Page 67 - Personality disorders and insecure attachment among adolescents
P. 67

Termination Status: Dropout and Completion of Therapy
In case premature termination was suggested by a patient, the patients family or the treatment staff, a supportive reassessment of treatment was organized. Only when both the therapist and the patient (and family) agreed that therapy goals had been reached, or when both agreed to terminate while therapy goals had only partly been reached, was the patient classified as a “completer.” When both stated that therapy was not completed yet, or only the patient or only the therapist stated that therapy was not completed, the exact reasons for termination were examined. In these cases, the patient was classified as a “dropout” when the patient prematurely terminated therapy but the therapist did not agree on this termination (i.e., according to the therapist the therapy should have been continued). The intention was to classify the patients as “unilaterally terminated by the therapist” when the therapist wished to terminate therapy while the patient wished to continue. Among the included 70 patients, there were no cases of “unilaterally terminated by the therapist.” Finally, 25 patients were classified as dropouts, and 45 patients were classified as completers.
Statistical analyses
All analyses were performed using the SPSS, version 25.0 (IBM, 2017). First, using a t-test dropouts and completers were compared on the C-SRS score of the first session and the last session. A mixed model analysis was performed with the C-SRS score as dependent variable and time and dropout as independent variables to see if dropout was related to C-SRS scores over time.
Second, the reliable change index (RCI) for the C-SRS was calculated using Jacobson and Truax formula (Jacobson & Truax, 1991), based on all questionnaires (N=2378) with a reliability (Cronbach’s Alpha) = .921 and SD = 8.15, the standard error was 3.24. The reliable change criterion was ( 1.96 * 3.24) 6.35.
Third, a Generalized Estimating Equations analysis (GEE) with an exchangeable working correlation matrix was performed to see if a descent in C-SRS score could predict dropout with the dichotomous variable significant decrease (RCI) in the C-SRS score between two consecutive sessions as independent and dropout within three sessions as dependent variable.
In a fourth step percentages of significant decreases (using RCI) during the last five sessions of therapy were compared between dropouts and completers.
61



























































































   65   66   67   68   69