Purpose: To improve DBN rate on a medicine unit through improved interdisciplinary communication.
Description: An interdisciplinary, multifaceted intervention was implemented on a 34-bed inpatient medicine teaching unit. The unit is delocalized, with 9 hospitalist teams and 2 subspecialty teams, and includes coverage by housestaff and nurse practitioners. Prior to our intervention, we had two interdisciplinary meetings during the day, at 10:00am to discuss care and disposition and 3:15pm to discuss early discharges.
In our intervention, we focused on concerted interdisciplinary communication and ensuring follow-through on DBNs. A daily email was sent after the 3:15pm check-in to the attending and interdisciplinary leads: nursing manager, clinical coordinator, social workers, case manager, and business associate. This included specific tasks that were required by discipline (social work, case management, physician) to achieve the DBN for the following morning. Additionally, the case manager visited each patient and caregiver flagged for next day’s DBN to confirm understanding of the following day’s discharge timing and plan. Lastly, each morning the team initiated communication to the hospitalist attending via page or phone as a reminder about DBN and address any barriers in real-time. The interdisciplinary team met bi-monthly to review cases and discuss and improve the process.
Baseline data (1/1/2014– 9/30/2015) were compared to post-intervention data (10/1/2015–9/30/2016). The main outcomes of interest were: 1) discharges before noon, 2) mean length of stay (LOS), and 3) occupancy rate. Significance of pre-post differences were assessed by Chi-square test for DBN and by t-test for LOS and occupancy. In addition, we applied a quasi-experimental interrupted time series design with segmented regression analysis to estimate changes before and after our intervention.
Conclusions: There was a significant increase in DBN rate from 8.9% to 29.8% (P<0.001) with no significant difference in LOS (7.6 vs. 7.7; P=0.929). There was a small increase in occupancy (96.2% vs. 98.3%; P<0.001). When adjusting for baseline trends, the segmented regression analysis showed an acute post-intervention increase in DBN (change in intercept +18.09%; P<0.001), which remained stable over the duration of the intervention period (Figure).
We have shown that the DBN rate can be significantly improved through enhanced communication and implementation of a highly structured interdisciplinary protocol.