Table 6.

Results from ANCOVA regression equations (differences between VIPI and TAU) with EAS score at T2 as dependent variable, VIPI, different values of EAS at T1 and different personality disorder Clusters scores and personality scores as covariates

EAS score T1/Percentile	97 (10th)	116.5 (25th)	143 (50th)	165 (75th)	172 (90th)
DIP-Q item
Cluster A	ns	ns	ns	ns	ns
Cluster B	ns	ns	ns	ns	ns
Cluster C	ns	ns	ns	ns	ns
Paranoid
Paranoid_0	15.56a	9.23a	0.64a	−6.50a	no data
Paranoid_2	–	–	11.22 a *	4.22a	no data
Paranoid_4	37.06 a ***	30.85 a ***	no data	no data	no data
Schizoid
Schizoid_0	ns	ns	ns	ns	ns
Cut-off (4)	ns	ns	ns	ns	ns
Schizotypal
Schizotypal_0	ns	ns	ns	ns	ns
Cut-off (5)	ns	ns	ns	ns	ns
Borderline
Borderline_0	ns	ns	ns	ns	ns
Cut-off (5)	ns	ns	ns	ns	ns
Histrionic
Histrionic_0	ns	ns	ns	ns	ns
Cut-off (5)	ns	ns	ns	ns	ns
Narcissistic
Narciss_0	35.10 b***	27.41 b ***	16.97 b **	8.30b	5.90b
Narciss_2	11.59b	3.90b	−6.54b	−15.21 b *	−19.10 b *
Antisocial
Antisocial_0	ns	ns	ns	ns	ns
Antisocial_2	ns	ns	ns	ns	ns
Avoidant
Avoidant_0	ns	ns	ns	ns	ns
Cut-off (4)	ns	ns	ns	ns	ns
Dependent
Dependent_0	17.63 c *	9.10c	−2.49c	no data	no data
Dependent_3	–	–	15.72 c **	no data	no data
Cut-off (5)	48.97 c ***	40.44 c ***	no data	no data	no data
Obsessive compulsive
Obs.comp._0	ns	ns	ns	ns	ns
Cut-off (4)	ns	ns	ns	ns	ns

In all analyses, (VIPI × EAS percentile score) is kept in the ANCOVA with p = 0.03–0.002. An abbreviation with _0, _2, and _4 means zero, two, or four traits in the respective personality disorder category, and the actual cut-off values for the respective personality disorder given in ( ).

VIPI = 1(0) for the treatment group (TAU). EAS: Emotional Availability Scales. DIP-Q: DSM IV and ICD-10 Personality Questionnaire.

_a = p (VIPI × Paranoid score) < 0.05.

_b = p (VIPI × Narcissistic score) < 0.01.

_c = p (VIPI × Dependent score) < 0.01.

*p < 0.05, **p < 0.01, ***p < 0.001.

ns = not significant.