Health Outcomes of Discontinuing Antipsychotics After Hospitalization in Older Adults

Importance Among hospitalized older adults, prolonged use of antipsychotic medications (APMs) following hospital discharge may increase the risk of APM-associated adverse events. There are limited data on whether early discontinuation of APMs is associated with reduced adverse clinical outcomes compared with APM continuation after discharge. Objective To compare clinical outcomes between discontinuation vs continuation of APMs initiated to manage hospitalization-related delirium. Design, Setting, and Participants This population-based cohort study examining nationwide US Medicare claims data from July 1, 2013, through December 31, 2018, and data from a large deidentified US commercial health care database (Optum CDM) from July 1, 2004, through May 31, 2024, included adults aged 65 years and older without psychiatric disorders or previous use of APMs who filled an APM prescription within 30 days of hospital discharge. Using incidence density sampling, APM discontinuers (gap ≥45 days) were matched with continuers based on the type of APM prescribed, the time since their first APM prescription, and whether they had been admitted to intensive care units prior to the first APM prescription. Data analysis was performed from July 12, 2024, to December 25, 2024. Exposure Discontinuation vs continuation of APMs. Main Outcomes and Measures Propensity score matching was applied to adjust for 162 covariates. Study outcomes included rehospitalization, specific rehospitalization reasons, and all-cause mortality. Hazard ratios (HRs) were estimated using the Cox proportional hazards model; estimates from the 2 databases were further pooled using the fixed-effects meta-analysis model. Results A total of 13 712 propensity score–matched pairs were included, for an overall sample of 27 424 adults (discontinuers: mean [SD] age, 81.86 [7.26] years; 7400 [54.0%] female; continuers: mean [SD] age, 81.86 [7.27] years; 7360 [53.7%] female). During the median (IQR) follow-up of 180 (87-180) days, APM discontinuation vs continuation was associated with significantly lower risks of rehospitalization (HR, 0.89 [95% CI, 0.85-0.94]), inpatient delirium (HR, 0.87 [95% CI, 0.79-0.96]), fall-related emergency department visits or hospitalizations (HR, 0.77 [95% CI, 0.67-0.90]), hospitalization with urinary tract infection (HR, 0.79 [95% CI, 0.66-0.94]), and all-cause mortality (HR, 0.77 [95% CI, 0.69-0.86]). There was no statistical difference in the risks of pneumonia (HR, 0.88 [95% CI, 0.73-1.06]) or stroke (HR, 1.22 [95% CI, 0.97-1.53]) between discontinuers and continuers. Subgroups by dementia status, type and dose of APM prescribed, and duration of APM exposure showed consistent results. Conclusions and Relevance Based on 2 nationwide US cohorts including older adults without psychiatric disorders, APM discontinuation was associated with reduced risks of all-cause rehospitalization and mortality, suggesting the importance of minimizing the duration of APM use after acute hospitalization.