Predictors of frequent emergency department utilization in Southeastern Pennsylvania

We sought to determine socioeconomic and demographic factors that predict frequent emergency department (ED) use among asthmatics in Southeastern Pennsylvania.

Retrospective cohort study using the 2004 Philadelphia Health Management Corporation's Southeastern Pennsylvania Household Health Survey.

Health behaviors were surveyed using telephone interviews (random-digit dialing) in Bucks, Montgomery, Delaware, Philadelphia, and Chester counties. Response rate was 30%.

Of 13,342 respondents, 1,799 (13%) stated that they had asthma. The mean reported visits in 12 months were as follows: clinic--1.2 (SD 2.7); doctor's office--4.7 (SD 5.6); ED--0.8; (SD 1.8); overnight hospitalizations--0.4 (SD 0.9). Most (91%) reported that a primary care doctor was their principal source of care, whereas 2% reported the ED as their principal source of care. Frequent ED use (>or=3 visits/12 months) was reported in 180 (10%). Frequent ED use was associated with race, education, diabetes, mental illness, smoking at home, and language barriers (p < 0.001) as well as indicators of severity of illness (clinic visits, office visits, prescription for asthma medication, p < 0.001). Using multivariable logistic regression, predictors of frequent ED use were: prescription for asthma meds (OR 2.3, 95%CI 1.4-3.7), having Medicaid or Medical Assistance (OR 1.7, 95%CI 1.1-2.6), having a high school education or less (OR 1.5, 95%CI 1.0-2.3), number of clinic visits (OR 1.1, 95% CI 1.1-1.1), office visits (OR 1.1, 95%CI 1.1-1.1), number of children living in the house (OR 1.2, 95% CI 1.0-1.3), and living in Philadelphia (OR 2.0, 95% CI 1.3-3.1).

Frequent ED use is associated with demographic factors, geographic factors, and markers of severity of illness. While care guidelines recommend prompt referral to a primary care physician to prevent ED visits, given the level of primary care use in this population, frequent primary care attendance alone may not prevent asthma-related ED visits.