Hospitals throughout the United States are using monitoring units located outside of the hospital ward to alert clinicians to changes in patients' vital signs that might otherwise go unnoticed, Casey Ross reports for STAT News.
The age of alarms—and the rise of the central monitoring unit
In today's hospital, devices connected to patients set off tens of thousands of alarms every day, Ross reports. While some of the alarms require a clinical response, most of them do not. Traditionally, it has fallen on the provider to determine when an alarm requires action.
But more than a decade ago, Nemours Children's Hospital developed a Central Monitoring Unit (CMU), becoming a "pioneer" of the patient monitoring model that helps providers cut through the alarms and respond when appropriate, Ross reports.
Under Nemours' setup, paramedics in a remote center monitor data feeds from EHRs and alert clinicians when there's an abnormal result.
Stephen Lawless, Nemours' chief clinical officer, explained, "If there is an abnormal laboratory result, the logistics center makes sure someone has looked at it."
The military's tactical command centers serve as the idea behind the CMU. The tactical command centers are located away from the chaos of the front lines, giving them a clearer, less chaotic picture of how to respond.
How Cleveland Clinic and others are using CMUs
Nemours Children's isn't the only hospital that's using a CMU. Other well-known institutions—such as Cleveland Clinic, Johns Hopkins, and Yale New Haven—are leveraging the technology as well.
For Cleveland Clinic, the CMU staff typically consists of trios of nurses and emergency medicine technicians. The staff watch computer screens that display data on patients' conditions and wait for signs of trouble.
A clinician-developed algorithm helps them gauge which patients are at highest risk. The algorithm accounts for a range of data, including blood pressure, heart rate, and oxygen saturation levels. The names of patients who are at highest risk appear in red at the top of the screen.
When CMU staffers notice a problem in the patient's report—for instance, a blood pressure drop or a concerning heart rhythm—they call the patient's nurses. According to Alicia Burkle, the unit's program manager, CMU staff made 77,000 calls to Cleveland Clinic nursing units in April 2019.
The majority of calls are routine, Ross reports. However, in some cases, technicians detect a change that prompts an emergency response.
That's what happened in the case of a patient dubbed "John S."—a pseudonym for a real patient at Cleveland Clinic. Over in the CMU, staff watching the patient's vital signs detected the onset of ventricular tachycardia, a serious heart rhythm disorder that requires an emergency response.
Kris Rhode, an RN who received the call about John S., recalled, "We looked (at his telemetry data) and said, 'Oh my gosh, this patient is in a lethal arrhythmia.'"
The patient was off the floor, over in imaging. The team was able to locate him and send an emergency response team into action. Rhode said, "We figured out where he was and got down there within two minutes."
John S. is one of many patients who've been helped by Cleveland Clinic's CMU. A paper published in JAMA in 2016 found that the CMU accurately alerted clinicians to 79% of heart rhythm changes that prompted an emergency response team. In addition, the CMU alerted clinicians to 27 cardiac arrests, with circulation restored in 25 of those cases.
Elsewhere, hospitals are using CMU for other purposes, such as monitoring patient flow. Johns Hopkins, for instance, uses a command center to manage hospital capacity and keep patients from spending too long in the ED, Ross reports. Yale New Haven Hospital also uses a command center for capacity.
Predicting problems before the start
As for next steps, Yale, Hopkins, and Cleveland all have their eyes on using machine learning to deliver care more quickly, Ross reports.
Cleveland Clinic is working toward a specific goal: Notifying clinicians of a cardiac event at least an hour before it happens. The current setup offers some advance notice, but much of it depends on clinicians' ability to dig through "massive data streams," Ross reports.
The key to success in that goal is identifying specific changes in a patient's data and connecting that to specific outcomes.
Daniel Cantillon, a cardiologist who serves as medical director of the CMU at Cleveland Clinic, said, " We have to weave [different measures] into an algorithm that is already very good at identifying sick patients and we have to do that in real time and validate it." He added, "It's a tremendous amount of work and development."
So far, the clinic has been doing much of the hard work itself but has started working with external artificial intelligence vendors to develop capabilities, according to Cantillon.
"What we're trying to do," Cantillon said, "is to make the machines work better for [workers], so they're able to more efficiently handle all of the data coming their way." He noted that the goal of the improvements is to boost, not replace, the efforts of human workers (Ross, STAT News, 5/13; Garrit, Becker's Hospital Review, 5/13).
Innovation 101: Cheat sheets for today's digital world
Download our cheat sheets so you can keep track of the fast-changing technologies and capitalize on opportunities for IT-powered innovation. Check out our guides for these topics and more: