Validation of the English version of the Parkinson Anxiety Scale in two Canadian Parkinson’s disease samples

Abstract

Introduction

The 12-item Parkinson Anxiety Scale (PAS) assesses the severity of anxiety symptoms in people with Parkinson’s disease (PwP). However, psychometric evidence supporting the English self-rated PAS remains largely limited to validation studies by the developers. This study aimed to reinforce validation of this scale through a comprehensive psychometric evaluation in PwP in Canada.

Methods

Data were drawn from two Canadian samples of 130 and 53 PwP, respectively, at a tertiary center. Acceptability, reliability, validity, and factor structure were assessed. The Mini International Neuropsychiatric Interview was administered to confirm anxiety diagnoses. Optimal cut-off scores suggesting the presence of an anxiety disorder were determined using receiver operating characteristic (ROC) curve analysis.

Results

The PAS and its three subscales demonstrated good acceptability and internal consistency reliability (Cronbach’s α > 0.70). Good convergent validity with the State-Trait Anxiety Inventory and divergent validity with scales assessing other non-motor symptoms were obtained. Confirmatory factor analysis supported the three-factor structure (comparative fit index = 0.987; Tucker-Lewis index = 0.983; root mean square error of approximation = 0.083; standardized root mean square residual = 0.087). Optimal cut-off scores were 12.5 for the total PAS (area under the ROC curve = 0.81, sensitivity = 0.89, specificity = 0.58), 8.5 for the persistent anxiety subscale, and 2.5 for the episodic anxiety and avoidance behavior subscales. Approximately 22 % of participants in the larger sample had a diagnosed anxiety disorder.

Conclusion

The English self-rated PAS is a valid and reliable instrument for screening and assessing severity and dimensions of anxiety in PwP.