I. Introduction
The United States is one of the most economically developed nations in the world, yet the health of its citizens is failing. Despite spending the most on health care, Americans “are the sickest and die the youngest compared with nine other high-income nations.”Footnote 1 While this tragic picture can be attributed to a slew of factors — economic, social, and environmental, to name a fewFootnote 2 — much of the blame lies at the feet of our diets.Footnote 3 Over half of U.S. adults suffer from chronic diet-related diseases, such as obesity, cardiovascular conditions, irritable bowel syndrome, and cancer,Footnote 4 and diabetes alone affects more than 11.5% of the population.Footnote 5 More than 80,000 new cancer cases a year are linked with suboptimal diet among U.S. adults.Footnote 6 Even maladies not typically associated with food, like chronic kidney disease, are caused to a large degree by nutritional factors.Footnote 7 While diet itself is a complex problem, encompassing malnutrition, nutrient deficiencies, irregular feeding, overconsumption, diet quality, low physical activity, and more,Footnote 8 the chemical load of our food supply is a major, yet often overlooked, culprit.Footnote 9 Food chemicals have been linked to neurological and immunological problems, obesity and cardiovascular disease, endocrine disruptions, impaired brain development, cognitive and behavioral issues such as ADHD, and cancers.Footnote 10
How did we get here? For millennia, human food came in whole form — the carrot in the ground, the apple from the tree, the milk from the cow — and was prepared at home.Footnote 11 This was not always an idyllic existence; food in those days rotted easily, carried pathogens, and was contaminated with heavy metals from soil, cookware, and even intentional use.Footnote 12 Synthetic chemicals, however, did not exist and were not a salient issue for food safety until the late nineteenth century.Footnote 13
With the dawn of industrialization, food manufacturers began using synthetic chemical compounds for the purposes of food preservation, shelf stability, flavor and texture enhancement, and nutrient supplementation.Footnote 14 While these developments provided some important benefits, they also introduced new dangers to human health.Footnote 15 Early examples of food additives include stimulants in drinks, formaldehyde in milk, and borax in meat.Footnote 16 Congress responded to this rapid proliferation of chemical food additives that could be injurious to consumers first with the passage of the 1906 Pure Food and Drugs Act,Footnote 17 and later with the 1938 Food, Drug, and Cosmetics Act (“FDCA”).Footnote 18 This early food safety regime placed the burden of proof for food safety on the FDA.Footnote 19 Manufacturers were allowed to introduce substances into the food supply without advanced notice and authorization, and the FDA had to affirmatively prove that these substances were harmful for human consumption before it could deem the relevant food products adulterated and remove them from the market.Footnote 20
The passage of the 1958 Food Additives Amendment ostensibly flipped the burden of proof for intentionally added ingredients onto food manufacturers,Footnote 21 who now had to affirmatively prove that their proposed additive was safe for human consumption before they could receive FDA authorization to include it in food products.Footnote 22 In theory, this burden shift should have ensured a much safer food supply. In practice, however, low statutory safety requirements and outsized exceptions allowed manufacturers to do the bare minimum and still placed a high load on the FDA to chase down and weed out unsafe substances.Footnote 23 In turn, skewed market incentives, a slim workforce, and a de-prioritization of food safety in FDA’s overall agenda meant that the agency was in no position to shoulder this functional burden.Footnote 24 Ultimately, the burden passed on to consumers to painstakingly weed out unhealthy foods or endure the increasing number of chronic illnesses caused by our vulnerable, polluted, and sometimes dangerous food supply.
Is there a way out? While food advocates and health care professionals have been sounding the alarm bells on these trends for decades,Footnote 25 these issues have finally become a mainstream topic of discussion.Footnote 26 Nutrition and food safety are increasingly a primary concern for consumers, voters, and politicians alike.Footnote 27 The recent rise of the “Food As/Is Medicine” movement illustrates the heightened awareness about the crucial role food plays in both causing and alleviating illness.Footnote 28 Similarly, the current administration’s focus on “Mak[ing] America Healthy Again” has led HHS Secretary Robert F. Kennedy Jr., FDA Commissioner Marty Makary, and USDA Secretary Brooke Rollins to make a significant push towards addressing dangerous food chemicals and additives.Footnote 29 Working within the existing statutory confines and practical realities of FDA’s limited capacity to police food safety, however, can only go so far. Therefore, despite enormous political will, the current “wins” in this area have come in the form of calling for voluntary commitments from industryFootnote 30 — gains that are neither permanent, nor binding,Footnote 31 and do not solve the safety problems that will inevitably arise with subsequent waves of new chemical additives.
This article offers a blueprint for reshaping food safety review in a more effective, enduring, and consistent way by requiring that food products meet rigorous safety standards akin (though not identical) to those required for pharmaceuticals. The basic premise of the argument is burden-shifting. The proposed reforms seek to realign statutory requirements and regulatory incentives in a way that lifts the burden from sickened consumers and an overworked FDA, and places it onto those who stand to gain the most from the introduction of food chemicals into the food supply — the manufacturers. The article proceeds in four parts and has both a descriptive and a normative component. Part I interrogates the central claim of the recent “Food Is Medicine” movement — that food is, or at least can be, medicine. This part reviews both the historical and contemporary links between food and medical drugs and posits that one implication of viewing food as medicine in terms of efficacy is that it should also be treated as such when assessing its safety. Next, the article offers a comparative analysis of two regulatory regimes. Part II examines the process for approving new drugs and the burden of proof on drug manufacturers to demonstrate safety before they are allowed to market their products to consumers. Part III shifts gears to food regulation and catalogues the many and significant ways in which food safety review lags immensely behind drug safety assessments and falls short of protecting consumers. Finally, Part IV puts forth a comprehensive package of proposed reforms that takes a page from the playbook of drug regulation and seeks to alleviate the administrative burden on the FDA and the health burden on consumers by holding food manufacturers responsible for the safety of their products.
II. Why Treat Food as Medicine?
The central claim of the Food Is Medicine (“FIM”) movement is simple: food is — or at least can be — medicine.Footnote 32 While this movement has tremendous intuitive appeal, public support, and strong scientific backing,Footnote 33 the discourse surrounding FIM often ignores a crucial link between food and health: food safety. This section examines this link by providing both a historical perspective and a modern nutritional take, and argues that food can only be medicine if we first ensure that, like any other medicine, it is safe.
A. The Link Between Food and Medicine
For most of human existence, and up until the middle of the nineteenth century, both food and medicine were exclusively sourced from plants, animals, and other natural sources,Footnote 34 and “the boundary between food and drug — and that between dietetics and pharmacology — was rather blurred.”Footnote 35 From the Egyptian Ebers Papyrus to Hippocrates’ Nutriment, early medical treatises readily linked proper nutrition with health and prescribed food as medication for various ailments.Footnote 36 Entire systems of medicine, like Chinese traditional medicine and Ayurveda, relied extensively on medicinal foods.Footnote 37 Likewise, Western medical tradition prior to the mid-to-late nineteenth century wholeheartedly adopted the use of food and food-derived compounds to ensure health and treat diseases.Footnote 38 These early traditions, however, were also acutely aware of the dangers that unsafe food could pose.Footnote 39 The knowledge of food toxicity, dose-dependent foods, and food pathogens was carefully passed down to the next generation through oral tradition and, later, scientific notes and records.Footnote 40
The modern-day Food Is Medicine movement embraces the interconnectedness between food and health but often lacks a food safety focus. The movement initially began in response to the HIV/AIDS epidemic in the 1980s, when volunteers brought food to afflicted people in the absence of formalized medical treatment.Footnote 41 In the 1990s, the use of medically tailored meals as a form of medical care increased with the passing of the Ryan White CARES Act.Footnote 42 These efforts eventually led to the creation of the Food Is Medicine Coalition, a group of nonprofits and other community-based organizations dedicated to addressing hunger.Footnote 43 The FIM movement in its present-day form gained nationwide recognition in 2022 after the U.S. Department of Health and Human Services (“HHS”) developed an initiative under the same name.Footnote 44 Today, HHS’s Office of Disease Prevention and Health Promotion coordinates federal and state government efforts to provide clinical care access, food, and educational support to medically vulnerable populations.Footnote 45 These include medically tailored meals, prepared and designed by registered dietitian nutritionists for patients with complex medical conditions; medically tailored groceries for patients with diet-related medical conditions; and produce prescriptions in the form of vouchers or debit cards.Footnote 46 An increasing number of states have also passed FIM legislation that provides coverage for direct food interventions and nutrition education.Footnote 47 FIM interventions not only increase the consumption of fresh and nutritious food, but also significantly reduce chronic disease risk factors and aid healing.Footnote 48
The FIM movement clearly recognizes the role that food plays in health — that an unhealthy diet (usually defined as one rich in sugar, fat, and sodium) can cause diet-related diseases and worsen other medical conditions, whereas a healthy diet (usually defined as fresh fruits and vegetables and clean protein) can have a tremendous medicinal effect.Footnote 49 What the FIM discussion often neglects, however, is that every medicine needs to be not only effective but also safe. Food — even prescription produce — can be a vector for unsafe agents, such as environmental chemicals and pathogens, and can have detrimental acute or chronic health effects on patients.Footnote 50 Food safety becomes an even more pertinent concern with the proliferation of synthetic chemicals in processed foods coupled with the virtual certainty that, even in the best of circumstances, patients with complex or diet-related medical conditions will rely on processed food for at least part of their diet.Footnote 51
B. The Link Between Synthetic Pharmaceuticals and Processed Foods
The significant advances in chemistry and manufacturing capabilities during the country’s science and industrial revolutions brought on rapid proliferation of synthetic pharmaceutical agents and processed food products that changed the landscape of both nutrition and medicine.Footnote 52 These developments, however, did not break the link between food and medicine; they provided an additional dimension to it.
