Linguistic Adaptation, Reliability, and Validation of the Turkish Version of the Reflux Symptom Index

Abstract

The reflux symptom index (RSI) has been designed to raise the clinical suspicion of laryngopharyngeal reflux (LPR) in patients presenting with ear, nose, and throat symptoms. To develop a Turkish version of the RSI (Tr-RSI) and to evaluate its internal consistency, reliability, and clinical validity METHODS: The RSI was translated into Turkish. A total of 44 asymptomatic control subjects without LPR and 62 patients with LPR were enrolled into this study. The patients were treated with diet and behavioral modifications combined with 20 mg of rabeprazole twice daily for 3 months. All individuals were asked to fill the Tr-RSI. Participants in the control group were asked to refill the Tr-RSI after 1 week, whereas those in the patient group after treatment. Test-retest reliability, validity, and internal consistency were computed. The Cronbach's alpha value of Tr-RSI was 0.959, indicating excellent internal consistency. The single-measurement intra-class correlation coefficient absolute agreement was 0.948 (with 95% confidence interval of 0.907-0.971), indicating good test-retest reliability. The Tr-RSI scores for total and each item were significantly higher in the LPR patient group than in the control group (P < 0.001). In the LPR patient group, the posttreatment Tr-RSI scores for each item and total were significantly lower than the pretreatment scores (P < 0.001). The Tr-RSI was successfully created. It can be used with strong internal consistency, high test-retest reliability, and optimal clinical validity.