The speech intelligibility of children with dysarthria and cerebral palsy has been observed to increase following therapy focussing on respiration and phonation. Aims: To determine if speech intelligibility change following intervention is associated with change in acoustic measures of voice. Methods and Procedures: We recorded 16 young people with cerebral palsy and dysarthria (9 girls; mean age 14 years, SD 2; 9 spastic type, 2 dyskinetic, 4 mixed; 1 Worster Drought) producing speech in two

more »

... nditions (single words, connected speech) twice before and twice after therapy focusing on respiration, phonation and rate. In both single word and connected speech we measured vocal intensity (RMS), period-to-period variability (Shimmer APQ, Jitter RAP and PPQ) and harmonics to noise ratio (HNR). In connected speech we also measured mean fundamental frequency, utterance duration in seconds and speech and articulation rate (syllables per second with and without pauses respectively). All acoustic measures were made using Praat. Intelligibility was calculated in previous research. Outcomes & Results: In single words statistically significant but very small reductions were observed in period-to-period variability following therapy: Shimmer APQ -0.15 (95% CI -0.21 to -0.09); Jitter RAP -0.08 (95% CI -0.14 to -0.01); Jitter PPQ -0.08 (95% CI -0.15 to -0.01). No changes in period-to-period perturbation across phrases in connected speech were detected. However, changes in connected speech were observed in phrase length, rate and intensity. Following therapy, mean utterance duration increased by 1.11 seconds (95% CI 0.37 to 1.86) when measured with pauses and by 1.13 seconds (95% CI 0.40 to 1.85) when measured without pauses. Articulation rate increased by 0.07 syllables per second (95% CI 0.02 to 0.13); speech rate increased by 0.06 syllables per second (95% CI <0.01 to 0.12); and intensity increased by 0.03 Pascals (95% CI 0.02 to 0.04). There was a gradual reduction in mean fundamental frequency across all time points (-11.85 Hz, 95% CI -19.84 to -3.86). Only increases in the intensity of single words (0.37 Pascals, 95% CI 0.10 to 0.65) and reductions in fundamental frequency (-0.11 Hz, 95% CI -0.21 to -0.02) in connected speech were associated with gains in intelligibility. Conclusions & Implications: Mean reductions in impairment in vocal function following therapy observed were small and most are unlikely to be clinically significant. Changes in vocal control did not explain improved intelligibility. What is already known on this subject: Motor disorders of cerebral palsy affect control and coordination of movements for respiration, phonation, resonance, articulation and prosody. Therapy focussing on respiration, phonation and rate has been associated with increases in speech intelligibility for children with cerebral palsy and dysarthria. What this study adds: Therapy focussing on respiration, phonation and rate may lead to a stronger vocal signal and increased coordination of speech movements. However, these acoustic changes are small and are unlikely to be perceived by listeners. Vocal changes have not been found to account for observed increases in intelligibility. Clinical implications of this study: Intervention targeting respiration, phonation and rate could be offered to children with dysarthria and cerebral palsy who use speech as their main means of communication but whose intelligibility is limited. Therapy outcome should be measured at the level of activity, that is, intelligibility of speech.