Chemically, pharmaceuticals and processed foods are closely linked. Indeed, many early pharmaceuticals (analgesics and antipyretics) and early food additives (dyes) originated from the same source — coal tar.Footnote 53 And often, both drugs and food products relied on the same chemical compounds: synthetically derived salicylic acid, for example, was widely used both as pain and fever reducers (and, in its esterified form, later became the active ingredient in Aspirin) and as a common food preservative.Footnote 54 Functionally, these powerful compounds held a lot of promise to alleviate various health-related threats. Pharmaceuticals, of course, do so by interacting with the body on a molecular level, while processed foods contribute to public health by staving off food rot, controlling pathogen proliferation, supplementing vital nutrients, and even providing sustenance for populations without access to fresh food.Footnote 55 Conditions such as botulism, scurvy, and neural tube defects are exceedingly rare today largely due to synthetic additives in our food supply.Footnote 56
Both synthetic pharmaceuticals and food additives, however, came with the potential to harm human health.Footnote 57 As a result, the predecessor to the modern FDA — established in 1862 and aptly named the Division of Chemistry (renamed the Bureau of Chemistry in 1901) — examined the safety of these potent chemical compounds in both medicine and food.Footnote 58 Harvey Wiley, Chief Chemist of the Bureau at the time and often credited with the passage of the first food safety laws, began testing the health effects of various food additives in human subjects because he firmly “believed that everyone in the country was eating chemically treated foods, that all consumers were receiving daily apothecary doses.”Footnote 59 In other words, as a matter of chemistry, the substances in processed food products were in many ways equivalent to those in drugs. And, although chemicals occurred in lower doses and concentrations in individual food products than in pharmaceuticals, early thinking within the FDA determined that consumers’ ingestion of chemicals in pharmaceuticals and their cumulative exposure to chemicals in food both deserved the same caution.Footnote 60
The landscape of synthetic chemicals and food products has changed significantly over the last 100 years, and the chemical load on consumers today is much higher than in Harvey Wiley’s time. More than 10,000 chemicals are used in processed food currently,Footnote 61 appearing in nearly all products from “junk foods” to “healthy” processed foods like almond milk and gluten free items, to even minimally processed foods like yogurt, meat, and dried fruit.Footnote 62 By latest estimates, between sixty and ninety percent of the typical American diet consists of processed food.Footnote 63 Thousands of these chemicals — including ninety-nine percent of chemicals introduced over the last twenty-five years — have never been subject to meaningful safety review.Footnote 64 Of the small minority of chemicals that have undergone some scrutiny, many have been linked to serious detrimental health effects, “including risk of cancer, developmental harm, and hormone disruption.”Footnote 65 And these numbers only cover the chemicals that are intentionally added to food.Footnote 66 Beyond these, our food supply also carries migrated substances, environmental contaminants, veterinary medicinal products, and pathogens, all of which can wreak havoc on human health.Footnote 67 More than ever before, consumers today are truly receiving “apothecary doses” of multiple chemicals with their every meal. Yet, at a time when our government, medical community, and public proudly announce that “food is medicine,” we treat food as anything but medicine when it comes to ensuring its safety.
The recent advances in science and medical knowledge,Footnote 68 combined with the practical reality of the role processed food plays in our diets, necessitate restructuring our food safety regime to mirror the pharmaceutical safety regime more closely. Indeed, although drugs contain chemicals in higher doses and concentrations, many attributes of food products make food safety review arguably more important. For one, most drugs are used for acute conditions and therefore consumers are likely to be exposed to them only for short periods at a time.Footnote 69 Even drugs for chronic conditions are typically taken once or twice daily, and only after onset, usually in late adulthood.Footnote 70 By contrast, most humans eat three to five times a day, every day for their entire life since infancy,Footnote 71 and most consumers tend to stick with the same types of foods over long periods of time, thus risking slow chronic exposure that is difficult to trace.Footnote 72 Second, medical drugs are prescribed by and taken under the supervision and care of licensed physicians, who are better equipped to understand safe doses, account for interactions with other chemicals, and track adverse reactions from the medication.Footnote 73 By contrast, a lay consumer has neither the knowledge nor training — much less the means — to test for and observe side effects from their food consumption or to consider chemical interactions and cumulative loads.Footnote 74 Finally, most medical drugs are taken out of medical necessity, and therefore even established side effects may be tolerable and the lesser of two evils.Footnote 75 By contrast, no single food is necessary for human thriving or healing, and therefore approving new food chemicals is never an emergent matter, nor is a somewhat unsafe product ever a worthwhile health tradeoff.Footnote 76
These differences between food and pharmaceutical intake arguably counsel in favor of stricter safety review for food products. This article does not go that far, nor does it advocate for food to undergo the exact same strict testing as drugs before going to market. That said, the gross disparity between how the FDA considers drug and food safety is illustrative of a foundational error, which must be remedied if the FDA is serious about protecting and enhancing Americans’ health.
III. New Drug Safety Review
New drugs go through “a lengthy, detailed, and costly approval process to determine [their] safety and effectiveness” before they are approved for market.Footnote 77 “A typical FDA new drug review involves hundreds of thousands of pages of data, and may require reviewers to rerun data analyses, to query companies for more information, and to closely scrutinize individual trial records.”Footnote 78 While this process is far from perfect and does not lack reasonable detractors,Footnote 79 it is hard to dispute the fact that safety review is an essential — indeed, paramount — component of developing, approving, and marketing new drugs.Footnote 80
Drug sponsors begin this lengthy process in the discovery or preclinical phase, usually conducting several rounds of laboratory, in-vitro, or animal testing to assess whether the product exposes humans to unreasonable risk and justifies commercial development.Footnote 81 In addition to pharmacology, toxicokinetic, and pharmacokinetic studies, the FDA also recommends several types of dose studies (e.g., acute, sub-chronic, and chronic toxicity, repeated-dose toxicity, microdose, etc.), genotoxicity, carcinogenicity, reproductive toxicity, immunotoxicity studies, and combination drug toxicity studies.Footnote 82 While none of these studies are strictly required, industry takes this guidance very seriously in the context of new drug development.Footnote 83 Through this process, the FDA can compel the generation of meaningful, balanced, and unbiased data.Footnote 84 New drug sponsors engage with agency representatives at the very beginning of their research and development process and continue to interact with them and receive guidance on additional information that would support a safety determination throughout the preclinical phase.Footnote 85 Thus, the preclinical phase typically takes at least eighteen months and generates numerous distinct toxicity and safety studies.Footnote 86 “Dozens, hundreds, or even thousands of biologically active agents may be tested in this initial stage of drug development for every drug that receives final approval.”Footnote 87 If any of these tests result in adverse reactions, the FDA requires that sponsors submit a safety report as part of their application.Footnote 88 In this phase, most proposed drugs go through a significant “winnowing” process, due in part to “the fact that it takes, on average, twelve years and costs more than $200 million to develop and obtain approval for a new drug.”Footnote 89
Only if this initial exploration shows promise and raises no significant safety concerns, a drug sponsor can file a formal Investigational New Drug Application (“IND”) with the FDA to proceed with human trials.Footnote 90 The IND substantially concerns safety, rather than efficacy, and requires that the sponsor include all previous test results and a plan of the proposed methods of investigation in humans.Footnote 91 Upon approval of an IND by the FDA and a local institutional review board (“IRB”), sponsors begin clinical trials that test for safety, effectiveness, side effects, unwanted interactions, and consistency.Footnote 92 Phase I of these clinical trials usually involves between twenty and eighty healthy human subjects and is designed to determine the safety, metabolic, and pharmacological profile of the drug and “the side effects associated with increasing doses.”Footnote 93 Sponsors can only move to Phase II efficacy trials if the drug does not reveal “unacceptable toxicity” levels.Footnote 94 Phase II trials, involving larger samples of patients with the targeted condition, mostly focus on drug efficacy.Footnote 95 However, in this phase, a sponsor also seeks to determine “the common short-term side effects and risks associated with the drug” and maximum safe dose of the drug.Footnote 96 Finally, Phase III trials occur in larger sample populations and focus on rarer side effects, drug interactions, and other safety and effectiveness information necessary to evaluate the “benefit-risk relationship of the drug.”Footnote 97
Throughout the clinical trial process, sponsors may only move to a later phase of testing if no safety concerns are raised in the previous phase.Footnote 98 Thus, drug safety is the foundation of new drug approval.Footnote 99 Indeed, “at any time during the clinical trials, a drug sponsor is required to notify the FDA of ‘[a]ny adverse experience associated with the use of the drug that is both serious and unexpected,’” and “the FDA may order a ‘clinical hold’ halting the trials if it determines that safety concerns so warrant.”Footnote 100
Due to these stringent safety requirements, about three-quarters of the drugs that reach clinical trials never make it past this stage.Footnote 101 For the remaining one-quarter of proposed drugs, after the conclusion of all three phases of clinical trials — a process that itself typically takes several yearsFootnote 102 — the drug sponsor submits a New Drug Application (NDA) to approve the product for U.S. marketing.Footnote 103 An NDA must contain (1) the full investigative reports of product safety and effectiveness, (2) the drug articles used for components, (3) the drug’s full composition, (4) the methods used in manufacturing, processing, and packaging, (5) any samples of the drug and the components used per the request of the Secretary of Health and Human Services (HHS), and (6) the proposed labeling of the product.Footnote 104 The NDA may also contain separate pediatric assessments, if applicable.Footnote 105 A typical NDA “frequently runs in the hundreds of thousands of pages, cataloguing the entire history of the drug’s development and reporting on all the premarketing studies.”Footnote 106 Despite this exhaustive record, during the review of the application, FDA’s Center for Drug Evaluation and Research (CDER) may require that the applicant provide additional drug information or even appear for a hearing before the Secretary of HHS.Footnote 107 The filing of an NDA entails hefty user fees for sponsors — for fiscal year 2025, the rates for an application with and without clinical data were $4,310,002 and $2,155,001, respectively.Footnote 108
Preclinical and clinical testing of new drugs is typically not separated by active and inert ingredients, so sponsors must ensure that the test substances used in preclinical and clinical testing are the exact combination substances they intend to market.Footnote 109 The drug substance consists of all the physical and chemical characteristics of a drug, as well as all the manufacturing information of the substance and the specifications needed to replicate the “identity, strength, quality, and purity” of the substance.Footnote 110 Therefore, the data filed with the NDA should reflect the overall safety, adverse reactions, and interactions of the entire drug, including all its active and inert ingredients.Footnote 111 The FDA decides on a case-by-case basis whether to require further testing to see if any of suspect components are unsafe.Footnote 112
