Bypass the ED for some admits?

John Nelson writes…

I think the medical system should find a way for some patients who “self-present” to the hospital, to safely bypass the ED and get admitted directly to the hospital.

This isn’t an idea that is ready for implementation on a large scale, and some might even say it is bad idea and not worth spending time on.  But I think it is worth thinking about for a minute.

One of the questions patients/families ask of me most often is whether they can bypass the ED if they get sick and require readmission.  I try to be as understanding as I can, but typically tell them that no, they will need to start off in the ED again if they have to return.  I tell them that I understand why they’d like to skip this step, but that it is necessary to ensure that a doctor sees them quickly and determines the best things to do initially.  Bypassing the ED would mean things would probably happen more slowly potentially reducing the chance of the best outcome for the patient.

I wonder when that will change.  Which will be the first hospital to institute a system in which the triage nurse in the ED can use a set of criteria to identify some patients who would go straight to a hospitalist and bypass the ED doctor?  Would this be a good thing?  Maybe such a system is worth thinking about if the criteria for deciding which patients can bypass the ED are narrow enough that both the really sick and not so sick patients (likely to be sent home without hospitalization) are kept in the ED and treated first by the ED doctor.  But maybe a frail elderly patient who was “found down” and has an altered mental status and fever (but reasonable vital signs and O2 sat) could go straight to the hospitalist.

Of course there are a lot of reasons that system of patients going straight from ED triage to the hospitalist might not be a good idea.  Even if we had robust research to define the set of patients who might be appropriately managed this way, we would still have to come up with a system that can produce a hospitalist as quickly as an ED doctor, and get ED turnaround time on labs and X-rays.  And after just a few hours in the hospital it will become clear that some of these carefully selected patients shouldn’t have been admitted and can go home after all.  And it will be tougher to get patients and families to accept this if they’re upstairs on the medical ward instead of still waiting in the ED.  And some patients will need to transfer to the ICU only hours after arrival in a “floor” bed.

Concerns like these may mean it is just unrealistic to conceive of a protocol allowing some patients to bypass the ED, or at least the ED doctor.  Sure there are costs and risks to such a system.  But aren’t there also risks and costs to the current practice of having most or all hospitalist admission seen first by the ED doctor?  It is easy to accept the current system as the only reasonable option and accept or ignore its costs.  When it is pretty clear to everyone that admission is the next step isn’t there something we can do to address patient’s unhappiness with having to stop in the ED?  And can’t we spare some patients the risk incurred by an ED doctor to hospitalist handoff, and reduce the sometimes unnecessarily broad workup and testing done in the ED? (I’m thinking of an emergent head CT in a demented patient who presents with fever and altered mental status.  If there is an obvious source of infection such as a UTI, the head CT is really low yield.  And did that patient really need the D-dimer and BNP ordered by the ED doctor?)

You may have seen Eric Howell’s article describing a system of involving hospitalists much earlier in the ED stay that got a lot of attention last year.  I’m just raising the possibility of continuing to involve the admitting hospitalist ever earlier after the patient arrives in the ED.

I’ve tried to stir up a little controversy here and I’d love to hear your thoughts about this.  I think most hospitalists and other doctors are likely to dismiss the idea as a non-starter, but maybe there are a few people who agree with me that it is worth thinking about.

9 Comments

  1. Jairy Hunter, MD, MBA, FHM on April 8, 2009 at 11:45 am

    It’s an innovative idea, but it seems to me the inefficiency of having to be available momentarily at the bedside, follow up on labs, not to mention to have to stabilize the patient would be time-prohibitive for the hospitalist assigned to that duty.

    Just thinking through how you *might could* do it (as we say in South Carolina), maybe it would work to have a dedicated ED-Hospitalist who would see the aforementioned patients (who met absolute admission criteria) from the get-go in the ED, without having them see an ED physician at all. Then you run the risk of discontinuity of care, unless you have a “new” physician embarking on a new block of shifts every day.

    You might get an argument from certain ED groups whose main compensation comes from patient volume….while it would essentially free them up to see more patients, they would lose out on some of the higher-yield encounters, but it might bring a small improvement in hospital efficiencies, LOS, core measures, etc.

    Would we start seeing combined ED/Hospitalist specialists? Interesting idea.

  2. Rick Harvey on April 14, 2009 at 3:44 pm

    This is nothing new, when I was in traditional practice our group did it all the time – direct admits. Get a call from a patient you know, hear their story and you know regardless of what the ED says you want them at least in observation. Satisfies patients and families, avoids unnecessary work up in the ED and yes, likely takes a little longer to get studies done and tests resulted.

    This is one of the failings of hospital medicine – we can not provide adequate continuity of care. I love the field, enjoy my practice but see the shortcomings and this is a major one. And a costly one.

    However, if the continuity is right between the outpatient PCP and the hospitalist this could be initiated by a phone call from the patient to the PCP then PCP to the hospitalist to discuss the direct admit. Makes sense but requires excellent communication and trust between the physicians.

