IMPAIRMENTS, BALANCE PERFORMANCE, AND FALLING STATUS IN COMMUNITY-DWELLING STROKE SURVIVORS. Beliz Belgin2; Marianne Beninato2; Patricia E. Sullivan*2 1. Physical Therapy, MGH Institute of Health Professions, Boston, MA; 2. Graduate Programs in Physical Therapy, MGH Institute of Health Professions, Boston, MA PURPOSE: The purpose of this study was to investigate self-report and performance-based measures of impairment and function in community-dwelling stroke survivors, and their relation to falling. Particularly: 1) to describe demographic and falling characteristics, 2) to determine how performance on the Berg Balance Scale (BBS), the Timed Up and Go (TUG), the Falls Efficacy Scale [FES(S)] and other subject impairments and characteristics are related to fall status, and 3) to examine whether performance on the BBS, TUG and FES(S) can be explained by subject characteristics and stroke-specific impairments. BACKGROUNDS/SIGNIFICANCE: Falling is a major complication of stroke during hospital stay and after discharge. The incidence, risk factors and consequences of falls in the elderly and in-patient stroke populations are well documented. However, it is unclear if these findings can be generalized to community-dwelling stroke survivors. Better understanding of the contribution of these various factors to balance and falls in persons post-stroke will enable more focused physical therapy intervention directed towards the goal of improving balance and reducing falls. SUBJECTS: A convenience sample of 50 community-dwelling stroke survivors were interviewed about health and falling status and assessed on impairment and functional measures in a single session. METHODS AND MATERIALS: Self-report measures included Stroke Impact Scale-mood (SIS-mood), balance confidence on FES(S)(Swedish version) and being afraid of falling (Y/N). Performance-based tests assessed motor impairment (Fugl-Meyer Assessment), composite strength (Timed-Sit-to-Stand), vision (field cut/extinction), vibratory sensation (tuning fork), balance performance (BBS) and functional mobility (TUG). ANALYSES: Descriptive analyses were performed on demographic and measurement results. Relative risk, chi-square and t-test analyses were used to associate significant characteristics with fallers and multiple fallers. Multiple regression analyses were used to explore the variables that explained the variance of the BBS, TUG and FES(S). Receiver Operator Characteristic (ROC) curves were generated to examine which variables best identified fallers from nonfallers and multiple fallers from non- and one-time fallers. Threshold values were also determined from the ROC curves. RESULTS: 50 community dwelling stroke survivors, with a mean age of 59.9 SD 11.9 years (range: 35 to 87 years), and a mean time since stroke of 62.2 SD 62.1 months (range: 3 to 312 months) participated. Subjects included 31 males and 19 females, 32 with right-sided and 18 with left-sided stroke. 20 subjects reported a fall history in the last 6 months (40%), of whom 11 (55%) had fallen more than once. Average scores of BBS (47.4 SD 7.7), TUG (17.6 SD 10.2) and FES(S) (23.7 SD 11.9) indicated a highly functional sample. Fallers had lower SIS-mood scores, less balance confidence and were 2.4 times more likely to be afraid of falling than non-fallers. Multiple fallers had poorer balance performance (BBS), were 5.6 times more likely to be afraid of falling, and 6 times more likely to use an assistive device. Of the self-report and performance-based measures, the FES(S) best differentiated fallers from non-fallers with an area under an ROC curve of .71 and threshold of 17.5. The BBS best differentiated multiple fallers from other subjects with an area under the ROC curve of .72 and a threshold of 52. The variance of the BBS score was explained by composite strength and age (Total R2: 0.52). Strength alone explained 63.6% of the variance of TUG. The variance of balance confidence was explained by knowing somebody who had a fall, alcohol use and composite strength (Total R2: 0.29). Previously reported thresholds of the BBS and TUG did not differentiate fallers from non-fallers. CONCLUSIONS: The self-reported rate of falling was relatively low for community-dwelling stroke survivors compared to other reports, although still high enough to be of major concern. Recovery, utilization of compensatory mechanisms, avoiding risky behavior and education on falls prevention may account for the difference. The ability of balance confidence to differentiate between fallers and non-fallers emphasizes the importance of including confidence rating as a standard part of falls risk assessment. Our finding that the composite strength measure explained balance performance, balance confidence and TUG emphasizes the importance of continued strength training as a component of stroke rehabilitation. Rehabilitating balance even in high functioning persons with stroke may reduce the risk of falls. Prospective studies are needed to further investigate these relationships. FUNDING SOURCE: N/A KEYWORDS: function, stroke, balance, falls, impairments Copyright 2009 by the American Physical Therapy Association. Requests for reprints should be directed to the corresponding author of the article. Educators, students, and other academic customers may receive permission to reprint copyrighted material from Physical Therapy (ISSN 1538-6724) by contacting the Copyright Clearance Center Inc, 222 Rosewood Dr, Danvers, MA 01923. Other types of customers who want permission to reprint should contact the APTA Editorial Office, Attn: Physical Therapy.