Delirium is a highly prevalent disease in the elderly and postoperative, cancer, and AIDS patients. However it is often misdiagnosed and mistreated. This may be partly due to the inconsistencies of the diagnosis itself. Delirium is best defined currently by an association of cognitive impairment and arousal disturbance. Three subtypes (hyperactive, hypoactive, mixed) receive a definition in the literature, but those definitions may vary from author to author according to the importance they give either to the motoric presentation of the delirium or to the arousal disturbance. Our aim is to point out the inconsistencies we found in the literature, but also to identify different paths that have been explored to solve them, that is, the suggestion to emphasize the arousal disturbances in defining the subtypes instead of the motoric presentations, which seem to be more fluctuating, and because of the fluctuating course of the disease to extend the observation over a period of time, which may improve the accuracy of the diagnosis. This is not without importance from a clinical standpoint. Subtypes of delirium may be explained by different pathophysiologic mechanisms, which remain partly unexplained, and may respond to specific treatments. There is a trend to isolate core symptoms (disorientation, cognitive deficits, sleep–wake cycle disturbance, disorganized thinking, and language abnormalities) so as to distinguish them from secondary symptoms that may be correlated with the different etiologies. Our contribution is also to challenge, with new data, the accepted belief that psychotic features are quite rare in the hypoactive type of delirium. We demonstrate that delusions and perceptual disturbances, although less frequent, are present in more than half of the patients with hypoactive delirium. The psychotic features are clearly correlated with a highly prevalent rate of patients', spouses', and caregivers' distress. The mixed subtype of delirium seems to have the worst prognosis, the hyperactive showing the best prognosis. The treatment of the agitated delirious patient is also more consensual. Haloperidol remains the gold standard in the treatment of delirium regardless of the clinical presentation, but the literature provides several alternatives that may prove more specific and have less adverse effects (atypical antipsychotics, psychostimulants, anesthetics).