Tell the FDA: Toxic chemicals shouldn’t be in our food.

Unleaded Juice: FDA’s challenge of continuous improvement and compliance assurance

EDF (October 2, 2022)

Originally published here.

This is the sixth in our Unleaded Juice blog series exploring how the Food and Drug Administration (FDA) sets limits for toxic elements like lead, arsenic, and cadmium in food and the implications for the agency’s Closer To Zero program. 

A core tenet of FDA’s Closer to Zero program is the “Cycle of Continuous Improvement” represented by the image below on the program’s webpage. The four-stage, outer ring represents FDA’s process for revising its action levels for food contaminants. The inner, grey ring describes the agency’s on-going monitoring, research, and compliance program.

This approach makes sense, and we fully support it. However, the success of this approach relies on FDA addressing several significant structural weaknesses.

  • Future funding is not guaranteed: In March 2022, Congress appropriated $11 million in Fiscal Year 2022 (FY22) funding for FDA’s maternal and infant health work—in part to support the agency’s efforts to reduce arsenic, lead, and cadmium in children’s foods. Last year’s request and appropriations were a significant increase over previous years, but that funding level is not guaranteed for future years.
  • Action levels are guidance—not legally binding requirements: FDA’s action levels for contaminants in food are established in guidance. The guidance introduction makes it clear that “The contents of this document do not have the force and effect of law and are not meant to bind the public in any way, unless specifically incorporated into a contract.” It assumes that the food industry—from the largest multinational corporation to the smallest entrepreneur—will comply.
  • The agency has limited means to monitor compliance: FDA largely relies on physical inspections and market sampling, supplemented by voluntary reporting, to assure compliance with action levels. Inspections at high-risk facilities must occur every three years (but likely have been delayed due to the COVID pandemic). We understand that most facilities will see an inspector once every eight years. This is particularly problematic because FDA says it lacks the authority to require food companies to provide requested documents without the physical inspection, and the agency does not require ongoing testing and reporting by companies for action levels.
  • Action levels must be consistently strong enough to drive research and impact markets: FDA correctly points to its success in setting an action level for inorganic arsenic in infant rice cereal as a model to lower contamination. Unfortunately, the model assumes the action level for a contaminant is set low enough to result in research investments and increased product and ingredient testing and to provide FDA with sufficient information to act on problems. This is not the case for lead in juice.
     
    We explore each of these weaknesses below.

Adequate funding is essential to success

In May 2021, FDA provided its FY22 budget request to Congress; it included $97 million in additional funds for core food safety programs. One element of the request was $18 million to hire 26 new staff members for maternal and infant health and nutrition—up from $4 million. Closer to Zero was one of five activities that funding would support. FDA provided a fact sheet detailing the request, which gave the request greater emphasis.

In March 2022, six months after the fiscal year started, Congress finally approved the federal appropriations. The law was accompanied by a report noting $11 million in new funding was intended for maternal and infant health and nutrition including Closer to Zero.[1]

That same month, in its FY23 budget justification FDA repeated its request for an additional $18 million. Currently, the House Appropriation Committee’s report that accompanies the appropriations bill says $7 million for the Closer to Zero program for FY23[2] as part of a larger appropriation  for maternal and infant nutrition.

Setting action levels in nonbinding guidance makes enforcement difficult

In 1994, FDA issued regulations for bottled water, setting enforceable limits on lead and other contaminants. If industry exceeds the limits, it must recall the product and is subject to fines for violating the law. In our experience, the risk of a recall and fines prompts companies to monitor their products and suppliers carefully and to set tight internal standards.

But for action levels (such as those for arsenic or lead in juice), the agency relies on guidance that “represents the [FDA’s] current thinking on this topic.” This guidance comes with a disclaimer: “The contents of this document do not have the force and effect of law and are not meant to bind the public in any way, unless specifically incorporated into a contract.”

