Objective: To discuss the literature for collection and analysis of speech data for patients with PD in order to create “best practice” guidelines.

Background: Speech can be affected in the majority of patients with PD. There has been an increased interest in measuring the effect of pharmacological, surgical and behavioural treatments on speech. While speech and voice researchers have consented on standardized clinical protocols (e.g. CAPE-V), these have not been used for patients with PD. There are unique aspects of parkinsonian speech and voice production that have to be considered for reliable and valid speech data.

Method: We undertook a literature review using the terms “speech” “voice” and “Parkinson’s disease” in Pubmed and identified 1,768 articles from 1931 to 2018. Critical review considered issues of “who, what, how and why” speech data was collected and analysed, in particular the recording setting and frequency; the tasks most appropriate for patients with PD; the variability of speech in PD; the need for multiple samples and the effect of cueing. Issues surrounding speech analysis involved most appropriate acoustic and perceptual measures; the use of participation outcomes; the use of visual analogue scales and the validity of patient reported outcome measures.

Results: The first article was by E Stolkind in 1931 who described perceptively “Six cases of parkinsonism, with investigation of speech defects and description of a graphic method of making speech records”. Neurosurgeons were the first to describe the effects of their procedures on speech. Neurologists rate speech mainly using the UPDRS-III item 18. Speech researchers and clinicians use a variety of scales ranging from sustained phonation to reading phrases, to connected speech. Due to the specific PD motor symptoms of tremor, rigidity, hypokinesia and dyskinesias the majority of the studies suggest a constant microphone-to-mouth distance; fatigue can limit the length of the speech examination; the effect of emotional load on speech in PD requires careful choice of topics. There is no agreement on the most appropriate scale or acoustic marker for speech analysis.

Conclusion: There is an imperative for informed consensus on speech data collection and analysis in PD with a careful consideration of disease specific issues and the knowledge already accrued.

« Back to 2019 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/best-practice-guidelines-for-the-collection-and-analysis-of-speech-data-in-patients-with-parkinsons-disease-pd/