Rounding first on discharging patients does not lead to earlier discharge times or shorter length of stay

Clinical Question

Does rounding first on patients with anticipated discharges result in earlier discharge and shorter length of stay?

Bottom line

Although considered a strategy to improve throughput, prioritizing rounding on discharging patients first thing in the morning did not result in earlier discharge times and did not reduce hospital LOS. Despite asking the intervention group to place a discharge order as early as possible, the average order placement time for both groups occurred in the afternoon. This may reflect other inefficiencies in the 3 hospitals in which this study took place. 1b

Study design: Randomized controlled trial (nonblinded)

Funding: Government

Setting: Inpatient (ward only)


Nita Shrikant Kulkarni, MD
Assistant Professor in Hospital Medicine
Northwestern University
Chicago, IL

Discuss this POEM



Impact assessment




Discharge is complex. Most patients being discharged have some idea by the night prior. Even if the doctor enters an order there are other limitations like nursing rounds. Removal of IV’s discharge instructions. Physio etc. Rides need to be available too. the order is only a small piece.



Makes sense



Makes sense


incomplete information

I was disappointed with the information provided. No strategies were identified. Was it at the discretion of the physicians or team? The conclusion stated could potentially cause teams to abandon a strategy that does have potential to help patient flow given appropriate physician buy-in and involvement

Pieter Richard Verbeek

Focus on discharging patients

Interesting study. Not too surprised by the results though. To me, I would have engaged the nursing staff to be participants since their typical daily work flow probablywouldn't change as a result of a physicians writing an earlier discharge order. Concerns expressed by the physicians e.g. risk in not seeing sicker patients earlier of course is simply conjecture likely without any evidence to support this notion.