Delirium is a severe psychiatric syndrome that is highly prevalent in elderly general hospital patients. However, the diagnosis of delirium is often missed. The use of rating scales can be helpful in detecting and measuring delirium symptom severity. This article reviews recent developments regarding psychometric qualities, measurement goals, content and rating procedures of some of the available rating scales in clinical practice. Literature from the Medline files up to 2008 were collected, using the following search entries: delirium, (acute) confusion, assessment/rating scale and screening. Articles were selected if their title or summary were related to the development or applicability of delirium rating scales. The reference lists of relevant articles were searched for additional references. The rating scales were split up according to their purposes, screening or severity rating and were discussed for the following aspects: content (theoretical background, rating domains and items), interview (duration, format and assessors’ expertise) and psychometric qualities (reliability and validity). There were eight screening scales (Confusion Assessment Method [CAM], CAM for the Intensive Care Unit [CAM-ICU], Intensive Care Delirium Screening Checklist [ICDSC], Delirium Symptom Interview [DSI], NEECHAM Confusion Scale [NEECHAM], Cognitive Test for Delirium [CTD], Delirium Observation Screening [DOS] and Nursing Delirium Screening Scale [Nu-DESC]) and seven severity rating scales (Delirium Rating Scales [DRS], Memorial Delirium Assessment Scales [MDAS], Confusional State Evaluation [CSE], Delirium Severity Scales [DSS], Delirium Index [DI], Delirium-O-Meter [DOM] and Delirium Detection Scale [DDS] ) selected for further research. The CAM, NEECHAM and DOS and the CAM-ICU for the ICU appear to be the most suitable as screening instruments, depending on the type of rater (trained) physician or nurse. The (revised) Delirium Rating Scale (DRS-R-98) and the DOM appear to be particularly useful for measuring delirium severity or monitoring change.
Read full abstract