Anxiety is a prevalent and disabling condition in Parkinson's disease (PD). The lack of anxiety rating scales validated for this population hampers research into anxiety in PD. The aim of this study is to assess the clinimetric properties of the Hamilton anxiety rating scale (HARS), the Beck anxiety inventory (BAI), and the hospital anxiety and depression scale (HADS) in PD patients.
Three hundred forty-two PD patients underwent a standardized assessment including a structured interview for diagnostic and statistical manual diagnoses of anxiety disorders and completion of the HARS, BAI, and HADS. Inter-rater reliability of the HARS was assessed in 60 patients; test-retest reliability of the BAI and HADS in 213 and 217 patients, respectively.
Thirty-four percent of patients suffered from an anxiety disorder, whereas an additional 11.4% had clinically significant anxiety symptoms in the absence of a diagnosis of anxiety disorder. Acceptability, score distribution, and known groups validity over different levels of anxiety were adequate. Inter-rater reliability for the HARS and test-retest reliability for the BAI and HADS were good. The HARS, but not the BAI and HADS, had a satisfactory inter-item correlation, convergent validity and factorial structure. For all scales, the positive predictive value was poor, and the negative predictive value was moderate.
Given the adequate known groups validity of all three rating scales, each of these scales is likely to be useful in clinical practice or research for evaluation of symptom severity. Limitations in the construct validity of the anxiety scales in this study raise questions regarding suitability for their use in PD. © 2011 Movement Disorder Society