After a recent staff meeting at Seattle Children’s Hospital, several doctors gathered around the coffeemaker to discuss sedation protocols used during a GI endoscopy.
One doctor said his technique and drug dosing was reducing young patients’ hospital stays. A lively debate ensued, filled with clinicians’ personal anecdotes of what worked best for them.
One physician decided to settle the question once and for all. She opened MDmetrix, a health data analytics tool a colleague had created to help with these exact care discussions.
Since its launch in 2016, MDmetrix has allowed anesthesiologists to detect variations in care and gauge which clinical practices might lead to better patient outcomes. The company on Monday is announcing a new software suite called OR Advisor, expanding its data tools to the operating room, in surgical specialties such as orthopedics, ENT and urology. At the same time, they’re looking beyond Seattle Children’s, shopping the product to hospitals and surgery centers nationwide.
MDmetrix raised a $758,000 seed financing round this year. The company is part of an early wave of startups using big data, machine learning and artificial intelligence to drive health-care decisions, said Asif Shah Mohammed, an associate principal with ECG Management Consultants, who specializes in healthcare technology.
What may give the Seattle-based company a competitive advantage, though, is that they’re coming from the health systems side, offering valuable physician-driven insights, Shah Mohammed said. Plus, they’ve already demonstrated they can successfully integrate their tool into a live, respected hospital system.
MDmetrix, which pulls data from patient electronic health records, provides a user-friendly tool — Ratliff compares the interface to Airbnb — to study variations in care. For example, Seattle Children’s physicians also recently looked at the impact of adding IV acetaminophen to patient recovery time and experience in eye muscle surgery – and found a counter-intuitive result. Based on this real-world evidence, the physicians discontinued the practice.
Before MDmetrix, they could have requested an analytics report, but that likely would have come months later, Ratliff said. The ability to quickly compare data means doctors can have spontaneous conversations around the coffeemakers, discussions that could impact thousands of patients who flow through the Children’s Hospital operating rooms annually.
The tool also appeals to administrators because it can lead to quality improvements, reductions in lengths of stays or free up bottlenecks in the hospitals, Ratliff said.
There are some limitations to the tool’s artificial intelligence. It can’t, for example, mine doctors’ notes to see if there were extenuating circumstances in a particular case. Still, it can use the structured data available in the electronic health record to say, see whether changes in care procedures impact a patient’s outcome such as pain level, recovery experience or the length of a hospital stay.
Ultimately, that means clinicians can delve into the data rather than share personal anecdotes.
As for the debate that day at Children’s Hospital, the original doctor was right. Within two minutes, a color-coded chart showed he was sending patients home 12 minutes sooner than other providers. Now, his colleagues have the data to potentially standardize care in those cases.
“That’s what all of medicine is working on,” Low said. “Your care shouldn’t be dependent on who takes care of you today.”
We told the story of MDMetrix on a past episode of GeekWire’s Health Tech podcast.