Diagnostic anticipation to reduce emergency department length of stay: a retrospective cohort study in Ferrara University hospital, Italy

Table 3 Multiple regression model predicting ED length of stay

	Chest Pain				Abdominal Pain				Non-Traumatic Bleeding
	β	95% CI for β		p-value	β	95% CI for β		p-value	β	95% CI for β		p-value
Diagnostic Anticipation	−28.9	−44.2	−13.6	< 0.001	6.1	−8.4	20.7	0.41	17.2	−8.9	43.4	0.20
Age, 5-year increase*	5.7	3.6	7.9	< 0.001	2.4	0.7	4.1	0.006	3.2	−0.2	6.6	0.066
Male gender	−8.1	−22.9	6.8	0.29	−24.1	−38.6	−9.5	0.001	8.3	−17.3	34.1	0.52
NEDOCS score, 10-point increase*	5.0	3.8	6.2	< 0.001	7.0	5.8	8.2	< 0.001	7.3	5.2	9.3	< 0.001
Imaging, 1 test increase*	17.1	9.6	24.6	< 0.001	28.2	21.2	35.2	< 0.001	56.6	41.3	71.8	0.006
Specialist consultations (vs no)	69.2	46.8	91.6	< 0.001	74.7	57.2	92.2	< 0.001	43.4	12.8	73.9	< 0.001
Hospitalization (vs discharge)	52.5	30.4	74.5	< 0.001	126.0	106.8	145.2	< 0.001	94.8	60.5	129.0	< 0.001
Yellow priority code (vs green)	41.2	20.5	61.8	< 0.001	50.6	35.1	66.2	< 0.001	5.8	−21.7	33.3	0.68

Abbreviations: NT = Non Traumatic; β = β coefficient; CI = Confidence Interval, DA = Diagnostic Anticipation; NEDOCS=National ED Overcrowding Study Score [15] at triage registration
*from the minimum value (0 for NEDOCS and Imaging Tests, 18 for Age)

ISSN: 1472-6963