RESUMO
OBJECTIVE: To validate the Spanish version of the CAM-ICU. DESIGN: Validation study. The inter-rater reliability and validity were evaluated using a blind comparison with the DSM-IVTR criteria (reference standard) in a group of patients hospitalized in an Intensive Care Unit (ICU). SETTING: Medical-surgical ICU. PATIENTS: 129 adults with RASS >-4 score, within the 24-48 h of ICU admission. Independent application of the CAM-ICU by two intensive care specialists and evaluation by a psychiatrist using the DSM-IVTR criteria for delirium. RESULTS: Prevalence of delirium in the sample was 26.4%, this being and 48.3% in the 29 patients who received mechanical ventilation. Inter-rater reliability in the whole sample according to the Kappa index was 0.792 (95% CI 0.662-0.922, p<0.001) and in those who received mechanical ventilation was 0.726 (95% CI 0.487-0.965, p<0.001). The sensitivity, specificity, positive predicted value (PPV), negative predicted value (NPV), positive likelihood ratio (+LR), and negative likelihood ratio (-LR) were 79.4% (95% CI 63.2-89.7%); 97.9% (95% CI 92.6-99.4%); 93.1% (95% CI 78.0-98.1%); 93.0% (95% CI 86.3-96.6%); 37.72 (95% CI 9.47-150.20) and 0.21(95% CI 0.11-0.41), respectively, in the whole sample. In mechanically-ventilated patients, sensitivity, specificity, PPV, NPV, +LR and -LR were 92.9% (95% CI 68.5-98.7%); 86.7% (95% CI 62.1-96.3%); 86.7% (95% CI 62.1-96.3%); 92.9% (95% CI 68.5-98.7%); 6.96(95% CI 1.90-25.51) and 0.08 (95% CI 0.01-0.55), respectively. CONCLUSION: The Spanish version of the CAM-ICU is reliable. Its sensitivity, specificity and predictive values should be considered good.