Predicting Kidney Injury Risk

According to a study published by Nature Reviews, acute kidney injury (AKI) is estimated to occur in seven to eight percent of patients in hospitals and approximately 50 percent of patients admitted to the intensive care unit.  

In a recent article published in the Journal of the American Medical Informatics Association, informaticians used four different clinical perspectives to build and evaluate prediction models for AKI in generalized hospital patients, based on electronic health records (EHR).

What did they discover? Models built at the time a patient gets admitted increases feasibility of predicting AKI will develop in the following day.

Why informatics? So, we can use EHRs to predict risk of acute kidney injury.