Working with an entrepreneurial startup company, Aurora Health Care in Milwaukee has developed an approach for leveraging the services of one provider who sees patients remotely during the triage process at multiple ED sites. The process has enabled the health system to accelerate throughput times while maximizing provider resources and boosting patient satisfaction.
- At Aurora Sinai Medical Center in Milwaukee, the approach has reduced door-to-provider times from 60 minutes to about 10 minutes, on average. In addition, the average length of stay has declined by 40 minutes, and the leave-without-being-seen rate has plummeted from 8% to 2%.
- Providers serving in the virtual triage role average 12-15 patient consults per hour, and the average length of these patient-provider interactions is 80 seconds.
- Developers say the key to the success of the approach is placing a technician in the ED who can execute the remote physician’s orders so that each patient’s workup is well under way by the time the patient sees the treating physician on site in the ED.
- Health system administrators are exploring other ways they can use telemedicine, perhaps to accelerate discharges from the ED and eventually putting remote physicians in charge of low-acuity cases.
Large, integrated healthcare systems have long turned to enterprise-level solutions to reduce costs and improve efficiency. Although most of these efforts have focused on administrative processes, some pioneering organizations are looking to make similar gains in the way clinical care is delivered. In particular, Aurora Health Care in Milwaukee has partnered with a Brookfield, WI-based startup, EmOpti, to design a way to leverage the services of one off-site provider in the triage process at multiple EDs.
“Effectively, we just took best practices from other industries and said, ‘OK, we should be able to manage this with a more centralized approach,’” explains Michael Rodgers, the director of strategic innovation at Aurora Health Care.
Further, rather than looking for a product, solution or vendor that was on the market already, Aurora decided to find an entrepreneurial partner that would codevelop a solution that was tailor-made to fit the health system’s needs, Rodgers observes. “We are actually building this together, which is a great augmentation of the expertise on both sides,” he says.
Early results are impressive. First launched at Aurora Sinai Medical Center in early 2016, there have been notable improvements, beginning with door-to-provider times. “We were at about 60 to 65 minutes at Aurora Sinai, and [this approach] has reduced that number to 10 to 15 minutes. And by doing that, it has also helped our overall length of stay [LOS], cutting that number by about 40 minutes,” explains Paul Coogan, MD, the president of Aurora Emergency Services and an emergency physician at Aurora Sinai, a busy, urban ED that sees 61,000 patients per year. “It has also cut our leave-without-being-seen [LWBS] rate from about 8% down to 2%, and patient satisfaction has improved.”
This past October, the virtual-provider-in-triage solution was expanded to include the ED at Aurora West Allis Medical Center in West Allis, WI, a department that sees 36,000 patients per year, and the ED at Aurora Medical Center in Kenosha, WI, which sees 30,000 patients per year. With this expansion, Rodgers anticipates additional gains. “There is a huge benefit to doing this at an enterprise level versus the local level ... because we can effectively scale up or down on providers at the [centralized hub] based on the volume we are seeing,” he says. “It is easier to scale up there than to get someone to a local site in a remote location when volume starts surging.”
One of the big motivators for taking advantage of telehealth during triage was the increasing focus on ED efficiency measures by the CMS. “We were never held to any standards [in this area before], and we certainly didn’t have any money at risk for things like door-to-doc times, LOS and all the other efficiency measures,” Coogan explains. “Those were all things that were developed in the past few years.”
Even without regulatory pressure, ED leaders always were trying to improve efficiency, Coogan stresses, but such efforts didn’t get much attention or support. “We were sort of under the radar as far as having to put resources toward solving problems in the ED,” he says. “It was kind of accepted that if someone showed up in the ED in a busy urban department, they were going to have to wait a while to be seen.”
Long waits in the ED mushroomed into other problems such as increased ambulance diversion and spikes in the LWBS rate, Coogan notes. He adds that although the virtual-provider-in-triage solution is not a silver bullet, it has helped address many efficiency-related problems.
The way the approach works is an off-site emergency provider is stationed in a room with a set of computers and a two-way hookup for audio and video. The provider then responds when a triage nurse at any of the participating EDs calls in for a patient consultation. “I literally get a ping at my workstation and I answer that,” Coogan explains, noting what a shift serving as the triage provider typically involves. “Immediately, it is like a Skype call. I see the patient’s face, the patient sees my face, and we begin a conversation.”
