Ires Ghielen

63 Replication factor analysis in neuropsychiatric PD patients Introduction Parkinson’s disease (PD) is classified as a movement disorder but is also associated with non-motor symptoms. These include neuropsychiatric symptoms such as depression, anxiety, cognitive decline, and psychotic experiences [1, 2]. A large international study of both motor and non-motor symptoms in 411 PD patients showed that non-motor symptoms have, as a whole, a greater impact on health related quality of life compared to motor symptoms [3]. Martinez-Martin and colleagues [3] assessed 545 patients and showed that 98.4% of PD patients reported one or more non-motor symptoms. Depression and anxiety symptoms were in the top three of most prevalent non-motor symptoms. In PD, motor and non-motor symptoms are highly intertwined and show reciprocal influences [3-7]. About 75% of patients with fluctuations in motor symptoms experience mood and/or anxiety fluctuations in parallel [8]. For example, PD patients can experience anxiety and panic attacks during an ‘off’ phase, in which motor symptoms are re-occurring [5]. We previously performed a factor analysis on the Beck Anxiety Inventory (BAI) to investigate the different symptom dimensions of anxiety in PD patients [9]. We uncovered four ‘somatic’ (containing mostly physical symptoms) and one ‘affective’ dimension (containing purely emotional and cognitive anxiety symptoms). A regression analysis demonstrated that depressive symptoms were significantly associated with the ‘affective’ dimension of the BAI, motor and autonomic symptoms were significantly related to the ‘somatic’ dimensions of the BAI. The main results of this study were replicated in an independent sample of PD patients that had been referred for specialist psychiatric evaluation (under review). These study findings demonstrate that some aspects of anxiety in PD are strongly connected to the motor symptoms of the disease. However, it is yet unclear which specific characteristics of anxiety are responsible for the significant associations with motor symptoms. Also, it remains to be determined whether PD patients with high levels of anxiety show stronger connections between anxiety and motor symptoms compared to PD patients with low levels of anxiety. An approach that is useful for investigating associations between symptoms is network analysis [10]. With this method, it is possible to explore and visualize partial correlations between symptoms. This method can establish which symptoms are correlated (strongly or weakly, positively or negatively) and can be used to compare symptom network architectures between groups [10]. A network analysis approaches the multifaceted and complex nature of the relationships between anxiety and motor symptoms in PD by calculating the associations between 4