Development and Validation of two Barthel Index-based Supplementary Scales for Patients with Stroke

• Main Authors: Ya-Chen Lee, 李雅珍
• Other Authors: I-Ping Hsueh
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/5uw3sy

博士 === 國立臺灣大學 === 職能治療研究所 === 103 === Background: Stroke is the most common cause of disability or dependence in activities of daily living (ADL) among the elderly. Three distinct constructs of ADL measures have been proposed: actual performance, self-perceived difficulty, and ability. Assessing these three ADL constructs can be useful for identifying disability in performing ADL and thus, also for intervention planning. However, according to literatures, there is still no ADL measure that assesses all three ADL constructs simultaneously which might overlook the relationships between constructs that affect patients’ ADL functions. The Barthel Index (BI) has been widely used in both clinical and research settings due to its advantages: it is quick and easy to administer, has sound psychometric properties, and is economical and practical to use. The BI is an appropriate instrument to be used as a basis to further develop other ADL constructs to comprehensively assess ADL functions. Objective: This dissertation aims to: (1) develop two BI-based Supplementary Scales (BI-SS), namely the Self-perceived Difficulty Scale and Ability Scale; (2) examine the construct validity of the BI-SS; (3) examine the reliability of the BI-SS; (4) investigate the responsiveness of the BI-SS in inpatients with stroke. Method: This dissertation consists of four studies. Study 1 was to develop the BI-SS. The development of the BI-SS had two stages. Stage one was to consult with experts to develop the draft version of the BI-SS. Stage two was to conduct pilot studies to examine the clarity of the administrative instructions and the feasibility of administration of the BI-SS. Study 2 was to examine the construct validity of the BI-SS. Patients undergoing outpatient or inpatient rehabilitation were recruited from the Department of Physical Medicine and Rehabilitation (PM&R). The construct validity of the BI-SS was investigated using the model of monotone homogeneity of Mokken scale analysis and analyzing associations between scales. Study 3 was to investigate the test-retest reliability of the Self-perceived Difficulty Scale and the Ability Scale. One convenience samples of outpatients with chronic stroke were recruited from the PM& R. Intra-class correlation coefficients (ICC2, 1) were calculated to examine the extent of agreement between repeated assessments. The minimal detectable change (MDC) was also calculated to determine whether the change score of an individual patient is real at the 95% confidence level. Study 4 was to examine: (1) The group-level responsiveness, including the internal responsiveness and external responsiveness. Three indices, standardized effect size (SES), standardized response mean (SRM), and paired t test, were used to examine the internal responsiveness. The Pearson correlation coefficient (r) was used to examine the association between the changes in scores on both scales and the changes in scores on the BI (treat as an external criterion). (2) The individual-level responsiveness (the number of patients whose change score on the BI-SS exceeded the respectively MDC). Results: Study 1: Based on the results of expert panel discussion and pilot testing, the Self-perceived Difficulty Scale consisted of 10 items with the total score of 20. The Self-perceived Difficulty Scale was assessed with a face to face interview format. The Ability Scale had 8 items with total score ranges of 18. The Ability Scale was assessed by observing patients doing a specific ADL task with assessment tools. Study 2: A total of 306 participants participated in this study. The results showed that items in each scale of BI-SS were unidimensional (Hi ≥ 0.3). The unidimensionality of each individual scale were strong (H ≥ 0.5). The scores of both scales were highly correlated with those of the BI (rho=0.78 and 0.90, respectively) and were significantly different from each other (p<0.001). These results indicate that the BI-SS each scale assesses unique construct. Study 3: A total of 84 patients participated in the test-retest reliability of the Self-perceived Difficulty Scale and the Ability Scale. The ICC values for the Self-perceived Difficulty Scale and the Ability Scale were 0.78 and 0.97, respectively. The MDC values were 5.5 and 1.9 points. Study 4: Fifty-seven patients completed both baseline and follow-up assessments. (1) The group-level responsiveness, the internal responsiveness of the BI-SS was moderate to large (0.78-1.56). For the external responsiveness, the change in score of the both scales had weak to moderation association with that of the BI (r=0.23 and 0.61, respectively). (2) The individual-level responsiveness, the numbers of patients having change scores exceeding the MDC of the both scales were 25 and 48, respectively. Discussion and Conclusion: The BI-SS was developed from the BI as supplementary scales in order to comprehensively assess ADL functions. The BI-SS had clear and understandable instructions and overall sufficient construct validity in patients with stroke. The results of reliability studies showed that the Ability Scale had appropriated test-retest reliability and reasonable responsiveness in patients with stroke undergoing inpatient rehabilitation. Further studies are needed to minimize the random measurement error of the Self-perceived Difficulty Scale. These results provide empirical evidence of the BI-SS in assessing stroke patients’ level of difficulty and ability in performing ADL tasks.