The Food and Drug Administration (FDA) recently completed its risk assessment of California Almonds, and their conclusions are similar to those made by Almond Board of California (ABC) after conducting its own risk assessments completed in 2006 and 2012 and performed by Linda Harris, Cooperative Extension Specialist in Food Science and Technology, UC Davis and her colleagues. Those risk assessments supported the requirement for a 4-log reduction process under the mandatory treatment rule, which became effective in 2007.
“FDA’s conclusions are similar to ours as to what level of treatment results in less than one case of salmonellosis per year in the U.S. population,” Dr. Harris said. That level of treatment is the 4-log reduction.
The FDA’s almond risk assessment was triggered by a request from ABC to recognize a 4-log treatment for Salmonella as being equivalent to a pasteurized product, which is normally considered a 5-log reduction. “Currently, almonds treated for 5-log can be labeled “pasteurized,” Dr. Harris said, “but in the almond mandatory Salmonella rule, a 4-log reduction in Salmonella is the standard.”
FDA began collecting data for this risk assessment in 2013, requesting data from the tree nut industry in general, including almonds, precans, pistachios and walnuts. Risk assessments for some of the other tree nuts are expected within a year.
Data supplied to FDA came from ABC and Harris, including some that had not previously been published.
“FDA took the framework of the risk assessment we did in 2012 and did a deep dive into what we had done, added the new data, and took a new look at analyzing the data,” Harris said. “They did a much deeper statistical, mathematical analysis than what we had done.”
One of the largest data packets submitted to FDA is nine years of Salmonella survey data, funded by ABC, consisting of over 15,000 samples. Those data include not only the prevalence of Salmonella in raw California almonds, but also the levels of Salmonella in those positive samples, according to Harris. “The more data the more accurate the risk assessment,” she said, “and having a data set that involves 15,000 samples is a very strong data set.”
FDA’s risk analysis verifies our conclusions and lends support to the 4-log process control target that is in the pasteurization rule, but it doesn’t answer the question about labeling product as “pasteurized.” However, ABC is now in a position to resubmit its request to FDA to consider a 4-log reduction to be pasteurized, using FDA’s own risk-assessment data as rationale. ABC plans to submit a 4-log process to FDA for consideration in the near future.
