Objective : The purpose of this study is to simplify the Postural Assessment Scale for Stroke Patients - 4Level (PASS-4L), and Berg Balance Scale - 5 Level (BBS-5L), into a 3 Level Assessment Scale. The data obtained from the simplified 3 Level Assessment Scale is then compared with those from both PASS-4L and BBS-5L in order to ascertain their reliability as well as validity characteristics.
Methods : This study used data from 62 patients in total. Data recorded in the form of PASS-4L(0-1-2-3)was converted into adjusted PASS-3L (ie, 0-1.5-3) in order to form PASS-3 Levels. Similarly, data recorded in the form of BBS-5L (0-1-2-3-4) was converted into adjusted PASS-3L (ie, 0-2-4). To examine the test-retest reliability, the PASS-3L and BBS-3L were administrated to each participant by rater A in 2 testing sessions 3 days apart. To allow the evaluation of inter-rater reliability, the score was rated simultaneously by raters B (PASS, BBS-3L) and C (BBS, PASS-3L) in the latter testing session. The reliability indexes (intra-class correlation coefficient, ICC3,1) of both simplified PASS-3L & BBS 3L were used for the intra-rater and inter-rater reliability, and Cronbach's αwas obtained. Standard Error Measurement (SEM) and Smallest Real Difference (SRD) methods were used to obtain the reliability index of measurement errors. We also made a correlation assessment using a Pearson's correlation coefficient to figure ou the concurrent validity between PASS (PASS-3L) and BSS (BSS-3L) as well as the convergent validity between Modified Barthel Index (MBI) and Fugl Meyer-Motor (FM-M) Scale.
Results : The PASS-3L and BBS-3L showed high intra-rater (ICC3,1 = .96, .96, respectively) and interrater reliabilities (.95, .94, respectively). The absolute reliability of the PASS-3L and BBS-3L released more accurate measurements (SEM = .99, 1.55; SRD = 1.74, 4.30, respectively). The internal consistencies of PASS-3L and BBS-3L are expressed using Cronbach's αcoefficient as .77 and .85, respectively. The PASS-3L and BBS-3L showed high concurrent validity (r=.75-.93, p<.01) and convergent validity with the MBI and FM-M scores (r=.42-.79, p<.01).
Conclusion : The PASS-3L BBS-3L showed high reliability and validity, and thus it can be stated that the proposed simplified 3-Level scale will be a useful clinical tool for evaluating balance in stroke patients.