Bed downtime: the novel use of a quality metric allows inpatient providers to improve patient flow from the emergency department.

Background: Emergency department (ED) boarding time is associated with increased length of stay (LOS) and inpatient mortality. Despite the documented impact of ED boarding on inpatient outcomes, a disparity continues to exist between the attention paid to the issue by inpatient and ED providers. A perceived lack of high yield strategies to address ED boarding from the perspective of the inpatient provider may discourage involvement in improvement initiatives on the subject. As such, further work is needed to identify inpatient metrics and strategies to address patient flow problems, and which may improve ED boarding time.

Methods: After initial system analysis, our multidisciplinary quality improvement (QI) group defined the process time metric 'bed downtime'-the time from which a bed is vacated by a discharged patient to the time an ED patient is assigned to that bed. Using the Lean Sigma QI approach, this metric was targeted for improvement on the internal medicine hospitalist service at a tertiary care academic medical centre.

Interventions: Interventions included improving inpatient provider awareness of the problem, real-time provider notification of empty beds, a weekly retrospective emailed performance dashboard and the creation of a guideline document for admission procedures.

Results: This package of interventions was associated with a 125 min reduction in mean bed downtime for incoming ED patients (254 min to 129 min) admitted to the intervention unit.

Conclusion: Use of the bed downtime metric as a QI target was associated with marked improvements in process time during our project. The use of this metric may enhance the ability of inpatient providers to participate in QI efforts to improve patient flow from the ED. Further study is needed to determine if use of the metric may be effective at reducing boarding time without requiring alterations to LOS or discharge patterns.