Lead
"Ham causes cancer." "Phosphates destroy kidneys." "Preservatives accumulate in the body." If you have spent any time in parenting media or on social feeds, you have almost certainly encountered these claims. They are not pure fabrications — each has a thread of real research behind it. The problem is that thread has been stripped of context, and a stripped-of-context finding produces fear that bears little relationship to actual risk.
This article takes four of the most common food-additive claims, traces each one back to the primary research that gave rise to it, and examines what that research actually says — and what it does not. The risk-assessment framework that provides the foundation for this discussion (ADI, NOAEL, safety factors) is covered in article 101 of this series.
Nitrites and IARC Group 1 — Strength of evidence is not size of risk
When processed meat is described as a "carcinogen," the reference point is almost always the 2015 assessment from the International Agency for Research on Cancer. IARC convened experts from 22 countries, reviewed more than 800 epidemiological: relating to the study of how diseases are distributed and what causes them in populations studies, and classified processed meat as Group 1 — "carcinogenic to humans" [1]. That announcement was widely reported, and the phrase "ham is in the same group as tobacco" circulated widely as a result.
What that classification actually means, however, is the strength of the evidence that processed meat causes cancer — not the magnitude of the risk at ordinary consumption levels. IARC's grouping system answers the question "how certain are we that this substance causes cancer?" It does not directly answer "how much does eating this raise my risk?" [1,2].
The same IARC evaluation estimated that each additional 50 g of processed meat consumed per day is associated with approximately an 18% relative increase: change expressed as a proportion of baseline risk, not as a real-world percentage-point increase in colorectal cancer risk [1]. Translated to absolute risk, that corresponds to the lifetime risk of colorectal cancer rising from roughly 5–6% to about 6% — a modest absolute change. Other Group 1 agents include solar ultraviolet radiation and alcohol, and yet the magnitude of risk associated with each differs substantially.
For families with young children: processed meat rarely appears in large daily quantities in an infant's or toddler's diet, and there is no evidence-based rationale for complete elimination driven by "carcinogen" labeling. For adults eating processed meat in large amounts several times a week, aiming for dietary variety has genuine, proportionate support in the literature.
Phosphates and kidney function — misapplying a finding to the wrong population
Phosphates are widely used as preservatives, binding agents, and pH regulators in products including ham, cheese, and carbonated beverages. The claim that "phosphates destroy kidneys" traces back to studies showing correlations between phosphorus intake, cardiovascular events, and mortality risk in patients with impaired kidney function [3].
The critical issue is the population to which that finding is being applied. Research conducted in adults with chronic kidney disease: long-term gradual loss of kidney function, requiring careful management of phosphorus and other nutrients cannot be extrapolated to healthy infants and toddlers — a straightforward principle of epidemiology that is routinely ignored when these claims travel through social media [3].
There is a further layer of nuance. Phosphates naturally present in foods (organic phosphorus) and phosphates added as food additives (inorganic phosphorus) differ in their intestinal absorption rates. The two are not interchangeable, and the distinction matters for assessing total dietary phosphorus load.
JECFA (the Joint FAO/WHO Expert Committee on Food Additives) sets the ADI: acceptable daily intake — the amount of a substance a person can consume daily over a lifetime without appreciable risk for phosphoric acid and phosphates at 70 mg per kilogram of body weight per day. In practice, an infant's or toddler's diet contains relatively limited quantities of phosphate-containing processed foods, and typical intake is not expected to approach this threshold.
Sodium benzoate + vitamin C = benzene — a question of scale
The observation that "the preservative sodium benzoate reacts with vitamin C to produce benzene" is chemically accurate. Under oxidative conditions in the presence of ascorbic acid, benzoate can form benzene — this has been confirmed experimentally. The question is: at what concentrations does this actually occur in beverages, and does that concentration represent a meaningful risk?
Between 2005 and 2007, the US Food and Drug Administration surveyed approximately 200 beverages and detected benzene above the EPA's drinking-water standard of 5 ppb in ten products [4]. Importantly, these detections often corresponded to conditions of high temperature or intense light exposure — worst-case scenarios rather than typical storage conditions — and the affected products were subsequently reformulated or withdrawn [4]. In the large majority of products tested, benzene was below 5 ppb, and the FDA concluded that detected levels "do not present a safety concern" [4].
