Better data key to improved food safety, says Yiannas

Article By Joe Whitworth Published June 14, 2022
Article Source: Better data key to improved food safety, says Yiannas | Food Safety News

Improving food safety will include using better data, according to the deputy commissioner for food policy and response at the U.S. Food and Drug Administration.

Frank Yiannas was speaking at a Health Talks webinar on digitalization, food safety and trade with other panelists from Ghana, India and Ireland.

“The world around us is changing rapidly. Part of this rapid change is that data and information have become digitized and can be shared at the speed of thought. And new and emerging technologies are increasingly taking big and real-time data and putting it to good use,” he said.  

“For example, advances in artificial intelligence, the Internet of Things, sensor technologies, and blockchain are improving business processes. And the food system is reshaping itself, using these technologies, to meet the expanding global supply chain and the changing needs of consumers.

“I believe the new digital technologies offer the potential to help us predict and prevent food safety problems and better detect and respond to problems when they do occur. Leveraging the power of data is going to allow us to solve some of our food safety challenges that I never thought we could solve.”

Traceability developments

People often think that food traceability is a reactive tool but this isn’t true, said Yiannas.

“In the event of a foodborne outbreak, better traceability leads to better safety by enabling rapid traceback to the source of a contaminated food, the ability to speed up recalls, and better fueling the root cause analyses to understand how the incident happened and prevent reoccurrences again in the future,” he said.

“A digital, traceable food system will be a safer food system. But we, as food safety professionals and regulators have to be very aware, we can’t create a digital divide. If we do this right, it will allow small and medium enterprises to compete better with large institutions.”

The FDA is set to publish a final rule on food traceability in November 2022 and held a no or low-cost technology traceability challenge in 2021 that received 90 submissions and had 12 winners from the United States, Canada, and New Zealand.

Yiannas also spoke about a data analysis tool called 21 Forward.

“Unleashing the power of data is an overarching goal in FDA’s work to modernize food safety. We are now using this tool in the infant formula crisis that we are seeing in the United States with shortages. Analyzing high volumes of data is enabled by the scalability of this platform. This in turn has helped guide discussions with industry on how to increase production of various types of infant formulas,” he said.

“Better food safety begins and ends with better data. We have a lot of food safety data, it was often kept on paper but now we have these new tools that can bridge the gap between data and converting that into information.”

Predictions and seafood pilot

FDA is working with the private sector to create public-private data trusts to share data better, said Yiannas.

“In the U.S., we’ve seen repeated outbreaks with fresh leafy greens and there’s an organization called Western Growers that is doing just that. Getting the private sector to share all the data they have on fresh leafy greens, anonymously, is a great example of work that will strengthen predictive capabilities and inform risk-management decisions,” he said.

“It is clear that the FDA and food producers should also be looking at ways to tap robust, high quality data sources to strengthen our predictive analytics. We are continuing to explore the use of artificial intelligence, specifically machine learning, in a pilot designed to strengthen our ability to predict which shipments of imported seafood pose the greatest risk of violation. We are in the third phase of the pilot.

“Initial findings suggest that machine learning could greatly increase the likelihood of identifying a shipment containing potentially contaminated products. Doubling or tripling the ability to predict which shipments potentially are violative through the screening process is expected to result in much more effective utilization of resources to examine, sample, and test products at the port of entry.”

Irish perspective

Orla Moore, from the Food Safety Authority of Ireland, said digitalization has increased since the COVID-19 pandemic.

“We’ve seen pop up kitchens, dark kitchens and new businesses, some registered and some unregistered, emerge during COVID and Brexit. We’ve seen changes in how people order food and get food delivered,” she said.

Applications of digital technologies include remote auditing, e-learning, horizon scanning, early warning systems, licensing and registration and e-certification for health certificates.

“Artificial intelligence makes for quick, easy and in some cases cheap, use of data and information. It reduces bias out of decision making quite often. A few examples are risk assessment, quality control, predictive modelling and hygiene monitoring of equipment,” said Moore.

“The key benefit with blockchain is the transparency, it is secure in that the files can’t be edited, so from an audit point of view it is a solid system. One of the main difficulties is that everyone along the supply chain has to adopt and use it.”

Moore also spoke about seeking out emerging threats and risks through horizon scanning.

“We want to be aware of the next thing that is going to happen. It boils down to data mining tools. We have an emerging risk screening group that meets quarterly, and there could be sub-groups if there are emerging threats. This is to acquire data, monitor social media platform and publications,” she said.

“The challenges of digitalisation are the financial cost or the perceived cost to businesses, a lack of information, training and resources, fragmentation in governance or regulatory frameworks, confidentiality and data protection, access, connectivity, data quality and ownership of the data. There is quite a difference in the level of adoption of digitalization across organizations, industries and countries. Whatever system is used by businesses they must be able to provide information to authorities on demand. Digitalization has already happened, everyone is trying to get on board, so they are not left behind.”