研究背景：影響機構老人跌倒的危險因素，在過去的跌倒研究中已被明確探究。然而，針對長期照護機構中的 住民發生跌倒事件與身體的平衡功能之間的關係，目前研究所知仍然有限。 研究目的：本研究旨在探討影響長期照護機構住民的身體平衡功能，跌倒發生事件之危險因素及其之間的關係。 研究方法：採前瞻性研究設計，以台北某機構中心老年住民為對象（N=133），採立意取樣原則收集資料。研究 工具主要為巴氏量表、Tinetti平衡步態評估表、簡易智能測驗、老人憂鬱量表、康乃爾失智憂鬱量表、 精神行為症狀、Hendrich跌倒危險評估量表、跌倒事件等，並以SPSS20.0套裝軟體進行階層性迴歸 與邏輯性迴歸分析。 研究結果：於階層迴歸分析結果發現「性別」（ß = .80，p < .001）、「平衡功能」（ß = -.10，p < .05）、「認知功能」 （ß = -.09，p< .001）、「NPI精神行為徵候」（ß = -.09，p < .05）之迴歸係數達顯著，共可解釋跌倒危 險因子47%的變異量。另透過一年期的追蹤，共有28人次發生跌倒，再以羅吉斯迴歸分析結果得知， 平衡功能中的起立、坐下的動作為跌倒發生的重要危險因素。 結論與建議：本研究發現機構住民的性別、健康狀況、疾病因素和機構治療模式皆為跌倒的危險因素。機構住 民的身體平衡功能為預測跌倒事件發生的重要變項，此發現可作為未來發展長照機構預防跌倒照 護策略之參考。

Background: The risk factors for falls happening to institutionalized elderly have been clearly explored. However, studies of the relation between the elderly residents’ balance function and fall events occurring in long-term care facilities are still limited. Objective: This study aimed to explore the balance function of elderly residents in a long-term care facility, the risk factors for the residents’ falls, as well as the relation between the balance function and the risk factors for the falls. Methods: This study was based on a prospective research design. Data were collected in a purposive sampling method from a long-term care facility/institution in Taipei. A total of 133 residents participated in this study (N = 133). Instruments for data collection included the Barthel index, Tinetti performance oriented mobility assessment scale, Mini-mental state examination, geriatric depression scale, Comell scale for depression in dementia, neuropsychatric inventory, and Hendrich II fall risk assessment tools and fall events. The data were analyzed with the use of hierarchical regression models and logistic regression models offered in SPSS 20.0 software package. Results: Firstly, the results of hierarchical regression analysis showed that gender (ß = .80, p <.001), balance function (ß = .10, p <.05), cognitive function (ß = .90, p <.001), as well as behavioral and psychological symptoms (ß = .09, p <.05) could predict the falls and explain 47% of the variance. Secondly, during one year’s follow-up in this research, 28 fall events happened and the results of logistic regression analysis showed that in terms of balance function, the action of standing up and the action of sitting down were important risk factors for the falls. Conclusion and suggestions: The results of this study revealed that the risk factors for falls happening to the institutional residents included gender, health status, disease factors, and treatment models provided in the facility, and that the residents’ balance function was an important variable to predict the fall events. It is suggested that the findings of this study can be references to the development of fall prevention strategies in long-term care facilities.