Today's hospital ICU is more challenging for CIOs than ever before. The increasing number of ICU patients, coupled with severe staffing shortages and increased focus on patient safety has driven many CIOs to search out IT solutions for hospital departments where care is complex.
“We were looking for a system for the same reason everybody else is looking,” says John Jenkins, CIO of four-hospital Moses Cone Health System in North Carolina. “Not enough intensivists and not enough ICU nurses.”
Jenkins has lots of company. Since most say the value of an ICU system lies in its ability to do more with less, one IT trend is remote monitoring — it allows a handful of clinical staff to monitor many ICU beds at once.
The other, more common solution in the ICU is system automation. The issue here is the decision to go best of breed or enterprise. So far, when it comes to automating the ICU, most core vendors don't have fully developed ICU offerings, so best of breed has been the chosen route for the majority of hospitals.
But for both remote monitoring and automation, interoperability issues loom large. And in the ICU, that can be a big problem.
Lynne Sterrett, a principal in New York-based Deloitte and Touche USA LLP's Healthcare Provider practice, says segregating the ICU episode of care can be a big safety and quality issue. “If you carve out this segment of care, which is often the most critical component, and it's never integrated with the rest of the care experience, it leaves open a lot of opportunities for failure points.”
She says hospitals need a way to tie together the clinical picture across the care venues, so the ICU is not an isolated segment of the patient experience. One common method for accomplishing this is a non-sophisticated printed “snapshot” of the patient's ICU record for the physician or next caregiver.
Sterrett says interoperability issues are increasingly influencing the ICU choices of CIOs. “I'm seeing a shift,” she says. “ICU was a niche product and the value of integration was not as high in the core vendors, but now they're starting to develop.” She believes that as the enterprise systems’ ICU offerings mature, a hospital should look to incorporate ICU functionality from its core system. “The niches are always out front, but the cores are beginning to take the market share as people see the value in integration.”
Sterrett also believes the use of remote monitoring is about to increase. “Many hospitals were waiting to see how the early adopters were doing.”
The number of those early adopters for remote monitoring is growing, and the vendor community has taken notice. In December, The Netherlands IT giant Royal Philps announced a merger agreement with Baltimore-based Visicu's remote monitoring system. Remote monitoring allows hospitals to increase the number of ICU patients monitored, usually at least by a factor of 10.
The Moses Cone Health system, a Greensboro, N.C.-based four-hospital system with more than 1,000 beds, has been on the Visicu system for a year, and has more than 100 beds on its “doc in a box” set up across the street from the hospital. This remote ICU is staffed with three ICU nurses 24/7, with an intensivist on duty 7:00 a.m. to 7:00 p.m. The system has cameras and loudspeakers in the ICU rooms. Patients (and their families) are informed so there are no privacy issues. Clinicians staffing the remote monitoring center can handle many ICU units, even those in multiple facilities. And advanced clinical decision support in the remote unit, via prompts and alerts, has been shown to reduce patient mortality, length of stay, and medical complications.
Jenkins says he chose Visicu because “aside from what Harry up in Lehigh Valley is using, Visicu is the only player out there.” (Harry Lukens, CIO of Lehigh Valley Hospital and Health Network in Allentown, Pa., uses Visicu's competitor, Israel-based iMdSoft. Read more about Harry Lukens at http://www.healthcare-informatics.com/lukens. During system-selection time, Jenkins says, Moses Cone was upgrading its GE Centricity system and an ICU module was not available. “There's always a tradeoff,” he says. “Do we wait for the GE product?”
At Moses Cone, ICU documentation is done in the Visicu product, with an interface to lab, pharmacy and registration. It's a standalone system, and Jenkins says he struggles with that. “They're always sold saying the interfaces are there but they're not very strong, and the documentation is not interfaced over to our Centricity so caregivers have to use both systems at the same time.”
Jenkins says when he made the decision to go with ICU documentation in Visicu, he believed industry pressure on the vendors would move interoperability along. In the meantime, users have two log-on IDs and passwords, one for Centricity and one for Visicu — and switch back and forth.
The ICU permanent record resides in Visicu, and if a patient transfers out of the ICU, caregivers log on to Visicu to see the prior ICU care. The Visicu log on, he says, is used most often (after the ICU) in the step down units. “Probably not that much of the Visicu record is needed in med surg.”
For Jenkins, being able to monitor more patients makes the interoperability challenges worth it.
Another area people are struggling with in the ICU environment is documentation. Today, a major player in that space is Wakefield, Mass.-based Picis. According to Sterrett, Picis ties together all the care plans, flow sheets and clinical decision tools into what she says is a nice niche package.
Lynn Vogel, CIO of 905-bed University of Texas M.D. Anderson Cancer Center in Houston, has been using Picis in his 50-bed ICU for three years, adding on to the peri-op and anesthesia products suite he already had from Picis. “It's a structured documentation, with billing notes and better revenue for billing charges,” he says. “We've been able to improve compliance for things like restraints and pain. The automation allows those triggers to be right in front of the nurse. In the ICU, it's real time data.” The Picis data is also available anywhere, as physicians can log on through a virtual private network.
For Vogel, the initial motivation was not only automation of a process, but using the system as a significant contributor to clinical research. “If there's a theme that comes out of this process, it's that the data that comes out of a system like Picis is not just used for the patient that happens to be in the room at the time,” he says. It's used to go back and review quality, outcomes and adding additional data points for the research-oriented hospital. “Picis is not just used to improve the care of the patients in the bed today, it's to improve the care of the patients we're going to have in the bed tomorrow.”
With all the interoperability issues on the table, is it worth it for a CIO to wait for the hospital's enterprise vendor to fully develop an ICU module? According to Sterrett, it depends. “If your core system is not putting development into an ICU module, you should not wait because the value of an electronic system far outweighs waiting and having a paper record.”
From a technical side, Jenkins says the most important factor is keeping the biomed or bioengineering departments in the loop. “This is truly a synergistic project, where IT, biomed and the ICU staff all have to work together.” For remote monitoring especially, which includes equipment like cameras, there is the question of ownership — is maintenance a biomed or IT issue? “It has to be defined,” says Jenkins. “Get them in early and define their responsibilities so there's no misunderstanding.”
Looking forward, Jenkins says there is a logical next step to remote ICU monitoring — offering the service to community hospitals that are understaffed in the ICU and typically don't have the capability for a remote system. “It's a service to our colleagues that need someone to monitor their ICU's.”
Sterrett has advice for the implementation and design itself in the nurse-heavy ICU. “Support your physicians by having the nurses go-live first. The best way to support the physicians is by having everybody else live,” she says.
Jenkins has one last piece of advice for CIOs, as the range for a remote project is $3-5 million, with another $3-5 million in annual costs for maintenance and staffing. “Bring a checkbook,” he says.