Giving young children large amounts of carbonated soft drinks is inadvisable on other grounds (sugar load, displacing nutrient-rich foods), but there is no demonstrated basis for treating "preservative plus vitamin C in a beverage" as an acute hazard.
Antifungal agents on imported citrus — reading a label correctly
Ortho-phenylphenol (OPP), thiabendazole (TBZ), and imazalil are antifungal agents applied to the skin of imported citrus fruit to prevent mold during transport. In Japan, these are classified and regulated as food additives — not pesticides — and their use must be disclosed on product labels.
EFSA and other regulatory bodies have established ADIs for each compound within the framework of their safety evaluations [5], and managed use under those ADIs is the regulatory objective. In practice, the primary site of exposure is the fruit's skin: when consuming the flesh, exposure is substantially lower. If the label indicates an antifungal agent has been used and you intend to use the peel (for marmalade, for instance), thorough washing is the standard precautionary recommendation.
The correct reading of a label that discloses antifungal use is not "this fruit is dangerous" but "this fruit is labeled and therefore regulated." The absence of labeling in unlabeled alternatives does not mean those fruits are antifungal-free; it may mean they were not treated, or it may mean they were not labeled when they should be.
Applying a sharper filter
Urban legends about additives share a structural pattern: they begin with a real piece of research, then strip out the population studied, the doses used, and the distinction between relative and absolute risk. Three questions that improve the resolution of any food-safety claim:
- Was the population studied infants and toddlers, or adults with a specific chronic condition?
- Does the reported risk figure represent relative risk or absolute risk?
- How does the dose used in the study compare to the amounts present in everyday food?
The foundation of infant and toddler nutrition is dietary variety and appropriate portion sizes. Targeting a specific additive for total elimination tends to exact a cost — reduced dietary variety, more restricted family meals — in exchange for risk reduction that is, on close examination, quite limited.
Summary
IARC's classification system expresses the certainty of evidence for carcinogenicity, not the size of everyday risk — and the two are routinely confused. Research linking phosphates to kidney harm was conducted in adults with chronic kidney failure; extrapolating it to healthy infants has no evidentiary basis. Benzene formation from preservative–vitamin C interactions is real chemistry, but real-world beverage concentrations are below the regulatory threshold in the great majority of products. Antifungal labels on citrus indicate regulated use, not a safety failure.
Reading risk accurately requires keeping the population, the dose, and the absolute numbers in view — not just the name of the substance.
References
- Bouvard V, Loomis D, Guyton KZ, et al; International Agency for Research on Cancer Monograph Working Group. Carcinogenicity of consumption of red and processed meat. Lancet Oncol. 2015;16(16):1599–1600. doi:10.1016/S1470-2045(15)00444-1. PMID: 26514947.
- International Agency for Research on Cancer. Red Meat and Processed Meat. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 114. Lyon: IARC; 2018. PMID: 29949327.
- Uribarri J, Calvo MS. Dietary Phosphorus Excess: A Risk Factor in Chronic Bone, Kidney, and Cardiovascular Disease? Adv Nutr. 2013;4(5):542–544. doi:10.3945/an.113.004234. PMC: 3771143. PMID: 24038251.
- US Food and Drug Administration. Questions and Answers on the Occurrence of Benzene in Soft Drinks and Other Beverages. FDA; 2007. https://www.fda.gov/food/environmental-contaminants-food/questions-and-answers-occurrence-benzene-soft-drinks-and-other-beverages
- EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS). Scientific Opinion on the re-evaluation of the safety of the food additive ortho-phenylphenol (OPP, E 231). EFSA Journal. 2012;10(3):2595. doi:10.2903/j.efsa.2012.2595.
- JECFA (Joint FAO/WHO Expert Committee on Food Additives). Summary and conclusions of the seventy-third meeting. WHO Food Additives Series. Geneva: WHO; 2011.