Objective measurement of speech has shown promising results to monitor disease state in Multiple Sclerosis (MS). Yet, it is not clear if changes in speech can be detected before overt dysarthria.
In this study, we characterize the relationship between disease severity and objective speech metrics exclusively in people with no perceivable dysarthria.
An acoustic composite score was calculated using regression modelling of speech data from 119 people with MS (pwMS, 75% female), irrespective of dysarthria presence. That score was then tested in pwMS without dysarthria, as determined by blinded perceptual rating, for correlations with the Expanded Disability Status Scale (EDSS), brain volume and lesion load from magnetic resonance imaging, and quality of life scores from the Multiple Sclerosis Impact Scale (MSIS-29) .
PwMS without dysarthria (n=77) were more likely to be female (82% vs 62%, p=0.017), were on average 5.7 years younger (age mean ± standard deviation 53.5±11.4, p=0.009), had MS for 2.5 years shorter (11±8.5 years, p=0.034) and scored EDSS 1.7 step lower (2.7±1.9, p<0.001) than pwMS with dysarthria (n=42). The acoustic composite score correlated with EDSS scores (r=0.45, p<0.001) and quality of life (r=0.4, p=0.01) in pwMS without perceivable dysarthria, but not with brain volume or lesion load.
Acoustic analysis offers a valuable insight into the subclinical development of speech impairment in MS. These results highlight the potential of automated analysis of speech to assist in monitoring disease progression and treatment response.