“Ultra-processed foods are as addictive as cocaine.”

• A 2023 BMJ analysis by Gearhardt and DiFeliceantonio argued that ultra-processed foods (UPFs) high in refined carbohydrates plus fat meet the standard criteria used to classify a substance as addictive — they deliver rapid, unnaturally high doses of reinforcing agents to the brain, trigger strong cravings and compulsive use, and are consumed despite harm — and estimated roughly 14% of adults and 12% of children show addictive-like consumption on the Yale Food Addiction Scale, a prevalence comparable to that of alcohol or tobacco use disorder in adults.
• Animal work provides the most-cited basis for the cocaine comparison: Lenoir et al. (PLoS ONE, 2007) found that the large majority of rats offered a choice between intravenous cocaine and sweetened (saccharin or sucrose) water chose the sweet taste, including rats already sensitized or dependent on cocaine.
• Neuroimaging and preclinical studies show that palatable, energy-dense foods engage the same mesolimbic dopamine/reward circuitry (nucleus accumbens, striatal D2 receptor changes) implicated in drug addiction, which underpins the 'food addiction' construct.
• Controlled feeding research (Hall et al., Cell Metabolism, 2019) showed people ate ~500 kcal/day more on an ultra-processed diet than on a matched unprocessed diet and gained weight, demonstrating that UPFs drive overconsumption beyond conscious intent — consistent with, though not proof of, addictive properties.

• The BMJ authors' own comparison is to tobacco, not cocaine; no peer-reviewed study has measured UPF and cocaine addictiveness on a common human scale, so a claim of equivalence has no direct empirical support.
• Pharmacologically, cocaine is a direct monoamine-reuptake blocker producing an acute intoxicating dopamine surge far larger and faster than any food response; UPFs produce no intoxication, no comparable overdose/lethality risk, and no well-established physiological withdrawal syndrome in humans.
• 'Food addiction' is not a recognized diagnosis in DSM-5 or ICD-11 (only binge-eating disorder is), whereas cocaine use disorder is an established, extensively validated diagnosis. Instruments like the YFAS measure addictive-like behavior, not a formally accepted disorder.
• Prominent critics (Ziauddeen, Farooqi & Fletcher, Nature Reviews Neuroscience 2012; Fletcher & Kenny, Neuropsychopharmacology 2018; Westwater et al., European Journal of Nutrition 2016) conclude the human evidence for substance-style addiction to food or sugar is weak, that overlapping reward circuitry is expected for any natural reinforcer, and that the behavioral pattern may better fit a behavioral/eating disorder model.
• The rat studies most often invoked used intensely sweet solutions (often non-caloric saccharin), not ultra-processed foods, and involved limited cocaine doses and delivery schedules; they do not establish that UPFs rival cocaine's dependence liability in humans.
• Expert drug-harm rankings (e.g., Nutt et al., The Lancet 2007) place cocaine among the substances with the highest dependence liability; nothing comparable has been demonstrated for food, and most people who eat UPFs daily do not develop compulsive-use pathology.

80/100 — the lab’s own confidence in the score above, not a second reading of the claim.

The BMJ (Gearhardt & DiFeliceantonio, 2023)journal · 92 · PLOS ONE (Lenoir et al., 2007)journal · 80 · Nature Reviews Neuroscience (Ziauddeen, Farooqi & Fletcher, 2012)journal · 93 · Neuropsychopharmacology (Fletcher & Kenny, 2018)journal · 90 · European Journal of Nutrition (Westwater, Fletcher & Ziauddeen, 2016)journal · 85 · Cell Metabolism (Hall et al., 2019, NIH)journal · 92 · National Institute on Drug Abuse (NIH)gov · 95 · The Lancet (Nutt et al., 2007)journal · 88

• Some peer-reviewed researchers argue that many ultra-processed foods—especially products high in rapidly absorbed refined carbohydrates and added fats—can produce addiction-like patterns in some people, including cravings, impaired control, continued use despite harm, and relapse-like behavior. A 2023 BMJ analysis estimated ultra-processed food addiction at about 14% of adults and 12% of children using Yale Food Addiction Scale criteria.
• A PLOS ONE study found that foods rated as more processed, higher in fat, and higher in glycemic load were more strongly associated with addictive-like eating reports, supporting the idea that not all foods are equally implicated.
• Animal research sometimes cited in support of the cocaine comparison found that intense sweetness could surpass cocaine reward in rats. This supports the narrower claim that sweet taste can be highly reinforcing in animal models, but it does not prove that ultra-processed foods are equally addictive to cocaine in humans.

• The statement says ultra-processed foods are “as addictive as cocaine,” which implies a demonstrated equivalence in addictive potential. That equivalence has not been established in human clinical evidence. Cocaine is a pharmacologically active stimulant with well-characterized addictive effects; ultra-processed foods are heterogeneous products, not a single drug-like substance.
• Food addiction or ultra-processed food addiction is not currently recognized as a formal substance-use disorder in the way cocaine/stimulant use disorder is recognized. The Yale Food Addiction Scale measures addiction-like eating symptoms, but it is not the same as proving that foods have cocaine-level addictive liability.
• Reviews skeptical of the food-addiction framing argue that evidence is stronger for addictive-like eating behavior than for a specific addictive substance in food. Unlike cocaine, food is necessary for survival, and the boundary between high palatability, compulsive eating, binge-eating disorder, and substance addiction remains scientifically contested.
• The rat sweetness-versus-cocaine finding is often overextended. It involved controlled animal models and sweet solutions, not real-world ultra-processed foods, human addiction rates, withdrawal syndromes, dose-response toxicity, or direct comparison with cocaine dependence in people.

