ORIGINAL RESEARCH
Ted Xiao, MD; James Miller, MD; Bradley Rowland, MD; Richa Bundy, MPH; Adam Moses, MHA; Lauren Witek, MStat; Corey Obermiller, MStat; Ajay Dharod, MD; Sean Rudnick, MD
Corresponding Author: James D. Miller, MD; Wake Forest University School of Medicine, Department of Medicine.
Email: jamesmil@wakehealth.edu
DOI: 10.3122/jabfm.2025.250261R0
Keywords: Best Practice Advisories, Clinical Decision Support Systems, Fibrosis, Elastography, Gastroenterology, Liver Diseases, Metabolic Dysfunction-Associated Steatohepatitis, Primary Health Care, Referral
Dates: Submitted: 07-08-2025; Accepted: 11-10-2025
Status: In production.
BACKGROUND AND AIMS: Metabolic-dysfunction associated steatotic liver disease (MASLD) is highly prevalent in primary care. The Fibrosis-4 (FIB-4) index noninvasively risk-stratifies patients needing confirmatory testing for advanced fibrosis. However, screening is underutilized, contributing to underdiagnosis and low rates of testing at-risk patients. Computerized clinical decision support (cCDS) tools can enhance recognition of MASLD and guide clinicians to appropriate care pathways. We aimed to determine if a cCDS tool identifying patients with MASLD at increased risk for advanced fibrosis improves rates of confirmatory testing and subspecialist referral.
METHODS: The cCDS tool was implemented in 3 primary care clinics for a 3 month period. The cCDS tool triggered if an encounter met criteria for high or indeterminate FIB-4 score (>1.3) plus diabetes, or ≥ 2 metabolic risk factors. The cCDS prompted testing with elastography and/or subspecialty referral. The primary outcome was proportion of encounters in which confirmatory testing/further evaluation was ordered.
RESULTS: In intervention clinics, the cCDS tool triggered during 1,410 encounters. Confirmatory testing increased from 0% to 2.1% (p<0.001) and subspecialty referral increased from 4% to 20% (p=0.12). The cCDS was ignored in 75.4% of encounters.
CONCLUSIONS: Implementation of a cCDS tool increased rates of confirmatory testing and subspecialty referrals in at-risk patients, though overall rates remained low, likely related to high rates of ignoring the tool. This study demonstrates the feasibility of utilizing a cCDS tool in primary care to screen for patients with MASLD in highest need of further evaluation. Further optimization of triggering criteria and user interface may increase use of the tool.