Page 87 - Personality disorders and insecure attachment among adolescents

‘Loving mother’ (p = .031), ‘Idealizing mother’ (p = .019), ‘Preoccupied anger mother’ (p = .003), ‘Loving father’ (p = .028), ‘Preoccupied anger father’ (p = .002), ‘Unresolved loss’ (p = .048), ‘Unresolved trauma’ (p = .040), ‘Coherence of transcript’ (p = .009) and ‘Coherence of mind’ (p = .009). Relating changes in attachment to changes in psychological distress at t-2 In the next step, treatment outcome groups were formed based on the assessment of severity differences between the five-way attachment classifications at the start and the end of treatment (see statistical analysis for more details). Either the pre- or post- total SCL-90 score was missing for three patients of the AAI-Unchanged, who were excluded from this outcome group. Of the three AAI outcome groups, the AAI-Improved (N = 16) differed significantly (p < .05) from the AAI-Unchanged (N = 9) and AAI-Deteriorated (N = 5) in changes on the ‘Rejecting mother’ (t = 3.620, p = .003, d = 3.979), ‘Rejecting father’ (t = 4.571, p = .000, d = 4.039), ‘Loving mother’ (t = -2.423, p = .029, d = 4.095), ‘Preoccupied anger father’ (t = 2.138, p = .049, d = 1.338), ‘Coherence of transcript’ (t = - 4.656, p = .000, d = 1.93), and ‘Coherence of mind’ scale (t = -3.982, p = .001, d = 1.799). Reciprocally, the AAI-Unchanged group differed significantly (p < .05) from the AAI-Improved group and the AAI-Deteriorated group in changes on the ‘Loving mother’ (t = -2.530, p = .028, d = 1.931), ‘Loving father’ (t = -2.347, p = .035) and ‘Preoccupied anger mother’ scale (t = 2.569, p = .026, d = 1.384). Finally, the AAI-Deteriorated group differed significantly from the two other AAI-outcome groups in changes on the ‘Metacognitive monitoring’ (t = 3.62, p = .034, d = 4.186) and ‘Involving/role reversing mother’ scale (t = -3.873, p = .018, d = 1.171). These groups were compared to each other on the basis of the total SCL-90 scores at the beginning and the end of treatment. While no significant differences were found on the pre SCL-90 scores (F = .214, p = .808), the total SCL-90 scores decreased significantly for both AAI-groups at the end of treatment (Table 4). The AAI- Improved group showed a medium symptom reduction according to the SCL-90 (N = 16, M = 72.75, SD = 68.01, t = 4.28, p = .001, d = .56). The AAI-Unchanged group also showed symptom reduction, although not as strong (N = 9, M = 48.11, SD = 56.10, t = 2.57, p = .033, d = 0.37). The AAI- Deteriorated group on the other hand, showed small symptom reduction (N = 5, M = 21.20, SD = 75.75, t = .63, p = .565, d = 0.14). Comparing the AAI-Improved group (N = 16, M = 72.75, SD = 68.01, t = 4.28, p = .001, d = .56) with the AAI-Unchanged combined with the AAI-Deteriorated group named the AAI-Non-improved group (N = 14, M = 38.50, SD = 62.30, t = 2.31, p = .038, d = .28), revealed that the changes toward increased secure attachment in the AAI-Improved group were associated with stronger reduction of psychological distress in comparison to the AAI-Non-improved group. 83