  3. Mark Thoelke MD FHM on April 14, 2009 at 3:52 pm

    I’m a bit surprised at this conversation. We currently accept a significant percentage of direct admits from our PCP colleagues at Barnes-Jewish hospital. We sometimes have to insist on ED visits, but most of the time it is appropriate. EDs in large academic facilities such as ours tend to have long wait times to be seen and to be admitted, so patients much prefer avoiding that step. We usually see the patient in a reasonable period of time, and can get treatment initiated promptly. We are always following up on labs and testing on all of our patients, so I don’t see that as an issue. Occasionally patients require prompt ICU transfer or transfer to a surgical service, but that happens with patients from the ED. I didn’t realize we were such an anomoly in this regard.

  4. Leonard Pollack on April 14, 2009 at 3:59 pm

    At our two inpatient pediatric sites in metro Detroit, we have been encouraging direct admits for appropriate patients. We have the primary care physician make the decision, call us and we arrange a direct admit when we feel it is appropriate to provide the best patient care. If the patient was a recent hospital discharge, the patient may be instructed to call the hospitalist directly to help with the decision making process. Finally, we encourage an almost direct admit process from the ER – as soon as the ER doctor can tell that an admission is warrented, they can start the process, even without lab or xray results as long as the results are not needed to guide the decision making process. This is notonly a patietn pleaser, but also very pleasing to the referring doctors.

  5. Michael Burton, MD on April 14, 2009 at 4:23 pm

    I had the same thought as Dr. Hunter — maybe the Hospitalists on duty could have designated times to be in the ED to see the patients who have a high likelihood of admission, bypassing the ED physician. The Hospitalist would keep the patient, so no discontinuity.

    My initial reaction to this thought, though, was that we would have a difficult time accurately predicting the admits vs the non-admits. You certainly wouldn’t want to spend your time seeing patients getting discharged from the ED.

    Also, as mentioned there is the time spent waiting for the workup, which we often take for granted, i.e. we get to walk into the room already knowing all the results. Starting from scratch takes a lot longer, and who of us has the time to add extra responsibilities?

  6. Joan Nagelkirk on April 15, 2009 at 6:25 pm

    Our service in in house 24/7. When I have the admitting beeper and get a referral from one or our PCP’s or a referral hospital I sometimes meet them in the ED and eliminate the ED doc portion but have them where I can get my labs and testing expeditiously. We would need addtional personel to be able to offer this routinely. To put another wrinkle in the scenario we and our ED group had floated the idea of a hospitalist stationed in the ED to see the ‘medical’ patients. We do find if patients are ‘direct’ admissions to the floor their initial workup is definitely prolonged and more cumbersome to the harassed hospitalist who has to follow-up the info as it trickles in.

  7. John Nelson on April 16, 2009 at 2:30 pm

    In my original post I was talking only about patients who self-present to the ED and have not been sent in by another doctor (such as their PCP) who has decided they need admission.

    It is already an established (if infrequent) practice at most institutions for a doctor such as the PCP to see a patient in their office and refer them for direct admission (bypassing the ED). In my view, bypassing the ED for patients who self-present to the hospital raises a number of additional concerns but is still worth thinking about.

  8. Andy Radvany on April 16, 2009 at 6:19 pm

    I have to agree with Joan Nagelkirk.
    I worked as a PCP and would often do direct admits on my own patients. I would require them to be able to come to my office to be “eyeballed”. If they were too sick to present in the office they were too sick to be directly admitted.
    Once a patient is “in the hospital” all the rules change. Even stat tests get delayed “because the ER was busy.” The ER gets priority even to the ICU in many instances. Unless a patient is very stable, the treatment plan is known, AND it is safe that treatment might be delayed 8 to 12 hours “because we couldn’t get an IV”, even direct admits from me to me are unsafe. Never mind an admit of an unkown patient and unkown MD to me.
    These are sad facts that cause me angst when I even consider “direct admits”. Unless the ER centered mindset changes routine direct admits will never be as safe as they need to be. If resources can be brought to bear on direct admits the same as in the ER, then we may be able to safely manage direct admits.

  9. Rob Bessler on April 20, 2009 at 1:00 am

    John,
    You certainly got some good responses. I think the take home is it will be very hospital dependant. We love the direct admit and when all the naysayers at the hospital say why it can’t happen, it is a great opportunity to identify the inefficiencies and begin to work on them. ie it takes 6 hours on the wards to get a ct for PE while it takes 30 minutes in the ED. Why? Fix it so the excuses for the direct admits go away.

    one thing we started doing that others might consider is carving out ED charges/costs when considering evaluating the utilization in the hospital. The most expensive day of hospitalization is the first day. The major costs on the first day are tests that r/o other etiologies from the patients true illness and many are done by someone other than the hospitalist. It can help identify which projects to work on like your CT head in the febrile elderly patient.

    Rob Bessler

Leave a Comment