The Validity and Reliability of the Reflux Finding Score

Laryngopharyngeal reflux (LPR) disease is common. The incidence of newly diagnosed cases has increased substantially due to awareness and development of new diagnostic measurements. The reflux finding score (RFS) and reflux symptom index (RSI) are believed to be useful in the assessment process, including after the initiation of therapy. However, many authors have suggested concerns about the reliability and validity of the RFS. OBJECTIVE: To evaluate the validity and reliability of the RFS. METHODS: Ninety-two patients diagnosed with LPR who had undergone 24-hour pH-Impedance tests were included. All patients underwent stroboscopy and 24-Hour pH-Impedance monitoring within thirty days. Fifty-nine patients filled out a RSI prior to stroboscopic exam. The RFS was determined by four blinded observers: one otolaryngology resident, two laryngology fellows, and one laryngologist. Stroboscopic images were reviewed again one year later to assess intrarater reliability. RFS and RSI were correlated with 24-hour pH Impedance testing. RESULTS: The Kappa value between reviewers was 0.479. The percent agreement of the four observers for total RFS was 74.04%.The percent agreement between reviewers for subglottic edema was 78.77%; for ventricular obliteration was 65.55%; for erythema/hyperemia was 69.62%, for vocal fold edema was 68.32%; for diffuse laryngeal edema was 66.86%, for posterior commissure hypertrophy was 73.54%; for granuloma/granulation was 96.80%; for thick endolaryngeal mucus was 72.81%. The intrarater reliability of the four observers for total RFS was 67.5% with an intrarater reliability range of 50%-90%. The intrarater reliability for subglottic edema was 85% with a range of 70%-100%; for ventricular obliteration was 77.50% with a range of 70%-90%; for erythema/hyperemia was 65.00% with a range of 50%-90%; for vocal fold edema was 52.50% with a range of 30%-70%; for diffuse laryngeal edema was 62.50% with a range of 20%-80%; for posterior commissure hypertrophy was 52.50% with a range of 10%-80%; for granuloma/granulation was 100%; for thick endolaryngeal mucus was 55.00% with a range of 10%-90%. There was no correlation between RFS and any parameter of the 24-Hr pH-Impedance Test. RSI had a significant correlation with number of upright events (r value of 0.271, R2 of 0.0733 and P-value of 0.037), total symptoms experienced (r value of 0.0.267, R2 of 0.0715 and P-value of 0.041), and symptom correlation score (r value of -0.297, R2 of 0.0884 and P-value of 0.022). CONCLUSION: Many authors have expressed concerns about the reliability and validity of the RFS. In our study we found a fair/substantial interrater reliability, and a modest intra-rater reliability. We found no correlation between the RFS and 24-Hr pH Impedance testing. This study suggests that the concerns about the validity and reliability of the RFS may be warranted. This widely used clinical score should be interpreted with caution and further research and refinement should be considered