Acoustic measures of essential vocal tremor compared to spasmodic dysphonia with co-occurring tremor

The purpose of this study was to compare acoustic measures hypothesized to measure vocal tremor and voice quality regularity differences between individuals with essential vocal tremor (EVT) and spasmodic dysphonia with co-occurring vocal tremor (SDVT). We hypothesized that those with SDVT would show greater irregularity across acoustic measures during sustained phonation than those with EVT. Individuals with a diagnosis of EVT and SDVT were identified from the Utah Health Voice Disorders Center medical records. Two speech-language pathologists with at least 5 years of experience with neurogenic voice disorders independently classified individuals as EVT or SDVT based on acoustic and endoscopic video recordings. Twenty-nine patients met inclusion/exclusion criteria (EVT = 22, SDVT = 7). The following acoustic measures were compared between study groups: average and standard deviation (SD) of fundamental frequency (fo) and sound pressure level (SPL) rate and extent, average and SD of fo and SPL modulation cycle duration, and the average and SD of cepstral peak prominence (CPP). Statistical comparisons between study groups were completed using the Wilcoxon rank-sum test and the Fisher's Exact Test. A significant difference between those classified as EVT (n=22) and SDVT (n=7) occurred (p<.01) with 11 of those identified from medical records with a diagnosis of SDVT being re-assigned to EVT by expert raters. A significant difference in females and males occurred across study groups (p<.003). More females than males represented the iv EVT group (18:4) compared to more males than females in the SDVT group (6:1). None of the acoustic measures demonstrated statistical differences between study groups. Population size estimates indicated that approximately 70 individuals would be needed in each study group to achieve significant differences between the majority of individual acoustic measures using an alpha of .05 and power level of .80. Outcomes of this study showed that the majority of individuals originally diagnosed with SDVT were re-assigned to the EVT study group (11/17) by expert listeners blinded to condition based on auditory-perceptual and nasoendoscopic recordings. Population size estimates from acoustic measure outcomes provided guidance regarding necessary group sizes before individual acoustic measure differences between groups would be shown.