3.1. Patient characteristics of CBT and PT groups
As shown in Table 2, the CBT cohort encompassed 20 participants (14 women and 6 men) with a mean age of 48.65±16.24. The median illness duration was 39.5 months, ranging from 8 to 180 months. Pre-treatment indices for the CBT group were as follows: DHI score of 49.60± 18.53, HADS anxiety score of 8.75±4.74, HADS depression score of 7.60±4.32, and NPQ score of 39.05±16.03.
As shown in Table 2, the PT group comprised 22 participants (15 women and 7 men) with a mean age of 48.55±12.31 years. The median illness duration was 25.73 months, ranging from 3 to 161 months. Pre-treatment indices for the PT group were as follows: DHI score of 51.64±21.94, HADS anxiety score of 8.77±5.04, HADS depression score of 7.82 ±3.23, and NPQ score of 36.23±14.78. Data are mean ± SD.
Independent t-tests revealed no significant differences in age, DHI score, HADS anxiety score, HADS depression score, and NPQ score between the CBT and PT groups. As normality was not observed for illness duration, the Mann–Whitney U test was employed, revealing a significant difference between the groups (U=284, p=0.02): the illness duration was significantly longer in the CBT group than in the PT group. According to Fisher's exact test, there was no significant difference in sex between the CBT group and the PT group.
As shown in Table 3, the precipitating diseases of the participants in the CBT group were as follows: acute attack of peripheral vestibular vertigo (n=6), no specific precipitants (n=5), BPPV (n=3), stress reaction (n=2), chronic anxiety disorders (n=1), Meniere's disease (n=1), mal de debarquement (n=1), and vestibular neuritis (n=1). In the PT group, the precipitating diseases were acute attack of peripheral vestibular vertigo (n=7), BPPV (n=5), stress reaction (n=3), no specific precipitants (n=2), vestibular neuritis (n=2), Meniere's disease (n=1), postoperative status of acoustic tumor (n=1), and vestibular migraine (n=1). There were no significant differences in the precipitating diseases between the CBT and PT groups.
As shown in Table 4, there were no significant differences in the vestibular function results, including, canal paresis (CP) %, vestibulo-ocular reflex-direction preponderance (VOR-DP) %, c- and o-vestibular myogenic potentials (VEMP), subjective visual vertical (SVV), and posturography, between the two groups. Moreover, all these data were within normal ranges in both groups.
3.2. Assessment of tolerability and safety in the CBT group
All patients completed the CBT program with no adverse events related to CBT.
3.3. Measurement of assessment indices
3.3.1. Results of Repeated Measures ANOVA
For the primary outcome, DHI, the main effect of time was confirmed: DHI (F(1,40) = 43.49, p < .001, ηG2 = 0.14, 95% CI [0.06, 0.23]). As secondary outcome, HADS and NPQ, a significant main effect of time was observed, indicating a significant decrease in index scores from pre- to post-treatment assessments: HADS anxiety (F(1,40) = 26.52, p < .001, ηG2 = 0.11, 95% CI [0.03, 0.19]), HADS depression (F(1,40) = 22.18, p < .001, ηG2 = 0.07, 95% CI [0.02, 0.14]), and NPQ (F(1,40) = 16.60, p < .001, ηG2 = 0.09, 95% CI [0.02, 0.18]).
However, no significant differences were observed between treatment modalities between the two groups: HADS anxiety (F(1,40) = 0.13, p = .72, ηG2 = 0.00, 95% CI [0.00, 0.02]), HADS depression (F(1,40) = 0.11, p = .74, ηG2 = 0.00, 95% CI [0.00, 0.03]), DHI (F(1,40) = 0.27, p = .60, ηG2 = 0.00, 95% CI [0.00, 0.06]), and NPQ (F(1,40) = 0.02, p = .88, ηG2 = 0.00, 95% CI [0.00, 0.00]).
A repeated measures ANOVA was conducted to examine the effects of time (pre-treatment vs. post-treatment) and treatment modality (CBT vs. pharmacological therapy) on the outcome measures (HADS anxiety, HADS depression, DHI, and NPQ). The results revealed a significant main effect of time for all indices, indicating a significant reduction in symptom severity from pre to post-treatment assessments. However, there was no significant main effect of treatment modality or interaction between time and treatment modality, suggesting that both CBT and PT were equally effective in reducing symptoms.
3.3.2. Corset plots and effect size
The corset plots in Fig. 1-2 visualize the respective scale measures before (Pre) and after (Post) the treatment. The half-violins show the distribution by treatment group at each time point, the box plots show the scatter of the data, and the circle and triangle shapes show the individual changes before and after the treatment.
Calculated within-group effect sizes for each treatment were as following (Table 5): DHI, CBT (Hedges’ g=0.929, 95% CI [0.268, 1.589]), PT (Hedges’ g=0.675, 95% CI [0.060, 1.289]); HADS anxiety, CBT (Hedges’ g=0.800, 95% CI [0.148, 1.452]), PT (Hedges’ g=0.557, 95% CI [-0.052, 1.166]); HADS depression, CBT (Hedges’ g=0.522, 95% CI [-0.117, 1.160]), PT (Hedges’ g=0.531, 95% CI [-0.077, 1.139]); NPQ, CBT (Hedges’ g=0.706, 95% CI [0.060, 1.353]), PT (Hedges’ g=0.482, 95% CI [-0.124, 1.088]). Table 5 presents the results for the CBT and PT groups.