COMPARISONS OF MENTAL HEALTH-RELATED MEDICAL CARE UTILIZATION AND COSTS FOLLOWING THE INITIATION OF ORAL VERSUS LONG-ACTING INJECTABLE RISPERIDONE IN PATIENTS WITH SCHIZOPHRENIA

OBJECTIVES: To assess the patterns of mental-health related medical care utilization and medical costs in patients with schizophrenia initiating on either oral or long-acting injectable risperidone. METHODS: Retrospective analysis (Jan 2006-Dec 2008) of Texas Medicaid data was conducted for patients >= 18 years old with schizophrenia (ICD-9: 295.xx). Patients who initiated either oral (oral cohort) or injectable risperidone (injectable cohort), had no prior risperidone for at least 12 months, and had at least one additional prescription following initiation were included. Medical care services and costs were compared within and between cohorts for 12 months pre- and post-initiation using paired-, independent t-tests and McNemar tests.   RESULTS: 1544 patients were included (oral cohort: n=1261, mean age = 38yrs, 54% male; injectable cohort: n=283, mean age=39yrs, 61% male). The injectable cohort (4.6, SD=2.9) had a significantly higher mean Chronic Disease Score (CDS) compared to the oral cohort (3.2, SD=2.9) (p<0.05). The percent of patients with at least one psychiatric-related hospitalization significantly decreased by 10.6% in the injectable cohort (25.0% to 14.5%; p<0.05) compared to 6.1% in the oral cohort (18.7% to 12.6%; p<0.05), and average hospital length of stay (LOS) was significantly reduced by 2.5 days (4.4 to 1.9 days) versus 0.9 days (2.6 to 1.7 days), respectively (p<0.05). Direct mental health-related medical costs significantly decreased by $1395 ($4968 to $3573; p<0.05) in the injectable cohort compared to $51 in the oral cohort ($1912 to $1861; NS) (p<0.05). CONCLUSIONS: For patients initiated on long-acting injectable risperidone, clinical utilization and economic outcomes improved one year post-initiation, as the number hospitalizations, LOS, and total mental health-related medical costs decreased. While injectable cohort patients may be a sicker population (as evidenced by higher CDS and higher levels of baseline and follow-up utilization and costs), their level of improvement was higher than that of the oral cohort.