老人跌倒預防介入計畫在預防跌倒相關傷害上之效果：探索性統合分析之應用

目的：在老人許多的健康問題中，跌倒是影響老年人健康及獨立生活之重要危險因子，過去關於跌倒預防介入計畫的研究已有證據指出實施多元計畫較單一計畫有效，但相關研究皆偏重在於降低跌倒發生率，忽略了實施老人跌倒預防計畫後是否有效降低跌倒發生後的相關傷害。本研究期望評值跌倒預防介入計畫對於預防跌倒後相關傷害之成效，並探討預防介入計畫的特徵變項對效果量的影響。方法：本研究運用探索性統合分析，自西元1996年至西元2016年以英文發表之期刊為研究對象，計算跌倒預防介入計畫的綜合效果及在相關傷害上的預防效果。結果：確認跌倒預防介入計畫對預防跌倒相關傷害有顯著效果(Effect size=0.855;95％ CI=0.753~0.971;Z=-2.417;p=0.016)，但對預防骨折卻無顯著效果（Effect size=0.802;95％CI=0.627~1.025;Z=-1.766; p=0.077)。其預防介入計畫的研究特徵中，高品質的研究有顯著效果量；樣本特徵中：男女皆含、跌倒史大於15%、跌倒傷害史小於30%、獨居大於50%的研究有顯著的效果量；介入特徵中：研究地點為社區、介入時數≦11小時及12~100小時、多元介入方式、有醫療專業背景介入皆有顯著效果量。特徵項目與相關傷害之相關性可提供給後續決策者、臨床工作者及研究者進行介入時的參考方向，並可減少相同研究的複製及節省介入規劃投入的成本、時間與精力。

關鍵字

跌倒預防介入計畫；統合分析；相關傷害；效果量

並列摘要

Purpose: Among various health issues, falls have become a critical risk factor that influence health and independent living for elderly. Previous studies suggested that multidisciplinary prevention programs be more effective than single dimensional ones. Reduction of fall incidence was generally emphasised in earlier research, the effectiveness of reducing fall-related injuries after the implementation of elderly fall-prevention intervention, however, was ignored. The objective of this study was to evaluate the effectiveness of fall-prevention intervention on related injuries cuased after falling, then to determine characteristics variables effects of prevention program on effect size. Method: An exploratory meta-analysis was applied on the English-published paper reported during 1996-2016. Summary effect of fall-prevention programs and preventive effect of on related injuries were computed. Results: This study confirmed that fall-prevention intervention programs had significant effect on related injuries caused after falling(Effect size=0.855;95％ CI=0.753~0.971;Z=-2.417;p=0.016), but no effect on fracture（Effect size=0.802;95％CI=0.627~1.025;Z=-1.766; p=0.077). Regarding the research characteristic of the fall prevention programs, the papers with high quality obtained significant effect size; in terms of sample characteristic, research with both gender included, a falling history larger than 15%, traumatizing falling history less than 30%, and independent living rate more that 50% were found to have significant effect size; concerning intervention characteristic, studies conducted in the communities, with intervention duration less than or equal to 11 hours or between 12 to 100 hours, multidisciplinary intervention approaches, and with medical background all secured a significant effect size. The relationship between characteristic variables and injuries followed by falling were found and could be served as reference for future intervention. Replication of similar studies could be reduced and cost and effort put into intervention planning could be saved.