Reliability and Validity of the Tamil Version of the Borg Scale of Rate of Perceived Exertion

Abstract

Background: The BORG Scale of rate of perceived exertion (RPE) is a valid, reliable, and widely used tool. Yet, there is no Tamil (Indian language) version of this tool. Hence, this study was conducted to translate, to find out face validity, and criterion validity of the Tamil version of Borg’s RPE, and to find out the reliability analysis of the Tamil version of Borg’s RPE. Methods: The original English version of the BORG scale of RPE was cross-culturally adapted to the Tamil version following the American Association of Orthopedic Surgeons’ guidelines. The process involved forward and back-translations, review by an expert committee, pre-testing, and cognitive debriefing with 60 subjects to produce the Tamil version of the BORG scale. Criterion validity was demonstrated by the correlation of the Tamil version of Borg’s RPE with VAS (dyspnea on exertion) at the same time, for 60 participants (18-25 years). As exertion is inducible with the Six Minute Walk Test, 60 normal subjects were made to do the six-minute walk test and the level of exertion was assessed using both the Tamil version of Borg’s RPE and VAS (dyspnea on exertion). The test-retest reliability was determined by administering the Six Minute Walk Test to 60 participants after an interval of one week. Results: The test-retest reliability was excellent ICC= 0.720(0.572- 0.823). Internal Consistency was found to be good (Cronbach’s alpha=0.837). The Tamil version of Borg’s scale of RPE showed significant correlation by comparing against VAS(dyspnea on exertion) and Spearman’s rho (0.849), p-value <0.001. The statistical analysis shows there is a relatively high positive correlation, test-retest reliability, with excellent agreement and an overall R-value of 0.812. Conclusion: The Borg’s scale translation and adaptation processes were successful. The Tamil translation of the Borg scale of RPE was found to have a relatively high positive correlation. The test- retest reliability showed excellent agreement, with an overall r value of 0.812 high positive correlation.