The Long-Term Effect of an Independent Capacity Protocol on Emergency Department Length of Stay: A before and after Study

Abstract

Purpose: In this study, we determined the long-term effects of the Independent Capacity Protocol (ICP), in which the emergency department (ED) is temporarily used to stabilize patients, followed by transfer of patients to other facilities when necessary, on crowding metrics.

Materials and methods: A before and after study design was used to determine the effects of the ICP on patient outcomes in an academic, urban, tertiary care hospital. The ICP was introduced on July 1, 2007 and the before period included patients presenting to the ED from January 1, 2005 to June 31, 2007. The after period began three months after implementing the ICP from October 1, 2007 to December 31, 2010. The main outcomes were the ED length of stay (LOS) and the total hospital LOS of admitted patients. The mean number of monthly ED visits and the rate of inter-facility transfers between emergency departments were also determined. A piecewise regression analysis, according to observation time intervals, was used to determine the effect of the ICP on the outcomes.

Results: During the study period the number of ED visits significantly increased. The intercept for overall ED LOS after intervention from the before-period decreased from 8.51 to 7.98 hours [difference 0.52, 95% confidence interval (CI): 0.04 to 1.01] (p=0.03), and the slope decreased from -0.0110 to -0.0179 hour/week (difference 0.0069, 95% CI: 0.0012 to 0.0125) (p=0.02).

Conclusion: Implementation of the ICP was associated with a sustainable reduction in ED LOS and time to admission over a six-year period.

Keywords: Crowding; clinical protocol; health resources.

Fig. 1. Conceptualized emergency department (ED) flow before and after introduction of the independent-capacity protocol. Dashed lines represent transfers from specific wards to other hospitals that were not analyzed in this study due to small numbers. Clear arrows represent discharges to home (from Cha WC, et al. Acad Emerg Med 2009;16:1277-83).

Fig. 2. Emergency department (ED) visits by year. After the intervention, the intercept changed from 119.1 to 115.1 (difference: 3.39, 95% CI: -1.43 to 8.21) (p=0.17) and was not statistically significant. The slope increased from 0.146 to 0.205 daily visits/week (difference: 0.060, 95% CI: 0.004 to 0.116) (p=0.04). IQR, interquatile range; SD, standard deviation; CI, confidence interval.

Fig. 3. Rate of transfer to community hospital from emergency ward by year (%). After implementing the ICP, the intercept increased from 3.21% to 5.50% (difference: 2.29, 95% CI: 0.27 to 4.31) (p=0.03). The slope increased from 0.0119% to 0.0194% per week (difference: 0.0075, 95% CI: -0.0159 to 0.0310) (p=0.53) and was not statistically significant. CI, confidence interval; ICP, Independent Capacity Protocol.

Fig. 4. Mean emergency department (ED) length of stay (LOS) by year. After the intervention, the intercept dropped from 8.51 to 7.98 hours (difference: 0.52, 95% CI: 0.04 to 1.01) (p=0.03), and the slope decreased from -0.0110 to -0.0179 hour/week (difference: 0.0069, 95% CI: 0.0012 to 0.013) (p=0.02). IQR, interquatile range; SD, standard deviation; CI, confidence interval.

Fig. 5. Waiting time of admitted patients by year. After the intervention, the intercept decreased from 33.6 to 31.1 hours (difference: 2.44, 95% CI: 0.84 to 4.05) (p=0.003). The slope decreased from 0.45 to -0.54 hour/week (difference: 0.100, 95% CI: 0.081 to 0.12) (p<0.001). IQR, interquatile range; SD, standard deviation; CI, confidence interval; ED, emergency department; LOS, length of stay.

Fig. 6. Comparison of trend of admission waiting time between study hospital emergency department and other level-1 emergency departments. The top line represents the trend of the study hospital, while the lower line represents the trend of the 15 other level-1 emergency centers.