Ancillary Analyses
There was no difference in the THI subscales changes (functional, emotional, and catastrophic) between groups (Table 2).
The covariate adjustment multivariate and regression model revealed that the educational level, guardian help to fill the THI score, mild hearing loss, and psychiatric illness did not interfere in the efficacy results.
Also, the sensitivity analyses with per-protocol participants did not impact the study group baseline characteristics and the efficacy results (eTable 4 in Supplement 1).