ASN's Mission

To create a world without kidney diseases, the ASN Alliance for Kidney Health elevates care by educating and informing, driving breakthroughs and innovation, and advocating for policies that create transformative changes in kidney medicine throughout the world.

learn more

Contact ASN

1401 H St, NW, Ste 900, Washington, DC 20005


The Latest on Twitter

Kidney Week

Abstract: FR-OR024

Variable Economic Impacts of Clinical Decision Support Algorithms That Predict AKI among Postoperative Patients: A Nationwide Analysis

Session Information

Category: Acute Kidney Injury

  • 101 AKI: Epidemiology, Risk Factors, and Prevention


  • Chang, Felix K., Uniformed Services University of the Health Sciences and the Walter Reed National Military Medical Center, Bethesda, Maryland, United States
  • Kirk, Allan D., Duke University School of Medicine, Durham, North Carolina, United States
  • Buchman, Timothy G., Emory University, Atlanta, Georgia, United States
  • Elster, Eric, Uniformed Services University of the Health Sciences and the Walter Reed National Military Medical Center, Bethesda, Maryland, United States

New clinical decision support (CDS) algorithms are being developed to predict the likelihood of AKI among patients who undergo operative treatment. Conceptually those predictions enable early prophylactic interventions that could prevent AKI and thereby avoid its adverse health effects and associated treatment costs. We sought to assess (1) how accurate such CDS algorithms and (2) how effective their paired prophylaxes would have to be to yield a net positive economic impact.


We searched the 2014 National Inpatient Sample using ICD-9 codes to identify patients ≥18 years old with one or more operative treatments. We combined our findings with data on the incidences and treatment costs of AKI at three RIFLE classification stages. We then performed a sensitivity analysis of the costs associated with the use of CDS algorithms designed to predict AKI-R, AKI-I, and AKI-F in concert with their paired prophylaxes.


We found that CDS algorithms designed to predict AKI-R yielded greater net economic benefits than those designed to predict AKI-I or AKI-F. With an algorithm accuracy of 0.75 and a prophylaxis effectiveness of 0.50, an AKI-R CDS algorithm could create an annual nationwide benefit of $11.6 billion, whereas AKI-I and AKI-F CDS algorithms could create benefits of only $8.3 billion and $4.7 billion, respectively. (Figure.) The area under each curve represents the combinations of algorithm accuracy and prophylaxis effectiveness that would produce a net economic benefit for each CDS algorithm.


Our study revealed that lower incidences of AKI-I and AKI-F relative to AKI-R in postoperative patients would likely result in higher false positive rates among CDS algorithms focused on AKI-I and AKI-F. Those higher rates, when combined with likely higher prophylaxis costs to treat potential AKI-I and AKI-F patients, add new costs that, in many cases, overwhelm the costs avoided through early intervention. Barring more effective AKI prophylaxes, future development of AKI-related CDS algorithms should focus on the prediction of AKI-R.


  • Other U.S. Government Support