Determinants of avoidable emergency department use at an urban, safety-net hospital
MetadataShow full item record
Nearly a third of all emergency department (ED) visits are for non-emergent conditions. Several factors contribute to non-emergent ED use, though the relative importance of these factors and how non-emergent ED users differ from those seeking similar care in primary care (PC) settings are less clear. Surveys were administered to adult, English-speaking, primary care patients seeking same-day, non-emergent care in the ED and PC clinic at an urban, safety-net hospital during normal clinic hours. ED patients were eligible if they had a primary care physician (PCP) located at the hospital and thus the ability to seek same-day care in the PC clinic. Surveys assessed sociodemographics and six major care-seeking factors (perceived urgency, cost, convenience, beliefs about alternative sites, access, and referral). Patient characteristics were compared between sites using t-tests and Fisher’s exact test, and multivariable logistic regression identified predictors of ED use. Compared to PC patients (n=61), ED patients (n=59) were significantly more likely to be male, middle-aged, homeless, Medicaid-insured, and unmarried with a trend toward being non-White and less educated. ED patients were significantly more likely to perceive the cost of an ED and PC visit to be the same/free (69.1% vs. 37.1%; p<0.01) and to believe the ED would provide higher quality care (24.6% vs. 3.6%; p<0.001). PC patients were more likely to consider their doctor’s office as their usual source of care (83.1% vs. 37.9%; p<0.0001) and believe it is easy to make PCP appointments on short notice (74.5% vs. 54.2%; p=0.04). In the adjusted model, patients whose usual source of care was not the doctor’s office had the highest odds of non-emergent ED use (aOR 4.25, 95% CI 1.28–15.20), and patients reporting ease of scheduling PCP appointments on short notice had significantly lower odds of ED use (aOR 0.19, 95% CI 0.04–0.67). Non-emergent patients in the ED and primary care clinic differed by sociodemographics, usual source of care, and perceptions of cost, care quality, and ease of scheduling appointments. Opportunities exist to reduce unnecessary ED use in similar populations by promoting the primary care clinic as a routine and easily accessible source of high-quality care.