Non-urgent use of emergency departments: populations most likely to overestimate illness severity

Patients' overestimation of their illness severity appears to contribute to the national epidemic of emergency department (ED) overcrowding. This study aims to elucidate which patient populations are more likely to have a higher estimation of illness severity (EIS). The investigator surveyed demographic factors of all non-urgent patients at an academic ED. The patients and physicians were asked to estimate the patients' illness severity using a 1-10 scale with anchors. The difference of these values was taken and compared across patient demographic subgroups using a 2-sample t-test. One hundred and seventeen patients were surveyed. The mean patient EIS was 5.22 (IQR 4), while the mean physician EIS was less severe at 7.57 (IQR 3), a difference of 2.35 (p < 0.0001). Patient subgroups with the highest EIS compared to the physicians' EIS include those who were self-referred (difference of 2.65, p = 0.042), with income ≤ $25,000 (difference of 2.96, p = 0.004), with less than a college education (difference of 2.83, p = 0.018), and with acute-on-chronic musculoskeletal pain (difference of 4.17, p = 0.001). If we assume the physicians' EIS is closer to the true illness severity, patients with lower socioeconomic status, lower education status, who were self-referred, and who suffered from acute-on-chronic musculoskeletal pain are more likely to overestimate their illness severity and may contribute to non-urgent use of the ED. They may benefit from further education or resources for care to prevent ED misuse. The large difference of acute-on-chronic musculoskeletal pain may reflect a physician's bias to underestimate the severity of a patients' illness in this particular population.