Association of Long-Acting Injectable Antipsychotics and Oral Antipsychotics With Disease Relapse, Health Care Use, and Adverse Events Among People With Schizophrenia

IMPORTANCE Evidence for improved clinical outcomes with long-acting injectable antipsychotics (LAIAs) vs oral antipsychotics (OAs) is limited in Asian populations and special patient groups, including older people (>65 years), people with substance use, and early initiators of LAIAs. OBJECTIVE To compare the risk of disease relapse, health care use, and adverse events associated with the use of LAIAs vs OAs among people in Hong Kong with schizophrenia. DESIGN, SETTING, AND PARTICIPANTS In this self-controlled case series study, individuals with a diagnosis of schizophrenia who were prescribed LAIAs and OAs between January 1, 2004, and December 31, 2019, were identified from the Clinical Database Analysis and Reporting System of the Hong Kong Hospital Authority. Data analysis was conducted from May to August in 2021. EXPOSURES Use of LAIAs vs OAs. MAIN OUTCOMES AND MEASURES Risk of disease relapse (hospitalizations for psychiatric disorders, hospitalizations for schizophrenia, and suicide attempts), health care use (all-cause emergency department visits and hospitalizations), and adverse events (hospitalizations for somatic disorders, hospitalizations for cardiovascular diseases, and extrapyramidal symptoms) between the period in which patients were treated with LAIAs and the period in which patients were treated with OAs were compared using Poisson regression. RESULTS Of the 70 396 individuals with schizophrenia (37 200 women [52.8%]; mean [SD] age, 44.2 [15.8] years), 23 719 (33.7%) were prescribed both LAIAs and OAs. Compared with OAs, LAIAs were associated with a lower risk of hospitalizations for any cause (n = 20 973; incidence rate ratio [IRR], 0.63 [95% CI, 0.61-0.65]), hospitalizations for psychiatric disorders (n = 19 283; IRR, 0.52 [95% CI, 0.50-0.53]), hospitalizations for schizophrenia (n = 18 385; IRR, 0.53 [95% CI, 0.51-0.55]), and incident suicide attempts (n = 1453; IRR, 0.56 [95% CI, 0.44-0.71]). During full treatment with LAIAs, there was a reduction in hospitalizations for somatic disorders (n = 15 396; IRR, 0.88 [95% CI, 0.85-0.91]), hospitalizations for cardiovascular diseases (n = 3710; IRR, 0.88 [95% CI, 0.81-0.96]), and extrapyramidal symptoms (n = 22 182; IRR, 0.86 [95% CI, 0.82-0.91]) compared with full treatment with OAs. No significant difference was found for emergency department visits. Similar associations were observed during the subsequent treatment periods (beyond 90 days) and among older people and those with substance use, except for an increased risk of extrapyramidal symptoms among older people when initiating LAIAs (first 90 days). Compared with late initiators, early LAIA initiators had a greater reduction in these outcome events. CONCLUSIONS AND RELEVANCE This self-controlled case series study of people in Hong Kong with schizophrenia suggests that LAIAs were associated with a lower risk of disease relapse and hospitalization than OAs, without an increased risk of adverse events. Clinicians should more broadly consider the long-term use of LAIAs for Chinese people with schizophrenia, especially early in the course of illness.