Epidemiologic analysis of an urban, public emergency department's frequentusers

Citation
Jh. Mandelberg et al., Epidemiologic analysis of an urban, public emergency department's frequentusers, ACAD EM MED, 7(6), 2000, pp. 637-646
Citations number
20
Categorie Soggetti
Aneshtesia & Intensive Care
Journal title
ACADEMIC EMERGENCY MEDICINE
ISSN journal
10696563 → ACNP
Volume
7
Issue
6
Year of publication
2000
Pages
637 - 646
Database
ISI
SICI code
1069-6563(200006)7:6<637:EAOAUP>2.0.ZU;2-A
Abstract
Objectives: To determine how the demographic, clinical, and utilization cha racteristics of emergency department (ED) frequent users differ from those of other ED patients. Methods: A cross-sectional and retrospective cohort s tudy was performed using a database of all 348,858 visits to the San Franci sco General Hospital ED during a five-year period (July 1, 1993, to June 30 , 1998). A "frequent user" visited the ED five or more times in a 12-month period. Results: Frequent users constituted 3.9% of ED patients but account ed for 20.5% of ED visits. The relative risk (RR) of frequent use was high among patients who were homeless (RR = 4.5), African American (RR = 1.8), a nd Medi-Cal sponsored (RR = 2.1). Frequent users were more likely to be see n for alcohol withdrawal (RR = 4.4), alcohol dependence (RR = 3.4), and alc ohol intoxication (RR = 2.4). Frequent users were also more likely to visit for exacerbations of chronic conditions, including sickle cell anemia (RR = 8.0), renal failure (RR = 3.6), and chronic obstructive pulmonary disease (RR = 3.3). They were less likely to visit for all forms of trauma (RR = 0 .43). Survival analysis showed that only 38% of frequent users for one year remained frequent users the next year. However, 56% of frequent users for two consecutive years remained frequent users in the third year. Conclusion s: Frequent use of the ED reflects the urban social problems of homelessnes s, poverty, alcohol abuse, and chronic illness. Frequent use of the ED show s a high rate of decline from one year to the next. This rate of decline sl ows after the first year and suggests the existence of a smaller group of c hronic frequent users.