Toward Sham Interventions for Behavioral Voice Treatment Outcome Research in Female Students Without Dysphonia.

OBJECTIVES/HYPOTHESIS: To develop sham voice treatment techniques to be used in voice treatment outcome research, and to investigate their effectiveness as sham. This entails that the techniques induce no changes in voice or voice physiology, yet still lead to a perception of efficacy.

STUDY DESIGN

Prospective randomized blinded controlled study.

METHODS

Three distinct sham intervention protocols (SIPs) were conceptualized as placebic comparators for three common voice treatment approaches with focus on vocalization (SIP1), respiration (SIP2), and manipulation (SIP3). Forty-eight female students participated in the study. Each participant attended ten 30-minute sessions over 5weeks, including a baseline evaluation, three sessions of one SIP, an inter-SIP voice assessment, three sessions of a second SIP, and a final post-SIP assessment. Auditory-perceptual and instrumental voice evaluations were used as voice treatment outcome measures. The participants' perception of voice-related quality of life was evaluated using the French Voice Handicap Index (VHI). Frequentist as well as Bayesian statistical methods were applied for group comparisons. The effects of combining two SIPs, the potential influence of SIP order, and experimenter/clinician effects, were also investigated.

RESULTS

The respiration-based SIP2 showed changes only in sound intensity level on a sustained vowel across the three sham intervention sessions. In contrast, the vocalization-based SIP1 impacted sound intensity level on a sustained vowel, sound intensity level on read text, and maximum phonation time. The manipulation-based SIP3 affected smoothed cepstral peak prominence on read text, Acoustic Voice Quality Index, and Dysphonia Severity Index. SIP2 thus demonstrated the highest alignment with the study's objectives, followed by SIP1 and SIP3. GRBASI ratings revealed no statistical differences for any SIP. VHI decreased significantly after all three SIPs. Combining the SIPs generally replicated the effects observed when each SIP was used individually. There was no order effect or experimenter/clinician effect on the results.

CONCLUSIONS

This study demonstrated significant changes in participants' perceived voice quality (measured with VHIFr) across various SIPs, despite minimal impact on objective voice function measures. Further investigation is necessary to establish one or more protocols as genuinely sham interventions.