Healthcare resource utilization and costs associated with first versus subsequent use of cariprazine for bipolar I disorder

Abstract

Aims To evaluate the healthcare resource utilization (HRU) and costs of patients who initiated cariprazine as their first versus subsequent atypical antipsychotic (AA) following a bipolar I disorder (BP-I) diagnosis. Methods Adults with a BP-I diagnosis (first claim = index), commercial, Medicare Supplemental, or Medicaid insurance, and ≥1 outpatient cariprazine dispensing were identified from Merative MarketScan database. Cohorts included patients who initiated cariprazine as either their first or subsequent AA after initial BP-I diagnosis. Characteristics were balanced between cohorts using inverse probability of treatment weighting (IPTW). Outcomes evaluated post-index included all-cause and mental health (MH)–related HRU (hospitalizations, emergency department [ED] visits, outpatient visits), total healthcare costs (medical + pharmacy), and treatment patterns. HRU and healthcare costs were reported per patient-year (PPY) and compared between cohorts using rate ratios and 95% CIs estimated using nonparametric bootstrap procedures. Treatment patterns were analyzed descriptively, with standardized differences ≥10% considered important. Results After IPTW, cohorts included 1,409 patients who initiated cariprazine first and 1,621 patients who initiated cariprazine subsequently; the average (standard deviation, SD) observation period was 678 (373) and 758 (389) days for first and subsequent initiators, respectively. Patients who initiated cariprazine first had 23% fewer all-cause hospitalizations and 28% fewer MH-related hospitalizations PPY (each comparison, p < 0.001). Rates of all-cause and MH-related outpatient visits were significantly lower in patients who initiated cariprazine first versus subsequently (each comparison, p < 0.001), while rates of ED visits were similar. Relative to subsequent initiators, first initiators incurred $2,587 and $2,130 lower all-cause and MH-related total healthcare costs PPY, respectively (each comparison, p < 0.05). Before starting cariprazine, first initiators used fewer BP-I–related medications on average than subsequent initiators (2.6 vs 3.9; standardized difference = 23.9%). Limitations Potential coding inaccuracies and residual confounding. Conclusions In this real-world database analysis, patients with BP-I who initiated cariprazine as their first AA had lower rates of HRU and incurred lower costs than patients who initiated cariprazine as a subsequent AA.