We read Dr Gusmao-Flores and colleagues’ correspondence with great pleasure [
1] and are delighted that our article on AI and delirium [
2], triggered such literary analogies. We think that, in this complex field involving theory of mind, Shakespearean insights have much to teach us as they describe Macbeth’s fluctuating delirious state. Indeed, we have recently focused on the role of hallucinations (much like the dagger appearing to Macbeth) as an early manifestation in the development of delirium [
3]. We agree that natural language processing (NLP) is currently only an epidemiological tool to study delirium [
4] but, even as such, it is incredibly powerful. It allows the analysis of millions of written thoughts by observers in only a few hours. No such similar powerful epidemiological tools exist for this condition. However, one does not require a leap of faith to imagine that, soon, NLP will be used to read notes in real time, scanning for words such as “delirium” and once detected, trigger a delirium team to attend the patient and initiate immediate treatment. …
