“Free from” is the most efficient marketing grammar in the food and supplement aisle. It costs four words, it is grammatically impossible to argue with at the shelf, and it tells you about the label rather than about the person holding it. This article is about the second layer: what those words are permitted to mean, and what an audit of real shelves found when the two were compared.
An absence claim is still a claim
Under the US nutrition labelling rules, a nutrient content claim is a regulated thing, and the “free” level of a nutrient is one of its defined categories. The 2003 FDA final rule on trans fatty acids, published in the Federal Register, shows how the definition is treated: when the agency was not satisfied with the proposed definition of the “free” level for trans fat, it withdrew the proposal entirely, along with the proposed footnote reading “Intake of trans fat should be as low as possible.”
That withdrawal is the useful precedent. An absence claim is not a free speech exemption. It is a defined level of a defined nutrient, and if the regulator cannot defend the threshold, the claim goes rather than the threshold being quietly relaxed.
The audit: 276 pouches
Brunacci and colleagues, in BMC Public Health in 2023, did the slow version of this. They analysed 276 ready-to-use complementary infant food pouches from 15 manufacturers sold in Australian retailers, for infants from 4+ to 12+ months, scoring nutritional content, texture, packaging labelling and serving size against a Nutrition Profile Index model and the Australian Infant Feeding Guidelines.
The numbers: total sugar ranged from 0.8 to 17.5 grams per 100 grams. Twenty per cent (n = 56) had added sugars, 17 per cent (n = 46) had added fruit juice, and 71 per cent (n = 196) had added fruit puree. Saturated fat ran from 0.0 to 5.0 grams per 100 grams, sodium from 0.0 to 69 milligrams per 100 grams, and dietary fibre from 0.0 to 4.3 grams per 100 grams. Only two products out of 276 were nutritionally adequate on the profiling tool.
And then the sentence that justifies the whole article: “no added sugar” was claimed on 59 per cent of pouches, despite the addition of free sugars.
That is not, strictly, a regulatory failure. Added sugar and free sugar are different definitions, and a product can honestly contain no added sugar while being sweetened entirely with fruit purée. It is a product-design failure, and the labelling is doing exactly what it is permitted to do. The authors’ own conclusion is blunt: there is an urgent need for improved regulation of product composition, serving sizes and labelling.
The marketing categories they found are worth listing, because they are the vocabulary of “clean label” in miniature: ingredient premiumisation, nutrient absence claims, claims about infant development and health, good parenting, and convenience.
Half of a country’s food supply
Franco-Arellano and colleagues, in BMC Nutrition 2017, measured claim prevalence across the Canadian supply using the University of Toronto Food Label Information Program, comparing 10,487 products in 2010/11 with 15,342 in 2013.
In 2013, 49 per cent of products displayed some type of claim and 46 per cent carried a nutrition claim of some kind. At least one approved nutrient content claim appeared on 42.9 per cent of products, down from 45.5 per cent in 2010 (p < 0.001). Disease risk reduction claims were on 1.5 per cent of products versus 1.7 per cent (p = 0.225). Front-of-pack general health claims rose to 20 per cent from 18.9 per cent (p = 0.020). Gluten-free claims appeared on 7.3 per cent.
The trend to notice is the shape of it: the legally regulated nutrient content claim is flat to slightly down, the general front-of-pack claim is up, and the disease claim is stuck at around one and a half per cent. Their conclusion — that claims related to nutrients of public health priority, such as sugars and sodium, were not commonly used, and that monitoring is essential to determine whether claims reflect public health objectives or are being used as marketing tools — is a fair characterisation of a supply that has been regulated for years and has not moved much.
“Free from” as a diet, not a label
Rej, Aziz and Sanders, in the Proceedings of the Nutrition Society 2019, look at what happens when an absence term migrates from the label to the diet. They note that coeliac disease now affects 1 per cent or more of all adults, with delays in diagnosis common enough that for every adult diagnosed roughly three to four cases go undetected. They also note a rising interest in the gluten-free diet as a “lifestyler”, “free from” or “clean eater” choice, which they say causes concern.
The clinical counterpart is Abdi and colleagues’ 2023 review in Nutrients. A strict gluten-free diet is the only available treatment for coeliac disease, where gluten triggers an immune reaction leading to enteropathy, malabsorption and symptoms. In non-coeliac gluten/wheat sensitivity, the mechanism leading to symptoms is unknown and neither wheat nor gluten triggers enteropathy or malabsorption — which is why a gluten-restricted diet may suffice for symptom control there. And for both populations, the risk of malnutrition and of macro- and micronutrient deficiencies is increased by adopting the diet, which is why the review calls for nutritional assessment and monitoring under multidisciplinary care.
The distinction that carries the article
An absence term on a label describes the product. An absence diet describes the person. The pouches audit and the coeliac review sit on opposite sides of that line, and confusing them is how a marketing grammar becomes a nutritional problem.
What this evidence does not show
It does not show that absence claims are deceptive. In the pouch audit the “no added sugar” claim is, on the regulator’s own definitions, accurate. It does not show that any particular “free from” product is harmful, and it does not show that gluten-free diets are a bad idea for anyone who chooses one. The Canadian data end at 2013, so nothing here describes the current supply. And the coeliac prevalence figure describes a diagnosed-enough population to be measured; undiagnosed disease does not appear in it.
This article summarises published audits, regulatory documents and reviews. It is not medical advice and it is not a recommendation for any individual.
