Integrating AI-powered clinical decision support into the EHR can help increase adoption and efficiency
Clinical decision support tools offer clinicians the ability to enhance the quality of care given to patients – lowering costs and improving patient outcomes. Yet, too often, the way such tools are implemented can interfere with the provider’s natural care delivery workflow. That’s one key reason why those applications are often underutilized at the point of care,” said Dr. Mark Weisman, Chief Medical Information Officer at Peninsula Regional Medical Center (PRMC), a 266-bed multi-service hospital in Salisbury, Maryland.
“These tools are frequently delivered as an invasive pop-up coming at clinicians when they weren’t asking for it,” he explained. “That’s why you’ll find that clinical decision support is not always so well received.”
It’s not that clinicians don’t see the benefits in such tools, Weisman said. But it’s vital to find a way to implement them so they enhance rather than detract from a user’s normal workflow. That’s why, when PRMC decided to deploy a new tool enabling clinicians to ask common pharmacologic questions at the point of care, they knew they’d have to find a way to make it easy to access and use at the point of care.
“Clinicians are under tremendous time pressure,” said Weisman. “It’s important to streamline efficiencies where you can – and keep clinicians inside the workflows they are used to using to do these sorts of tasks. If they have to go outside those workflows to find information they need, that’s often a barrier to entry that may prevent them from getting it done.”
Weisman and colleagues opted to deploy IBM Micromedex with Watson – a novel clinical decision support tool powered by artificial intelligence (AI) – that allowed doctors and nurses to look up information regarding drug compatibility, adverse effects and special dosing requirements within the hospital’s electronic health record (EHR) system. Rachel Cordrey, PharmD, supervisor of in-patient pharmacy at PRMC, said the tool permits clinicians to type conversational questions directly into a chat window.
“Time and clicks were the two barriers we really had to overcome,” she said. “But using this tool, clinicians can just use natural language to ask what they want to ask and get rapid turnaround on any drug information questions from a trusted database of evidence-based information.”
Weisman said previous research had uncovered that 92% of the hospital’s clinicians would be more likely to use such a tool if placed directly in the EHR. And, when placed on an easy-to-access “Watson” tab on the main interface, they did just that.
“Once we rolled out the tool hospital-wide, we saw it was used 489 times in just a single month,” he said. “In comparison, when it wasn’t integrated into the EHR, they only used it 275 times for the prior six months before that. That’s a huge difference.”
When surveying end users, the team found that clinicians overwhelmingly said IBM Micromedex with Watson saved time, was easy to use and made their jobs easier. Cordrey said that was exactly the result they were hoping for.
“This is a valuable tool for pharmacists and other clinicians alike,” she said. “It makes it easy for a doctor or nurse to ask a question in natural, conversational language right at the point of care to get the information they need when they need it. By working with an AI platform and then integrating this tool into our EHR, we’ve seen great benefit for patient care at PRMC.”
To learn more about how AI-based technologies can help deliver fast answers to the point of care, visit IBM Watson Health https://www.ibm.com/watson-health/solutions/clinical-decision-support.