Emergency department use disparities in schizophrenia: Mediating role of antipsychotic availability and administration

Racial disparities in emergency department (ED) use among patients with schizophrenia may reflect differences in treatment access and continuity after antipsychotic initiation. We examined whether oral antipsychotic availability or administering long-acting injectable antipsychotics, measured by proportion of days covered (PDC) using claims data, mediated the association between race and ED use and whether mediation differed by antipsychotic class and route of administration. Using South Carolina Medicaid claims from 2004 to 2019, we identified 1666 adults with schizophrenia or schizoaffective disorder initiating antipsychotic treatment with 12-month washout period. PDC was measured during the first 6 months after treatment initiation. The outcome was any ED visit during the subsequent 6 months. Mediation analysis was conducted overall and by antipsychotic subgroup. Black beneficiaries had higher ED use than White beneficiaries (33% vs. 27%, p = 0.03) and lower PDC (mean 0.45 vs. 0.55, p < 0.0001). In unadjusted analyses, PDC mediated 23% of the association between race and ED use (95% CI 8%-94%, p = 0.03). After covariate adjustment, PDC mediated 25% of the association (95% CI 10%-95%, p = 0.006). Significant mediation was observed among oral antipsychotic users (28%, p = 0.02), second-generation antipsychotic users (29%, p = 0.008), and most prominently in oral second-generation antipsychotic users (30%, p = 0.006). No significant mediation was observed among first-generation antipsychotic or long-acting injectable users. These findings suggest that antipsychotic availability and administration may be one pathway contributing to racial disparities in acute psychiatric care.