[A proposal to adapt the voice handicap index to the singing voice]

The voice handicap index by Jacobson et al doesn't render the difficulties experienced by singers. The aim of this study is to adapt the V.H.I. to the singing voice and evaluate it's reliability and coherence. In order to obtain this first scale, we went through three stages. MATERIAL ET METHOD: The first consisted in the modification of the existing items and the creation of new ones. The second was to measure the stability by a test - retest and the coherence of the adapted and the new items. This was done on groups of dysodic singers and professional voice users. The third stage consisted in evaluating the reliability and the coherence of 31 items on a homogeneous population of 132 classical singers with and without vocal impairment. The subjects were divided in three groups. The first group was composed of 95 normophonic singers, the second included 37 dysphonic singers and the third group was the control group, composed of 20 non singers without vocal complaints.

The statistical results show a good stability of the test - retest (ICC=0.949; Rho: 0.912; Bland et Altman [-010.19, +13.23]). The internal coherence of each sub scale (Cronbach's alpha) is good (>0.70). The correlations between the three sub scales are moderate between the functional and the emotional sub scale: 0.58; between the emotional and the physical: 0.64; and between the functional and the physical: 0.78). Each sub scale is measuring a specific dimension; still they are homogeneous between one another. Four variables have a significative influence on the results: the pathology (p<0.001), being a professional singer (p<0.001), the repetition (p=0.003), being a chorister (p=0.002).

The VH.I adapted to the singers, appears to be reliable, valid and adapted to the population of dysodic classical singers.