Objective: To propose a short instrument for the screening of dysphonia in the Brazilian population through the investigation of traditional voice self-assessment instrument items.
Methods: We analyzed the medical records of 139 individuals with an average age of 37.4 years and a minimum and maximum age of 18 and 77 years, respectively. The participants were classified as dysphonic (D) or non-dysphonic (ND) according to an analysis of the combination of vocal complaints and laryngological reports. Responses to the items of the following self-assessment instruments were collected: the Questionário de Qualidade de Vida em Voz - QVV (Voice-Related Quality of Life - V-RQOL), the Índice de Desvantagem Vocal - IDV (Voice Handicap Index - VHI) and the Escala de Sintomas Vocais - ESV (Voice Symptom Scale - VoiSS). These items were analyzed regarding their predictive capacities for dysphonia through logistic regression models.
Results: The model containing items of the QVV was not observed to be valid. The model for the IDV produced a set of three items (10, 13, and 14), and the ESV model showed two items (4 and 20) to be significant. A Global model combining the previous models shows that items I feel as though I have to strain to produce voice from the IDV and "Is your voice hoarse?" from the ESV are the most significant in the classification of the presence of dysphonia. This decision-making model was considered the most efficient to identify the dysphonia, with the highest level of accuracy compared to the other models investigated (83.4%).
Conclusion: Dysphonia screening can be performed using a simple, rapid protocol with a high-efficiency index that includes two items taken from traditional voice self-assessment instruments.
Keywords: Clinical protocols; Dysphonia; Outcome assessment; Regression analysis; Self-assessment; Voice.
Copyright © 2021 The Voice Foundation. Published by Elsevier Inc. All rights reserved.