The Food and Drug Administration’s PREDICT tool for estimating the risk of imported food products is generally working as intended but the FDA could take some steps to improve its effectiveness, according to a recent report from the Government Accountability Office.

The GAO states that the volume of FDA-regulated imported food continues to grow but that the agency physically examines only about one percent of food shipment entry lines annually. In response, the Predictive Risk-based Evaluation for Dynamic Import Compliance Targeting tool uses a variety of data to identify food shipments deemed higher risk and subject them to additional scrutiny. The GAO’s analysis of FDA data from fiscal years 2012 through 2014 shows that PREDICT is generally working as intended: imported food with higher-risk scores is more likely to be physically examined and to be found in violation of food safety standards or labeling requirements.

Nevertheless, the GAO found several opportunities for improving PREDICT. The FDA’s Office of Regulatory Affairs does not have a documented process for identifying the type of open source data to collect, obtaining such data and determining how PREDICT is to use it. In addition, the programs provided for under the Food Safety Modernization Act that are likely to provide additional data for estimating the risk of imported food are not yet fully implemented and the details of how PREDICT will use that data have not been worked out. The FDA developed 24 recommendations for improvement following a May 2013 evaluation and has fully implemented 15, and the GAO urged the FDA to establish a timeline for completing the rest as resources become available.

Copyright © 2021 Sandler, Travis & Rosenberg, P.A.; WorldTrade Interactive, Inc. All rights reserved.

ST&R: International Trade Law & Policy

Since 1977, we have set the standard for international trade lawyers and consultants, providing comprehensive and effective customs, import and export services to clients worldwide.

View Our Services 


Cookie Consent

We use cookies on our website. By continuing to use our website, you agree to the Privacy Policy and Terms of Use.