University of Arizona College of Nursing Associate Professor Dr. Sheila Gephart had an epiphany when she realized that unwieldy electronic health systems were impeding the quality of work for nurses on the job. In the midst of creating an international quality registry for babies born with low birthweight, she discovered that a large amount of patient data was slipping unused through the cracks. If the point of electronic health records is to provide quality outcomes for patients, she concluded, then there was clearly room for improvement to support clinical decision-making, especially for nurses.
Last year, Dr. Gephart and her co-author, Dr. Karen Dunn Lopez, published a paper in the Journal of the American Medical Informatics Association (JAMIA) that reported on the state of the science of clinical decision support (CDS) for hospital bedside nurses. Their conclusion contained mixed news: CDS that targets decisions made by bedside nurses improved the processes and effectiveness of care, but the number of nurse decision-making research studies lagged behind physician decision-making CDS studies of CDS, obscuring the evidence to show positive change in patient outcomes.
How do you define nursing informatics?
Nursing informatics lies at the intersection of information science, nursing science and cognitive science. One component of informatics is the study and use of clinical decision support systems. The purpose of CDS is to make it simple for people to have what they need at the time and place of decision-making to deliver the best possible patient care.
Tell us about your clinical decision support study.
When I was in practice, I noticed gaps in how we were making decisions in the neonatal intensive care unit and I wanted to help clinicians by providing them better systematic and convenient information about the context of care and the factors that put a baby at risk to develop complications. CDS has been shown to improve the process and outcomes of care delivered by physicians but we need more studies that target the decisions that nurses make. For the paper published in JAMIA, we wanted to discover what the state of the science was for bedside nursing CDS.
Tell us more about CDS for nurses.
Simply put, CDS matches patient characteristics to best recommended care and delivers reminders, alerts or advisory messages to support nursing care that aligns best with expert recommendations. The best and most effective CDS is built on data that comes directly from the electronic health record. It generates context and communicates a recommendation through an alert. So instead of communicating ‘This patient has an early warning score of nine and it’s high-risk,’ the message would say, ‘This patient’s early warning score is nine and you should do this.’ Even better, the message would be delivered directly to the clinician or set of clinicians who can act on the information.
What are some of the decisions nurses make that this work would benefit?
Nurses make decisions all the time. They’re assessing how at-risk a patient is for a fall, to develop an adverse complication, or to develop sepsis. They’re deciding how to prevent a patient from getting a complication they shouldn’t. One critical decision nurses make is about monitoring and communicating signs of a patient getting worse. CDS is especially good at making the decision about care more systematic, expert-informed and consistent. It keeps nurses from missing the simple signs, which is important because their work environments can be very fragmented, interruptive and rapidly changing.
What makes your study particularly unique?
Nurses put more information into the electronic health record (HER) than anybody else in the hospital and their needs are the least well met. So far, nurses get very little value for what they put in. In some situations, it has not been designed or adapted to help them do their job easier or more efficiently. With our paper, we showed how little evaluation of nursing CDS is available and how underdeveloped it is to meet nurses’ decision-making and ultimately patients’ needs. The message we came away with is that there’s a huge need for people to study this, and to develop CDS that meets CDS design best practices.
What’s the future of this project?
We proposed some very specific recommendations about what needs to be done to help design better decision support that is aligned with people’s workflow. We recommend expansion of nursing CDS, to integrate it into workflow and to allow shared views of CDS delivered across the health-care team (e.g. nurses to be able to see what CDS was offered to the physician for their patient and how they responded to it). Dr. Dunn-Lopez and I are also both individually developing CDS that targets nurses.
How will the results of your study impact nursing education?
It brings more visibility to the decisions that nurses make and how they need technology to support them to add value to the EHR for them. I’ve taught the methods for systematic reviews to my students and conducted others since then. Systematic reviews are important to make sense of a body of research and to identify next steps. This helps both PhD and DNP nursing students. They need to be able to take in the whole breadth of what’s going on, find the gap, and decide what to do next.
How is this project helping to build better futures?
When we support nursing decisions, we help them to do their work better, which can keep a patient healthy or even save their life. This project showed that more decision support is needed in nursing, especially in the pediatric and maternal child settings, where my work fits. It highlights opportunities to improve the value of electronic health records for nursing. As nursing informatics and CDS develop, trends are pointing toward better integration of EHRs with decision support – making their structure more open. The future is bright for nursing informatics!