Article
Author(s):
Predictive model uses routine lab values and machine learning method.
Predictive model uses routine lab values and machine learning method.
With price tags that can exceed $84,000 for a 12-week treatment course, deciding which patients infected with hepatitis C are most in need of these antiviral drugs has become more important than ever.
As a result, investigators from the University of Michigan Health System created a risk prediction model to evaluate which patients have the most urgent need for hepatitis C drugs in a study published in the June issue of Hepatology. While hepatitis C can remain stable without treatment for a long duration, approximately one-third of these patients face a high risk for complications, which requires immediate treatment to stop liver damage.
The predictive model utilizes routine lab values and machine-learning methods that aid providers in forecasting the health outcomes of hepatitis C patients.
"Offering immediate treatment to patients identified as high risk for poor health outcomes would allow these patients to benefit from highly effective treatments as other patients continue to be monitored and their risk assessment updated at each clinic visit," lead author Monica Konerman, MD, MSc, said in a press release.
The researchers examined clinical data from a National Institutes of Health study that included age, body mass index, type of virus, and routine lab measurements to predict the risk of progression for liver disease. The study noted that the use of significantly more lab values in the new model than most traditional models are able to use provides a substantial benefit.
Furthermore, machine learning methods were able to evaluate how these lab values change over time. The values include the slope and acceleration of platelet count, hepatic panel, and AST to platelet ratio index, which indicate liver injury and health.
In patients who the model deemed as low risk, just 6% were predicted to develop cirrhosis over the next year, compared with 56% in patients from the high risk group.
"Ideally we would treat all patients. Until logistic and financial barriers are solved, clinicians and policy makers are faced with trying to target these therapies to patients with the most urgent need," Dr. Konerman said. "The model allows us to identify these patients with greater accuracy."