Related Experiment Video
Updated: Apr 8, 2026
Computerized Adaptive Testing System of Functional Assessment of Stroke
Published on: January 7, 2019
Reliability and validity of a vertical numerical rating scale supplemented with a faces rating scale in measuring
Li-Ling Chuang1,2, Keh-Chung Lin3,4, An-Lun Hsu5
1Department of Physical Therapy and Graduate Institute of Rehabilitation Science, College of Medicine, Chang Gung University, Taoyuan, Taiwan. lchuang2@gmail.com.
Background:
Poststroke fatigue is a persistent and distressing symptom among stroke survivors. In this study, we investigated the reliability and validity of a vertical numerical rating scale supplemented with a faces rating scale (NRS-FRS) in measuring poststroke fatigue.
Methods:
The fatigue intensity of 106 individuals with stroke was measured twice, 1 week apart, using a vertical NRS-FRS to measure test-retest reliability. The intraclass correlation coefficient, a relative reliability index, was calculated to examine the degree of consistency and agreement between the two test occasions. Absolute reliability indices, including the standard error of measurement, minimal detectable change, and Bland-Altman limits of agreement, were used to quantify measurement errors and determine systematic biases of the two test occasions. We also administered the vertical NRS concurrently as a comparator measure for assessing fatigue in 50 consecutive patients with stroke who were recruited later in the study period. The Spearman rank correlation coefficient (ρ) was used to examine the concurrent validity of the NRS-FRS. Discriminant validity was assessed by means of receiver operating characteristic curves, sensitivity, and specificity.
Results:
The intraclass correlation coefficient was 0.95 for the NRS-FRS. The standard error of measurement and the minimal detectable change at the 95 % confidence interval of the NRS-FRS were 0.50 and 1.39, respectively. The Bland-Altman analyses showed no significant systematic bias between the repeated measurements. A narrow range of the limits of agreement was shown on the Bland-Altman plot, indicating the NRS-FRS had high stability and low variation between the two test occasions. The correlations between the NRS-FRS and NRS were good at test (ρ = 0.85) and retest (ρ = 0.84). Compared with the NRS cutoff value of ≥1, sensitivity with the NRS-FRS at test and retest was 94 and 92 % and specificity was 79 and 90 %, respectively.
Conclusions:
This study provides further evidence of the reliability and validity of the NRS-FRS in measuring fatigue intensity in patients with stroke. The NRS-FRS had high sensitivity and specificity. The NRS-FRS may be a reliable and valid measure for clinicians and researchers to assess fatigue and determine whether a real change has occurred in groups and at the individual level of patients with stroke.
