Background:
Several studies have presented clear evidence that emergency department (ED) crowding contributes to poor patient outcomes. Recent efforts to reduce ED crowding have focused on decreasing the ED boarding time of admitted patients, which will be used by the Centers for Medicare and Medicaid Services as a measure in pay‐for‐reporting in 2013. Hospitalists play an important role in safely and expeditiously admitting patients from the ED, and the aim of our study was to evaluate whether hospitalist‐led efforts to improve the time from ED admission decision to written inpatient orders had an impact on ED boarding times.
Methods:
Since January 2010, the Division of Hospital Medicine has focused on decreasing the time from the ED's decision to admit the patient to the admitting team writing inpatient orders at a large, 600‐bed urban academic medical center. Educational interventions to hospital medicine house staff and attendings included lectures on the dangers of ED crowding and monthly feedback of ED throughput times to inpatient teams. System changes like a 1‐page order set that could be completed by either the emergency physician or the admitting team were also implemented with widespread adoption. Using a retrospective cohort design, we compared ED decision to admission orders times (D2O) with ED boarding times among 9665 patients admitted through the ED to a medicine floor from January 2010 to May 2012. Linear regression was used to determine the effect of ED D2O on ED boarding times over the study period.
Results:
Among 9665 patients admitted from the ED to the medicine ward, we found that ED D2O time accounted for 31% (SD ± 21%) of the overall ED boarding time (Figure). Over the 29‐month study period, mean ED D2O time decreased by 28% (53 minutes; 95% CI, 45–61 minutes; P < 0.001) from 186 to 133 minutes. Concurrently, mean ED boarding time decreased by only 4% (23 minutes; 95% CI, 6–39 minutes; P = 0.008) from 555 to 532 minutes (Figure). After accounting for changes in ED D2O time, we found that every subsequent month in the study was associated with an average increase of 0.7 minutes in ED boarding time, for an estimated cumulative increase of 19 minutes (95% CI, 3.7–35 minutes; P = 0.02) over the study period. Neither ED census nor hospital census changed appreciably (P > 0.25), and adjustment for these factors did not change our results.
Conclusions:
Our results demonstrate that ED D2O time represents a sizable proportion of the overall ED boarding time, and hospitalist‐led efforts to decrease this time were successful. However, an equivalent decrease in overall ED boarding time was not observed, underscoring the observation that prolonged ED boarding times are multifactorial in etiology and that substantive reductions will require a multidisciplinary approach.
