1E-06 - Effects of Discontinuation and Tapering of Antipsychotics in Older Adults Living with Dementia: Target Trial Emulations using Electronic Health Records

Background: Prior evidence has found that antipsychotic use by those with dementia is associated with increased risk of adverse outcomes. It remains unknown whether antipsychotics can be discontinued safely and whether method of discontinuation impacts risk of adverse outcomes.

Objectives: To evaluate the effect abruptly discontinuing or tapering antipsychotic use has on the risk of stroke, pneumonia, fracture, hospitalisation, and death in those living with dementia.

Methods: We conducted target trial emulations using UK primary care data from the Clinical Practice Research Datalink linked with Hospital Episode Statistics and Office for National Statistics from January 1998 to March 2021. Patients aged ≥65 at diagnosis were included if they had no history of severe mental illness or stroke, had not had a fracture or pneumonia in the prior 3 years, and had continuous antipsychotic treatment for at least 12 weeks. We emulated hypothetical pragmatic trials for two discontinuation strategies: (1) continuing versus tapering (dose reduction of ≥50%) and (2) continuing versus abrupt discontinuation, after 12 weeks and 24 weeks of continuous antipsychotic treatment. The study outcomes were death, hospitalisation for delirium, stroke, pneumonia, or fracture within 2 years. In all emulated trials, the clone-censor-weight approach was used with a 6-month grace period. Patients were replicated and each set of clones assigned to each treatment strategy. Clones were artificially censored if they deviated from the assigned treatment. Inverse probability of censoring weights were estimated to account for selection bias due to censoring. Weighted pooled logistic regression models were used to estimate hazard ratios (HRs) of each discontinuation strategy on the risk of each outcome compared with continuation.

Results: 24,822 patients were included in the 12-week trials and 16,795 were included in the 24-week trials. Compared with continuation after 12 weeks of use, both tapering and abrupt discontinuation were associated with lower risk of delirium with HRs of 0.84 (95% CI: 0.65 to 0.94) and 0.88 (95% CI: 0.80 to 0.96) respectively. Tapering was associated with a lower risk of fracture with a HR of 0.81 (95% CI: 0.66 to 0.95). Discontinuation by either method was not associated with a difference in risk for any other outcome. Similar results were found for the 24-week trials.

Conclusions: Tapering or abrupt discontinuation after 12 or 24 weeks of continuous use was found to be associated with a lower risk of hospitalisation for delirium and tapering was found to be associated with a lower risk of fracture. Discontinuation by any method was not found to be associated with a higher risk of stroke, pneumonia, or death. Our results suggest that antipsychotics can be safely discontinued in those living with dementia.