Because no amount of testing could ever rule out all possible side effects and concerns, absolute safety is not an achievable standard.Footnote 113 Additionally, depending on the intended benefits and efficacy of the drug, the FDA may be willing to tolerate a higher degree of safety risk to ensure that critically ill patients receive the care that they need.Footnote 114 In recognition of this reality, the FDA also engages in a robust labeling process. As part of the NDA examination, the drug sponsor must submit proposed labeling for review, which must include warnings, indications, contraindications, dosage recommendations, and interactions.Footnote 115 The FDA then reviews the proposed labeling and engages in sometimes lengthy negotiations with the drug sponsor over adding more thorough information.Footnote 116 The labeling process remains active after market approval, with the manufacturer remaining responsible for updating the labeling as new information about adverse reactions and contraindications becomes available.Footnote 117 For instance, the FDA allows drug manufacturers to change their labeling without approval, if such changes serve to strengthen safety warnings and add precautions or adverse reactions.Footnote 118 The FDA can further order updates to the labeling post approval.Footnote 119
After market approval and distribution, manufacturers are required to maintain records and reports such that the Secretary can determine withdrawal or immediate suspension upon a finding of an imminent hazard to the public health.Footnote 120 Sponsors are required to submit annual reports that include significant new information from the previous year affecting the safety, effectiveness, or labeling of the drug product and a description of actions the sponsor has taken or intends to take as a result.Footnote 121 Sponsors are also required to update the drug label to include warnings about clinically significant hazards as soon as there is reasonable evidence of a causal association with the drug, even if a definitive causal relationship has not been established.Footnote 122 The FDA may also require further post-approval clinical trials if new safety information arises after the drug is marketed.Footnote 123 Such information could come in the form of a clinical trial, an adverse event report, a post-approval study, peer-reviewed biomedical literature, data derived from the post-market risk identification and analysis system,Footnote 124 or other available scientific data.Footnote 125 If the FDA determines that there is a safety concern with an approved product, it may require specific risk evaluation and mitigation strategies to ensure safe use.Footnote 126 Failure to comply with the imposed strategies may result in civil or criminal penalties, withdrawal of drug approval, or other enforcement actions.Footnote 127
Beyond ongoing safety testing and post-market reporting, drug manufacturers are also required to use current Good Manufacturing Practices (“GMPs”)Footnote 128 and conduct continual purity testing. Manufacturers must have laboratory controls in place to ensure conformity to identity, strength, quality, and purity standards of components, in-process materials, and final drug products.Footnote 129 Each lot of components, drug product containers, or closures must be tested for contamination and microbiological purity,Footnote 130 while each batch of product must be tested for conformance to final specifications, including the identity and strength of active ingredients and, where applicable, for sterility and absence of objectionable microorganisms.Footnote 131
This approval process is not beyond criticism. Drugs with problematic safety profiles have previously gained FDA approval and have been marketed to the public with detrimental health effects and little to no post-market intervention by the FDA.Footnote 132 The cost and length associated with this approval process also create unnecessary barriers to entry and discourage pharmaceutical innovation for rare diseases, smaller patient populations, or by new market participants.Footnote 133 These and many more concerns are beyond the scope of this article. The point of surveying the regulatory safety requirements for new drugs here is not to raise them as a paragon of regulatory perfection but simply to point out that legislators and regulators alike take the safety of drug chemical formulations that would be ingested by the public seriously.Footnote 134
IV. Food Safety Review
The process of approving new food ingredients is “much weaker” and ineffectual in ensuring safety than FDA’s oversight over prescription drugs.Footnote 135 Unlike medical drugs, which are tested for safety as a complete formulation, without reference to active, inert, or unintended ingredients, whole food products do not undergo any safety review.Footnote 136 Instead, it is only individual ingredients that may require safety approval for specific uses. For ease of conceptualization, this article will review the safety requirements of such food ingredients in three groups: (1) intentionally added food and color additives; (2) expected-to-migrate food contact substances; and (3) unintended environmental contaminants. With every subsequent category, meaningful safety review precipitously declines.
A. Food And Color Additives
Food and color additives have a safety review process that, at first blush, appears similar to the safety evaluation performed for new drugs and has been described by some as “a rigorous statutory scheme.”Footnote 137 The Food, Drug, and Cosmetics Act requires the FDA to “protect the public health by ensuring that … foods are safe, wholesome, sanitary, and properly labeled,” by assessing the safety of food and color additives.Footnote 138
Food additives refer collectively to “any substance the intended use of which results or may reasonably be expected to result, directly or indirectly, in its becoming a component or otherwise affecting the characteristics of any food.”Footnote 139 Typical food additives include flavors and flavor enhancers, fortifiers and nutrient substitutes, emulsifiers, preservatives, stabilizers, binders, acidulants, anti-caking agents, humectants, conditioners, leavening agents, and sweeteners.Footnote 140 The FDA is granted broad authority to evaluate the safety of any food additive by exempting an ingredient from regulation, promulgating “a food additive regulation governing use of the additive,” either temporarily or permanently, requiring “discontinuation of the use of the additive,” or “any combination thereof.”Footnote 141
Color additives, on the other hand, refer to any “dye, pigment, or other substance made by a process of synthesis or similar artifice” or derived “from a vegetable, animal, mineral, or other source,” which is added to food products for the purpose of “imparting color thereto.”Footnote 142 These are regulated separately from any other food additives under Section 721 of the FDCA.Footnote 143 While the FDA regulates all color additives, it only requires certification for synthetic dyes, lakes, and pigments, sourced from petroleum or coal.Footnote 144 By contrast, color additives derived from natural sources, such as plants, minerals, and insects, are exempt from certifications and manufacturers must only comply with identity and purity specifications.Footnote 145
Congress enacted the Food Additives and Color Additives Amendments “in response to public concern about the increased use of chemicals in foods and food processing.”Footnote 146 As is the case with new drugs, the statute requires that, for both food and color additive certification, the sponsors file a petition with the FDA.Footnote 147 This petition must include the “chemical identity and composition of the … additive, its physical, chemical, and biological properties, and specifications prescribing its component(s) and identifying and limiting the reaction byproducts and other impurities.”Footnote 148 It must also include stability data, proposed amount of use, probable exposure, as well as “full reports of investigation made with respect to the safety of the [food/color] additive.”Footnote 149 The regulations further clarify that the safety reports “ordinarily should include detailed data derived from appropriate animal and other biological experiments in which the methods used and the results obtained are clearly set forth,” and “shall not omit without explanation any data that would influence the evaluation of the safety of the … additive.”Footnote 150 Although no specific tests are required, FDA’s Redbook lists a variety of possible preclinical studies, which somewhat overlap with the types of toxicity studies recommended for preclinical drug investigations.Footnote 151 Based on the application and attached comprehensive safety report, the FDA seeks to evaluate whether the color or food additive would be safe, meaning that “there is convincing evidence that establishes with reasonable certainty that no harm will result from the intended use of the … additive.”Footnote 152
The similarities between drug safety review and food/color additive safety review end there. Differences, on the other hand, abound.
1. A Markedly Lower Burden of Proof
The scheme for food and color additives requires a notably lower burden of proof to establish safety for food ingredients than for new drugs. Even a cursory comparison between the food additive regulations (which total eight pages) and the new drug regulations (which clock in at 128 pages) amply demonstrates the significantly laxer safety process for food ingredients.Footnote 153 But the differences run deeper than the page count.
First, unlike new drug regulations, which specifically require a “summary of the pharmacological and toxicological effects” and “pharmacokinetics and biological disposition” of “the drug in animals and, to the extent known, in humans,”Footnote 154 the color and food additive regulations only refer in general terms to any safety information gathered during “appropriate animal and other biological experiments.”Footnote 155 Therefore, although in theory a food or color additive manufacturer can conduct many of the same FDA-recommended studies, there is no expectation that they should. Typically, food additive petitions include substantially less data than a new drug application, and their focus is almost exclusively limited to the chemical’s toxicology and toxicity, if that.Footnote 156 Indeed, a 2013 study of all approved food additives in FDA’s database revealed that a shocking 63.9% of them lacked any feeding toxicity studies, only 6.7% had reproductive toxicology and teratology data, and two had developmental toxicology measurements.Footnote 157 Even looking specifically at chemicals with concern level of II or III — for which the FDA recommends reproductive and developmental toxicity studies — a mere twelve percent of these food additives had such data.Footnote 158
Second, and more significantly, unlike the required three phases of clinical studies in humans for new drugs,Footnote 159 there are no required human safety studies for food or color additives. Instead, the regulations only mention that if “clinical investigations involving human subjects are involved,” the petition should include a statement that each such clinical investigation was “conducted in compliance with the requirements for institutional review … or was not subject to such requirements.”Footnote 160 Indeed, back in 1993, the FDA considered adding a separate chapter to its Redbook to provide guidance to industry on human clinical testing for long-term physiological responses to food or color additives, if preclinical data suggests human subjects will not be put at risk from extended exposure.Footnote 161 However, the agency decided to exclude this chapter from its final publication, further deepening the impression that human studies are not expected for food or color additives. This distinction alone means that food and color additives require years less of safety investigations and likely no data for safety in humans specifically. This is particularly problematic because some food and color additives can cause neuro, cognitive, and behavioral issues in humans that cell or animal-based studies often miss.Footnote 162
And even when food manufacturers do conduct clinical studies for food ingredients, those on balance tend to be much more limited.Footnote 163 “Typically, the sample sizes and objectives of food trials are much smaller and more exploratory than the sample sizes and primary objectives of drug trials” and food trials are primarily focused on substantiating specific labeling claims than assessing safety.Footnote 164 “Food trials do not typically require a safety review by the FDA before the trial can commence,” unless the IRB has specific concerns, and, unlike drug trial sample selection, which is “rigorous,” “[s]ubject selection in food trials is broad and varied, often enrolling healthy persons or those with self-described mild health conditions.”Footnote 165 Unsurprisingly, food trials are also significantly cheaper to conduct, involve fewer placebo and blind presentations, and rely more on “[c]ase studies and anecdotal evidence.”Footnote 166 In sum, what little clinical data is generated in support of a food additive’s safety is hardly sufficient to uncover potential long-term carcinogenic, neurocognitive, autoimmune, or other chronic exposure consequences.
