Setting the Record Straight on GRAS: Part 1

April 15, 2014

By Ricardo Carvajal & Diane B. McColl

When GAO issued its 2010 report criticizing FDA’s oversight of the GRAS exception, we were motivated to respond first in short form in a blog post, and then in long form through a Washington Legal Foundation Legal Backgrounder.  In the years since, attacks on the GRAS exception have grown increasingly shrill and detached from reality, so we thought it would be a worthy exercise to begin cataloguing the principal errors and mischaracterizations underpinning those attacks.  Even if one presumes that FDA’s oversight of the GRAS exception needs to be improved, knowing which improvements would be beneficial would require an accurate diagnosis of where problems might lie.  Many of the diagnoses offered thus far are based on unwarranted assumptions, erroneous analyses, and factual misrepresentations.  The catalogue of such flaws has grown so vast that we don’t aim to complete our task in a single blog posting – thus the designation of this posting as Part 1 of a series that we’ll continue in the coming weeks.

We begin with the report that the Natural Resources Defense Council ("NRDC") released earlier this week, Generally Recognized as Secret: Chemicals Added to Food in the United States.  That report contends that the GRAS exception is “the loophole that swallowed the law.”  According to the revisionist history of the Food Additive Amendments of 1958 ("FAA") offered in the report:

The 1958 law exempted from the formal, extended FDA approval process common food ingredients like vinegar and vegetable oil that are “generally recognized as safe” (GRAS). It may have appeared reasonable at the time, but that exemption has been stretched into a loophole that has swallowed the law. . . .

Congress concluded that the FDA would make all safety decisions, except in the most obvious situations in which a chemical’s use in food was “generally recognized as safe.”  This is known as the GRAS exemption.  Examples include such common food ingredients as oil and vinegar.

The suggestion that the GRAS exception was intended only for common household food ingredients like oil and vinegar is repeated on NRDC’s website, both in text and video formats (see here).  That suggestion is, at the least, misleading.

In passing the FAA, Congress was well aware that numerous substances other than common food ingredients would be eligible for GRAS status.  That is evident in FDA’s first GRAS list, which was published shortly after the law’s passage.  In publishing that list, FDA stated: 

It is impractical to list all substances that are generally recognized as safe for their intended use. However, by way of illustration, the Commissioner regards such common food ingredients as salt, pepper, sugar, vinegar, baking powder, and monosodium glutamate as safe for their intended use.  In addition, the following lists include some substances that, when used for the purposes indicated, in accordance with good food manufacturing practice, are regarded by the Commissioner as generally recognized as safe for such uses. [(Emphasis added.)] 

The lists compiled by FDA included dozens of substances in the categories of buffers and neutralizing agents, sequestrants, antimycotics, antioxidants, nonnutritive sweeteners, and emulsifying agents, among others.  The existence of FDA’s first GRAS list is known to the authors of the NRDC report because they cited to it in one of their earlier PEW publications (see Navigating the U.S. Food Additive Regulatory Program, at 347).  Evidently, acknowledging the variety and nature of the substances on the first GRAS list has grown inconvenient, as it undermines one of the NRDC report’s central claims – namely that both FDA and industry have perverted the original purpose of the law.

Even a cursory reading of the FAA’s legislative history makes clear that the law was intended to achieve two principal purposes, only one of which was to ensure the safety of substances added to food.  The other principal purpose, increasingly ignored by critics, was to avoid unnecessarily impeding innovation in the food supply.  In that respect, the GRAS exception has played a pivotal role, and its continuing utility should come as no surprise given the increasingly sophisticated analytical tools available to experts in the field of food toxicology.  The flexibility afforded by the GRAS exception is such that FDA relied on it with great success in developing its regulatory paradigm for bioengineered foods – a technological advance that certainly was not on the Congressional radar in 1958.  In summary, continued reliance on the GRAS exception is perfectly consistent with the twin purposes of the FAA, and has yielded the added benefit of enabling both industry and FDA to direct their food safety resources to areas where those resources can yield a greater return on investment – a significant benefit given the anticipated costs of implementing FSMA.

Another of the report’s central claims is that GRAS determinations are plagued with conflicts of interest because companies have a financial incentive to sell their products, and the fear of legal liability and damage to reputation are insufficient to counter that incentive.  In support of that contention, the report states:

The FDA has acknowledged that a company’s potential liability and its interest in protecting its brand are insufficient to ensure that food is safe.  In 2013 the agency said, “Because the demand for many manufactured or processed foods may not be sufficiently affected by safety considerations, incentives to invest in safety measures from farm to fork is diminished. Consequently, the market may not provide the incentives necessary for optimal food safety.”  “Even in cases where consumers are aware that their illness was contracted from a specific food,” the FDA explained, “it is often difficult to determine who is ultimately responsible for their illness, since the particular source of contamination is not known in many circumstances.”  It concluded that “it is unlikely that the existence of brands in the food sector creates the optimal level of safety for society.”

As a supporting reference, the report cites to FDA’s Preliminary Regulatory Impact Analysis for the Proposed Rules for Current Good Manufacturing Practice and Hazard Analysis and Risk-Based Preventive Controls for Human Food.  However, a review of that document reveals another inconvenient fact.  The quoted statements were made in the context of unintentional contamination of food.  In that context, it stands to reason that tracing an illness back to a specific product manufactured by a specific company is very challenging, and therefore the likelihood that the responsible company would be held accountable is diminished.  That logic does not apply in the context of the intentional use of an ingredient that is declared on a product’s label.  Indeed, NRDC had no difficulty identifying specific ingredients, products, and companies to target in its current report.

The relative ease with which a product can be identified is not the only factor that distinguishes the food ingredient context from the unintended contaminant context.  In the food ingredient context, FDA doesn’t have to show that the ingredient actually caused harm in order to challenge its use as an unapproved food additive; it is enough for FDA to show that general recognition of safety of the use of the ingredient is lacking.  The power of FDA’s enforcement tools in that context is amply demonstrated by some of the agency’s recent activities.  For example, FDA has issued warning letters to distributors of foods that contain melatonin (see here and here). The letters contend that FDA is unaware of any basis to conclude that the use of melatonin as a food ingredient is GRAS, and therefore the product in question is adulterated by virtue of containing an unapproved food additive.  FDA does not assert that the products in question actually caused injury because the agency doesn’t have to make that showing.

Similarly, when FDA announced its intent to investigate the safety of caffeine in food products, the mere expression of FDA’s concern was sufficient to prompt a number of companies to stop distribution.  That announcement followed on the heels of FDA’s prior warning letters targeting certain caffeinated alcoholic beverages, which the manufacturers subsequently stopped distributing.  These actions and industry’s response thereto give the lie to any claim that FDA’s ability to address potentially unlawful uses of food ingredients is hampered by the agency’s inability to show harm.  If anything, the challenge FDA faces in exercising its food additive authority lies in doing so judiciously, as illustrated by the agency’s recent, sweeping proposal to revoke the GRAS status of partially hydrogenated oils.

In closing out Part 1, we note that, whereas some critiques of the GRAS exception have been intended to enlighten, the current critique appears primarily intended to frighten.  The narrative is simple – profit-hungry companies are using unsafe chemicals in food to pump up their bottom lines, under the lazy eye of an agency that relies on outmoded science and an antiquated law.  While that narrative might provide a nice fund-raising platform for an NGO, it contributes little to the reasoned discourse needed to evaluate whether and how the existing system might be improved, and which improvements might yield benefits commensurate with their costs.  In that vein, we note that industry is stepping up to the plate with the establishment of a Center for Research on Ingredient Safety ("CRIS") in conjunction with Michigan State University.