Disagreement Between Clinicians and Score in Decision-Making Capacity of Critically Ill Patients*

{Objectives: To compare the assessment of decision-making capacity of ICU patients by attending clinicians (physicians, nurses, and residents) with a capacity score measured by the Mini-Mental Status Examination, completed by Aid to Capacity Evaluation if necessary. The primary outcome was agreement between physicians' assessments and the score. Secondary outcomes were agreement between nurses' or residents' assessments and the score and identification of factors associated with disagreement. Design: A 1-day prevalence study. Setting: Nineteen ICUs in France. Subjects: All patients hospitalized in the ICU on the study day and the attending clinicians. Interventions: The decision-making capacity of patients was assessed by the attending clinicians and independently by an observer using the score. Measurements and Main Results: A total of 206 patients were assessed by 213 attending clinicians (57 physicians, 97 nurses, and 59 residents). Physicians designated more patients as having decision-making capacity (n = 92/ 206 [ 45%]) than score (n = 34/206 [17%]; absolute difference 28% [95% CI, 20-37%]; p = 0.001). There was a high disagreement between assessments of all clinicians and score (Kappa coefficient 0.39 [95% CI, 0.29-0.50] for physicians; 0.39 [95% CI, 0.27-0.52] for nurses; and 0.46 [95% CI, 0.35-0.58] for residents). The main factor associated with disagreement was a Glasgow Coma Scale score between 10 and 15 (odds ratio, 2.92 [1.18-7.19]