Abstract
Delirium is highly prevalent in the general medical setting, and is associated with significant morbidity and mortality. In spite of this, under-recognition of delirium is a significant barrier to improving patient outcomes. This pilot study looked at validating a screening tool for delirium in hospitalized oncology patients. The screening tool consisted of a single question ‘Has the patient been confused lately?’ which was put to a relative or friend of the patient. The results of this study demonstrate that this question was both sensitive (80%) and specific (77%) in detecting the presence of delirium when compared with the psychiatric diagnosis of delirium according to DSM-IV. In conclusion, this single-item screening tool has utility in the prompt recognition of delirium, and is worthy of further investigation.