PubMed2002
This study determined the accuracy of diagnosis and documentation of delirium in the medical and nursing records of 55 elderly patients with hip fracture (mean age = 78.4, SD = 8.4). These records were reviewed retrospectively on a patient's discharge for diagnosis of delirium, and for description of clinical indicators or symptoms of delirium. Additionally, all patients were monitored by one of the research members on days 1, 3, 5, 8, and 12 postoperatively for signs of delirium, as measured by the Confusion Assessment Method (CAM). Clinicians were blinded to the purpose of the study. According to the CAM criteria, the incidence of delirium was 14.5% on postoperative Day 1; 9.1% on postoperative Day 3; 10.9% on postoperative Day 5; 7.7% on postoperative Day 8; and 5.6% on postoperative Day 12. For those same days, no formal diagnosis of delirium or a description of clinical indicators was found in the medical records. In the nursing records, a false-positive documentation of 8.5%, 4%, 4.1%, 4.2%, and 5.9%, respectively was noted. False-negative documentation was found in 87.5%, 80%, 66.7%, 75%, and 50% of the cases on the respective days. Documentation of essential symptoms--namely onset and course of the syndrome--and disturbances in consciousness, attention, and cognition, were seldom or never found in the nursing records. However, behaviors of the hyperactive variant of delirium and which are known to interfere with nursing care were documented more often (e.g., 13.4% restless, 10.3% fidget with materials, 7.2% annoying behavior). Both medical and nursing records showed poor documentation and under-diagnosis of delirium. However, a correct diagnosis and early recognition of delirium may enhance the management of this syndrome.