(P-634) Utilization Pattern of Antipsychotics in Patients with Dementia

Background: Patients with dementia often suffer from behavioral and psychological symptoms (BPSD) like agitation, aggression, and psychosis, thereby impairing patients’ quality of life. Evidence on the utilization pattern of antipsychotics remains limited in patients with dementia.

Objectives: To evaluate the pattern of antipsychotic prescription, persistence, switching, and the incidence of death in patients with dementia.

Methods: We conducted a retrospective cohort study using data of one million beneficiaries randomly sampled from Taiwan’s National Health Insurance Database, which includes records of diagnoses, medications and procedures from outpatient, inpatient, emergency rooms, and contracted pharmacies. We included patients newly received antidementia medications (donepezil, rivastigmine, galantamine, and memantine) from 2004 to 2013. Index date was defined as the first prescription date of antidementia medications. Patients were followed up for five years after the index date. To ensure that antipsychotics were used for BPSD, patients with a record of schizophrenia, bipolar disorder, and depression were excluded from the analysis. We calculated the rates of antipsychotic prescription, persistence, switching, and the incidence of death by year. We used Sankey diagram to present the dynamic utilization pattern of antipsychotics in patients with dementia.

Results: We identified 1853 patients with a mean age of 76.8 (SD: 8.0) years, of whom 61.7% were female. The annual prescribing rate of second-generation antipsychotics (SGA) was around 10.2% to 14.3%, while that of first-generation antipsychotics (FGA) was about 2.1% to 2.7%. We found that 46.0% to 54.5% of SGA users and 14.6% to 28.1% of FGA users persisted to the regimen in the next year. The switching rate from FGA to SGA ranged from 21.9% to 27.1%, which was much higher than that from SGA to FGA (3.4% to 6.2%). The mortality rate was 1.5% in the first year and overall 8.5% within 5 years. The mortality rates were higher in patients initiated FGA (4.2% to 9.1%) than those initiated SGA (0% to 2.5%).

Conclusions: The findings indicated that clinicians tend to prescribe SGAs rather than FGAs in patients with dementia. Patients initiating SGAs had better persistence, fewer switching, and lower incidence rates of mortality compared with those initiating FGAs. Understanding the utilization pattern of antipsychotics in patients with dementia lays a solid foundation for clinical decision making and comparative studies regarding specific outcomes like effectiveness or safety.