Effectiveness of medication treatment is determined by three components: treatment efficacy (symptom reduction), tolerability/safety, and adherence. Compared with efficacy and safety, research into adherence has been lacking. Nevertheless, medication non-adherence is a risk factor for relapse and for aggressive behavior in association with substance abuse in schizophrenia patients. Non-adherence has been estimated to cause approximately 40% of relapses in patients with schizophrenia. High rates of treatment discontinuation in all arms of the CATIE study illustrate the widespread nature of non-adherence. Most of previous research has defined non-adherence as a complete discontinuation of medication. However, many schizophrenia patients show partial adherence: they do not completely discontinue their medication, but they do not take all that has been prescribed. Partial adherence is more difficult to define and study than complete non-adherence.

e had the opportunity to study partial adherence in the context of a randomized, double-blind, 8-week, fixed-dose study comparing olanzapine 10mg/d, 20 mg/d and 40 mg/d for patients with schizophrenia or schizoaffective disorder (N=599). Medication non-adherence was measured by pill counts. Baseline characteristics including demographics, illness history and symptom severity were investigated as potential risk factors for treatment non-adherence.

Approximately 1/3 of patients were non-adherent with their medication at least once during the 8-week study. These non-adherent patients had significantly less improvement compared to adherent patients. Adherent patients had greater weight gain than the non-adherent ones. Among the available baseline measures, greater baseline depression severity appeared to be a significant risk factor for non-adherence.