Researchers from Harvard have developed a predictive model, called MELD-Plus, capable of identifying patients at high-risk for developing negative outcomes following a hospital admission for cirrhosis. Findings were published in PLOS One.

The ability to identify and predict negative outcomes after hospital admission could reduce readmission and give healthcare providers a chance to personalize treatment plans. To develop the post-discharge mortality prediction model, researchers analyzed 314,292 patients from 1992 to 2010 who received care for cirrhosis at Massachusetts General Hospital (MGH) and Brigham and Women’s Hospital (BWH). From the sample population, 68 variables from patient electronic health records (EHRs) were included in measuring patient risk.

Overall, nine variables that were most effective in predicting 90-day mortality were included in MELD-Plus. In testing the models’ predictive accuracy, researchers applied MELD-Plus to 18 million cirrhosis-related admissions from 2010 to 2015. Results showed that the model was able to improve on current predictive methods for 90-day and one-year outcomes.

“We developed a new risk score, MELD-Plus that accurately stratifies the short-term mortality of patients with established cirrhosis, following a hospital admission,” concluded first author Uri Kartoun and colleagues. “Our findings demonstrate that using a small set of easily accessible structured variables can help identify novel predictors of outcomes in cirrhosis patients and improve the performance of widely used traditional risk scores.”