Diagnostic accuracy of screening tools for depression and anxiety in cervical dystonia

Despite their high prevalence and impact, depression and anxiety are not routinely screened for, and accuracy of screening procedures is unknown in adult-onset dystonia. We evaluated accuracy parameters of selected self-rated scales for depression and anxiety in patients with idiopathic cervical dystonia (CD).

Two-hundred-and-ten patients with idiopathic CD were recruited from 10 movement disorders centers from the US, Canada, Australia, and UK. At the end of each botulinum toxin cycle, participants were administered the Adult Standard Mini-International Neuropsychiatric Interview (MINI) as reference standard for depression and anxiety. Participants completed 8 self-administered index instruments (2 for depression, 2 for anxiety, and 4 combining screening for both). Sensitivity, specificity, positive and negative predictive values, covariate-adjusted area under the receiver operating characteristic curve (AUC), and likelihood ratios were calculated for all instruments.

On the MINI, 8.6 % (100 % female) fulfilled criteria for current major depressive disorder and 10.5 % (91 % female) fulfilled criteria for any current disorder amongst panic, social anxiety or generalized anxiety disorders. For depression screening, all tools had an AUC higher than 0.80, with the two most accurate being the BDI-II (AUC 0.91, sensitivity 87.5 %) and the HADS-Depression (AUC 0.88, sensitivity 93.7 %). For anxiety screening, the only instrument showing clinical usefulness was the HADS-Anxiety (AUC 0.82, sensitivity 86.3 %).

Current major depression can be screened in CD with high degree of accuracy using different self-administered scales, whereas existing screening tools for anxiety perform worse. Dystonia-specific instruments are less accurate than scales developed for the general population.