Objective: Observation units may exclude geriatric patients (defined as age >or=65) due to the high rate of observation failure (admission to an inpatient unit) among these patients. We evaluated whether geriatric patients on a chest pain protocol are admitted to an inpatient unit from an emergency department (ED) observation unit at a higher rate than nongeriatric patients.
Methods: This was a retrospective chart review of all patients placed in the ED observation unit at the University of Utah Medical Center over a 14-month period from April 2006 to June 2007. The observation unit did not exclude geriatric patients nor did it exclude patients with a history of coronary disease; patients were admitted per the discretion of the attending ED physician. Patient information, including age, date of admission, history of coronary disease (defined as a previous myocardial infarction, stent, or coronary artery bypass graft), and admission to an inpatient unit from the observation unit, was recorded. Decision to admit to an inpatient unit was made by the consulting cardiologist. Results were analyzed using chi2 statistics.
Results: One hundred thirty-four geriatric patients were admitted to the observation unit under the chest pain protocol during the study period. Seventeen percent of these patients were admitted to an inpatient unit from the observation unit versus 10.7% of the 394 chest pain patients who were under age 65 (P = 0.048). Geriatric patients were more likely to have coronary disease (31.3%) than nongeriatric patients (20.8%; P = 0.013). We then performed a subanalysis on the 404 patients (92 geriatric, 312 nongeriatric) who had no history of coronary disease. Geriatric patients without a history of coronary disease had a 12% inpatient admission rate from the observation unit versus a 7.7% admission rate for nongeriatric patients without a history of coronary disease (P = 0.2).
Conclusion: Geriatric patients without a history of coronary artery disease were admitted to an inpatient unit at a rate consistent with a generally accepted observation failure rate of 10%. When screened appropriately, these patients may be appropriate for chest pain evaluation in the ED observation unit.