The Centers for Medicare & Medicaid Services (CMS) have traditionally utilized the emergency department (ED) metrics to gauge operational performance. CMS has identified timeliness and efficiency as core attributions for the ED, specifically length of stay (LOS) and boarding time. Prolonged times reduce the quality of care and can lead to increased inpatient length of stay and poor clinical outcomes. Furthermore, lengthy LOS has been associated with decreased patient satisfaction. Prior literature reports that higher staffing levels may result in earlier discharges and improved metrics. However, financial barriers may prevent hospitals from hiring more staff. Given the current allocation of staffing, we hypothesize that our novel queueing algorithm can reduce metrics times without increasing staffing. The intervention was based on a novel fluid queueing model, which quantifies the impact of the assignment of available ED and ED Inpatient (EDIN) nursing staff to the different areas of an urban quaternary ED at the beginning of each shift. The algorithm accounts for two stages of care (in-treatment and boarding) and two types of nursing providers, ED and EDIN nurses, and is based on the current number of patients at each ED area at the beginning of the shift, boarding times, fraction of admitted patients, average treatment times, and the time-dependent arrival rates with the goal to minimize nurse idleness. The study involved two 2-week periods; the period before the intervention (Off Period), March 5, 2018, to March 18, 2018, and the period after the intervention (On Period), March 19, 2018, to April 3, 2018. The data contains operational level information (times stamps for arrival, first treatment, discharge, change to admit status as well as assigned area and disposition) and patient-level demographic and medical information (age, sex, race and ethnicity, insurance type, ESI level, and Elixhauser score). We measured the impact of the intervention on patient waiting times, evaluation to disposition decision, evaluation to discharge, and boarding times for admitted patients. Our data consists of 2383 Off Period and 2617 On Period patient visits, stratified by ED disposition. In both groups, “evaluation to discharge”, “disposition decision to admission,” and “ED arrival to discharge from ED”, the On Period was associated with statistically significant reductions in these time metrics. On the other hand, “ED arrival to evaluation” and “evaluation to disposition” demonstrated no statistical difference. Our novel fluid queueing model demonstrated improvements in ED metrics without showing a clinically significant increase in “ED arrival to evaluation time.” The algorithm had successfully shown that operationally based metrics can improve without increasing nurse staffing.TableAdmitted PatientsDischarged PatientsOff Period (n = 720)On Period (n = 765)p-valueOff Period (n = 1586)On Period (n = 1755)p-valueCharacteristicsAge63.907 (18.818)64.442 (18.507)0.58149.070 (18.614)50.901 (19.334)0.005ESI3.00 [2.00-3.00]3.00 [2.00-3.00]0.5083.00 [3.00-4.00]3.00 [3.00-4.00]0.015Elixhauser4.00 [2.00-7.00]4.00 [2.00-7.00]0.5741.00 [0.00-2.00]1.00 [0.00-3.00]0.005Sex (M)364 (50.6%)377 (49.3%)0.623676 (42.6%)741 (42.2%)0.815Race0.5870.438White384 (53.3%)409 (53.5%)712 (44.9%)781 (44.5%)Black110 (15.3%)125 (16.3%)261 (16.5%)323 (18.4%)Asian37 (5.1%)37 (4.8%)71 (4.5%)88 (5.0%)Pacific Islander8 (1.1%)4 (0.5%)20 (1.3%)22 (1.3%)Other99 (13.8%)117 (15.3%)272 (17.2%)301 (17.2%)Unknown82 (11.4%)73 (9.5%)250 (15.8%)240 (13.7%)Ethnicity0.3970.249Hispanic/Latino90 (12.5%)88 (11.5%)229 (14.4%)267 (15.2%)Not Hispanic/Latino353 (49.0%)402 (52.5%)631 (39.8%)735 (41.9%)Unknown277(38.5%)275 (35.9%)726 (45.8%)753 (42.9%)Insurance0.1460.005Private146 (20.3%)134 (17.5%)767 (48.4%)767 (43.7%)Medicaid119 (16.5%)105 (13.7%)395 (24.9%)433 (24.7%)Medicare365 (50.7%)406 (53.1%)320 (20.2%)441 (25.1%)Self-Pay9 (1.2%)15 (2.0%)104 (6.6%)114 (6.5%)Unknown81 (11.2%)105 (13.7%)0 (0.0%)0 (0.0%)Operational MetricsWaiting Time (min)ED arrival to Eval17.162 (21.140)19.086 (26.865)0.12715.719 (20.885)16.888 (24.572)0.141Treatment Time (hrs)Eval to Disposition5.064 (3.328)5.214 (3.819)0.423Eval to Discharge21.700 (16.494)19.158 (13.092)<0.0016.360 (15.023)5.282 (3.449)0.004Boarding Time (hrs)Dispo to Admit16.717 (15.712)14.069 (12.283)<0.001Total LOS (hrs)ED Arrival to DC22.010 (16.539)19.468 (13.147)0.0016.619 (15.007)5.552 (3.486)0.004Mean (standard deviation) or median [interquartile range] of the outcome measures during the OFF and ON periods. Open table in a new tab