背景與目的:台灣人口快速老化，跌倒盛行率與死亡率隨年齡增加而提升，造成龐大的醫療支出與社會負擔。文獻指出具跌倒風險者其未來…年跌倒發生率高達45-68%。本研究探討多面向介入對於不同跌倒風險老人之跌倒預防成效。方法:自台北、台南社區及醫院門(急)診，篩檢具跌倒風險之老年人參與多面向防跌介入，依據跌倒風險等級分層隨機分為介入組及對照組。介入組接受為期八週、每週一次的小團體運動指導課程、防跌衛教課程、居家環境改善建議及用藥、視力轉介;對照組財給予防跌衛教及居家環境改善手冊。身體功能成效評量於八週介入前後執行，包含問卷及跌倒風險評估(含視差、肌力、反應時間、平衡和本體戚覺)等檢測，且每月電訪追蹤一年內跌倒情形。使用Kaplan-Meier存活分析比較兩組年內累積跌倒發生率，並採用重複量數二因子變異數分析比較兩組身體功能變化(α-level=，(5)，所有資料均採用意向分析(intention-to treat analysis )處理。結果:共蒐集有效樣本411位、平均年齡77±7歲。介入組與對照組之一年累積跌倒發生率無顯著差異皆為29%。以跌倒風險進行次族群分析，發現兩組之跌倒機率在各風險等級亦無顯著差異:低風險( 19%: 27%， p=.358)、中風險(40%: 24%， p=.078)、高風險( 28%: 35%， p=.327 )。介入絕追蹤期間的跌倒機率由過去年的48.1%顯著降低為28.6%(p<.OOI ) ;間對照組白過去一年的36.3%微幅降低為28.9%(p=.101 )。身體功能結果顯示:不論是低、中、高跌倒風險者經過八揖介入後兩組在多數項居都有明顯進步(p<.05)，尤其是中、高風險族群;但介入與對照時組之進步幅度大多未里明顯組閣差異。結論:本研究介入與對照絕之一年累積跌倒發生率皆為29% ;但介入組追蹤一年的跌倒機率由過去的48.1%顯著降低為28.69毛，顯示多面向介入計畫仍有潛力可降低老年人跌個機率。具跌個風險老人可避攝跌倒風險篩檢、衛教諮詢建議與每月電話追蹤提醒，改善其跌倒機率及身體功能，但若要突顯兩組差異可能須增加運動介入之劑量、強度或頻率。(物理治療2011;36(4):3的“330)

Background: Falls are the leading cause of morbidity and mortality among people aged 65 years and older、resulting in marked costs and social burden. The purpose of this study was to evaluate the effect of multifaceted fall prevention program on community-dwelling elderly of ditferent tàll risks. Methods: Community-dwel1ing elderly aged 65 and older who met our fall screening criteria were recruited to participate in a stratified randomized controlled fallprevention trial. Multifaceted intervention included exercise program令fall prevention education courses， recommendations for home hazards modification， as well as referrals for medication or vision check-up. The control group got recommendation and referrals without direct exercise intervention. Primary outcome was incidence of falls in 12 months. Secondary outcomes were PPA battery (including vision、muscular strength， reaction time， balance and proprioception)、 timed up-anιgo (TUG)， Taiwan version oflntemational Physical Activity Questionnaire (IPAQ)，EuroQoL-5D、and Fall EfficacyScale (FES). Results: This study recruited 411 older participants with mean age of 77±7 years from hospital outpatient c1inicsor communities. The fall incidence within I-year for IG (29%) and CG (29%) showed no significant ditì'erence. Subgroup analyses stratified by PPA risk-level found that there was no significant difference in fall incidences of both groups among any stratum of fall-risk: low risk (IG:CG= 19%: 27%,p=.358)， moderate risk (IG: CG=40%﹒24%， p=.078)， and marked risk (IG:CG=28%: 35%， p=.327). The fall incidence during the I-year follow-up period significantly decreased from 48.1 % to 28.6% (p<.001)in IG， while no significant decrease in CG (36.3%→28.9%， p=.101). Both IG and CG improved significantly in most secondary outcomes after intervention， especial1y for those with moderate or marked fall risk; however， it revealed no significant within-group difference. Conclusion: The findings of 29% fal1 incidence in both groups during the l-year fol1ow-up period along with the significant decrease of faJ1 incidence from 48.1 % to 28.6% in IG might suggest the potential role of the multifaceted fal1-prevention program. Interventions including recom-mendation and referrals seemed to take some effects on improving physical functions of the elderly， especial1y to the moderate or marked risk group. Whether higher-intensity of longer du-ration exercise program is required to significantly decrease the fal1 incidence warrants further study. (FJPT 2011 ;36( 4 ):315-330)