Harvard, IBM researchers develop prediction model for cirrhosis outcomes

A team of researchers from Boston-based Harvard Medical School, Boston-based Massachusetts General Hospital and IBM Research developed a risk score to predict patient outcomes after cirrhosis-related admissions, according to a study published in PLOS One.

To develop the post-discharge mortality prediction model, the researchers analyzed a population of 314,292 patients who received care at Mass General or Boston-based Brigham and Women's Hospital between 1992 and 2010. They determined nine of 68 variables identified in patient EMRs were effective predictors of 90-day mortality, including total cholesterol, white blood cell count and age.

With the nine variables, the researchers created a risk score, called MELD-Plus. To evaluate MELD-Plus, they applied the risk score to all cirrhosis-related admissions in an independent EMR database of more than 18 million patients between 2010 and 2015. They determined the risk score resulted in improved performance over various existing prediction models.

"Accurate assessment of the risk of mortality following a cirrhosis-related admission can enable healthcare providers to identify high-risk patients and modify treatment plans to decrease the risk of mortality," the study authors wrote.

More articles on data analytics & precision medicine:
Johns Hopkins, Northwestern among institutions leading new National Center for Data to Health
Duke appoints leadership team to health data sciences center
KLAS: 14% of providers report 'deep interoperability' among disparate EMRs

© Copyright ASC COMMUNICATIONS 2018. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Top 40 Articles from the Past 6 Months