Treatment compliance and early termination of therapy: a comparative study

Compliance in therapy appears to be linked to the process of therapeutic alliance. A positive patient-therapist relationship usually leads to successful completion of therapy. The aim of this study was to compare evidence, collected at two time periods in a Community Mental Health Center in Athens, on factors affecting treatment compliance. The hypothesis was that by modifying the therapeutic team's functioning, noncompliance could be reduced.

Epidemiological data were collected from child and adolescent out-patient files at two time periods: time period A, 1990-1994, n = 455 (sample A) and time period B, 2000-2002, n = 476 (sample B). Variables pertaining to the patient, his family and recommended treatment were examined. Student's t test and the Pearson chi2 test were used in order to explore the correlation of the variables with treatment completion in each sample. Logistic regression analyses were used to determine whether the effect of each variable on treatment completion differed between the two samples.

In sample A, most patients (58.6%) did not comply with therapy. The type of recommended treatment, the number of sessions, season of admission and the type of presented problem were found to correlate with treatment completion. Between the two time periods, certain modifications were implemented in the team's functioning (less diagnostic sessions, focused psychotherapy techniques, less time interval between referral and first diagnostic appointment). A significant reduction in the early termination rate (45.7%) was noted in sample B. The type of recommended treatment, the number of sessions, the family's situation, the mother's educational level and the patient's gender were related to treatment completion in sample B. The effect of the type of proposed treatment and the mean number of sessions of the completed treatments differed significantly between the two samples.

Early termination rates in therapy decreased between two time periods. This decrease may be attributed to modifications in the team's functioning, aimed at improving the therapeutic relationship. The limitation of this study is that only those factors pertaining to the service's organization and functioning were investigated.