Treatment outcomes
The repeated-measures ANOVA revealed a significant main effect of time on the primary outcome, GAF. A post hoc analysis revealed that there were significant differences between pre-and post-treatment (p < 0.001, d = 3.30) and between pre-treatment and follow-up (p < 0.001, d = 3.35) (Table 2). Additionally, the main effect of time for WHO QOL-BREF was significant; however, post hoc analyses result did not remain significant between each time point after the correction.
Table 2
Changes in each evaluated item before ST, after ST, and at follow-up
|
Pre-ST
|
Post-ST
|
Follow-up
|
|
Cohen's d
|
Variable
|
Mean (SD)
|
Mean (SD)
|
Mean (SD)
|
F
|
Pre-ST vs. post-ST
|
Post-ST vs. follow-up
|
Pre-ST vs.
follow-up
|
GAF
|
45.90 (10.82)
|
76.00 (7.06)a
|
76.20 (6.83)c
|
F (1.3, 11.67) = 151.65**
|
3.30
|
0.03
|
3.35
|
QOL
|
65.20 (16.03)
|
75.70 (14.28)
|
78.50 (12.77)
|
F (1.21, 10.87) = 9.93**
|
0.69
|
0.21
|
0.92
|
BDI-Ⅱ
|
27.70 (17.54)
|
16.70 (13.94)
|
13.00 (11.76)
|
F (1.16, 10.4) = 9.64**
|
0.70
|
0.29
|
0.99
|
STAI (state)
|
45.20 (9.47)
|
49.70 (8.21)
|
41.50 (6.92)b
|
F (2, 18) = 3.93*
|
0.51
|
1.08
|
0.45
|
STAI (trait)
|
55.80 (14.16)
|
53.00 (10.53)
|
47.50 (8.82)
|
F (2, 18) = 3.56*
|
0.23
|
0.57
|
0.70
|
OCI
|
67.70 (30.94)
|
59.40 (27.39)
|
49.70 (23.15)
|
F (1.17, 10.49) = 3.98
|
0.29
|
0.38
|
0.66
|
LSAS_Total
|
62.70 (22.60)
|
62.80 (27.53)
|
50.70 (18.64)
|
F (1.2, 10.81) = 2.71
|
0.00
|
0.52
|
0.58
|
LSAS_fear/anxiety
|
35.90 (13.54)
|
35.40 (15.18)
|
27.20 (9.32)
|
F (2, 18) = 5.22*
|
0.04
|
0.65
|
0.75
|
LSAS_escape
|
26.80(13.23)
|
27.40(13.70)
|
23.50 (11.21)
|
F (1.27, 11.46) = 0.59
|
0.05
|
0.31
|
0.27
|
Note: **p < 0.01, *p < 0.05 calculated using analysis of variance |
a: Significant difference between pre- and post-intervention values (p < .05, Bonferroni corrected), b: Significant difference between post-intervention and follow-up values, c: Significant difference between pre-intervention and follow-up values (p < .05, Bonferroni corrected). |
BDI-II, Beck Depression Inventory-II; STAI (State), State items from the State-Trait Anxiety Inventory; STAI (Trait), Trait items from the State-Trait Anxiety Inventory; OCI, Obsessive Compulsive Inventory; LSAS (Total score), Total items from the Liebowitz Social Anxiety Scale; LSAS (Fear/Anxiety), Fear and anxiety items from the Liebowitz Social Anxiety Scale; LSAS (Avoidance), Avoidance items from the Liebowitz Social Anxiety Scale |
Regarding the secondary outcomes, the repeated-measures ANOVA revealed a significant effect of time on the YSQ-SF (Total score), Disconnection and rejection, Impaired autonomy and performance, Impaired limits (Over-vigilance and inhibition), Social Isolation/Alienation schema, Failure schema, Dependence/Incompetence schema, Emotional Inhibition schema, Approval-Seeking/Recognition-Seeking schema, Negativity/Pessimism schema, and Punitiveness schema scores, but not on other subscales (Additional File 3). The Bonferroni post hoc test showed a significant difference between pre-treatment and the follow-up (p = 0.00048, d = 1.01) for the YSQ-SF (Total score). Similarly, for Disconnection and rejection, a significant difference was found between post-treatment and follow-up (p = 0.00026, d = 0.34). There was also a significant difference for Impaired limits (Over-vigilance and inhibition) between pre-treatment and follow-up (p = 0.00002, d = 0.63). In addition, for the Emotional Inhibition schema and the Negativity/Pessimism schema, although the ANOVA showed a significant main effect of time, the Bonferroni post hoc test did not show any significance.
Regarding the SMI, the repeated-measures ANOVA revealed a significant main effect of time for the Adaptive mode, Maladaptive mode, Vulnerable child mode, Angry child mode, Enraged child mode, Impulsive child mode, Undisciplined child mode, Happy child mode, Compliant surrenderer mode, Detached protector modes, Self-aggrandizer mode, Punitive parent mode, Demanding parent mode, and Healthy adult mode scores (Additional File 4).
Conversely, there was no significant difference in the Detached self-soothing or Bullying and attack modes. Bonferroni post hoc tests showed a significant difference between pre-treatment and follow-up for the Maladaptive mode (p = 0.00051, d = 1.57), Enraged child mode, (p = 0.00012, d = 1.86), Undisciplined child mode (p = 0.00066, d = 0.97), and Demanding parent mode (p = 0.00010, d = 1.10).
Regarding other outcomes, significant effects of time were observed for BDI-II, STAI-state and -trait, and LSAS fair/anxiety; however, post hoc comparison with Bonferroni correction did not remain significant between each time point. No aversive reports or effects were observed; thus, ST can be implemented and adapted for short-term use in outpatient settings for adults with HF-ASD.