The management of and prognosis for delirium are affected by its subtype: hypoactive, hyperactive, mixed, and none. The DMSS–4, an abbreviated version of the Delirium Motor Symptom Scale, is a brief instrument for the assessment of delirium subtypes. However, it has not yet been evaluated in an intensive care setting.
We performed a prospective/descriptive cohort study in order to determine the internal consistency, reliability, and validity of the relevant items of the DMSS–4 versus the Delirium Rating Scale–Revised-98 (DRS–R-98) and the original DMSS in a surgical intensive care setting.
A total of 289 elderly, predominantly male patients were screened for delirium, and 122 were included in our sample. The internal consistency of the DMSS–4 items was excellent (Cronbach's α = 0.92), and between the DMSS–4 and DRS–R-98 the overall concurrent validity was substantial (Cramer's V = 0.67). Within individual motor subtypes, concurrent validity remained at least substantial (Cohen's κ = 0.65–0.81) and sensitivity high (69.8 to 82.2%), in contrast to those of the no-motor subtype, with less validity and sensitivity (κ = 0.28, 22%). Similarly, total concurrent validity between the DMSS–4 and the original DMSS reached perfection (Cramer's V = 0.83), as did agreement between the subtypes (κ = 0.83–0.92), while sensitivity remained high (88.2–100%). Only in those with delirium with no-motor subtype was agreement moderate (κ = 0.56) and sensitivity lower (67%). Specificity was high across all subtypes (91.2–99.1%). The DMSS–4 yielded very sensitive ratings, particularly for hypoactive and hyperactive motor symptoms, and interrater agreement was excellent (Fleiss's κ = 0.83).
We found the DMSS–4 to be a most reliable and valid brief assessment of delirium in characterizing the subtypes of delirium in an intensive care setting, with increased sensitivity to hypoactive and hyperactive motor alterations.
* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.
Usage data cannot currently be displayed.