A major dilemma facing operating rooms is that historically, no one was measuring time. Documentation takes up a huge part of a clinician’s day. And when it comes to the OR, that documentation also is cumbersome. Inaccurate reporting can lead to delayed surgeries: In fact, more than 70% of surgeries don’t start on time, according to a Duke University study.
THE CHALLENGE
OR efficiency is essential to maximize the available surgery time for surgeons and patients. Like most hospitals, Houston Methodist often relied on estimation – human-entered data to the EHR or charge nurse knowledge – for decisions that affected efficiency in the operating room, not only in terms of planning and scheduling, but also throughout the day as things change.
This meant that operating rooms veered toward either underutilization, meaning too few patients were being seen, or overutilization, meaning cases were running beyond the schedule and staff was working overtime.
“We also struggled with coordination, meaning information flowed slowly such that surgeons, anesthesiologists, perioperative teams and others sometimes were not in the right place at the right time, causing further delays,” said Roberta Schwartz, chief innovation officer at Houston Methodist Hospital. “And we rarely knew what to prioritize for improvement and how to do so, given everyone already was working their hardest, but with too little timely information.
“A challenge in the OR is quality, reliable operational data,” she continued. “Historically, we counted on the EHR for all of our data. But that’s meant for record-keeping and billing, not for management and analysis. While efficiency data such as time in and out of the OR can be extracted from the EHR, documentation of timing is impacted by simultaneously occurring patient care events that results in documentation delays, also known as latency.”
The data relies on a nurse documenting each data point by hand, which is not meant to be their core job. In addition, data was kept on large tasks (such as wheels in), but not intermediate OR steps that would give an update on when the patient was expected to discharge from the OR, such as closing or extubation. Similarly, if staff ever made a change, they had little insight into whether it impacted outcomes (either improved or worsened).
And, generally, perioperative teams face challenges related to capacity, utilization, coordination and worker burnout, which have led to reduced case productivity and higher costs.
PROPOSAL
With the complex nature of the surgical suite, perioperative teams often have limited insight into what just happened, what’s happening and what’s going to happen. Advancements in surgery are only as good as the whole team’s ability to keep the operating room running as smoothly as possible: on time, efficient and agile.
“Real-time insights fed by AI computer vision and live visuals from cameras on the ceiling had the prospect to help surgeons prepare and keep their caseloads on track,” Schwartz explained. “AI can forecast scheduling changes throughout the day and look for opportunities to improve resource productivity and team coordination to significantly increase case volumes without compromising quality of care and decreasing the burden on perioperative staff.”
Houston Methodist wanted to use innovative technology like AI to help surgical teams stay more in control, spend less time dealing with logistics and, ultimately, enhance the overall patient experience all while giving high-quality care.
MEETING THE CHALLENGE
Houston Methodist adopted Apella, which uses ambient sensor technology and AI to provide a 360-degree view of the operating rooms. The vendor’s computer vision and deep learning create a real-time data feed from secure and compliant OR video that surgical teams use to drive efficiency throughout the day, improving scheduling, planning, coordination, utilization and patient outcomes, Schwartz said.
“This technology seamlessly integrated with the existing EHR to provide precise data to staff about things like wheels-in and wheels-out times – as well as steps between – the statuses of turnovers and cleanings, and schedule optimizations – during and planning and throughout the day,” she continued.
“In collaboration with this technology vendor, our OR events data has been much more accurate and timely,” she noted. “That meant we had the confidence to use this data to inform our schedules and make updates in real time as new information surfaced.”
Staff generate texts to let the team know what’s happening and to manage schedules so they are not running into rushes or lags, meaning OR utilization goes up, along with patient throughput.
“We also found that, with this technology, Houston Methodist was able to use well-trained artificial intelligence to identify new key surgical events essential to analyzing OR methods but that had never been captured before,” she explained.
“For example, the data recorded the additional event Patient Draped between the existing events Anesthesia Ready and Case Start,” she continued. “Where Anesthesia Ready and Case Start had historically been paired, which prevented easy comparison across unlike surgeries, the new Patient Draped event split those events into two distinct phases of the process.”
As a result of having more and better data that tracked the entirety of all perioperative steps, the team used these insights to segment surgeries into shorter, more definitive phases that made benchmarking and improvement much more straightforward.
RESULTS
Houston Methodist worked with this technology vendor to deploy cameras in the hospital’s Walter Tower’s ORs, paired with computer vision software to automatically capture, classify and compare every event in the video feeds.
While hospital staff are typically skeptical of EHR data, both the surgeons and perioperative teams were able to see how Apella’s AI passively and objectively recorded events as they happened much more reliably.
“The initial OR pilot launch at Houston Methodist Hospital coincided with a 10% increase in monthly case volume site-wide, equating to an additional 33 cases per month,” Schwartz recalled. “This data helped solidify rollout plans system-wide across Houston Methodist’s 14 surgical facilities around the Houston metropolitan area.
“To date, our ambient intelligence project in our operating rooms has seen a 15% increase in OR capacity without adding additional staff members, based on data from an initial pilot in 23 orthopedic and cardiovascular OR rooms,” she added. “This increased capacity allows more OR time to better care for our patients, and we have seen significant quality improvements in the cardiovascular ORs in particular.”
ADVICE FOR OTHERS
The biggest challenge is change management, Schwartz advised.
“It’s never easy to get surgeons, nurses, directors and others to agree on and embrace a new technology, especially when it comes to AI and sensors, particularly those fed through cameras,” she said. “As new technologies, especially those focused on operations and productivity, are introduced into care settings, it’s essential to provide education that identifies value to drive buy-in and enables the desired impact and outcomes.
“In our experience, demonstrating results – as we did with our initial OR pilot – is the best way to gain and reinforce adoption,” she continued. “Our teams saw how much better, more accurate and credible the data had become with the technology compared to the EHR alone.”
And it meant that everyone was able to focus on what they do best – treating patients and driving health outcomes – instead of the sometimes-cumbersome task of manually entering information into the digital record, she concluded.
Follow Bill’s HIT coverage on LinkedIn: Bill Siwicki
Email him: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication
WATCH NOW: Chief AI Officers require a deep understanding of the technologies and clinical ops