Adherence to istradefylline in patients with Parkinson's disease: A group-based trajectory analysis

Background: Understanding the different patterns of adherence to istradefylline treatment is essential to identifying Parkinson's disease (PD) patients who might benefit from targeted interventions. Objectives: This descriptive study aimed to identify longitudinal istradefylline adherence patterns and to characterize factors associated with them. Methods: We identified PD patients aged 21-99 years who initiated istradefylline treatment in a Japanese hospital administrative database. Group-based trajectory modeling was used to model the monthly proportion of days covered over time to identify distinct 360-day adherence patterns. Factors associated with each adherence pattern were assessed using univariable multinomial logistic regression models. Results: Of 2088 eligible PD patients, 4 distinct adherence groups were identified: consistently high adherence (56.8%); rapidly declining adherence (25.8%); gradually declining adherence (8.5%); and gradually declining and then recovering adherence (9.0%). Compared to the consistently high adherence group, the other groups had the following characteristics associated with a likelihood of lower adherence: the rapidly declining adherence group received fewer dopamine agonists (63.8% vs. 69.4%), monoamine oxidase B (MAO-B) inhibitors (26.8% vs. 31.6%), and catechol-O-methyl transferase inhibitors (31.6% vs. 37.0%) and had a higher prevalence of anxiety/mood disorders (29.9% vs. 24.6%); the gradually declining adherence group received fewer MAO-B inhibitors (22.5% vs. 31.6%) and amantadine (8.4% vs. 16.1%) and had a higher prevalence of mild cognitive impairment/dementia (27.0% vs. 18.8%); and the declining and then recovering adherence group had a higher prevalence of anxiety/mood disorders (34.2% vs. 24.6%). Conclusions: Clinicians should be aware of the heterogeneous patterns of adherence to istradefylline.