ED Throughput: A Fixable Problem


billboard RMWe put a man on the moon; surely we can shorten the wait in the emergency department. Two recent studies give practical steps for changing a department’s throughput trajectory.

We put a man on the moon; surely we can shorten the wait in the emergency department. Two recent studies give practical steps for changing a department’s throughput trajectory.


billboard W

As I travel the freeways of the various cities I visit, I often come across billboards announcing waiting times at the local ED, or billboards promising no wait at all. Sometimes they just advertise “FASTER” care, whatever that means. These promises, which hospitals pay thousands to advertise, demonstrate just how backwards our EDs have become.

Do you know of any other business that has a universal reputation for making people wait? Sure, the DMV and at the Social Security office rival the emergency department for perpetual long lines . . . but they don’t charge you $1000 to treat a cold.


Emergency department throughput. It’s a perennial problem that seems to impact every aspect of ED activity (not to mention hospital economics). But why, oh why, is this such a difficult nut to crack? Come on – we put a man on the moon! We created the iPhone! Surely we can solve the problem of ED throughput.

It’s actually not that complicated. Having been the medical director of a community ED for 25 years in Los Angeles, it became very clear how to make a great ED – have a CEO who wants to have a great ED. Honestly, it is that simple. And if the CEO needs some substantial additional motivation, tie their bonuses to ED throughput measures. Done and done.

Not so fast. What if yours is the only game in town? Fixing ED throughput assumes that there is some competition in your neighborhood and your hospital wants to distinguish itself from the others. If you are in a town where yours is the only hospital, one would think that throughput wouldn’t matter. After all, where else are patients going to go? What does it matter if you make patients wait? They have no choice. Turns out this logic doesn’t work.

A consistently crowded ED is an engraved invitation for the local entrepreneurial physicians in the community to start an urgent care center that will skim off all of your ED’s insured easy patients, leaving the ED with the rest. And also watch out for your local CVS – they have about 880 NP-staffed Minute Clinics and will double that number shortly. CVS even takes Medicaid, plus over 200 other insurance plans. So the competition is out there whether you see it now or not.


Thankfully, you can actually change the trajectory of your ED. I like to see success stories that prove you can make a dysfunctional ED a lot better. As an aside, these stories are sadly under reported. Most of our journals focus on clinical care and are interested in ultrasensitive troponins and the like. But what about making the ED work better? Many of us are so used to slugging it out every day that we start to think there are no viable solutions.

Well here are two studies that you need to read – one is particularly inspiring. They both discuss how they turned their departments around dramatically. Yes, there have been lots of ideas regarding how to make EDs flow better, but implementation is the challenge. Here are a few of the ideas that get thrown around:

  • Provider at triage
  • PAs, NPs
  • Scribes
  • Triage only when patients are waiting
  • Fast track
  • RVU-based compensation
  • Patients don’t own their beds
  • Internal waiting rooms for lab/discharge
  • Chairs over gurneys for minor cases
  • Point-of-care lab testing
  • Physician staff trained in ultrasound
  • Move admitted patient to ward hallways
  • Order sets (not one of my favorites)

The first study is one from Kaiser (a great lab rat given that everyone is similarly motivated). Although it wasn’t heavy on specifics, the goal was to decrease the holding of admitted patients (even Kaiser struggles with this problem). Based on a collaboration between the ED, administration and a jillion departments inbetween they were able to establish two goals – a 30-minute goal for a patient to be assigned a bed and a 30-minute goal to move the patient to the bed once assigned.

Although they didn’t do fabulously well, they did get admitted patients out within an hour 58% of the time (vs 35% pre-intervention). The fascinating part of the study was to look at the other very positive effects this moderate improvement was associated with in the rest of the department. Length of Stay (LOS) for discharged patients was substantially improved, from a baseline mean of almost four hours (3:53) to 3 hours (3:03). This is a significant improvement. Although the length of ED stay for admitted patients was nasty at baseline (8:47) it improved substantially to 6:49. The left-without-being-seen rate, which wasn’t bad to begin with (1.2%) decreased to 0.6%. Given these improvements, the most surprising result of the study was that their patient satisfaction scores (which started at 67% rating care as very good to excellent) only improved to 74%. 

Although Kaiser EDs are certainly not particularly representative of community hospital EDs with regard to back-up staffing and access to follow-up care, this enlightening report reflects that substantial changes can be made when there is the will and the motivation of administration.

Patel, P.B., et al, Acad Emerg Med 21(3):266, March 2014
METHODS: This retrospective observational study, from Kaiser Permanente in California, was conducted at the Kaiser hospital in Sacramento. ED and hospital leadership formulated a process improvement plan to reduce wait times once patients were admitted to a goal of 60 minutes from the time of the ED bed request. The program involved meetings with relevant personnel, development of a plan to limit the intervals from an inpatient bed request to acceptance of the patient and from acceptance to transport out of the ED to 30 minutes each, and regular dissemination of real-time interval information to encourage progress. The study compared data during a two-year pre-intervention and a three-year post-intervention period.

