Factors associated with ambulance use among patients with low-acuity conditions

The use of ambulances for low-acuity medical complaints depletes emergency medical services (EMS) resources that could be used for higher-acuity conditions and contributes to emergency department (ED) overcrowding and ambulance diversion. Objective. We sought to understand the characteristics of patients who use ambulances for low-acuity conditions. We hypothesized that patients who arrive to the ED by ambulance for low-acuity conditions are more likely to be members of vulnerable populations.

A secondary analysis was performed on the National Hospital Ambulatory Medical Care Survey (NHAMCS). We included only patients aged 18 years or older who were triaged to the "nonurgent" category upon presentation to the ED. To compare patients who arrived by ambulance with those who arrived by all other modes, multivariate logistic regression was performed using a generalized linear model, and adjusted relative risks (ARRs) were calculated.

A total of 16,109 records from 1997 to 2008 (excluding 2001-2002) were included in the analysis. Significantly higher rates of ambulance use for low-acuity conditions were associated with: 1) older age (ARR 1.30, 95% confidence interval [CI]: 1.18-1.43; per 10 years); 2) Medicare or Medicaid insurance (ARR 1.81, 95% CI: 1.36-2.41, and ARR 1.46, 95% CI: 1.12-1.91, respectively); 3) homelessness (ARR 3.30, 95% CI: 1.61-6.78); 4) arrival between 11 pm and 6:59 am (ARR 1.80, 95% CI: 1.43-2.27); and 5) certain chief complaint categories: psychiatric (ARR 1.78, 95% CI: 1.03-3.07), toxicologic/poisoning (ARR 3.26, 95% CI: 1.85-5.76), and neurologic/psychological (ARR 1.71, 95% CI: 1.34-2.18). Patients who arrived by ambulance were more likely than nonambulance patients to receive laboratory diagnostic tests (ARR 3.50, 95% CI: 2.80-4.39), radiographic imaging (ARR 2.26, 95% CI: 1.91-2.68), and admission to the hospital (ARR 3.99, 95% CI: 3.03-5.27).

Our study builds on a body of work highlighting the factors associated with ambulance transport to EDs, confirms that certain vulnerable populations disproportionately use ambulances, and may inform interventions aimed at increasing access to nonambulance transportation and urgent care for these patients.