of retinol, vitamin D, calcium, years past menopause, years on hormone replacement therapy, and physical activity on BMD variables. Together, these variables explained 14% of the variance in total hip BMD at follow-up (R��=0.14, SEE=0.12, p=0.020), 26% of the variance in the anterior-posterior spine BMD at follow-up BMD (R��=0.26, SEE=0.17, p=0.051), and 33% of the variance in lateral spine BMD at follow-up (R��=0.33, SEE=0.10, p=0.009). Two-year changes in hip BMD were poorly predicted using the model with only 5% of total hip BMD variance being explained by the six independent variables (R��=0.05, SEE=0.03, p=0.558). Logistic regression was used to determine whether the likelihood of being a faller vs. a non-faller could be predicted from a model using retinol, vitamin D, average hip strength, and physical activity. It was shown that 11.5% of the variability in fall status could be explained by the model (Cox & Snell's R��=0.115). Using an ROC curve analysis, the model correctly classified 69% of the individuals into the correct "fall category". We conclude that retinol, although not an independent predictor of BMD or fall status, is an important component in the prediction of both BMD and falls. Further interventional research is needed to determine the effects of retinol on BMD and falling. / Graduation date: 2004

published_or_final_version / Medical Sciences / Master / Master of Medical Sciences

published_or_final_version / Medical Sciences / Master / Master of Medical Sciences

published_or_final_version / Sports Science / Master / Master of Science in Sports Science

In six experiments, the relationships between history of falls, reinvestment, fear of falling, perception, balance ability and walking ability of community-dwelling older adults was investigated. In addition, the Movement Specific Reinvestment Scale (MSRS) (Masters, Polman, & Hammond, 1993; Masters, Eves, & Maxwell, 2005) was further validated, using a Chinese version (MSRS-C). In the first experiment (Chapter 2), it was shown that elder fallers scored significantly higher than non-fallers on both the movement self-consciousness and the conscious motor processing components of the MSRS-C. The conscious motor processing component of the MSRS-C was found to discriminate previous faller from non-faller status. In the second experiment (Chapter 3), findings demonstrated that internal focus of attention was greater in elder repeat fallers and increased as task demands increased. However, external focus of attention increased in both elder repeat fallers and elder non-fallers as task demands increased. Elder repeat fallers scored significantly higher than elder non-fallers on the MSRS-C. In the third and fourth experiments (Chapter 4), it was revealed that elder fallers demonstrated greater fear of falling and a higher propensity for movement specific reinvestment than non-fallers. Elderly people perceived stairs as steeper than they were and judged stairs as steeper when making visual-matching and verbal-report estimates that required conscious involvement compared to haptic estimates that are thought to require little conscious involvement. Overestimations when making explicit, conscious judgments were reduced by carrying out a concurrent secondary task during estimation, but there was little effect of the secondary task on implicit, non-conscious judgments of steepness. In Chapter 5, focus group work was conducted to investigate whether (1) elderly people respond differently when asked to complete the MSRS-C in respect of contexts that are not directly related to balance or locomotion and (2) elderly people are better able to differentiate a 4-point Likert response format when completing the MSRS-C than the original 6-point format. Experiment five (Chapter 6) further validated the MSRS-C based on the findings from Chapter 5. Results revealed that both the MSRS-C (general) and MSRS-C (walking) can be used with a six-point or a four-point response format to differentiate elderly Chinese fallers from non-fallers. The overall findings were discussed in the context of theories of motor learning and reinvestment. Implications for rehabilitation training were elucidated. / published_or_final_version / Human Performance / Doctoral / Doctor of Philosophy

Objective: To evaluate the effectiveness of exercises on fall prevention among community-dwelling elderly. Design: Systematic review of randomized controlled trials. Method: The literature search of articles was conducted through the electronic databases of PubMed, Medline and EMBASE and manual search, and was confined to articles in English language with full text and publications from 2002 to 2012. Randomized controlled trials with exercise as the only intervention which aimed at reducing falls in older people aged 60 or above in community i.e. community-dwelling elderly were included. The primary outcomes were number of fallers, fall rate, time to the first and subsequent falls. Data Extraction: A total of 145 articles were retrieved through the electronic data bases (137 articles) and manual search (8 articles) of which 10 were selected after applying the inclusion criteria. According to the checklist developed by National Institute for Health and Clinical Excellence (NICE) for randomized controlled trials, the overall methodological quality of the 10 studies was rated as good as they had fulfilled 79%-93% of the assessment criteria in the NICE checklist. Results: The studies involved a sample size of 3,138 at the median age of 69 to 83. All the subjects were ambulatory and able to mobilize independently. The duration of exercise interventions ranged from 1.5 months to 12 months at a total of 11-156 hours. Eight studies showed that exercises were effective in fall prevention whereas two studies found no evidence of such effectiveness. The effect of exercises on fall prevention was multifactorial including the duration, frequency and continuity of the exercise programs, the health status of the target population and the individuals’ adherence to the exercise programs. Shorter exercise programs and make-up class arrangement resulted in higher adherence rate. Tailor-made (based on the functional capabilities of the subjects) and progressive (gradual increase of the intensity and challenge of the exercises during the intervention period) nature of the exercise programs also enhanced their effectiveness on the elderly. For the studies in which exercise did not have any positive effect on fall prevention, the subjects were generally older (median age at 81) and frailer (with 3-7 frail attributes). Conclusions: This review suggests exercise may be an effective intervention for fall prevention among community-dwelling elderly. Different types of exercise including strengthening, balance, endurance and weight-bearing exercises as well as Tai Chi may prevent falls. The elderly’s age and health status must be taken into consideration when designing exercise intervention programs for the elderly. Further researches are recommended to determine the optimal type, intensity, frequency and duration of exercises in fall prevention. Observations from the studies provide insights for future researches, such as the fall prevention effect of group and home exercises, tailor-made and progressive exercise programs. In Hong Kong, no randomized controlled trial studies have been conducted to examine different intervention programs on fall prevention. To achieve generalizabilty of the studies in the review, further local researches, particularly well designed and powered randomized controlled trials have to be conducted to assess the effect of different kinds of exercise interventions on fall prevention among the community-dwelling elderly. / published_or_final_version / Public Health / Master / Master of Public Health

published_or_final_version / Gerontology / Master / Master of Social Sciences

Thesis (Ph.D.)--University of Wollongong, 2005. / Typescript. Includes bibliographical references: leaf 251-287.

Thesis (Ph. D.)--Victoria University (Melbourne, Vic.), 2008. / Includes bibliographical references.

Thesis (Ph. D.)--Oregon State University, 2010. / Printout. Includes bibliographical references (leaves 112-121). Also available on the World Wide Web.