Dropout rates with olanzapine or risperidone: a multi-centre observational study

Abstract

Objective: In patients with schizophrenia, risperidone and olanzapine are the two most commonly used atypical anti-psychotics. A recent meta-analysis based on randomized trials suggests that, in the long term, olanzapine can have a lower frequency of treatment discontinuation (or dropout) in comparison with risperidone. To better test this hypothesis, our observational study was aimed at assessing whether or not this advantage of olanzapine versus risperidone could be confirmed in a patient series examined in an observational setting.

Methods: Our study was based on a retrospective multi-centre observational design. We collected the following information from each patient: demographic characteristics; current anti-psychotic treatment (olanzapine or risperidone, under the condition of a stable therapy over months -1 to -4); cumulative dose of the drug; previous anti-psychotic treatment (during months -5, -6, -7 and/or, when available, also before month -7); daily dose and treatment duration. Our primary analysis traced back the patient's history from the date of enrollment retrospectively up to month -7. The secondary analysis followed-up the patient's history prior to month -7, thus extending this retrospective recording as long as possible (depending on what information was actually available for individual patients).

Results: The patients were enrolled from 31 institutions. In our primary analysis (months -1 to -7), a total of 144 patients were included; in this subgroup treated with olanzapine or risperidone as initial drug ( n=94), we observed 4 of 54 switches from olanzapine to risperidone and 11 of 40 switches from risperidone to olanzapine ( P=0.01). A total of 454 patients were enrolled in our secondary analysis (from month -1 up to month -73); the same comparison showed 9 of 236 switches from olanzapine to risperidone and 17 of 150 switches from risperidone to olanzapine ( P=0.004).

Conclusion: Our analysis confirms the results of a recent meta-analysis and shows that olanzapine might imply a lower risk of dropout than risperidone.