SAN FRANCISCO — Managing patients who present to the emergency department with a transient ischemic attack in an ED observation unit rather than admitting them to the hospital reduced costs and lengths of stay for the initial visit, preliminary results of a prospective, randomized study showed.
Investigators randomized 46 patients with a transient ischemic attack (TIA) to be admitted to a hospital bed under the care of their primary physicians. Fifty-three patients were randomized to management by an emergency department physician in the ED observation unit, which was considered an accelerated diagnostic protocol.
The need to admit these patients to inpatient wards has been controversial, with some physicians suggesting that management in an ED observation unit might be more cost effective.
All patients underwent four diagnostic tests: carotid imaging, cardiac ultrasound, cardiac monitoring, and serial clinical evaluations. If all tests were negative, patients were discharged home on appropriate medications. Patients in the observation unit with positive diagnostic test results were considered for admission to a traditional hospital bed.
The ED observation unit group averaged 41 hours from arrival in the ED to discharge, a length of stay 23 hours shorter than the average 64 hours seen in the control group, Michael A. Ross, M.D., said in a poster presentation at the annual meeting of the American College of Emergency Physicians.
Seven patients managed initially in the ED observation unit were admitted to the hospital. Even though the length of stay for these patients averaged 167 hours, dramatically shorter stays (23 hours, on average) by the 46 patients who were discharged directly home from the observation unit lowered the overall mean length of stay in that group.
Mean total direct hospital costs (not including professional costs) were lower in the ED observation unit group ($1,392 per patient), compared with the inpatient group ($1,871 per patient), said Dr. Ross of William Beaumont Hospital, Royal Oak, Mich. Again, it was the dramatically lower costs for patients discharged from the ED observation unit ($767 per patient, vs. $5,038 per patient admitted to the hospital from the ED observation unit) that lowered overall costs in that group.
Four patients who were discharged home from the observation unit returned to the hospital within 30 days, compared with no return visits by patients admitted to the hospital either directly or from the observation unit. Costs of return visits were not included in the study.
The incidence of stroke within 90 days of the initial visit was similar between groups, however, with three strokes in the control group and four in the observation-unit group (two each among patients discharged or admitted).
The current study included patients with an emergency physician-confirmed TIA with a resolved deficit, not a crescendo TIA.
Head computerized tomography showed no acute infarct, bleed, or other acute pathology. These or a number of other conditions excluded patients from the study, including a possible embolic source, known carotid stenosis, nonfocal symptoms, severe headache or evidence of cranial arteritis, fever, previous stroke, severe dementia, history of intravenous drug use, residence in a nursing home, or other factors.
The investigators borrowed the accelerated diagnostic protocol concept for TIA from studies of similar ED protocols used to rapidly assess patients with chest pain who are at low to intermediate risk of acute cardiac ischemia.