Safety of Atypical Antipsychotics in the Elderly with Parkinson’s Disease

Objectives: According to the 2015 American Geriatrics Society (AGS) Beers criteria, except for aripiprazole, clozapine, and quetiapine, antipsychotic medications are considered generally inappropriate in PD. However, limited data exists regarding safety of atypical antipsychotics in general and inappropriate atypical antipsychotics in specific in patients with PD. This study evaluated the incidence and predictors of inappropriate atypical antipsychotic agents among older patients with PD; and the risks of pneumonia and mortality in older patients with PD using inappropriate atypical antipsychotic agents. Methods: A retrospective design involving Minimum Data Set (MDS) linked Medicare claims data was used to examine incidence and predictors of inappropriate atypical antipsychotic agents and to evaluate the study hypotheses that there are higher risks of pneumonia and all-cause mortality among older patients with PD using inappropriate atypical antipsychotics when compared to the three selected atypical antipsychotic agents (i.e. aripiprazole, clozapine, and quetiapine). The inappropriate atypical antipsychotics encompassed olanzapine, asenapine, brexpiprazole, iloperidone, lurasidone, paliperidone, risperidone, and ziprasidone. The study sample was selected from a cohort of older adults with depression. Multivariable logistic regression was used to examine various sociodemographic and clinical factors associated with inappropriate antipsychotic use in PD based on Andersen’s Behavioral Model. Safety evaluation involved a propensity-matched approach to adjust for the selection bias across antipsychotics within the multivariable context of the Andersen Behavioral Model. Cox proportional hazards regression model stratified on matched pairs was used to evaluate the safety profile of antipsychotics in PD. Results: There were 13,352 patients aged 65 years or older with PD diagnosis and without schizophrenia/ bipolar disorder who started one atypical antipsychotic agent in 2008-2009. The incidence of atypical antipsychotic use was 17.50% in 2-year follow-up. The most frequently used inappropriate antipsychotics were risperidone (22.95%) and olanzapine (11.25%). The likelihood of inappropriate antipsychotic use was higher for patients who had dementia or Chronic Obstructive Pulmonary Disease (COPD). Conversely, patients who were taking levodopa, dopamine agonists, Catechol-O-methyltransferase (COMT) inhibitors, Monoamine Oxidase (MAO) inhibitors type B, or amantadine were less likely to receive inappropriate antipsychotics. For the second and third objectives, the analysis involved 6-month washout and follow-up periods. There were 12,076 patients in the matched propensity score cohort. The Hazard Ratio (HR) of pneumonia was 1.23 (95% CI: 1.10 – 1.36) and the HR of all-cause-mortality was 1.13 (95% CI: 1.01 - 1.28) for patients who used inappropriate vs. appropriate atypical antipsychotics. There was a significant association between pneumonia and death. Conclusions: More than one-third of PD patients received inappropriate agents among those who were treated with atypical antipsychotics in this study. Various socio-demographics and clinical factors were associated with inappropriate antipsychotic use among older patients with PD. The risks of pneumonia and all-cause-mortality were significantly higher for inappropriate atypical antipsychotic users in comparison to appropriate antipsychotic group. This study provided a strong evidence base regarding the safety of atypical antipsychotic use in older patients with PD. The study findings can help in optimizing the use of these medications to improve quality of geriatric care in PD.