What is it about?
Whenever something new is proposed to detect a disease or predict something in the future it is important to understand not simply that it is related to the event, but that it adds value to the prediction tools already available. This paper reviews some recently proposed statistical methods, namely net reclassification improvement (NRI) and Integrated Discrimination Improvement (IDI) and gives examples of how they may be applied using biomarkers in Acute Kidney Injury.
Featured Image
Why is it important?
Perhaps the most unique aspect is the introduction and description of a graphical technique which we call a Risk Assessment Plot (RAP) to visualise the new metrics (NRI and IDI). The RAP enables the researcher to see beyond the numbers. The manuscript also strongly argues that when presenting the NRI and IDI that normally the IDI is best and that one should present IDIs separately for those with and without the event.
Perspectives
The manuscript was originally written because I thought the NRI was being misused and was giving the wrong impression about the added value of some biomarkers. I find in my own work the RAP an important step. Indeed, it has shown me that sometimes a biomarker adds value for patients where the initial risk profile was low risk, but not for those where it was high or vice versa. Only a plot and not the raw numbers can show this.
Dr John W Pickering
University of Otago
Read the Original
This page is a summary of: New Metrics for Assessing Diagnostic Potential of Candidate Biomarkers, Clinical Journal of the American Society of Nephrology, June 2012, American Society of Nephrology,
DOI: 10.2215/cjn.09590911.
You can read the full text:
Contributors
The following have contributed to this page