RESULTS: The intervention was associated with a significant increase in the percentage of patients admitted within 60 minutes (from 35% to 58%), and significant decreases in boarding time per admission (from 82 to 26 minutes), lengths of stay for both admitted and discharged patients (8 hours, 47 minutes vs. 6 hours, 49 minutes for admitted patients, and 3:53 vs. 3:03 for discharged patients), the percentage of patients who left without being seen (1.2% vs. 0.6%), monthly ambulance diversion hours (10 vs. 1) and patient satisfaction (67% vs. 74% who rated their care as very good or excellent) (p<0.001 for all comparisons).

CONCLUSION: A goal-directed partnership between hospital and ED leadership to reduce waiting times prior to admission from the ED can significantly improve ED flow. 51 references. ([email protected] for reprints) 
Copyright 2014 by Emergency Medical Abstracts – All Rights Reserved 9/14 – #10

The next study has some absolutely remarkable results. It also reports on a redesign process at one of the five Scripps hospitals in San Diego. Although this paper lists more of the specifics to obtain their improved throughput, most are on the list noted above. A comparison of similar periods before and after the redesign process were reported. Admission rates remained stable during both periods at about 18% (typical for a community hospital). But look at these remarkable improvements – door-to-provider went from 127 minutes (embarrassingly bad) to 26 minutes (very good); overall ED length of stay dropped dramatically (5.5 to 3.6 hours) and the LWBS rate went from a whopping 8.7% (nasty) to an extraordinary 0.2%. As with the Kaiser study, patient satisfaction did not increase to a clinically significant degree (Press Ganey 77 to 81).

Sharieff, G.Q., et al, J Emerg Med 45(3):426, September 2013
BACKGROUND: A 32% increase in ED visits from 1996 to 2006 was accompanied by a decrease in ED capacity. ED crowding and decreased access to care have resulted in a Joint Commission mandate for streamlining of ED operations and patient flow management interventions and a CMS requirement for reporting of various quality measures.

METHODS: The authors, from Scripps Health in San Diego, report on selected effects of a redesign process in a single ED in a large five-hospital system. Key elements of the redesign included implementation of a quick-screen triage protocol, zoning of patients into low- and high-acuity areas, the use of recliner gurneys for low-acuity patients to increase bed capacity, development of a separate area for patients awaiting test results, procedures and formal discharge, and staffing adjustments to accommodate added bed space and increased patient volume. The study outcomes were compared for a six-month period in 2009, before the redesign project, and a comparable period in 2010 after the project was completed.

RESULTS: There was an 11% increase in daily patient volume from 2009 to 2010 (110 vs. 122 patients/day, p<0.05). The average admission rate during both periods was 18%. From 2009 to 2010, there was a significant decrease in the time to be seen by a provider (from 126.7 minutes to 26.3 minutes, p<0.001), in the overall ED length of stay (from 5.5 to 3.6 hours, p<0.01), and in the percentage of patients who left without treatment (from 8.7% to 0.2%, p<0.005). There was an increase in average Press Ganey patient satisfaction scores from 76.8 to 81.4 (p=NS).

CONCLUSIONS: This ED redesign initiative was associated with marked changes in the time to be seen by a provider, the ED length of stay and the percentage of patients who left without being treated. 14 references
Copyright 2014 by Emergency Medical Abstracts – All Rights Reserved 2/14 – #13

All of these authors are to be commended for not only being willing to show their less-than-stellar “before” numbers, but on the feats they and their colleagues accomplished. I have to believe that these changes occurred because the hospital administration (along with the ED staff) wanted them to occur. Although we all can make improvements internally in our ED’s performance, it sure helps to have the hospital administration greasing the skids to make things happen.

Richard Bukata, MD is the Editor of Emergency Medical Abstracts at cc:Me


  1. If you want to reduce patient waiting time, the worst thing you can do is to encourage patients who can be taken care of at lower cost by primary providers to come to the ER. The best way to avoid being saturated with uninsured patients is to use our lobbying muscle to ensure every American has insurance, even if it means that those with higher incomes have to pay more.

    On the other hand, if our primary goal is to increase ER income then we want to market intensively and get all the insured patients to come to the ER instead of going to primary providers. Our goal should be to advertise selectively to people with insurance and discourage people without insurance by advertising our higher rates.

    In this case the most critical change that was made was the “staff adjustments”. What exactly were these adjustments? If staff was increased during busy periods, this could have been a large part of the reason for the improvement in waiting time.

    In summary, we cannot interpret the effectiveness of any strategy without a clear indication of our overall goals. Diverting patients who can be seen by a primary provider to the ER is counterproductive and should not be our objective. If we are losing money because we are seeing uninsured patients, then we need to support political leaders who will implement effective universal health care.

  2. Martin I Herman on

    Universal healthcare is not going to solve the problem of long waits and it definitely will not correct the financial stress felt. Think about it, did ACA bring health insurance with reimbursement rates that are better than Medicaid or medicare? No! it did not. can providers and hospitals survive financially on Medicaid and medicare alone? NO!. so asking for Universal healthcare is not the panacea some think it will be..

Leave A Reply