86/100 — the lab’s own confidence in the score above, not a second reading of the claim.

BMJjournal · 90 · PLOS ONEjournal · 83 · PLOS ONEjournal · 82 · National Institute on Drug Abusegov · 95 · American Psychiatric Associationother · 86 · Obesity Reviews / PubMedjournal · 82 · Neuroscience & Biobehavioral Reviews / PubMedjournal · 82 · National Institutes of Healthgov · 94

• Ultra-processed foods (UPFs), particularly those high in refined carbohydrates and added fats, stimulate the brain's dopamine reward system in a manner similar to addictive substances.
• The claim often stems from a heavily publicized 2007 animal study published in PLOS One, which found that rats given a choice between sweet water and intravenous cocaine overwhelmingly preferred the intense sweetness, popularizing the narrative that sugar is more addictive than cocaine.
• A 2023 analysis published in The BMJ argued that UPFs meet the established scientific criteria for addiction, estimating that 14% of adults exhibit addictive eating behaviors toward these foods (a rate comparable to tobacco addiction).

• Major psychiatric and diagnostic guidelines, including the DSM-5, do not recognize 'food addiction' as a clinical diagnosis, drawing a line between compulsive eating behaviors and true substance use disorders.
• Neuroscientists and behavioral experts note that animal studies showing sugar preference over cocaine rely on strict intermittent fasting protocols; when rats have continuous, free access to sugar, the severe addiction-like behaviors and neurochemical changes do not manifest.
• Equating food to cocaine creates a false equivalence that ignores the profound pharmacological toxicity, extreme physiological withdrawal symptoms, and acute life-destroying behavioral disruptions uniquely caused by illicit drugs.

95/100 — the lab’s own confidence in the score above, not a second reading of the claim.

The BMJjournal · 95 · PLOS Onejournal · 85 · European Journal of Nutrition / PubMed Centraljournal · 95

• Some ultra-processed foods can produce addiction-like patterns—craving, loss of control, compulsive consumption, and continued use despite harm—measured using the Yale Food Addiction Scale. A BMJ-linked analysis of 281 studies estimated clinically significant food-addiction symptoms in about 14% of adults and 12% of children, though this is not equivalent to proving cocaine-level pharmacological addiction.[4]
• A peer-reviewed review reports that highly processed foods may engage reward, craving, and impulse-control mechanisms that overlap with substance-use disorders, particularly when foods combine refined carbohydrates and fat and are rapidly absorbed.[5]
• The comparison has some behavioral plausibility: both drug addiction and food addiction can involve craving, impaired control, and persistent use despite adverse consequences. However, behavioral similarity does not establish equal addictive potency or biological equivalence.[5]

• The claim says ultra-processed foods are as addictive as cocaine, but available evidence generally concerns addiction-like eating or a proposed food-addiction construct—not a head-to-head demonstration that UPFs produce cocaine-equivalent dependence, reinforcement, withdrawal, or harm.[5][6]
• A peer-reviewed commentary explicitly rejects comparing these foods directly with cocaine or amphetamine, noting that those drugs produce much larger dopamine increases; the cited estimate was roughly 300%-1000%.[3]
• Food addiction is not currently an official psychiatric diagnosis, and the Yale Food Addiction Scale applies substance-use-disorder criteria to eating behavior; a positive score therefore indicates clinically significant symptoms, not proof that a food is pharmacologically equivalent to cocaine.[4][6]
• Ultra-processed food is a broad category that includes nutritionally and pharmacologically diverse products. Evidence suggesting higher addictive potential for some foods—especially highly rewarding foods rich in refined carbohydrate and fat—cannot validly be generalized to every ultra-processed food.[5]
• The strongest evidence is largely observational, self-reported, or based on animal and mechanistic research, which limits causal conclusions and direct comparisons with cocaine.[2][5]

91/100 — the lab’s own confidence in the score above, not a second reading of the claim.

BMJjournal · 94 · National Institutes of Health / Current Neuropharmacologyjournal · 91 · University of Michiganedu · 88 · McGill University Office for Science and Societyedu · 83 · Springer Nature / Current Addiction Reportsjournal · 70

• A 2007 rat study found intense sweetness (saccharin) was preferred over cocaine, showing highly palatable substances can outcompete cocaine in some reward paradigms.
• Highly processed foods rich in fat, sugar, and salt can activate overlapping dopamine/reward circuitry and are more strongly linked to addiction-like eating (craving, loss of control) than minimally processed foods on tools such as the Yale Food Addiction Scale.
• Some addiction researchers argue ultra-processed foods meet formal scientific criteria for addictive substances due to rapid nutrient delivery, hyper-palatability, and compulsive use in a subset of people.

• Cocaine is a potent pharmacological stimulant with high addiction liability, rapid dependence, severe withdrawal, overdose risk, and DSM-recognized cocaine use disorder; ultra-processed foods do not produce equivalent clinical, pharmacological, or population-level effects.
• Food addiction is not an official substance-use diagnosis in DSM-5 or ICD; most people regularly eat ultra-processed foods without developing cocaine-comparable compulsive disorders.
• Headline comparisons typically over-interpret limited animal preference studies and do not establish equal addictiveness, potency, or harm in humans.
• Prevalence of clinically significant food-addiction scores is far lower than would be expected if ultra-processed foods matched cocaine’s addiction potential.

78/100 — the lab’s own confidence in the score above, not a second reading of the claim.

PLOS ONEjournal · 88 · PLOS ONEjournal · 88 · Harvard T.H. Chan School of Public Healthedu · 90