A longitudinal study of medication nonadherence and hospitalization risk in schizophrenia

Previous cross-sectional studies have suggested an association between medication nonadherence and hospitalization for individuals with schizophrenia. However, such analyses typically measure adherence averaged over long time periods. We investigated the temporal relationship between nonadherence and hospitalization risk using a daily measure of medication availability.

Our observational cohort included 1191 patients with schizophrenia (ICD-9 criteria) enrolled in Maine and New Hampshire Medicaid programs who initiated atypical antipsychotic therapy between January 1, 2001, and December 31, 2003. Pharmacy claims were used to define days with gaps in medication availability. We tested the association of gaps in medication availability with all-cause, mental health, and schizophrenia-specific hospitalization using a Cox regression model.

Compared to individuals with available medication, individuals in the first 10 days following a missed prescription refill had a hazard ratio of 1.54 (95% CI = 1.02 to 2.32) for mental health hospitalization and 1.77 (95% CI = 1.16 to 2.71) for schizophrenia hospitalization. Similarly, medication gaps of more than 30 days were associated with 50% increased hazard for all 3 hospitalization outcomes. Switching and augmenting therapy, previous hospitalization, and clinical severity measures were also associated with substantially increased hazard of hospitalization.

Our findings indicate that patients may be at significantly increased risk for hospitalization as early as the first 10 days following a missed medication refill. Patients who switched or augmented medications or were previously hospitalized also demonstrated increased hospitalization risk. Clinicians and Medicaid programs should consider using pharmacy claims to monitor medication use and target adherence interventions to reduce relapses in this vulnerable population.