Focus on Lower-acuity Patients
There is always a nurse with the patient at the local ED site, and he or she will record vital signs and input the patient’s chief complaint. The virtual triage provider has access to the patient’s medical record for review. “We will talk with the patient, and then when we conclude the interaction, we thank the patient and explain to them that one of our techs will be drawing blood, collecting a urine sample, performing an EKG,” or perhaps that the provider has ordered some X-rays, Coogan explains.
In other instances, the triage provider may note that the nurse will be administering medication to help relieve the patient’s symptoms while he or she waits to see an on-site provider who will perform a full history and examination. “The virtual provider will then put in the orders, and the tech or nurse on site will get the orders started,” Coogan notes. “Even if the patient goes on to spend another 40 minutes in the waiting room, by the time he or she gets back to a treatment room, hopefully [most of] their tests will be done.” Then when the treating provider goes into the room, the patient’s workup is nearly complete.
Coogan acknowledges that before the approach was implemented, there was some concern that older patients especially might not like interacting with a provider via video hookup and that they might find the approach impersonal, but this concern was dispelled quickly. “They really seem to have embraced it the most,” he says. “I don’t know if they find it cool or what, but they have really appreciated [the approach] and like the interaction.”
Not all patients who present to the participating EDs are seen by the virtual provider during the triage process, Coogan observes. “There are five triage levels. The ones and the twos are the most sick, and those patients are taken directly back” to see an in-person provider, he explains. However, Coogan adds that when the ED is very busy, he has had occasion to see triage level two patients through this process virtually.
“If someone were showing stroke symptoms, [the virtual triage provider] could quickly see the patient, call a stroke alert, put in stroke orders, and then [emergency staff] would quickly make room for that patient. It would be a very brief [virtual] interaction,” Coogan says. “The [virtual-provider-in-triage] approach is really meant for triage levels three, four and five, and the nurse primarily establishes the triage level.”
Certainly, deploying the centralized, virtual-provider-in-triage approach in the health system’s EDs has required resources in terms of technology and software tools, but it has not required an influx of personnel. However, participating EDs have found it necessary to redeploy a technician from the back of the department to the triage area. This is a critical step, according to Coogan.
“Unless you are going to dedicate a tech to triage and carry out the provider’s orders, then it is not really worth the investment” to implement the approach, he says. “If you are just going to wait until the person gets back to a room to do the orders, then you really have missed the opportunity to save time. And if you are telling patients you are going to get things started and then you don’t, they are going to get frustrated.”
However, when executed properly, the health system actually maximizes provider resources, Coogan notes. “What we used to do is have what we called the pit shift, which is a provider in triage where we were on site and doing the same job. But we found if we used [telemedicine] we could cover multiple sites doing the same thing,” he says. “Now, we are able to cover three sites with that single provider.”
Emergency providers typically spend time both working on site in the EDs and taking shifts in the virtual triage role. “Those shifts are seven hours long, and, routinely, we are averaging about 12 to 15 consults per hour,” Coogan explains. “The most consults I have done is 22 in one hour, which was a little taxing, but it can be done.”
Coogan notes that the average length of the virtual provider-patient interaction is about 80 seconds. The time between a nurse calling in and the virtual provider answering is averaging about 15 seconds, he says. However, during busier times, there can be one or two patients waiting to interact with the virtual provider, he says.
Interestingly, some of the older emergency physicians on staff suggest that the virtual triage role actually may extend their careers a few years, Coogan notes. “Instead of retiring, they could use their lifelong knowledge to do a few of these shifts per month along with some shifts in the ED instead of having to hang it up,” he says. “Maybe someday physicians will be able to do this out of their homes.”
Fully Leverage Approach
With more time, Coogan believes additional benefits from the approach will become apparent. “By decreasing your LWBS rate and your bed turnaround time, you are really creating space without adding beds,” he says. “Certainly, there is a maximum amount that you would be able to accomplish, but our hospital [Aurora Sinai] has not been on ambulance diversion in over a year.”
Coogan emphasizes that the virtual-provider-in-triage approach is not the only reason for the reductions in ambulance diversion. He notes that the Aurora health system decided to adopt a no-diversion policy in November 2015. One year later, Milwaukee County decided that all hospitals would adopt a similar no-diversion policy. “That has been successful,” and the telemedicine solution has certainly helped achieve that, he says. “But you have to do a lot of other things too.”
Although the virtual approach is part of the triage process, administrators are looking at additional ways to leverage the telemedicine technology. For example, Coogan notes that it could be used to facilitate discharges or to handle patient encounters for low-acuity conditions. “I think that is where we should be headed for minor complaints,” he says.