Algorithm spots people with MASLD, MASH using health records
Goal is to open treatment to people who might not know they've a problem
Researchers have developed an algorithm to identify people with metabolic dysfunction-associated steatotic liver disease (MASLD), a type of fatty liver disease, based on information in their electronic health records.
The approach also spots people with the liver scarring, or fibrosis, that characterizes a more severe stage of MASLD known as metabolic dysfunction-associated steatohepatitis (MASH).
“Our machine-based algorithm is flagging patients who have scar tissue and don’t know it, patients who may have the disease and don’t know it, and patients with atypical (or no) risk factors that put them into a high-risk category,” Julia Wattacheril, MD, the study’s senior author and an associate professor at Columbia University Irving Medical Center in New York, said in a university news story.
Scientists hope that their algorithm might be used to more efficiently diagnose fatty liver disease and stratify patients for optimal disease treatment.
MASLD is the most common form of liver disease worldwide
The algorithm’s design and validation were detailed in the study “Rapid identification and phenotyping of nonalcoholic fatty liver disease patients using a machine‐based approach in diverse healthcare systems,” published in Clinical and Translational Science.
MASLD, previously referred to as nonalcoholic fatty liver disease, is thought to affect up to 45% of people worldwide, and is the most common form of chronic liver disease. It’s characterized by liver fat accumulation in the presence of cardiometabolic risk factors such as type 2 diabetes, high cholesterol, and obesity.
While MASLD typically doesn’t cause health problems, it progresses in some cases to MASH, where excessive fat deposits start to cause inflammation and fibrosis. Eventually, MASH can lead to permanent liver scarring and damage (cirrhosis) or liver failure, with a possible need for a liver transplant.
Early management of risk factors — including lifestyle changes or the use of newer weight loss medications — can reverse or stop MASLD progression. In the U.S., Rezdiffra (resmetirom) also is approved to ease fibrosis in people with MASH. But once patients have cirrhosis, interventions are lacking.
Prompt disease recognition is key, but many people are not aware of MASLD and its risk factors.
“It’s the most common liver disease in the world, but it’s still surprising how underrecognized this disease is among people on the street, patients, and even doctors,” said Wattacheril, who also directs the MASLD program at Columbia.
MASLD can be hard to recognize in early stages because disease symptoms usually are not present until the liver is substantially damaged.
“Tons of patients make their way to us too late and already have cirrhosis,” Wattacheril said, adding that there is a need for “a proactive approach towards health,” where patients are identified before a liver transplant becomes their only option.
Algorithm scans electronic records for information relevant to a diagnosis
The algorithm essentially works in the same way as a liver specialist. That is, it scans a patient’s electronic health records for information relevant to a possible diagnosis — but it does so much more quickly.
“It’s raising the antenna on patients who have risk factors but whose diagnosis remains hidden to provider recognition, often buried in an enormous chart that’s hard to find in a time- and resource-limited practice,” Wattacheril said.
The algorithm was developed using health records covering more than six million patients at the Irving Medical Center, dating back to 1985. It identifies candidates based on risk factors and relevant diagnostic codes, rules out people with exclusion factors such as alcohol use or other liver diseases, and then verifies MASLD based on laboratory or imaging findings.
Among more than 840,000 people with MASLD risk factors and/or diagnostic codes in the center’s database, the algorithm identified 16,006 considered to have MASLD. Two-thirds of this group (67%) had no indication of such a diagnosis in their health records.
Clinical records indicated liver biopsy-confirmed MASH in 356 of the total MASLD group. An additional 943 people with likely advanced fibrosis were identified by applying common fibrosis scoring metrics, which predict the degree of liver scarring without a biopsy.
Algorithm identified up to 91% of MALSD cases across 3 US healthcare systems
Researchers validated their algorithm by applying it to patients seen at two other U.S. healthcare systems: the University of Pennsylvania and Vanderbilt University. They also compared its results with manual chart review results conducted by clinical experts, finding the new algorithm correctly identified 75%-91% of MALSD cases in the three healthcare systems.
They plan to continue developing the algorithm, which may eventually incorporate artificial intelligence.
“Future research should focus on optimizing this framework … paving the way for a more efficient and effective translation of knowledge into improved patient care,” the researchers wrote.
Meanwhile, the scientists are gearing up for an initial clinical trial aimed at validating the software’s ability to diagnose MASLD.
“We’ll test patients for MASLD with non-invasive tests, determine if they really have the disease identified by the software, and if so, connect them to clinical care and other research studies if they are interested,” Wattacheril said. “Then we’ll know if our technology is ready to be deployed for multisite validation.”
The study was supported by Janssen Research and Development, and two of its 16 authors work for Johnson & Johnson Innovative Medicine, formerly known as Janssen.
About the Author