Finally, unlike drugs, which are evaluated as complete products, color and food additives are reviewed for safety only as individual ingredients for a specific intended use. This has several implications. For one, approval of an additive for a particular use does not simply usher in a single product, but hundreds or potentially thousands of food products on the market.Footnote 167 Further, examining the safety of individual ingredients in vacuum prevents an evaluation of adverse interactions with other chemicals in the same food product or across the typical consumer diet.Footnote 168 Nor are reviewers able to properly ascertain the “cocktail effect” of additives and the cumulative exposure to chemicals with common mechanism of toxicity, even though this is an express statutory requirement.Footnote 169 Additionally, although color additives contemplate a dose-dependent approval, the reality of food consumption — which, unlike drugs, is not prescribed by a trained medical professionalFootnote 170 — makes any such dosing restrictions illusory.
2. Exceptions that Swallowed the Rule
What low-level food safety review is contemplated by the statute, and regulations get further gutted by exceptions that have grown large enough to swallow the rules.
In the food additive context, an exception which makes meaningful review of additive safety nearly impossible goes by the acronym of “GRAS,” or Generally Recognized as Safe. During the Congressional debates on the 1958 Food Additives Amendment, legislators discussed the fact that substances such as salt, pepper, sugar, and vinegar, while technically “food additives,” have been used for a long time, and their safety is so widely recognized, that they should be exempt from regulation.Footnote 171 The Amendment, therefore, granted the FDA authority to exempt these substances from regulation when there was a consensus among “experts qualified to evaluate the safety of substances” that they are GRAS.Footnote 172 This rule served to expedite FDA’s review of all food additives on the market following passage of the Amendment and, at the time, seemed to be a limited exception. For the next decade, manufacturers would routinely request FDA opinion letters, hoping for their sponsored ingredient to be included on the published list of GRAS substances.Footnote 173 Over the next few years, however, the FDA, with the help of the Select Committee on GRAS Substances (“SCOGS”), had managed to review only slightly more than 400 substances.Footnote 174
Eventually, the FDA abandoned its review and instead went on to significantly streamline the process.Footnote 175 First, in 1974, the agency clarified that GRAS status could be obtained either through a scientific determination or through a demonstration of “a general recognition of safety” “by experience grounded in the common usage in food for substances used in food before 1958.”Footnote 176 Because many substances had been used in food prior to 1958 without any meaningful safety review, this clarification opened a wide door for food manufacturers. Subsequently, in 1997, the FDA issued a proposed rule that established a voluntary GRAS notification process, through which manufacturers could, but did not have to, notify the FDA of their own private GRAS determinations.Footnote 177 Under this rule, both the safety assessment and the ultimate GRAS determination would be entirely the responsibility of the manufacturer.Footnote 178 If a manufacturer chose to notify the FDA, the agency could engage and either send a ”No Questions Letter" (a type of soft approval), or a “No Basis Letter” that there is not enough evidence in the notification to establish GRAS status, or acknowledge that the manufacturer has voluntarily withdrawn the notification and cease review.Footnote 179 If the FDA sent a “No Basis Letter," however, nothing prevented the manufacturer from withdrawing the notification and continuing to market the food product with the proposed ingredient.Footnote 180
The rule was finalized in 2016.Footnote 181 Ostensibly, under the new regime, a GRAS determination requires the same scientific rigor as that “required to obtain approval” for a food additive under the petition process, except that the manufacturer need not share the actual data on which their claims are based.Footnote 182 In reality, however, the current regulatory scheme allows a manufacturer to (1) obtain a private determination from a third-party scientific body based on “stale, conflict-ridden, and often unpublished, non-peer-reviewed science,” (2) never make the actual data public, (3) not notify the FDA at all, and (4) proceed to market its product for public consumption.Footnote 183 Proponents of the GRAS rule point out that it is in the manufacturers’ interest to engage with the voluntary notification program and receive a “No Questions Letter" from the FDA. That may well be true for those manufacturers who have no concerns about the safety of their ingredients, but the sheer opportunity to evade review is too tempting for industry and too scary for consumers.
Since the FDA instituted the voluntary notification program, it has taken no steps to re-evaluate the safety of GRAS substances or attempt to determine what and how many self-proclaimed GRAS ingredients exist in our food supply. What we do know is that, for the period between 2000 and 2022, there have been only ten new food additive petitions filed with the FDA and only one of these was filed after the 2016 rule was finalized — a petition for Vitamin D mushroom powder.Footnote 184 By contrast, 756 GRAS notices were filed through the voluntary notification program for the same period.Footnote 185 In other words, 98.7% of all new substances registered since the turn of the millennium were self-declared safe by the industry, not tested and reviewed by the agency.Footnote 186 The data that these determinations relied on, to the extent available, has been deemed “inadequate” — of all GRAS notices filed until 2013, 61.8% lacked feeding studies and 31.4% lacked any data.Footnote 187 Some of those ingredients have subsequently raised significant safety concerns: BHA, for example, has been classified as “reasonably anticipated to be a human carcinogen” by the National Toxicology Program; BHT may disrupt endocrine function by causing thyroid changes and affect development; green tea extract EGCG may increase risk of cancer; and carrageenan is “associated with the occurrence of intestinal ulcerations and neoplasms,” as well as precancerous lesions.Footnote 188 Yet, these substances still enjoy their undisturbed GRAS status.Footnote 189 Indeed, the FDA has been notoriously slow to revoke GRAS determinations, even with strong evidence of negative health effects — partially hydrogenated oils (also known as trans fats), for example, maintained their GRAS status from 1958 until 2015, and were permitted in foods until 2021, despite strong evidence since at least 1990 that they had negative health effects.Footnote 190
Even more disturbingly, experts theorize that around 1,000 undeclared ingredients are masquerading as GRAS substances for the purpose of evading meaningful safety review, and neither the FDA nor consumers have any way of identifying them.Footnote 191 As former FDA Deputy Commissioner for Food Michael Taylor stated, “[w]e simply do not have the information to vouch for the safety of many of these chemicals.”Footnote 192 It seems that the only times undeclared GRAS substances come to the attention of the regulators is when one of those ingredients causes an acute health scare for consumers. For example, in 2022, there were nearly 500 unexplained cases of liver toxicity, gastrointestinal pain, gallbladder issues, and other serious medical complications that eventually were linked to a French lentil and leek crumbles product made with tara flour.Footnote 193 Tara flour was neither approved as a food additive nor was there a GRAS notice filed for it.Footnote 194 Instead, the manufacturer had made a private determination of GRAS status based on the ingredient’s use in food products in South America and a single six-week study in rats on general and hepatic toxicity conducted in Peru.Footnote 195 Until the company’s voluntary product recall and identification of tara flour as the culprit ingredient, the FDA was not even aware of this additive’s existence on the U.S. market.Footnote 196 Following this massive health scare, the agency initiated its own GRAS review and concluded — two years after the fact — that there is not enough evidence to deem tara flour as generally recognized as safe and that, therefore, to continue using the ingredient in food products, the manufacturer must file a food additive petition for full review.Footnote 197 Such acute health crises, however, are rare with food ingredients — most additives cause health issues only after a prolonged chronic exposureFootnote 198 — and therefore the hundreds or thousands of undisclosed and untested GRAS additives continue to enter our food supply under the FDA’s radar.
GRAS does not apply to color additives.Footnote 199 Colors, however, along with flavors, spices, and chemical preservatives, can nonetheless evade scrutiny by not being listed on the label. The FDCA generally requires that all ingredients in a food product must be listed on the label by their common name, or else, the FDA must consider the product misbranded.Footnote 200 The statue carves out an exception to this express listing requirement, however, for “spices, flavorings, and colorings,” which can be listed as a group without naming individual ingredients.Footnote 201 While color additives subject to certification must be listed by their proper ingredient name, those that are not subject to certification can simply be designated as “artificial color” or “color added.”Footnote 202 Such group listing, however, makes it impossible for consumers to be on notice about ingredients in their food that they may be sensitive to or that may interact with other chemicals in their diet or medication.Footnote 203 It also makes it difficult to track whether a manufacturer has actually complied with the required FDA certifications or has instead included in their product formulation untested and unregulated ingredients.
A similar problem exists for spices and flavorings. FDA regulations state that flavors only have to be designated as “natural” or “artificial.”Footnote 204 “Natural” flavors can include any essence or extract that starts with a natural product and undergoes a series of complex chemical reactions and may, in its final form, be a mixture of multiple — as many as 250 — undeclared chemicals at a time.Footnote 205 Likewise, artificial flavors, spices, and “chemical preservatives,” can be labeled as such without a specific mention of the exact ingredients.Footnote 206 And an incidental additive originating in the spice or flavor does not need to be listed at all.Footnote 207 These relaxed labeling requirements compound the problem created by GRAS because a manufacturer is not only under no obligation to notify the FDA that they are using an ingredient they deem to be GRAS, but they also do not need to notify anyone that the ingredient is in the product in the first place.Footnote 208 In essence, these exceptions create a complete black box for what untested ingredients may be hiding in our food, in what quantities, and with what adverse effect to our health.Footnote 209
3. Practical Difficulties in Food Safety Review
The safety review gap between new drugs and new food or color additives grows even wider when considering the practical realities of FDA’s review. Although the FDA has oversized responsibilities for food safety, its food budget pales by comparison to its budget for medical devices and pharmaceuticals.Footnote 210 The budget for the Center for Food Safety and Nutrition — the body within the FDA responsible for food safety — was a mere 1.8% of the FDA’s total budget for 2023, and the entire food budget accounted only for eighteen percent of the pot.Footnote 211 The agency’s food budget is also entirely dependent on Congressional appropriations as only one percent of it comes from nominal user fees for color additives ($0.45 per pound of color additive to be certified with a minimum of $288).Footnote 212 Neither food additive petitions nor GRAS notifications generate any user fees, despite the fact that food additives are an increasingly lucrative market — with an estimated market size of over $23 billion in the U.S. alone and $123.98 billion globally.Footnote 213
By contrast, sixty-six percent of FDA’s Human Drug Program’s $2.166 billion budget for 2022 was made up of user fees (between $2 and 4.5 million per application), allowing for meaningful engagement with drug sponsors and robust safety review.Footnote 214 These stark differences in disposable funds result in a much thinner and more overwhelmed food safety workforce, one which often must triage and can focus only on the most urgent matters, such as pathogen outbreaks. Unsurprisingly, then, while new drug review (including study follow-ups and hearingsFootnote 215) takes three to ten years to complete and involves numerous engagements between drug sponsors and FDA reviewers, a typical food additive review takes less than twenty-four months and relies on much more cursory interactions.Footnote 216
4. Lack of Post-Market Action
Finally, significant differences exist in the FDA’s post market review for approved drugs and approved food or color additives. Whereas drug manufacturers have annual reporting obligations, requirements to report any adverse reactions, and a duty to update their labeling for new contraindications — all while facing the looming threat of an FDA-imposed hold or approval cancelation if any source puts the drug’s safety in questionFootnote 217 — food and color manufacturers have no general post-approval reporting obligations.Footnote 218 The only reporting obligation imposed on food manufacturers is if they learn of “a reasonable probability that the use of, or exposure to, such article of food will cause serious adverse health consequences or death to humans or animals”Footnote 219 — an acute situation that typically does not arises in the context of food additives.
