Objective: Delirium has been characterized into its subtypes-hypoactive, hyperactive, mixed, or no motor subtype-along with the use of the Delirium Motor Symptom Scale (DMSS). The German version of this scale (DMSS-G), however, has not yet been validated. Method: We determined internal consistency, reliability, and validity of the DMSS-G in the surgical intensive care unit, using DSM-IV-TR criteria and the Delirium Rating Scale-Revised-98. Results: In total, 289 patients were included, and out of these, 122 were delirious. The DMSS-G showed excellent internal consistency (Cronbach's α = 0.92) and interrater reliability (Fleiss κ = 0.83). Additionally, the overall concurrent validity was substantial (Cramer's V = 0.69); within subtypes, hyperactive, hypoactive, or mixed, the concurrent validity remained at least substantial (Cohen's κ = 0.73-0.82) and the sensitivity ranged from 60% to 97%. In contrast, in those with no motor subtype, we found the concurrent validity (Cohen's κ = 0.31) and sensitivity to be low (22%). Overall, specificity for all individual subtypes was high (82% to 100%). The DMSS was very sensitive in both rating hyperactive and hypoactive motor symptoms of delirium. Conclusion: The DMSS-G is a highly reliable and valid instrument for detecting motor symptoms in delirium, which provides an accurate instrument to classify the motor subtypes of delirium.
Keywords: German version; ICU; delirium; intensive care unit; motor subtypes; validation.