Validation study of the European Portuguese version of the Confusion Assessment Method (CAM)

Background: The Confusion Assessment Method (CAM) is the most widely used delirium screening instrument. The aim of this study was to evaluate the reliability and validity of the European Portuguese version of CAM. Methods: The sample included elderly patients (>= 65 years), admitted for at least 48 h, into two intermediate care units (ICMU) of Intensive Medicine and Surgical Services in a university hospital. Exclusion criteria were: score <= 11 on the Glasgow Coma Scale (GCS), blindness/deafness, inability to communicate and to speak Portuguese. For concurrent validity, a blinded assessment was conducted by a psychiatrist (DSM-IV-TR, as a reference standard) and by a trained researcher (CAM). This instrument was also compared with other cognitive measures to evaluate convergent validity. Inter-rater reliability was also assessed. Results: In this sample (n = 208), 25% (n = 53) of the patients had delirium, according to DSM-IV-TR. Using this reference standard, the CAM had a moderate sensitivity of 79% and an excellent specificity of 99%. The positive predictive value was 95%, indicating a strong ability to confirm delirium with a positive test result, and the negative predictive value was lower (93%). Good convergent validity was also found, in particular with Mini-Mental State Examination (MMSE) (r(s) = -0.676; p <= 0.01) and Digit Span Test (DST) forward (r(s) = -0.605; p <= 0.01), as well as a high inter-rater reliability (diagnostic k = 1.00; single items' k between 0.65 and 1.00). Conclusion: Robust results on concurrent and convergent validity and good reliability were achieved. This version was shown to be a valid and reliable instrument for delirium detection in elderly patients hospitalized in intermediate care units.