Dr. Alison Fox-Robichaud - Christine Probst
Dr. Alison Fox-Robichaud, Director – Medical Education & critical care physician - Christine Probst, Director of Informatics, Hamilton Health Sciences Centre
Supernova Award Category
Data to Decisions
The Problem
Code blues are called when a patient needs immediate resuscitation. Our goal was to reduce the occurrence of code blues. Acute care hospitals need to consistently and actively monitor patient vital signs. Hospital teams must recognize patient earlier deterioration, as failure to do so lead to code blues and death. It is challenging for traditional software applications to track this due to the sheer amount of data produced by IoT-enabled medical equipment every day. Nurses can typically record and monitor patients’ vital signs, but often not in a timely manner, and may need to leave the bedside to do it.
Hamilton developed an algorithm, the Hamilton Early Warning Score (HEWS) to support the existing rapid response team and preempt code blues by analyzing patient vitals data and other information, resulting in a 10-15% reduction in code blues overall. Despite having the HEWs algorithm, which was calculated in our EMR, and displayed to nurses with guidelines for action, the operational response still failed to preempt and prevent code blues. We wanted to see true transformation from our predictive analytics and found our existing technology incapable of delivering on this. We needed a product that was electronic, fast-moving and able to contact the right people to put them in the right place at the right time.
The Solution
We embarked on a joint project with ThoughtWire and the Department of Informatics. ThoughtWire’s EarlyWarning application utilizes machine intelligence to enable automation and real-time interactions between medical staff, systems and devices. At a patient's bedside, the nurse scans the patient's wristband barcode to verify his or her identity, and then enters vital signs into the mobile device. ThoughtWire’s system then calculates the HEWS score with the help of predictive analytics algorithms and produces a score that is simultaneously transmitted to the patient’s electronic health record (EHR) and smart phones carried by the charge nurse and the rapid assessment of critical events (RACE) team. If the score is six or above, the RACE and the attending physician are notified. This effectively automates nurses’ workflows and shortens notification time.
The results
Our information system previously had many disparate sources of information in the public domain. We were searching for a way to integrate and present information in non-traditional, interfacing ways. HL7 provided point-to-point data-based information, but we couldn’t take advantage of the cross-platform usage of mobile devices which enable high-res images in color, for example. We needed something more modern and flexible, which would allow for more creativity.
ThoughtWire’s EarlyWarning application, we finally have the right tools in place through the process of implementation, have witnessed a change in behavior. Nurses are now able to enter vitals in the least amount of time possible with a handheld device. Everybody carries at least one phone in their pocket and mobile phone usage is very intuitive, especially with triggered and targeted notifications to provider teams.
People have been comfortable with the shift, which has been supported by the organization and enables them to reduce lag time through a sensible process. 100 percent of staff in the adult acute care areas are using ThoughtWire’s EarningWarning app and on board with continued usage.
Metrics
Following the implementation of ThoughtWire’s EarlyWarning application, we experienced a 50 percent reduction in code blues called. This has helped us save lives.
In addition, by receiving this information sooner than usual and therefore enabling care givers to take preventative measures, we’ve been able to reduce the total length of stay for certain patients, prevent ICU admissions in many of these cases and ultimately saved the hospital in ballooning costs due to critical care stays. This ultimately saves on overall costs and allows us to allocate beds to patients who are most critically in need. We have also seen a 17 percent decrease in the number of Critical Care Response Team consults requiring patients to be admitted to the intensive care unit.
Prior to ThoughtWire’s EarlyWarning app, the response to a HEWS event required 23 steps and took on average 2 hours and 21 minutes. EarlyWarning was able to consolidate the workflows and remove bottlenecks to get that down to 13 steps and approximately 16 minutes on average, making the entire process 8.8x faster.
The Technology
EarlyWarning serves as an automated alerting system for rapid response teams that orchestrates the right people and equips them with the right data to respond before a high acuity event occurs. It automates vitals capture and calculation of the HEWS score to quickly display this information for nurses on their mobile device. This triggers the appropriate operational response to alert hospital teams to patient conditions and nudges them to take the appropriate actions.
Disruptive Factor
Part of our mission and vision is to innovate. We are starting to think of information-sharing as more of a process than a repository, as hospital information systems get smarter and can be more strategically designed. Thought leaders who are trying to do the right thing for the patient will look to adapt these kinds of innovative approaches, which don’t take extra money or time once the tools are set in place.
Shining Moment
This new technology has sparked a culture change as our organization continues to push for quality improvement. Staff are recognizing holes in reporting and proactively troubleshooting solutions with this more process-driven and automated approach. We have received several awards for its collaborative work with ThoughtWire, including two awards from the Intelligent Health Association – for Improving Patient Care and Health Delivery as well as the 2017 Intelligent Health Grand Award.
