CONCERN initiative combines nurses insight with predictive tech

It baffles Kate Judge, executive director and executive vice president of the American Nurses Foundation (ANF), that nurses-—the single largest contingent of providers in health care; the patient’s point person—keep getting overlooked.  

“Why aren’t we investing in the people that are taking care of our loved ones and ourselves to do this work?” Judge says. 

Hence the Reimagining Nursing Initiative (RNI),10 projects led by nurses and funded by a $14 million grant, which aims to improve “access, care, and outcomes.” Technology is a key part of RNI’s slate, including the years-long CONCERN (COmmunicating Narrative Concerns Entered by RNs) pilot.  

CONCERN is a predictive tool that extracts nurses’ “expert and knowledge-driven behaviors within patient health records,” turning them into observable data that support early prediction of critical conditions in hospitalized patients.  

“As this project has evolved, we’ve consistently had nurses say to us, ’Of course, this makes sense. That’s how I document.” says Sarah Rossetti, RN, Ph.D., the pilot’s leader and creator of CONCERN.  “…In their practice, they identify patient risk early, before physiological changes, and that’s what we’ve been able to tap into.”  

Judge has seen “a huge waste of money in creating tools that actually don’t help patients’ care and nurses.” For example, nurses have told Judge that too many tools and devices simply capture the same data.  

“Why aren’t nurses at the table?” Judge says, on the development of digital clinical tools. “They’re not invited. They’re often discounted, which is kind of a gender issue sometimes. It’s an invisibility issue.” The result is patients, nurses, and the healthcare system suffer while manufacturers keep making devices that don’t really help anyone.  

For Rossetti, an assistant professor of biomedical informatics at Columbia University Medical Center, it’s natural that nurses’ feedback gets entered into electronic health records—that is already a big part of their daily responsibilities. CONCERN, Rossetti says, helps “to parse through that information overload that’s in the records,” and focus on “the useful information that nurses generate based on their own decisions and the patient.”  

That information—e.g., what nurses choose to document; what drives patient care—is then analyzed. “That is essentially the signal that we’ve found can be found used for prediction,” Rossetti says. “We think that approach can be used around ways in which we can better design our system so that we don’t burden our nurses with non-value– added documentation.” 

A 2020 survey by the American Nurses Association (ANA) revealed that almost two-thirds of nurses (62 percent) experience burnout. In a 2022 survey from ANF, the philanthropic arm of ANA, nearly half of nurses under age 35 said they had sought professional mental health support since March 2020, the start of the COVID-19 pandemic. Sixty-nine percent of nurses under the age of 25 reported burnout. Nearly 12,000 nurses nationwide participated in ANF’s survey.   

One of the causes of burnout, according to the World Health Organization (WHO), is reduced workplace efficacy. Another arm of RNI examines the use of robots—or Cobots—at ChristianaCare (Newark, Del.) so nurses can take care of patients instead of performing tasks such as dropping off lab specimens.  

“If we don’t create solutions that help the workforce better use their expertise, the workforce is going to step away and say, ‘Enough. Enough of this trying to [bandage] a broken environment,’” Judge says. “Let us create the technology that will actually will make care better and make what we do the best that it can be. That’s what this is all about.” 

Columbia University Irving Medical Center has partnered with three hospital systems—Mass General Brigham in Massachusetts, Vanderbilt University Medical Center in Tennessee, and Washington University School of Medicine/Barnes-Jewish Hospital in St. Louis—to test the effectiveness of CONCERN’s 10-year-long implementation toolkit, developed to support large-scale adoption.  

“We’re ready to scale and spread this intervention,” Rossetti says. “We’re early on in the sense of, I think, really showing and demonstrating getting value out of this across sites.” The goal is to get CONCERN used nationally. Its value, Rossetti believes, is expansive. “There’s the prediction and there’s the way that it can improve patient outcomes.” It also allows “how we can use and understand nursing data and the valuable nursing signals.”   

Adds Judge, “This really is a small investment in this critical element of care. If you look at the role nurses have in all of healthcare and how you invest with R&D, this would just be a logical place. As Sarah said, they’re now taking this data [and looking at] how does this work in another situation so it’s constantly informing better care and better patient-centered care, which is what nursing is all about.”  

It all comes down to not pushing nurses aside or relegating them as a cog in AI’s ever– increasing encroachment.  

“If you ask the right questions, if you embrace nurses to ask the right questions of that data, we can learn and do so many more things more things with it,” Rossetti says.