Even when the FDA is made aware of safety concerns, it has been notoriously slow in using its post-market authority to revoke food and color additive approvals. The most recent example of Red Dye 3 is instructive. The FDA approved Red Dye 3 in 1969.Footnote 220 Due to mounting evidence of safety concerns throughout the 1980s, including thyroid tumors in rats, the agency banned the use of Red Dye 3 in cosmetics in 1990.Footnote 221 Yet, it took an additional thirty-five years before it finally revoked its approval of the color additive in use in food products in January 2025, to go into effect in January 2027.Footnote 222 Likewise, the FDA took over fifty-four years to revoke its authorization for the use of brominated vegetable oil (BVO) as a food additive.Footnote 223 In 1970, after significant evidence surfaced that BVO could harm vital organs, such as the liver and heart, the FDA downgraded the ingredient from a generally recognized as safe status to that of an approved food additive.Footnote 224 Despite strong scientific evidence, including studies supporting an analogous ban enacted in the UK in the 1970s, the FDA revoked its own approval only in July 2024 — and only after industry had voluntarily stopped using the ingredient for years.Footnote 225
These examples might sound extreme, but sadly, they are representative of the FDA’s slow response to food safety concerns.
Currently, the FDA’s color additives inventory lists nine approved main synthetic color additives (not counting the now-banned Red Dye 3 but including titanium dioxide).Footnote 226 Of these, at least seven have been associated with negative health effects in humans.Footnote 227 In general, consuming artificial food colorings has been proven to increase the amount of manganese and worsening ADHD symptoms in susceptible individuals.Footnote 228 Further, Yellow No. 5 may be linked with negative neurobehavioral changes in children, and several microbiological and rodent studies of Yellow 5 were positive for genotoxicity.Footnote 229 The FDA’s “acceptable” level for Yellow 5 is set at more than sixty times higher than the level that researchers have identified as triggering neurobehavioral effects in a double-blinded, placebo controlled, study of young children.Footnote 230 Blue No. 1 and its analogs have been shown to affect neurodevelopment and hyperactive behavior.Footnote 231 Blue 1 has also been shown to cross the blood-brain barrier even in adults.Footnote 232 Three dyes (Red 40, Yellow 5, and Yellow 6) have been found to be contaminated with benzidine or other carcinogens.Footnote 233 At least four dyes (Blue 1, Red 40, Yellow 5, and Yellow 6) cause hypersensitivity reactions.Footnote 234 Red Dye 40 is also associated with allergic reactions.Footnote 235 Titanium dioxide, first approved by the FDA as a color additive in 1966 on the belief that it does not accumulate in the body, is now known to build up in tissues and to promote precancerous changes and alter the absorption of nutrients from food.Footnote 236 Toxicity tests on Citrus Red 2 and Orange B — two dyes that the FDA recently indicated it would reexamineFootnote 237 — likewise suggest general safety concerns.Footnote 238 Alarmingly, the effects of these dyes are even more pronounced in conjunction with other approved food dyes — a scenario that most closely replicates real-world consumption patterns.Footnote 239 Consumption of over-the-counter medications and vitamins, which likewise include food dyes, further increases exposure to these substances.Footnote 240
Despite many of these ingredients being banned or reexamined in European and other countries,Footnote 241 they are perfectly legal for use in food products in the United States, including for products specifically marketed to children. Indeed, while their consumption here has increased more than fivefold since the 1950s,Footnote 242 the FDA has taken almost no actions to re-examine their approvals. In fact, Red Dye 3 is the only color additive that the FDA has banned for use in food products since 1976.Footnote 243
Likewise, a staggering number of food additives remain approved for use in the U.S. despite significant concerns about their safety. While the exact number is hard to determine due to overlap between the various databases, currently, there are approximately 4,000 food additives approved for use in the United States.Footnote 244 Nearly 1,000 of these are not allowed for use in food in the European Union and other developed countries, yet each of these are found in hundreds to several thousand different U.S. food products.Footnote 245
Several categories of food additives have raised significant safety concerns in the scientific community. Nitrates and nitrites, for example, have long been criticized in the U.S. for use as preservatives in meats and other animal products, due to their significant gastrointestinal impact, carcinogenic potential, and thyroid blocking functions.Footnote 246 Emulsifiers directly affect the intestinal microbiome and increase bacterial genes’ pro-inflammatory activity.Footnote 247 Individual additives have also raised significant safety concerns. Sucralose, for example, has been proven to promote inflammation.Footnote 248 Maltodextrins can lead to increased susceptibility to intestinal damage.Footnote 249 Trichloroethylene was banned by the Environmental Protection Agency in December 2024 as “an extremely toxic chemical known to cause liver cancer, kidney cancer, and non-Hodgkin’s lymphoma.”Footnote 250 Potassium bromate was banned in the European Union, the UK, Canada, India, and Peru and has been classified as a “genotoxic carcinogen” by a joint committee of the United Nations and the World Health Organization.Footnote 251 Yet, these ingredients continue to be widely used in food in the U.S.Footnote 252 Indeed, in the past twenty years, the FDA has only revoked eight additive approvals: bromated vegetable oil plus a batch of seven synthetic flavoring substances, one of which had long been abandoned by industry.Footnote 253
This dynamic highlights the much lower safety threshold that food manufacturers are expected to cross before they can freely market their chemical compounds to consumers and the much higher safety burden that consumers bear each time they purchase a candy bar.
B. Food Contact Substances
The burden of proof on manufacturers decreases, and the safety burden on consumers increases even further with the safety review contemplated for food contact substances. Food contact substances are defined as any substances “intended for use as a component of materials used in manufacturing, packing, packaging, transporting, or holding food if such use is not intended to have any technical effect in such food.”Footnote 254 Under the Food Additives Amendment, food contact substances (“FCS”) were considered “indirect food additives” and were regulated in the same way as direct food additives, through a food additive petition.Footnote 255 An exception was created under the FDA’s Threshold of Regulation rules for any substance with a minimal daily exposure level, which does not present any other health or safety concerns.Footnote 256
A third, more expeditious and direct approval mechanism was introduced in 1997 (effective in 2000) by the Food and Drug Administration Modernization Act.Footnote 257 The Act specifically carved out food contact substances as a separate group of additives.Footnote 258 It also created the Food Contact Substance Notification Program, under which an FCS manufacturer could submit a food contact notification (“FCN”) to the FDA for each new additive and, if the FDA does not object within 120 days, the substance gains automatic market approval.Footnote 259 Due to its lower requirements and more predictable timeline, the FCN program is the preferred route of FCS approval for both manufacturers and the agency. Under this abbreviated safety review, the FDA nonetheless claims to conduct rigorous review of the scientific data and to only approve a substance if “new scientific information” demonstrates that the substance has a “reasonable certainty of safety” for the intended useFootnote 260 — in other words, that there is a “reasonable certainty in the minds of competent scientists that a substance is not harmful under the intended conditions of use.”Footnote 261
The reality is different. Because a filed FCN automatically effectuates within 120 days unless the FDA requests more data or otherwise objects, the FDA’s understaffed workforce matters even more than in traditional food additives review.Footnote 262 Moreover, the level of data required for an FCN “depends on the estimated daily intake of the [substance],” an estimate that the sponsor determines on their own.Footnote 263 Indeed, a self-determination of “no migration” completely eliminates the need for filing a notice because, technically, the substance does not fall within the definition of a “food additive.”Footnote 264 Given that different testing conditions can produce vastly different levels of migration from food contact materials to food, it is not surprising that most manufacturers greatly underestimate the actual exposure of consumers to their proposed substances.Footnote 265
The negative effects of these (already low) statutory requirements for FCNs are compounded by the same practical realities that weaken food and color additive review in general. Food contact substance notifications come with no user fees and often involve very little, if any, meaningful engagement between sponsors and overburdened FDA reviewers — a feature that has been generously termed “one of the most efficient notification processes within the FDA.”Footnote 266 Similarly to the direct food additive context, here too, GRAS presents an oversized loophole that allows manufacturers to introduce food contact substances into their products on the basis of self-certifications and pinky promises, without any meaningful review.Footnote 267 Indeed, many “[l]eading food and drug attorneys long have advised against seeking the agency’s sanction when marketing without it is equally legal.”Footnote 268 Furthermore, post market review and enforcement is virtually nonexistent for food contact substances.Footnote 269 The FDA has gone decades without rescinding FCN status despite mounting evidence of negative health consequences, ordinarily reclassifying FCNs as ineffective only after manufacturers voluntarily phase out the concerning substances.Footnote 270
The low safety burden for approving FCS may not sound particularly troubling when thinking about these substances in the abstract because of the general (wholly inaccurate)Footnote 271 expectation that estimated consumer exposure from migration between food contact materials and foods must be miniscule. The issue becomes much more problematic when examining the outsized detrimental impact on human health of specific FCSs, such as phthalates, parabens, bisphenols (like BPA), perfluorinated compounds (like PFAS), and perchlorate.Footnote 272 Each of these categories of substances has been proven to migrate in significant quantities from food contact materials, causing serious negative health effects.Footnote 273 BPA, for example, is linked with endocrine disruptions, reduced fertility, altered puberty timing, and development of neoplasia; yet, while currently banned for use in infant feeding bottles, it is widely used as a permitted ingredient in plastic bottles and containers and as a polymeric resin to prevent metal corrosion.Footnote 274 Similarly, Perchlorate, which has been linked to major thyroid disruptions, is used as an antistatic agent for plastic packaging and as a cleaning solution in food manufacturing.Footnote 275
Many of these FCS exist in functional groups of thousands of chemicals, all of which have the same or substantially similar effects. The FDA, however, currently regulates FCSs as individual compounds only.Footnote 276 In other words, even if the FDA were to go through the laborious exercise of reviewing and ultimately rescinding an effective FCN — which has not happened in decades — that decision would only apply to a single chemical formulation, allowing a manufacturer to immediately file an FCN for any of the related substances with the same or substantially similar effects. This game of whack-a-mole effectively negates the FDA’s capacity to conduct meaningful post-market review, leading to substances proven to cause health issues for consumers to be used and marketed with impunity. PFAS, for example, are a group of tens of thousands of chemicals, yet only a handful of these have been voluntarily phased out of select food contact materials (mostly, grease-proof paper).Footnote 277 Thousands of other PFAS remain widely in use in food contact materials like cookware and HDPE plastic containers despite alarming evidence of their link to a panoply of cancers.Footnote 278 Phthalates are another functional group containing many different substances linked to testicular toxicity, childhood obesity, insulin resistance, arrhythmia, and general inflammation.Footnote 279 While some of those compounds are getting phased out of food products, others — like diisodecyl and diisononylphthalate — are increasingly prevalent in both products and humans.Footnote 280
C. Environmental Contaminants
Environmental contaminants come in many shapes and forms — pesticide chemicals, legacy chemicals banned for use but persisting in the environment, heavy metals, microplastics, pathogens, and mycotoxins, to name a few.Footnote 281 The problems with environmental contaminants in food are plentiful and could be the subject of several law review articles. Because this area involves joint jurisdiction with both the United States Department of Agriculture and the EPA, FDA’s authority to regulate environmental contaminants in food is significantly curtailed.Footnote 282 In essence, the FDA attempts to regulate environmental contaminants in two ways: (1) by requiring Good Manufacturing Practices to prevent unintentionally added contaminants; and (2) by regulating poisonous and deleterious substances that are not food additives.
