Voice disorders and treatment effects in Parkinson’s disease studied with a portable voice accumulator

Abstract

Background: Information about voice use outside the clinical setting has traditionally consisted of subjective data collected via interviews or questionnaires. Quantitative, objective information about voice use in daily life could increase our understanding of the challenges faced by patients outside the controlled clinical environment. There is a high prevalence of hypophonia in individuals with Parkinson’s disease (PD). Objective information regarding voice use would be of value, not the least for this population considering the changes in perception of speech production and cognition found in PD, and may be collected with the use of a portable voice accumulator.

Objectives: The aim of this thesis was to study voice use and communication in daily life in individuals with PD and healthy matched controls using a portable voice accumulator. A further aim was to assess the outcome following biofeedback intervention provided with a portable voice accumulator with the goal to increase voice sound level in daily life for individuals with hypophonia related to PD.

Materials and methods: In study I, the portable voice accumulator VoxLog was used to monitor voice use in daily life regarding voice sound level and phonation ratio in different levels of environmental noise for twenty-one individuals with PD and twenty-one matched healthy controls during a week-long registration period. In study II, treatment outcome following Lee Silverman Voice Treatment (LSVT LOUD®) was assessed in a pseudo-single case study through monitoring of voice use in daily life before, during, and after treatment as well as at follow-up one year after treatment. In study III, the portable voice accumulator VoxLog’s biofeedback function was used to provide biofeedback intervention during one week for six individuals with PD with the goal to increase voice sound level in daily life. Study IV focused on experimentally testing the biofeedback function of the VoxLog to assess how different feedback frequencies can be achieved by adjusting the available settings. Six different biofeedback settings were assessed during semi-structured conversations with twenty participants. In study V, a four-week long biofeedback intervention period was evaluated for eight individuals with PD. Voice use was monitored the week before intervention, during the four-week intervention period, the week after intervention and one week at follow-up, three months post intervention.

Results: Seventy-eight participants were included in the analysis and field registrations were performed for a total of 127 weeks in the project. The week-long registrations of voice use in study I showed that individuals with PD use their voice 50-60% less than their matched healthy controls. The difference increased in situations with loud environmental noise, which in many cases are social situations. A difference in mean voice sound level was found where individuals with PD used a voice sound level 6-8 dB lower than their matched healthy controls. Voice sound level used in daily life differed significantly from the voice sound level during controlled studio registrations for both the male and female groups. Study II showed that the changes in voice sound level after LSVT LOUD® when studied in daily life with a portable voice accumulator reflect earlier reported changes registered during controlled studio recordings. In study III, a significant increase in voice sound level in relation to the environmental noise was found during the intervention week but the effect did not remain when the biofeedback was removed. Based on the findings in study IV, a systematic procedure to configure the biofeedback settings in the VoxLog was recommended to achieve a low feedback frequency, known to facilitate retention after training in the motor learning literature. The results from study V showed an increase in voice sound level in relation to the environmental noise following a four-week long biofeedback intervention and the effect was still present at the three-month follow-up. Results from screening with the Montreal Cognitive Assessment showed that individuals who scored above the cut-off for mild cognitive impairment had more positive and lasting outcomes compared to the group who scored below the cut-off for mild cognitive impairment.

Conclusions: Objective assessment of voice use in daily situations outside a controlled clinical setting is a valuable complement to traditional methods and can increase the ecological validity of our assessments. Providing biofeedback to help individuals with hypophonia secondary to PD increase their voice sound level may be an alternative or complement to voice therapy performed at a clinic. Further research with more participants is however needed to determine the efficacy of this intervention approach, and development of the biofeedback capabilities of portable voice accumulators are needed to improve the method.