Differences in long-term mortality for different emergency department presenting complaints

Abstract

Objectives: To characterize long-term mortality based on previous emergency department (ED) presenting complaints. Methods: The authors followed, for 10 years, all of the 12,667 nonsurgical patients visiting an ED during 1995/1996. Differences in standardized mortality ratio (SMR) depending on presenting complaints were then investigated. Results: During follow-up, 5,324 deaths occurred (mortality rate 6.6 per 100 person-years at risk), giving a SMR of 1.33 (95% CI = 1.30 to 1.37, p < 0.001) when compared with the expected mortality in the catchment area. Different presenting complaints were associated with different long-term mortality rates, independent of age and gender (p < 0.0001). The subjects with seizures had the highest SMR (2.62, 95% CI = 2.13 to 3.22) followed by intoxications (2.51, 95% CI = 2.11 to 2.98), asthmalike symptoms (1.84, 95% CI = 1.65 to 2.06), and hyperglycemia (1.67, 95% CI = 1.42 to 1.95). The largest complaint group, chest pain, had a 20% higher mortality rate than the background population (95% CI = 1.13 to 1.26). Patients with a discharge diagnosis of myocardial infarction, but without chest pain as the presenting complaint, had an increased long-term mortality (hazard ratio [HR] 1.70, 95% CI = 1.15 to 2.42) compared to the group with chest pain. In contrast, stroke patients without strokelike symptoms had a reduced mortality (HR 0.74, 95% CI = 0.65 to 0.84) compared to patients with strokelike symptoms. Conclusions: Long-term age- and gender-adjusted mortality is the highest with seizures out of 33 presenting complaints and differs markedly between different ED admission complaints. Furthermore, depending on the admission complaint, long-term mortality differs within the same discharge diagnosis. Hence, the presenting complaint adds unique information to the discharge diagnosis regarding long-term mortality in nonsurgical patients. © 2008 by the Society for Academic Emergency Medicine.