The Food Safety Modernization Act of 2011 (the “FSMA”) instituted multiple safety requirements for food manufacturing facilities that aimed to minimize the risk of production contamination.Footnote 283 The Act also augmented the FDA’s enforcement authorities in case of noncompliance, including providing for mandatory recalls,Footnote 284 administrative detention,Footnote 285 and import certification requirements.Footnote 286 Finally, the FSMA also required the FDA to increase its rate of inspections, imposing a minimum inspection frequency of three and five years for high-risk facilities and non-high risk facilities respectively.Footnote 287 However, the FDA’s finalized rule implementing the FSMA’s requirements — originally set to go into effect in January 2026 — has been postponed to mid-2028 due to industry pressure.Footnote 288 It therefore remains to be seen whether the FDA would in fact use its authority to recall products from noncompliant manufacturers.
For poisonous or deleterious substances that are not a food additive and not a pesticide (regulated separately largely under EPA authority), the FDA in turn has the power to: (1) declare a product adulterated, if the substance can be avoided through good manufacturing practices, or (2) set a tolerance, regulatory limit, or an action level, if contamination with the substance is unavoidable.Footnote 289 In theory, if a tolerance, regulatory limit, or action level is exceeded, the product qualifies as adulterated, and the FDA has the right to recall or seize it.Footnote 290 The agency has set tolerances for a handful of contaminants, including nitrites and benzene in bottled waterFootnote 291 and aflatoxins in food.Footnote 292 However, the FDA has not set specific levels for most environmental toxins of concern — including PCBs, PFAS, dioxins, perchlorate, radionuclides, and microplastics.Footnote 293 Instead, its entire engagement is monitoring consumer exposure through the Total Diet Study and targeted field tests.Footnote 294 The FDA publishes this data but, other than issuing occasional import alerts that permit (but do not require) the detention of a good without inspection,Footnote 295 it does not otherwise act on the information.
The most involvement the agency has with this type of contamination is with pathogens and heavy metals. For pathogens, the agency takes a strong stance towards Listeria monocytogenes in Ready-to-Eat Foods.Footnote 296 For heavy metals, the FDA established the Closer-to-Zero initiative in 2021, through which it is currently finalizing action levels for lead, arsenic, cadmium, and mercury.Footnote 297 In the meantime, however, recalls for pathogen outbreaks abound;Footnote 298 institutions like Cleveland Clinic note that “[h]eavy metals like arsenic, lead, mercury and cadmium routinely taint baby food,”Footnote 299 and consumers are currently suing Amazon over “alarmingly high” levels of toxic metals found in their food products.Footnote 300
V. A Comprehensive Food Safety Reform
Consumers, advocates, and policymakers increasingly recognize that our food system is broken, polluted, and deleterious to public health,Footnote 301 creating strong momentum for change.Footnote 302 Proposed solutions abound.Footnote 303 What this article adds to the conversation is a comprehensive, concrete, and complete blueprint for how to strengthen food safety in a systematic, predictable, and science-based way. It does so by categorizing existing proposals, adding novel suggestions, and examining the pathways and authority necessary to achieve an effective food safety system overhaul in the current political climate at both the congressional and agency levels. Mindful of the economic, logistical, and administrative hurdles that stand in the way of the agency, the proposals aim to realign incentives and flip both the legal and practical burden of proving safety at all stages of the process from an overworked and understaffed agency and a sickened public to the food manufacturers.
A. Congressional Actions
A legislative reform that strengthens the current food safety requirements and brings them more in line with the process for approving new drugs offers a direct route to robust and effective food safety review. Ordinarily, proposing a legislative solution to a pressing problem is a near-illusory approach to a crisis because passing legislation in general, and doing so in an expeditious manner in particular, has been exceedingly difficult and unlikely in recent decades.Footnote 304 However, for the first time in nearly a century, political circumstances make meaningful legislative reform possible (and, indeed, likely). Not only are the House, Senate, and White House politically aligned with each other, but they are also aligned with the Make America Healthy Again movement, which aims to radically remake food safety review and to ensure a cleaner food supply.Footnote 305 Indeed, FDA Commissioner Marty Makary has previously signaled that the Administration is “exploring every tool in the toolbox,” including potential “statutory and regulatory changes.”Footnote 306 Beyond that, these goals also receive significant bipartisan support at both the federal and state level and are strongly favored by a majority of voters, with many states pursuing localized reforms.Footnote 307 Harnessing the power of this political and public support and directing it towards meaningful, effective, and permanent reform, therefore, is critical. This section outlines a packet of legislative actions that could accomplish these goals.
1. Close the GRAS Loophole
The most widely supported and impactful legislative change would be to close what many call the GRAS “loophole” to food safety review.Footnote 308 This can happen in several ways.
First, Congress can scrap GRAS altogether. Following the logic of the current GRAS exception for food additives, the FDCA originally offered GRAS status for all drugs that had been on the market prior to 1938 and had retained their exact formulation.Footnote 309 By 1962, however, Congress realized that drugs should not simply be grandfathered in and should instead be subject to rigorous review. Thus, it amended the FDCA to require that all drugs must be proven both safe and effective prior to being marketed.Footnote 310 The FDA created the Drug Efficacy Study Implementation to review the safety of all drugs introduced on the market between 1938 and 1962.Footnote 311 And, while some early drugs were still grandfathered in, at no point after 1962 were drug manufacturers ever allowed to self-certify their products as safe, let alone to do so without even notifying the FDA.Footnote 312 The lack of such an exception did not lead to the sky falling for the pharmaceutical industry;Footnote 313 it merely allowed the FDA, based on actual clinical evidence, to determine whether a pharmaceutical compound is safe for human consumption.
The same should be true for food. The mistaken legislative assumption that food additives in use prior to 1958 were generally safe has now been proven categorically false.Footnote 314 Many of the chemical substances synthesized in the 1940s and 1950s are the very scourges that the EPA, USDA, and FDA now understand cause significant adverse health effects.Footnote 315 Likewise, the assumption that food additives occur in low doses and concentrations and are therefore innocuous is demonstrably untrue.Footnote 316 We know that these substances lead to low-dose chronic exposure that, years or decades down the line, manifests in a variety of chronic illnesses.Footnote 317 Many of these chemicals also can bioaccumulate in human tissues — once ingested, they never leave the body and multiply the negative health effects with every subsequent exposure.Footnote 318 The rapid proliferation of food additives in processed food has also led to the same type of information asymmetry between consumers and food producers that exists in the pharmaceutical space, necessitating closer FDA scrutiny and a more robust information-gathering function.Footnote 319 A categorical removal of the GRAS exception therefore is supported by both logic and science.
Second, if Congress is not inclined to end the GRAS exception, it should honor the original purpose behind it by making GRAS applicable only to substances that have a proven long-term and safe history of use (minimum twenty years) and precluding its application to new substances.Footnote 320 For substances that currently have GRAS status and are already recorded in the FDA’s database — 1,234 substances, as of the date of this writingFootnote 321 — Congress should order a comprehensive safety reevaluation by the FDA within five years,Footnote 322 akin to the reevaluation it ordered for pesticides under the Food Quality Protection Act.Footnote 323 For any future GRAS notifications, Congress should mandate that they be accompanied by all applicable and available safety data, as is true for new drug applications,Footnote 324 to allow for agency safety assessment or, at a minimum, “judicial and public oversight.”Footnote 325 For these more robust future filings, the FDA should be required to conduct only a focused review to (1) verify that there is evidence of long-term use of the ingredient, and (2) do a literature review for any reported adverse reactions or negative safety determinations.
