OBJECTIVES: To investigate the relationship between gaps in oral antipsychotic therapy and risk of hospitalization in the outpatient treatment of schizophrenia within a managed-care environment. METHODS: A retrospective evaluation of schizophrenia patients from a mental health subset of managed care (PHARMetrics) data was conducted. Patients were required to be continuously eligible for 13 months following their index date (i.e., first claim date during the enrollment period between January 1, 2000 - December 31, 2001), and were followed for 12 months post-index. Inclusion criteria included a diagnosis of schizophrenia and at least one claim of an oral antipsychotic in 1999 and two in 2000 or 2001. Exclusion criteria included bipolar disorder, claims for long-term care, and long-acting antipsychotics. Medication compliance measures included maximum gap in therapy, medication possession ratio (MPR), consistence, and persistence. Logistic regression was used to analyze the odds of being hospitalized. Independent variables included compliance, age, and sex. RESULTS: Of the 1499 subjects included in the study (mean [SD] age = 45.1 ± 12.4 years; 46.6% were male), 5.9% were hospitalized at least once. The odds of hospitalization in this population rose by 2.1% for every 5-day increase in therapy gap (p = 0.0004) based on a continuous model. Patients with > 30-day maximum gap in their medication had higher odds (4.66, p <0.0001) of hospitalization than the 0 to 10 day reference group. The 11 to 30 day gap group approached significance with an odds ratio of 2.10 (p = 0.05769). MPR and consistence models indicated that the odds of hospitalization decrease by a factor of 0.831 (p <0.0001) and 0.812 (p <0.0001) for a 10% increment in MPR and consistence respectively. Age was a significant predictor of hospitalization in all models. Persistence and gender were not significant predictors. CONCLUSIONS: Gaps in oral antipsychotic treatment are associated with an increased risk of hospitalization in schizophrenic patients.