The agency relies on guidance rather than rulemaking because it sees the former as faster and less burdensome. Guidance can be effective if cautious companies treat it as an enforceable limit. But not all companies are cautious—especially those that lack the resources or sophistication to recognize the risks from exceeding the action level, pay for sufficient sampling and appropriate analysis at credible labs, and actively manage their suppliers.

FDA lacks the means to ensure compliance with action levels

Most environmental regulations require the regulated community to do extensive reporting of test results and evidence of potential problems. In contrast, FDA lacks a similar regulatory infrastructure for foods—relying instead on physical inspections, market sampling, and voluntary reporting.

While physical inspections are essential, they are conducted every 3 years for high-risk facilities: it can be as long as every 8 years for others. This is too infrequent to spot problems with food before consumers are harmed.

Most agencies, including FDA’s drugs program and EPA, also have the authority to require that the regulated community provide compliance information without a physical inspection. However, FDA has concluded that it lacks those authorities for food. Therefore, a food inspector must physically visit a facility to review compliance programs and testing results. The pandemic has made the problems with this shortcoming abundantly clear, prompting FDA to ask Congress for specific authority earlier this year. We support this request.

Just as we rely on EPA to test for air and water quality, we know that having FDA testing foods in the marketplace and monitoring for disease outbreaks is critical. But FDA requires very little regular reporting—leaving the agency to react to rather than prevent problems. For substances like lead and arsenic—where the harm can be subtle and may take years to manifest—it has serious limitations.

Requiring more timely reporting would be best. However, another mechanism to strengthen compliance with action levels could be FDA’s new Laboratory Accreditation for Analysis of Foods (LAAF) program. The December 2021 rules establishing that program provide a means for FDA to ensure laboratories are conducting food testing and analyzing samples properly. It also requires companies to submit their food testing results and supporting information to FDA.

But there is yet another drawback: Companies are only required to use the labs in the program in certain circumstances, such as when there is an import alert or a regulatory limit. And there is no option for voluntary reporting of test results.

Strong action levels drive research and testing to ensure compliance

Action levels should prompt industry to conduct testing. But if the limits affect only a small percentage of the market – as is the case for FDA’s proposal for lowering lead levels in juice – companies have little incentive to invest in driving continuous improvement.

By contrast, strong action levels will drive companies and federal agencies like USDA’s Agricultural Research Service to make investments in research to identify best practices and to manage their supplies more tightly. An excellent example is FDA’s 2016 proposed inorganic arsenic limits for infant rice cereal. When the agency proposed the limits, it said 53% of the products sampled from retail stores in 2014 exceeded the limit. Four years later, less than 24% exceeded the limit – still too much, but significant progress.

The proposed limits also drove research investments by food companies, growers, and USDA to understand best practices to grow, harvest, process, and prepare rice to reduce inorganic arsenic.

Despite sustained attention on this issue, we have yet to see this result for other crops. In August 2022, a consortium of universities led by Purdue requested $8 million from USDA to research best practices to reduce lead, arsenic, and cadmium in carrots, sweet potatoes, squash, and leafy greens. The project just missed the funding cut because 25% of grant reviewers scored the issue as a low priority—despite public announcements by both FDA and USDA that toxic element contamination of children’s food is a high priority.

Even with strong limits, continuous improvement will require FDA to have a stronger mechanism to drive research and testing. It should consider the option of using guidance to establish current action levels—as well as future targets—to help industry prepare. For example, a certain percentage reduction in the action level every several years that could be adjusted based on compliance information as the deadline approaches.

Conclusion

For FDA’s Closer to Zero program to drive continuous improvement, the agency needs stable funding from Congress. It also needs more robust mechanisms to ensure compliance, including the authority to secure relevant information without a physical inspection. And finally, it must set more protective action levels and clearly signal to the public and to industry that the agency is likely to tighten action levels in coming years.

[1] See page 33 of 98 in the PDF.

[2] See page 117 and 125. The $7 million is part of the $8.5 million for Maternal and Infant Health.