Disease relapse, all-cause mortality, and adverse events associated with long-acting injectable antipsychotics versus oral antipsychotics in older people with schizophrenia in Hong Kong: a population-based within-subject analysis

Abstract

BACKGROUND: Older individuals (aged ≥65 years) with schizophrenia are at increased risk of treatment non-adherence due to cognitive decline and complex polypharmacy. Use of long-acting injectable (LAI) antipsychotics in this population has not been systematically investigated. We aimed to compare risk of disease relapse, all-cause mortality, and adverse events associated with LAI versus oral antipsychotics among older people with schizophrenia. METHODS: In this population-based within-subject analysis, we used territory-wide electronic health records from the Clinical Database Analysis and Reporting System of the Hong Kong Hospital Authority. Individuals diagnosed with schizophrenia (ICD-9-CM code 295) in inpatient, outpatient, or emergency department settings between Jan 1, 1993, and Dec 31, 2023, were identified, and people aged 65 years and older who were prescribed LAI or oral antipsychotics between Jan 1, 2004, and Dec 31, 2023, were considered for inclusion. Individuals with incomplete demographic information were excluded. The primary outcome was disease relapse, defined as hospital admission for schizophrenia. The secondary outcomes comprised all-cause mortality and adverse events, including cardiovascular hospital admission, extrapyramidal symptoms, acute liver injury, and acute kidney injury, which were identified using ICD-9-CM codes or laboratory tests. A self-controlled case series study using Poisson regression was conducted to compare the risk of disease relapse and adverse events between the treatment periods of LAI and oral antipsychotics. An individual-stratified Cox regression was performed for all-cause mortality. Time-varying confounders, including age, season, class of antipsychotics, COVID-19 stringency index, and concomitant treatment with antidepressants, benzodiazepines or Z-drugs, mood stabilisers, and antiparkinsonian drugs, were adjusted for. People with lived experience of schizophrenia were not involved in the design or conduct of the study. FINDINGS: Of 24 985 older individuals with schizophrenia, 4696 (18·8%) were prescribed LAI or oral antipsychotics. 10 655 (42·6%) of 24 985 individuals were male and 14 330 (57·4%) were female. Data on ethnicity were not available. Compared with oral antipsychotics, LAI antipsychotics were associated with statistically significantly lower risk of hospital admission for schizophrenia (incidence rate ratio [IRR] 0·71 [95% CI 0·64-0·78], p<0·0001) and all-cause mortality (hazard ratio [HR] 0·23 [95% CI 0·12-0·44], p<0·0001). No statistically significant difference was found in cardiovascular hospital admissions, acute liver injury, and acute kidney injury. LAI antipsychotics were associated with increased risk of extrapyramidal symptoms (IRR 2·17 [1·24-3·80], p=0·0068), but only with first-generation (2·86 [1·41-5·84], p=0·0038) not second-generation LAI antipsychotics. INTERPRETATION: In older individuals with schizophrenia, LAI antipsychotics were associated with lower risk of disease relapse and mortality, without increased risk of adverse events, than oral antipsychotics, except for elevated risk of extrapyramidal symptoms specifically related to first-generation LAI antipsychotics. Overall, LAI antipsychotics, especially second-generation medications, could be more broadly considered for long-term use among this population, particularly at an early stage of disease. FUNDING: National Natural Science Foundation of China.