AI in the NICU to improve neonatal nutrition

Neonatal Intensive Care Units (NICU) care for some of the most fragile patients, including both premature and full-term infants.  Providers must take into account each baby’s unique clinical needs, such as breathing support and nutrition supplementation, to help them grow and thrive.   

 But in such a highly personalized environment, deciding on the optimal course of care for babies can be challenging – especially when providers lack access to actionable insights for decision making in areas like nutrition.   

Large-scale research on such topics is relatively limited, says Enrique Gomez, MD, MsC, a neonatologist at St. Bernards Medical Center in Jonesboro, Arkansas, and busy NICU physicians don’t always have time to attend the latest conferences or dive into journal articles to keep up with every new development. Instead, they need technology to help surface relevant recommendations for feeding patterns and nutrition supplementation. 

“Good nutrition is one of the most important things we can do for a NICU baby,” he told Digital Health Insights. “How, when, and how much a baby eats is critical for their growth – and if there’s something not quite right about their eating, it can help us identify underlying issues early enough to address them.” 

“However, as an industry, that information doesn’t always get returned into the system to help us continuously learn from past results and refine our care for the next patients that come our way. It takes expertise in data analytics, and it takes time that we don’t always have as frontline providers. We need to do a better job of making those insights accessible at the point of care so we can make the most appropriate decisions.” 

Artificial intelligence is helping to simplify that process and equip providers with the data they need to tailor nutrition to each infant and optimize their care. St. Bernards has implemented a NICU nutrition clinical decision support (CDS) dashboard, created by Astarte Medical, that intuitively surfaces relevant information about feeding, nutrition, and growth measurements for patients. 

“It’s very important to have eyes on more than what happened that day or that week, especially for longer-term NICU patients,” explained Gomez. “Being able to open a dashboard and see the details of what happened over the past month, or longer, can give us a different perspective on how a baby is doing without the need to spend hours digging through my EMR to find the relevant stats.” 

With the tool, NICU providers can track adherence to organizational feeding protocols and receive automated alerts if a patient isn’t hitting growth targets despite established care procedures, he added. “Being able to get feedback when something isn’t right and take corrective action helps us deliver better, safer care for patients who can’t communicate their needs on their own.” 

Access to key patient data also supports broader-scale learning for St. Bernards and other health systems, especially in what is a relatively new specialty, Gomez noted.   

“Neonatology has only really been around for 50 or 60 years, so there’s a lot of research still to be done on specific topics around early life care,” he said. “Unfortunately, with the lack of standardized data in EMRs and limited resources to aggregate data at scale, we aren’t making as much progress as we could.” 

“We need to invest in standardizing and aggregating data across disparate systems so that we can look at more than just a few hundred babies at a time. It isn’t until we can analyze real-world data on thousands or tens of thousands of cases that we can start to see meaningful correlations and draw conclusions. We aren’t there yet as a specialty, but analytics technologies that extract and normalize data will help us get where we need to go.” 

Collaboration between health systems, payers, and technology companies will be key for creating an environment in which evidence-driven insights are more accessible at the point of care, Gomez concluded. 

“The tools we’re using now provide an excellent foundation for supporting better NICU nutrition and overall care, but we need to think bigger about how we’re going to continue advancing. Data accelerates innovation, and innovation saves lives in the healthcare setting.  I would love to see more efforts to aggregate datasets across disparate systems so we can get the volume we need to further refine our approaches to NICU nutrition and other aspects of critical care.” 

Jennifer Bresnick is a journalist and freelance content creator with a decade of experience in the health IT industry.  Her work has focused on leveraging innovative technology tools to create value, improve health equity, and achieve the promises of the learning health system.  She can be reached at jennifer@inklesscreative.com.