Clinical and economic impact of long-acting injectable antipsychotics in patients previously treated with short-acting oral antipsychotics

BACKGROUND: Available literature supports the use of long-acting injectable (LAI) antipsychotics over short-acting oral (SAO) formulations. The majority of evidence is centered on patients with schizophrenia insured under the Medicaid benefit. OBJECTIVE: To assess real-world clinical and economic outcomes of LAI compared with SAO antipsychotics in patients with psychiatric conditions insured under commercial, health care exchange, or Medicare plans. METHODS: In this exploratory treatmenteffectiveness study, a retrospective, before and after study design was used to evaluate differences in clinical and economic outcomes in patients switching from SAO to LAI antipsychotics. Patients who had at least 1 claim for an LAI antipsychotic and a psychiatric diagnosis were considered eligible for the study if they were continuously enrolled in a commercial, health care exchange, or Medicare plan for 12 months before (preperiod) and 12 months after (postperiod) their first LAI antipsychotic claim. During the preperiod, patients were required to have filled at least a 30-day supply of any SAO antipsychotic medication. Clinical outcomes included health care resource utilization (eg, hospitalizations per member per year [PMPY]), adherence measures, and medication switch trends. Economic outcomes included total per member per month (PMPM) spending across the medical benefit alone, the pharmacy benefit alone, and the combined spending across both benefits. Additionally, we examined patient costs and health plan spending within each of these categories. RESULTS: There was a significant decrease in overall hospitalizations PMPY (1.80 vs 0.88; P < 0.001) and psychiatric hospitalizations PMPY (0.65 vs 0.20; P <0.001) when comparing patients treated with SAO antipsychotics in the preperiod to the same patients treated with LAI antipsychotics in the postperiod, respectively. No differences were observed in the percentage of days covered with SAO and LAI agents (87.4% vs 85.8%; P=0.312). More patients switched between SAO antipsychotics during the preperiod, as compared with the number who switched between LAI antipsychotics during the postperiod (57.4% vs 10.3%; P < 0.001). On average, patients switched medications sooner in the preperiod vs the postperiod (114.50 vs 211.26 days; P < 0.001). No difference was observed between the preperiod and postperiod in total spending PMPM ($3,798.76 vs $3,702.63; P = 0.826) CONCLUSIONS: Patients switching from SAO to LAI antipsychotics experienced fewer hospitalizations, with no change in overall spending. Adherence was similar, though fewer medication switches occurred and there was a longer time before switching medications with LAI compared with SAO antipsychotics