At a bare minimum, Congress must unequivocally end the voluntary notification program and instead mandate that all GRAS certifications be logged in with the agency before a manufacturer begins to use an ingredient in their food products.Footnote 326 For any currently undeclared substances that have not been logged through the notification system and enjoy under-the-radar GRAS status, Congress should mandate that manufacturers must file a notice, accompanied by evidence of long-term use and all safety data — akin to an INDFootnote 327 — within one year of the effective date. Because these self-certified substances should at least in theory have gone through internal safety review, manufacturers should already have all safety data on hand to file and need time only to put together their paperwork.Footnote 328 In the (likely) event that manufacturers do not have the necessary safety data within that first year, this provision would effectively make the use of such unauthorized and unexamined substances illegal and their products adulterated, unless and until manufacturers decide to file a fully supported GRAS notice or a food additive petition.Footnote 329
Logistically, of course, such changes would pose some difficulty for food manufacturers who have relied heavily on GRAS self-certifications.Footnote 330 They would also temporarily burden the FDA because every single food chemical in use in the food supply would need to be logged in with comprehensive safety information.Footnote 331 This added burden is not a sufficient reason to shy away from reform. First, as a principled matter, the purpose of the FDA is to ensure food safety, not to help manufacturers more easily get a product to market or to maintain a light workload.Footnote 332 Indeed, several former FDA commissioners in charge of food safety have lamented the fact that the FDA currently does not have the authority to fully investigate the safety of food substances.Footnote 333 Jim Jones, the former Deputy Commissioner of the FDA’s Human Foods Program, for example, testified before a Senate committee in December 2025 that “[w]e are several decades behind Europeans and our Canadian counterparts because they have legal mandates to reevaluate chemicals that have been authorized at some point in the past.”Footnote 334 Many of the additional actions proposed in this package — such as user fees, mandatory purity testing, and mandatory post-market reporting — would also significantly lighten the load for the FDA and would instead shift the work onto the food manufacturers. In addition, neither Congress nor the FDA are new to this issue and have previously aptly utilized phasing out periods, tiered review, and graduated tolerances to allow the market the time to adjust to stricter regulation.Footnote 335 Perhaps most importantly, as a practical matter, because GRAS self-certification does not exist in other more health-conscious markets like the EU and the UK,Footnote 336 food manufacturers have either already conducted safety studies for these jurisdictions or have readily available safer alternative ingredients that would allow them to reformulate their U.S.-based products within a reasonable time frame.Footnote 337
2. Repeal the Food Contact Notification Program
In a similar vein, Congress should also repeal the Food Contact Notification Program.Footnote 338 As discussed earlier, this Program was instituted for the purpose of efficiency, but its efficiency has now vastly overtaken its efficacy. The automatic “approval” of a food contact notice, combined with the strain on FDA resources virtually guarantees that any filed notice would get “approved” and that no amount of evidence would be sufficient for the FDA to revoke it post-market.Footnote 339 Given evidence of the massive detrimental effect of food contact substances like BPA, PFAS, and phthalates on human health,Footnote 340 it is imperative that the FDA re-review the safety of all previously noticed substances and comprehensively evaluate the safety of any new proposed substances through a full petition process.Footnote 341 In this context too, Congress should clarify that no food contact substances could be self-certified as GRAS. Any added burden on the FDA through these actions should be mitigated through the same phase-out periods and graduated review expectations outlined for GRAS notices above.
Relatedly, Congress should also clarify the language of the so-called Delaney Clause. Named after Representative James Delaney, this clause was introduced with the 1958 Food Additive Amendment and provides that “[n]o additive shall be deemed to be safe if it is found to induce cancer when ingested by man or animal.”Footnote 342 Read literally, the provision empowers the FDA to reject any food or color additive if there is any evidence of carcinogenicity.Footnote 343 Despite this sweeping language evincing a seemingly zero-tolerance policy towards carcinogens in food,Footnote 344 the Delaney Clause is “surely the most discussed, but perhaps the least used, provision of the FD&C Act.”Footnote 345 Subsequent court decisions have significantly curtailed the applicability of the clause and the FDA’s own reticence to use the clause has rendered it virtually illusory.Footnote 346 One uncertainty about the Delaney Clause is whether it applies to GRAS substances. As written, the provision applies only to “food additives” and GRAS substances are exempt from the regulatory definition of “food additives.”Footnote 347 Congress should step in to clarify that, to the extent the GRAS exception remains in any form, the Delaney Clause should apply to such GRAS substances with the same force as it applies to direct and indirect food additives and color additives.Footnote 348 Congress should also resolve the debate as to whether a de minimis exception to the Delaney Clause exists, particularly for food contact substances.Footnote 349 A de minimis exception sounds eminently reasonable at first blush given the low concentrations and seemingly low levels of migration for many food contact substances.Footnote 350 However, such an exception has proven problematic considering both the chronic exposure to these substances across products and the compounding effect of all food additives in our diet.Footnote 351
3. Institute User Fees for All Safety Assessments
A natural corollary to requiring the FDA to both reevaluate over a thousand currently permitted additives and conduct more thorough pre-market safety review is a proposal that the FDA institute user fees for all food safety petitions — including direct and indirect food additive petitions, food contact notices, GRAS notifications, and color additive petitions (the last of which currently have only nominal fees). This idea has been explored sporadically in the past: a 1993 Report by the Government Accountability Office, for example, suggested user fees as a potential strategy to regulate novel ingredients and technologies.Footnote 352 The idea has a particular appeal today, however, considering the financial and staffing realities at the FDA and across the federal government. The FDA has recently been sharply criticized over its “inability to fulfill critical food safety activities related to its pre- and post-market review of substances added to food.”Footnote 353 Even more recently, the FDA’s Human Foods Program lost eighty-nine of its employees as part of the federal reduction in force, further reducing the agency’s capabilities.Footnote 354 An influx of new safety assessment petitions, therefore, would leave the agency’s food arm overburdened.
Increased congressional appropriations are not a viable solution. As the United States Supreme Court recently noted, the congressional appropriations process forces agencies “to regularly implore Congress to fund their operations for the next year.”Footnote 355 The process’s unpredictability prevents the agency from charting a long-term course towards increased workload and more meaningful safety review. Beyond that, although the Make America Healthy Again movement has tremendous political backing and the latest iteration of the federal government’s budget does not contemplate any cuts to FDA’s food budget (in fact, it includes some increases to advance MAHA’s agenda),Footnote 356 the level of funding that the FDA would need to fill in the gaps is incompatible with the overall goal of decreasing the federal budget deficit.
User fees for all applications requiring food safety review would solve this issue. They would provide the FDA with much needed resources not only to handle any increased workload from the reforms proposed in this article, but also to meet “statutory and regulatory deadlines in its premarket review of petitions and notifications and in creating a more efficient and effective regulatory process for its post market review of substances in the food supply.”Footnote 357 As the FDA itself has recognized, in the pharmaceutical context, user fees permit the “timely availability of innovative FDA-regulated products without compromising the agency’s commitment to scientific integrity, public health, regulatory standards, patient safety, and transparency.”Footnote 358 Even industry groups, which originally opposed user fees for generic drugs, ultimately came to appreciate the predictability, uniformity, and general improved functioning of the process after the fees.Footnote 359 Congress can institute fees for all food safety assessments through an act, as it has done in the past — through the Prescription Drug User Fee Act of 1992, the Generic Drug User Fee Amendment in 2012, or the Food Safety Modernization Act, which authorized the FDA to collect fees for imports, facility inspections, and third-party accreditations.Footnote 360
User fees for food safety review could theoretically lead the FDA to become more beholden to the food industry,Footnote 361 but ultimately such concern is misplaced for three reasons. First, paying authorization application fees is a landmark feature of our administrative system — we pay fees for zoning permits, professional licenses, and citizenship applications, to name just a few.Footnote 362 In none of those contexts is the deciding body beholden to the applicant or more likely to grant the request because of the fee.Footnote 363 Second, given that the FDA’s modus operandi for decades has been to seek voluntary actions by food manufacturers rather than to engage in enforcement,Footnote 364 it is difficult to see how user fees can make the agency any less engaged than it is at present. Finally, the current system of GRAS self-determinations by paid third party scientific bodies is already ridden with express conflicts of interest.Footnote 365 Any concern about user fees indirectly clouding the agency’s judgment pales by comparison.
4. Require Clinical Studies
As demonstrated by both medical studies and clinical experience, food additives could have negative long-term impacts on human health through low-dose chronic exposure that does not always show up in animal studies.Footnote 366 To ensure food safety, Congress should require clinical studies for any new food or color additives seeking approval through a petition process at a concern level II or above. Under the current regime, where clinical studies are never required, food manufacturers are incentivized not to track such data “due to concern about future litigation.”Footnote 367 Indeed, many food manufacturers do not even conduct basic toxicity testing.Footnote 368 Mandatory clinical evaluations would resolve this concern. The required studies do not need to mirror the lengthy and involved process of clinical pharmaceutical studies, because in this context there is no need to assess medical efficacy.Footnote 369 Rather, the studies should focus only on safety, and the FDA should be empowered to promulgate a rule requiring studies that assess the most common adverse health effects from food and color additives: carcinogenicity, neurocognitive effects in children, developmental and reproductive toxicity, endocrine disruptions, and microbiome impact.Footnote 370 The FDA should also require at least one long-term chronic toxicity study of cumulative effect of the proposed food substance, administered as a cocktail with other food additives typically found in the American diet.Footnote 371
One of the FDA’s most important functions in assessing the safety of pharmaceuticals is precisely this role of “information production” — something the FDA can accomplish only if it has the statutory authority to require this type of data, rather than recommend it and hope for the best.Footnote 372 To be sure, required clinical studies for food additives would lengthen the period of safety review for individual manufacturers, would entail a more significant investment in safety assessment, and would delay introduction of new ingredients and products on the market.Footnote 373 Unlike a delay in life-saving vaccines or other medication, however, a delay in introducing a novel flavor of chips or cookies is not a significant burden on society and the benefit of having comprehensive safety information vastly outweighs the minor inconvenience of delayed marketing campaigns.Footnote 374
5. Require Post-Market Reporting
Finally, Congress should empower the FDA to conduct a more meaningful post-market review of food additive use and safety. In addition to the current manufacturer obligations to report acute serious adverse health effects,Footnote 375 Congress should allow the FDA to implement an annual reporting obligation, like the one required for pharmaceutical companies.Footnote 376 In this annual submission, food manufacturers should be required to list: (1) any new internal safety studies conducted by the manufacturer; (2) any adverse safety findings by external peer-reviewed studies; (3) any reported adverse reactions or effects (even if they do not fall within the statutory definition of “reportable foods” requiring immediate notification); and (4) total annual additive usage. These obligations would shift the onus of tracking ongoing safety findings and conducting necessary follow-up studies to the manufacturer. They would also allow the FDA to track annual food additive usage, which would in turn help it prioritize review and re-review for chemicals with higher occurrence in the typical diet.Footnote 377
Additionally, Congress should authorize the FDA to create a reporting database for any third-party adverse testing or lab results, and an adverse report should trigger mandatory safety review and an automatic manufacturer obligation to submit additional safety information or conduct follow-up investigation, as appropriate for the type of reported event. Currently, the agency operates three adverse event reporting databases — one for pharmaceuticals (FEARS), another for dietary supplements (Safety Reporting Portal), and one for acute serious adverse events for food manufacturers.Footnote 378 The food registry, however, is deficient in several respects. First, it is open only to manufacturers rather than to consumers or researchers.Footnote 379 Second, it is intended only for acute serious health events,Footnote 380 like pathogen outbreaks, but is otherwise irrelevant to food additive safety or environmental contamination. Instead of (or in addition to) this registry, Congress should authorize the creation of a database through which any third party can report an adverse health determination made by another health agency or body, such as the European Union or the World Health Organization, a peer-review study, or certified lab results showing that a food product is unsafe for human consumption. Such findings could be due to food additive toxicity, high levels of heavy metals, pathogens, environmental contaminants, production impurities, or another concern. Many such studies and tests are conducted regularly by scientific bodies, consumer protection groups, and even concerned citizens.Footnote 381 At present, however, unless these results become the basis of prolonged litigation or a lengthy and futile petition process,Footnote 382 they will end up in the black hole of the internet.
