Hospital uses early warning system technology to help save lives

Code blues in a hospital often mean death for patients. One hospital is using an early warning system to intervene before a patient reaches code blue status.

Editor's Note: Each month, the editors at SearchHealthIT recognize an innovative software, service or technology approach. The early warning system created by ThoughtWire's Ambiant is our February 2017 selection.

Product: Ambiant by ThoughtWire

Release date: 2009

In medicine, code blues -- which are used in hospitals to indicate that a patient needs immediate resuscitation, usually due to a heart attack or respiratory collapse, for example -- often mean death for the patient despite the medical staff's best efforts to save them.

Mark Farrow, CIO of Hamilton Health Sciences and St. Joseph's HealthcareMark Farrow

Mark Farrow, VP HITS and CIO of Hamilton Health Sciences and St. Joseph's Healthcare Hamilton in Hamilton, Canada, decided to explore using early detection technology to notify medical staff when a patient is heading for a code blue so that the care team could intervene and potentially save that patient's life.

Nurses at Hamilton would record and monitor a patient's vital signs, "but a lot of times they wouldn't be entering it into the system in a timely manner," Farrow said. "It really doesn't help to put it into the system two or three hours later and say, 'Oh, your patient is going to have a code blue two hours ago.'"

To try to fix the problem, Farrow and Hamilton turned to ThoughtWire's Ambiant software, which utilizes machine intelligence to enable automation and real-time interactions between medical staff, systems and devices. Hamilton worked with ThoughtWire and used Ambiant to create the Hamilton Early Warning System (HEWS).

What Ambiant does

Jason O'Neill, director of client delivery at ThoughtWire, which is located in Toronto, Canada, explained that it was ThoughtWire's job to translate the design Hamilton wanted in their early warning system onto the Ambiant platform.

Jason O'Neill, director of client delivery at ThoughtWireJason O'Neill

In the case of Hamilton, Ambiant runs on a Honeywell Dolphin CT50, an Android smartphone. At a patient's bedside the nurse uses the Honeywell phone to scan the barcode on the patient's wristband to verify the patient's identity. Once that is done, the nurse uses the phone to enter the patient's vital signs such as blood pressure, oxygen levels and respiratory rate. Once those vital signs have been entered, Ambiant calculates the HEWS score with the help of predictive analytics algorithms and machine intelligence.

If the score is five or above, the rapid assessment of critical events (RACE) team is notified.

"We automatically direct to the correct role depending on the score," Farrow said.

Furthermore, the information the nurse enters into the Honeywell device is also automatically entered into the patient's electronic health record (EHR). This means two things for nurses: One, they don't have to input all that information manually into the EHR, and two, they don't have to use their own judgment to assess the patient's vital signs and decide what action to take.

Why Ambiant matters in healthcare

Intervening with a patient before they reach code blue status has the potential to save lives, both Farrow and O'Neill said.

"It starts to trigger a change," Farrow said.

In addition to this, ThoughtWire's Ambiant not only helps Hamilton generate early warning scores and alert the appropriate staff to attend to the patient, but it can also be used to facilitate communication among the staff.

"Previously a page would go out or a call would go out to people on the RACE team and there was no confirmation whether it [was] received or not," O'Neill said. "Whereas now everyone in the system, the bedside nurse, the charge nurse and the RACE team can all see who the alert went to [and whether] they acknowledged it yet."

What a user says

Farrow explained that resuscitating a patient puts a lot of stress and strain on the body in order to revive the patient. Ultimately, he said, the goal should be to avoid a patient getting to that point in the first place.

[A] patient on a ward shouldn't be heading to a code blue if we're monitoring them appropriately.
Mark FarrowCIO at Hamilton Health Sciences and St. Joseph's Healthcare Hamilton

"[HEWS] is a real example of taking data and turning it into knowledge and information and [then] into action," Farrow said. "We're actually starting to show people that there's value in collecting all of this information they've been collecting … for years and years and being able to turn it into real patient outcomes."

Farrow said ever since HEWS was implemented, "it is very rare to hear a code blue called in our hospitals."

In fact, HEWS has allowed Hamilton to reduce Critical Care Response Team consults requiring ICU admission by 17% in the four months it has been using the system.

"[A] patient on a ward shouldn't be heading to a code blue if we're monitoring them appropriately," Farrow said.

Not only does this help Hamilton save lives but by receiving this information sooner than usual and therefore enabling care givers to take preventative measures, total length of stay for certain patients is reduced.

Farrow said Hamilton decided to use handheld mobile devices so that the visual component of the nurses' job -- observing whether the patients' behavior has changed, for example -- was maintained as opposed to just purely automating everything. While many believe technology in healthcare is removing the personal touch, Farrow believes that with HEWS that is not the case.


O'Neill said the initial implementation at Hamilton took six weeks, noting that the implementation process varies depending on a hospital's specific needs.

"We would work with the hospital, install ThoughtWire's Ambiant platform, deliver the product that is the early warning system, and then tailor the rules of that unique team with the unique characteristics of the RACE team to suit the need of their hospital," he said.


ThoughtWire declined to provide pricing information.

Next Steps

With clinical analytics automation, human intervention still needed

Predictive analytics helps with OR utilization

Predictive analytics can help better manage high risk patients

Dig Deeper on Patient safety and quality improvement