Validity and reliability of the Persian version of the confusion assessment method for intensive care units

Delirium is prevalent in the intensive care unit (ICU), especially among mechanically-ventilated patients. Delirium is associated with a significant increase in adverse outcomes but it usually remains undiagnosed, making it necessary to develop and validate diagnostic tools. This study determined the validity and reliability of the Persian version of the Confusion Assessment Method for Intensive Care Units) CAM_ICU( in Iran.

 This cross-sectional study was conducted in open-heart ICU of three university hospitals of Tehran, Iran. After piloting the translated CAM-ICU on 10 patients and refining the translated scale accordingly, 40 ventilated patients were consecutively selected and screened for delirium by two independent evaluators (applying CAM-ICU), and one psychiatrist (using DSM-IV criteria for diagnosis). Inter-rater reliability between the two evaluators was assessed by the Kappa coefficient. Validity indices (i.e., sensitivity and specificity) of the Persian-CAM-ICU and 95% confidence intervals were calculated, given the psychiatrists’ diagnosis as the reference standard. Data were analyzed in Stata software (v. 11).

Of 40 selected patients, CAM_ICU detected delirium in 30%. The Persian-CAM-ICU had a sensitivity and specificity of 75% and 96%, and a positive and negative predictive value of 92% and 85%, respectively. Youden’s J statistic of the scale was 71%. Each of the four domains of the CAM-ICU showed a sensitivity and specificity of more than 69% and 90%, respectively, suggesting acceptable construct validity. There was good agreement between the two evaluators in terms of delirium diagnosis with the Persian-CAM-ICU (kappa coefficient = 0.74, P<0.0001).

The Persian version of the CAM_ICU is an effective, valid and reliable diagnostic tool in critically ill ICU patients. Application of the scale is recommended for the prompt diagnosis and prevent potential delirium in ventilated patients.