Barthel Index in a Middle-East Country: Translation, Validity and Reliability

Abstract

Background: In Iran, a Middle-East country, no disability scale has been translated and validated for use in stroke clinical trials. This study was designed to translate the Barthel Index and make its Persian translated form valid and reliable. Methods: All items of the Barthel Index were translated into Persian. Also, the Modified Rankin Scale (MRS) was also translated to Persian. Telephone interview was used as the method of data acquisition. Two interviewers were chosen for this means in order to accelerate data gathering and measure interrater agreement. Samples were selected from Isfahan Cardiovascular Research Center’s Stroke Registry Unit, a WHO collaborating center in the center of Iran. All the patients were registered as stroke or intracerebral hemorrhage (ICH). These patients were inhabitants of Isfahan Province who had suffered from stroke or ICH between 12 and 24 months before data acquisition. Chronbach’s alpha, test-retest reliability, concurrent validity with MRS, interrater agreement and item analysis were done for the translated questionnaire. Results: Translated questionnaires were filled by interview from 459 stroke patients. Their mean age was 68.11 ± 11.59 years. 243 of them were male (52.9%). Cronbach’s alpha was 0.9354. Spearman’s correlation coefficient between translated Barthel Index scores and MRS scores was –0.912. Spearman’s correlation coefficient between 2 scores, to determine test-retest reliability was 0.989. Concordance correlation to determine interrater agreement was 0.994. All corrected item–total correlations were greater than 0.5. Conclusions: The Persian translated version of the Barthel Index is a reliable and valid questionnaire for use in stroke clinical trials.