Are the Acoustic Measurements Reliable in the Assessment of Voice Quality? A Methodological Prospective Study

Acoustic parameters are widely used as voice quality therapeutic outcomes in many laryngological diseases. The aim of this study is to explore the impact of changes in the nature and duration of the analyzed time interval and the vowel choice on the significance of the acoustic measurements used as therapeutic outcomes in two different diseases.

A prospective case series.

From September 2013 to January 2018, patients with laryngopharyngeal reflux (LPR) disease were recruited and treated with pantoprazole, diet, and behavioral changes for 3 months. The reflux symptom index and reflux finding score were used for both diagnosis and assessment of treatment effectiveness. Simultaneously, patients with early idiopathic Parkinson's disease (IPD) were enrolled and benefited from a levodopa challenge test. An Iowa Oral Performance Instrument was used for objective outcomes in the assessment of levodopa effectiveness on muscular strength of IPD patients. Acoustic measurements were performed in both groups pre- and postmedication intake at different time intervals, including the "most stable" time intervals of 1 second, 2 seconds, 3 seconds, 4 seconds, and 5 seconds and a 1 second-time interval positioned at mid-production. We also measured acoustic parameters on the entire signal of three vowels and on the signal of each vowel being taken separately.

A total of 80 LPR and 19 IPD patients met our inclusion criteria and completed the study protocol. LPR and IPD patients had significant clinical improvements throughout treatment, according to reflux symptom index, reflux finding score, and Iowa Oral Performance Instrument scores. The acoustic analysis revealed that acoustic parameters significantly improved from pre- to post-treatment and varied across methods used for measurement. The duration and position of the analyzed time interval in the production and the vowel on which the acoustic measures were made yielded considerable differences in the results.

Depending on the time interval over which the acoustic parameters are measured, the clinically demonstrated effect of the medication may or may not be statistically demonstrated irrespective of the disease. According to the results of this study and regarding the lack of standardization of acoustic measurement methods, a line of thought is proposed to bypass the interval selection problem.