Background: Delirium is a complex medical disorder associated with high mor
bidity and mortality among elderly patients. The goals of our study were to
determine the prevalence of delirium in emergency department (ED) patients
aged 65 years and over and to determine the sensitivity and specificity of
a conventional clinical assessment by an ED physician for the detection of
delirium in the same population.
Methods: All elderly patients presenting to the ED in a primary acute care,
university-affiliated hospital who were triaged to the observation room on
a stretcher because of the severity of their illness were screened for del
irium by a research psychiatrist using the Mini-Mental State Examination an
d the Confusion Assessment Method. The diagnosis of "delirium" or an equiva
lent term by the ED physician was determined by 2 methods: completion of a
mental status checklist by the ED physician and chart review. The prevalenc
e of delirium and the sensitivity and specificity of the ED physician's cli
nical assessment were calculated with their 95% confidence intervals. The d
emographic and clinical characteristics of patients with detected delirium
and those with undetected delirium were compared.
Results: A sample of 447 patients was screened. The prevalence of delirium
was 9.6% (95% confidence interval 6.9%-12.4%). The sensitivity of the detec
tion of delirium by the ED physician was 35.3% and the specificity, 98.5%.
Most patients with delirium had neurologic or pulmonary diseases, and most
patients with detected delirium had neurologic diseases.
Interpretation: Despite the relatively high prevalence of delirium in elder
ly ED patients, the sensitivity of a conventional clinical assessment for t
his condition is low. There is a need to improve the detection of delirium
by ED physicians.