Diagnosing Delirium

To the Editor: In their systematic review, Dr Wong and colleagues¹ supported the use of the CAM because of its brevity and ease of use. The article indicated that the CAM takes 5 minutes to administer; however, this does not include the time required to perform the formal cognitive assessments on which the validity of the CAM relies.²

Although the extra 10 minutes required to perform a formal cognitive assessment (such as the Mini-Mental State Examination [MMSE]) is unlikely to be burdensome when diagnosing a sporadic case of delirium, evidence indicates that the CAM is not reliable for bedside nursing surveillance in a real-world clinical setting.³ The good CAM test characteristics presented in the article by Wong et al reflect the ability of highly trained, specialized research nurses to perform the CAM accurately. When the CAM is used as part of a bedside nursing delirium surveillance system for at-risk patients, the amount of time and training required to replicate a valid CAM assessment is likely to be onerous for bedside nursing. In a study in which the CAM was used by bedside nurses for surveillance, the sensitivity rate was only 66%.⁴ Moreover, 15 or more minutes multiplied by the number of patients a nurse is monitoring can result in an unacceptable burden for nurses with other patient care responsibilities during a typical shift. The time pressures of clinical care compromise the ability of a clinician to correctly assess a sophisticated psychometric test like the MMSE and to apply its results to the scoring of the CAM algorithm.