Simple Reading Task May Help Detect Early Disease

Assessing vowel percentage, or duration, during a simple reading task is an effective method to detect changes in speech rhythm at early stages of Parkinson’s disease, according to a small study from Italy.

Notably, such rhythmic differences between early-stage Parkinson’s patients and healthy individuals were not as pronounced during spontaneous speech.

These findings point to vowel percentage in a reading task as a potential marker of Parkinson’s and suggest it can be used to monitor the progression of speech difficulties in these patients.

Larger, long-term studies, including in other languages, are needed to confirm these findings, the researchers noted.

The study, “Speech Rhythm Variation in Early-Stage Parkinson’s Disease: A Study on Different Speaking Tasks,” was published in the journal Frontiers in Psychology.

Parkinson’s affects not only the muscles needed for movement, but also those necessary for oral communication, resulting in changes in voice, speech rhythm, and articulation. Articulation refers to the coordinated movements of the lips, tongue, teeth, and lungs to produce speech sounds, both in isolation and within words and sentences.

Since speech changes can occur before motor symptom onset, identifying objective measures of such mild disturbances may help to diagnose the disease early, as well as to monitor progression and treatment response.

However, there is a “lack of a widely accepted and shared methodology for speech data collection,” and uncertainty regarding the best speaking task to detect rhythm changes in Parkinson’s patients, the researchers wrote.

Now, researchers in Italy set out to assess whether two vowel-related acoustic parameters could effectively detect rhythm variation in early-stage Parkinson’s speech and whether these changes were more obvious in a simple reading task, or in a more complex spontaneous speech task.

They analyzed the reading and spontaneous speech samples of 20 adults (12 men and eight women) with early-stage Parkinson’s and a mean age of 63.8 years (range, 41–81), and of 20 age- and sex-matched healthy individuals.

Patients had been living with the diagnosis for less than four years, and 13 had mild-to-moderate speech difficulties, while seven had no speech problems.

The evaluated acoustic parameters included vowel percentage (%V), or vowel duration, and VtoV — the mean interval between two consecutive vowels — that is dependent on articulation rate.

Results showed that people with early-stage Parkinson’s disease had a significantly increased vowel duration than healthy individuals in both speaking tasks, with this difference being more pronounced in the reading task.

These results confirmed that %V is a useful parameter for characterizing early-stage [Parkinson’s] speech, in which vowels are sustained for longer, and the passage to consonants is delayed.

“The alteration of this acoustic parameter even at the initial stages of PD [Parkinson’s disease] may potentially reflect [neuronal] changes that occur very early in the disease course,” the researchers wrote.

“Conversely, the articulation rate, expressed by VtoV mean values, does not seem to unambiguously distinguish [Parkinson’s-associated impaired speech] from healthy speech,” the team added. Particularly, significant group differences in VtoV were detected only in spontaneous speech.

Regarding differences between patients with and without clinically significant speech difficulties, significantly higher VtoV values were detected in monologues of patients with speech problems.

Additional analysis highlighted that %V values in the reading task are more effective at distinguishing Parkinson’s patients from healthy people than either VtoV values or spontaneous speech tasks.

The threshold %V value for discriminating between groups was about 48%, with Parkinson’s patients having vowel percentages always above it and unaffected individuals mostly below it.

These findings suggest that both spontaneous and nonspontaneous speech are “suitable for the assessment of early changes in speech rhythm associated with PD,” but the “reading passage seems to be the task that maximizes the difference” between patients and healthy people in terms of %V, the researchers wrote.

Still, the fact that spontaneous speech data also can be used to detect rhythmic differences in Parkinson’s patients allows “future larger pooling of data from various sources, with the involvement, for example, of low-educated people,” the team added.

“Longitudinal studies on larger PD Italian samples and also on other languages, with different rhythmic patterns, are needed to support and enlarge our observations,” the researchers concluded.