Improving diagnostic acoustic measurements of vocal dysfunction in voice clinics

Current assessment guidelines recommend computer-assisted instrumental acoustic measurements for a comprehensive examination of voice patients. The most widely recommended parameters jitter, shimmer, Cepstral Peak Prominence (CPP) and Smoothed Cepstral Peak Prominence (CPPS) are used to objectively describe vocal dysfunction and treatment effects. However, under current standards to phonate at “comfortable loudness”, individual differences in speaking characteristics between healthy adults influence assessment results. Specifically, habitual speaking voice sound pressure level (voice SPL) and pitch (fundamental frequency, fo) significantly affect jitter, shimmer and CPPS in vocally healthy adults in vowel phonation. To date it is unclear, if similar effects apply to children and adults with voice disorders, or analysis samples excised from speech. Further, natural ageing and voice training effects may interfere with measurement results. Therefore, main aims of the here presented works were to a) investigate the effects of speaking voice SPL, gender, age, height and weight on jitter and shimmer in prepubescent children, b) assess the effects of voice SPL on jitter and shimmer in voice disordered adults, c) investigate the combined effects of speaking voice SPL and fo variations on CPPS in vowel phonations and samples from speech in vocally healthy and voice disordered adults. First, in a single cohort study, 68 vocally healthy prepubescent children (39 female/29 male) between 5;0 and 9;11 years were examined. All children sustained the vowel /a/ at subjectively ‘‘medium’’ (modelling clinical practice), ‘‘soft’’ and ‘‘loud’’ voice intensity levels, and a prescribed intensity level of “>80 dBA’’. Gender, age, height and weight had no significant effects on speaking voice SPL, fo, jitter and shimmer. However, there were significant jitter and shimmer differences between all voice tasks (p 80 dBA’’. Thereafter, in a matched case-control study, 58 vocally healthy and 58 voice disordered women matched according to approximate age and occupation were investigated. Vowel phonations were recorded at “soft”, “comfortable” and “loud” levels. In both patient and normative control groups, increasing voice SPL correlated significantly with decreased jitter (patients r2=0.27, controls r2=0.18) and shimmer (patients r2=0.35, controls r2=0.46, p.05). Finally, in a retrospective cross-sectional study, the effects of prosody related SPL and fo variations on CPP and CPPS in speech from 27 voice disordered Brazilian Portuguese adults (19 women, 8 men) were investigated. Speaking voice SPL as single factor and combined with fo had a highly significant effect (p≤ .001), while fo alone had no significant impact on both CPP and CPPS (p≥ .77). Voice fo, SPL, CPP, and CPPS of the first vowel were all significantly lower than of the last excised vowel from the recorded sentences (p≤ .03). The present studies in children and adults show that habitual differences in speaking voice intensity alone and in combination with fundamental frequency significantly influence standard diagnostic acoustic measurements of jitter, shimmer and CPPS. In all studies, jitter, shimmer and CPPS unequivocally showed improved values in louder phonations of vowels and speech in both vocally healthy and voice disordered individuals. Further, in patients with voice disorders, higher SPL alone and combined with higher fo were related to better CPP and CPPS in vowels from connected speech samples. Also, a later vowel position in the text was associated with systematically higher speaking SPL and CPPS. Thus, acoustic measurements of human voice quality are substantially influenced by natural prosodic variations of speaking voice SPL alone and combined with fundamental frequency, and the speaking task. Currently, these factors are not sufficiently controlled for in clinical voice assessments. This may lead to a significant underestimation of vocal pathology in naturally louder speaking individuals, and a misjudgment of treatment effects. Therefore, it is recommended that the clinical assessment of voice control for variations in voice SPL. Such control may be performed by applying SPL-corrected values for jitter, shimmer and CPPS, or by taking care to elicit similar voice SPL levels throughout a patient’s stage of treatment. Other potentially confounding factors such as voice task type, language structure, gender, age, and voice training status should be investigated in the future on patients across a larger spectrum of voice disorders