Help with throughput

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

thorg12

Junior Member
15+ Year Member
Joined
Nov 19, 2005
Messages
122
Reaction score
15
Hello all
I'm interested in your thoughts on improving throughput.
I'm on a hospital and hospital system wide committee that is interested in improving "landing day" efficiency. The biggest aspect being admit order being placed ie "care complete," to the time they are upstairs in their inpt bed.
Some numbers
-small community hospital affiliated with big academic center
-36k Ed volume
-19 beds total 5 are fast track
-22% admission rate
-admitted pt arrival in Ed to in their bed upstairs 300 minutes (arrival to departure for admitted pts)
-admit order placed to in their bed upstairs 115 minutes (care complete to departure for admitted pts)

One of the ideas proposed was to have an admit hold area, with an "admitting team."
Consisting of case management, 2 nurses, and the house physician.
Theory is
-admit order placed pt immediately taken to admit hold area(only for remote tele, no drips or icu pts), which is out of Ed, nurse does admission intake, house doc writes order, case Managment determines inpt vs obs, it goes to inpt bed when it opens up.
This is to cut down on the 115 min time of admitted pt sitting in our Ed.
Are there benchmark numbers out there that are easily accessible?
Does this idea seem feasible?
What are your thoughts it suggestions?
Thanks for your input

Members don't see this ad.
 
I'll be interested to hear others thoughts on this as this is the classic EM problem.

Overall,your numbers don't seem bad. I don't have a reference but benchmarks I've heard are:
90 minutes - door to dispo decision for admitted patients (your shop is 185)
30 minutes - decision to transport for admitted patients (your shop 115 minutes)

For discharged patients - 90 minutes door to door time.

For admitted patients probably ANY strategy will be effective if you get the collective buy in from all involved groups (usually hospital administration, then hospitalists and lab and rads). I've worked places with a admitting unit - it worked when we or the hospitalists weren't busy but when we filled up either the hospitalists were too busy to churn people through there or we were boarding people in it(essentially our ER got bigger). Really, the biggest problem at that place was that the hospitalists weren't staffed to handle surges (they were staffed to handle the bare minimum of averages) and they had no incentive to work harder. They didn't really care about the safety of the patients in our waiting room or our boarded patients. They just wanted what qualified as "their problem" to be as small as possible. I'm not saying they were lazy at all, they were getting killed. They just weren't set up with a system that allowed them to care about anything other than what was right in front of them.

I've worked other places where there is collective buy in and trust and we just wrote admitting orders (short, simple, such as admit to Dr X and call Dr X when patient reaches the floor and inpt vs obs) the same time we sent the page out. They were pretty motivated to call us back quickly bc the patient was coming either way. To make this safer for all involved, we wouldn't write orders on the minority of patients where we thought they were on the fence about flor vs step down or hospitalist vs nephrologist. This was a minority of patients and more than 70% went right through. Those 70% were the main drivers of our numbers and the exceptions didn't hurt the numbers too much bc there weren't that many of them.

Lastly, it is worth pointing out that 78% of your patients are discharged. Small gains (30 minutes) on your door to door time for them equates to big gains for ED bed availability. Getting these patients out quick requires lab, rads, and nursing buy in and sometimes an extra provider in fastrack or a provider in triage set up. Some of these will also help your overall door to decision for all patients.

Most ED docs are not super interested in this stuff and see it as metrics over patients. While I understand that mindset, some of these metrics mater at a lot of shops. I don't like untriaged patients in our waiting room or triaged, known sick patients out there bc of a lack of beds. Fixing these metrics can fix some of these issues.

Nursing staffing affects this as well. I worked at one shop where the hospital administration set goals for these metrics and based incentives off the metrics. The group petitioned successfully for the hospital to owe all of the incentives for any month when there was more than a 2% drop in nursing staffing. This worked very nicely and the hospital became very motivated in providing good staffing and we were given the tools (nurses and techs) to adress the metrics. This is all to say, it should be a two way street. If the admin is asking you to fix numbers, they need to put their money where their mouth is (just as you are).
 
It's hard to give useful advice without knowing what's driving that 115 time frame. 30 min seems a reasonable and unreachable benchmark for admit to depart but 45 minutes should be doable with admin buy-in. Where is the delay coming from?

Possible sources:

1) Docs not sending patients up because they want to f/u on tests prior to patient going to floor - if you ask nurse managers this is a huge problem and if you ask anyone else they'll say it happens every once in a while. It's an easy fix if this is your problem. Just make the doc wait to call the admitting doc until everything's back. Of course this lengthens your door to dispo time but fewer peoples' bonuses are tied to that metric.

2) In-patient doc not calling back in a timely fashion - make the med staff enforce some sort of call to response metric.

3) Floor nurses refusing to take report - give ED nurse authority to send patient up with faxed report. Make the floor responsible for any safety events this generates.

4) Lack of personnel to transport patient once report is given - hire more techs (they're cheap)

5) Lack of in-patient beds - build more in-patient beds, allow boarding of stable patients in hallway upstairs, cancel elective surgeries when certain capacity benchmarks are reached, transfer to other hospitals with more capacity (preferably in-system), get housekeeping to actually staff for when patients' rooms need cleaned, make PACU and cath lab hold their post-procedure patients, etc.
 
Members don't see this ad :)
Thanks for all the feedback. I will bring all these things up.
Are biggest problem is admitting doc call back times and nursing not taking report.
Thanks again
 
The mantra you must repeat:
ED THROUGHPUT IS A HOSPITAL PROBLEM
ED THROUGHPUT IS A HOSPITAL PROBLEM

It is a very complex topic, and requires SERIOUS buy-in from not just the ED (RN, Uco, MD, PA, etc) staff, but from multiple OTHER departments and hospital administration.

I agree a bed request-to-leave the ED time of 115 minutes is too high. I've seen 60 minutes tossed around as a reasonable goal for a community ED, and 30 minutes (median, not mean) as an aggressive goal.

Thankfully, I work in a place where administration has bought into the idea of the entire hospital needing to help with ED flow. We are slightly smaller than you (30k/year). A few years ago, our request-to-rolling time was hovering around 90 minutes. Administration (meaning registration, senior nursing, etc) made changes to the way they physically processed bed requests, and just by doing that dropped times to 75 minutes. Free. Just by educating hospitalist, etc and changing the way we page, and having a cell-phone number available for "not returned" pages we dropped it to about 65 minutes. Again, free. Then, just in the past couple of months nursing switched from a phone report to a faxed/computer report which is based on the charting they have already done. This required a great deal of work on their part, but they had by in from their counterparts on the floor. Signout is well-hated by EVERYONE. Anyway, this change moved us down to 40-50 minutes median times... again basically free.

So you CAN do this, and it won't necessarily cost your hospital a ton of money. It will require support from people aside from the ED MDs...
 
Top