Ironically, the FDA bemoans its limited resources to conduct long-term safety studies or to test food products on the market.Footnote 383 Allowing such external reporting would alleviate some of the burden currently placed on the small force of FDA researchers and field inspectors and would ensure greater transparency.
B. Agency Actions
In lieu of (or in addition to) congressional reform, the FDA can independently make several improvements through its existing rule-making authority.
First, even without congressional abrogation of the GRAS exception, the FDA can resolve the central issues with self-certification. Indeed, on March 10, 2025, Secretary Kennedy directed Commissioner Makary to explore potential rulemaking to revise GRAS.Footnote 384 The FDA should take this opportunity to convert the notice system from voluntary to mandatory.Footnote 385 It should also require that manufacturers file all available data with their notices. Currently, “[t]he food industry does massive amounts of research that we have no access to.”Footnote 386 That outcome should be unacceptable, and both the agency and the public should have ready access to all studies conducted on a particular substance. Lastly, the FDA should issue guidance to industry that novel substances cannot be subject to GRAS determination.Footnote 387
Because the FDA does not have a statutorily imposed deadline for reviewing filed GRAS notices,Footnote 388 these changes would not impact the agency’s workload significantly but would offer vastly increased transparency. To help with the eventual work of reviewing these submissions, the FDA could consider engaging artificial intelligence software to pre-assess substances and pre-categorize them in tiers of concern — such as those that pose “risks to children’s health, endocrine disruption, biomonitoring data” or have been identified as high priority by “authoritative bodies like U.S. Environmental Protection Agency (EPA), International Agency for Research on Cancer (IARC) and the National Toxicology (NTP) Program.”Footnote 389 The same type of software could help by mining scientific databases for all studies conducted on a given substance.Footnote 390
Second, the FDA should enforce in practice the manufacturers’ legal burden of proving safety. For both additives approved through petition and GRAS substances,Footnote 391 the statute and regulations require the manufacturer to prove consensus or reasonable certainty of safety, not the FDA to prove lack of safety.Footnote 392 In practice, however, the agency often conducts both its initial and any post-market review as if it has the burden to show that a substance is harmful to human health.Footnote 393 The FDA revoked Red Dye 3’s authorization, for example, only after it affirmatively found that it “can induce cancer in male rats, through a rat specific hormonal mechanism,”Footnote 394 even though studies casting doubt on the additive’s safety have existed for over thirty-five years. This de facto burden of proof that the FDA has taken upon itself taxes the agency’s resources and makes it unable to fulfill its “pre- and post-market” statutory obligations.Footnote 395 To fix this, the agency should reverse course, demanding that manufacturers provide robust safety data with their filings or risk an automatic denial of their petition or GRAS certification. The FDA should examine all filed GRAS notices and approved food and color additives and should summarily revoke those that are filed without sufficient toxicology, feeding toxicity, and other relevant data.Footnote 396 Likewise, the FDA should use Artificial Intelligence to identify any approved additives or noticed GRAS substances for which there are reputable, peer-reviewed studies with adverse safety results, including “findings from other authoritative institutions such as the National Institutes of Health (NIH) and the European Food Safety Authority (EFSA).”Footnote 397 The very existence of such studies should sufficiently demonstrate that the manufacturer failed to demonstrate reasonable certainty of safety.
For future petitions and GRAS notices, the FDA should issue guidance to industry strongly recommending that manufacturers conduct long-term tests on the carcinogenic, neurocognitive, endocrine, and microbiome effect of their proposed substances to provide a more robust safety profile. The agency should also make cumulative exposure to a cocktail of food additives an explicit factor in its assessment,Footnote 398 as it does for new drug applications.Footnote 399 To decrease its burden of reviewing the data from these studies, the FDA should institute a graduated approach by assigning specific point values, with the highest value assigned to clinical studies in humans conducted by authoritative scientific bodies, and the lowest assigned to small-scale, short-duration lab tests funded by the manufacturer. The agency should also be more open to partnering with external reviewers; the recently announced collaboration between the FDA and NIH on the causes of “the diet-related chronic disease crisis” is a welcome start.Footnote 400 The FDA should continue collaborating with additional research agencies to obtain similarly “‘critical information’ to inform policy,”Footnote 401 while creating a body of external peer reviewers who could help analyze incoming safety assessments.Footnote 402 This would simultaneously ease the agency’s workload and ensure a broader view of available scientific approaches is represented in its safety determinations.
The FDA should also promulgate rules allowing it to consider chemicals in functional groups. While the statute already mandates consideration of the cumulative dietary effect of chemically related substances in assessing safety,Footnote 403 the agency should begin considering chemically related substances for purposes of making initial batch safety determinations and streamlining its review. In other words, if a given substance (say, PFOA) has been shown to have detrimental health effects, the FDA should consider all chemically related substances with common mechanism of toxicity (in this example, all long-chain PFAS) to be presumptively unsafe. The burden should then switch to manufacturers of individual substances to demonstrate that their sponsored substance behaves differently than the functional group. This burden-shifting would avoid the current whack-a-mole approach to functionally equivalent chemicals, where the FDA takes years to assess safety of an individual substance only to have the manufacturer alter a single atom and restart the clock — all while consumers continue to get sicker.
Third, the FDA should expand its mandate for lot testing to require that manufacturers conduct purity testing for environmental contaminants.Footnote 404 Under existing regulations, environmental contaminants in food products are considered unsafe for human consumption, with food products considered “adulterated” if contaminants occur in excess of an action level — a limit set by the FDA, “at or above which FDA will take legal action to remove products from the market.”Footnote 405 However, the FDA’s limited resources guarantee that most products never undergo testing unless a series of health events triggers a recall.Footnote 406 While the FDA is currently considering partnering with states to conduct more of these safety inspections,Footnote 407 a more efficient way to ensure that products are free from heavy metals, environmental toxins, and pathogens would be requiring manufacturers to test every lot, as is required for pharmaceuticals.Footnote 408
Fourth, the FDA should mandate stricter labeling requirements.Footnote 409 Several state bills can provide helpful insights. Texas’s House recently passed a food label bill requiring food products with ingredients banned in other countries to contain a warning, akin to the warnings for pharmaceuticals: “WARNING: This product contains an ingredient that is not recommended for human consumption by the appropriate authority in Australia, Canada, the European Union, or the United Kingdom.”Footnote 410 A bill in New York, in a similar vein, would mandate that manufacturers disclose all GRAS ingredients on their label.Footnote 411 The FDA should likewise require that all food and color additives be expressly listed on the label and that the manufacturer’s labeling information (on their website or otherwise available) contain data on the relevant safety testing that supports the determination that the ingredient is safe for use.Footnote 412 This would allow not only increased agency, public, and judicial oversight, but would also help vulnerable consumers with various medical conditions, chronic medications, or other possible causes of adverse interactions to tailor their diet as medically appropriate.Footnote 413
VI. Conclusion
Food is both a hero and a villain in our public health story. Food is powerful medicine; it has been sustaining, nourishing, and healing people for generations. And yet, food is also the main culprit for our epidemic of diet-related diseases, cancers, childhood neurological deficits, staggering autoimmune conditions, unprecedented levels of hormonal disruptions, and shattered microbiomes. Consumers do have autonomy and bear personal responsibility to make healthier dietary choices. But the current lack of transparency and oversight has made that effectively impossible for the average consumer. With the advent of processed and ultra-processed foods, novel, lab-created, and genetically engineered commodities, dietary supplements, wellness and specialty foods, and products marketed specifically to infants and children, information asymmetry — and therefore the power dynamic — between consumers and food manufacturers has drastically shifted. To many consumers, the ingredients on product labels sound like a foreign language, making it impossible to fully appreciate the health consequences of consuming these foods. Our regulators and food manufacturers thus bear a greater responsibility to ensure that food marketed to consumers is at baseline safe. Both have increasingly failed in that task over the last century. Deceptive marketing strategies, superficial safety assessments, powerful industry lobby, and an ask-for-forgiveness-rather-than-permission mentality on the part of the manufacturers have impeded both consumers’ ability to make intentional food choices and the market’s ability to effect positive change.
The tide is turning, and food safety regulation is having its moment on the national stage. We know now more than ever that food, with its chemical complexities and capacity to cause chronic cumulative exposure, plays a vital role in public health. Legislative and regulatory reform, therefore, must increase the baseline requirements for food manufacturers, create transparency for consumers, and allow regulatory agencies to exercise robust safety review and control. It is time to ensure that our food is medicine, not poison, and that burden must rest on those who profit from food, not those who need it to live.
Open access
