The Food Standards Code puts a legal ceiling on the sodium in food for infants, and it puts three different ceilings in the one paragraph. Which one applies to a pack depends on two words the Code’s definitions standard doesn’t define. The paragraph is 2.9.2-3(1)(c) of Standard 2.9.2.
The paragraph opens “Food for infants must not contain”, and paragraph (c) then sets out “more than the following amounts of sodium”:
| Food class | Ceiling, per 100 g |
|---|---|
| Rusks | 350 mg |
| Biscuits | 300 mg |
| Flours and pasta; ready-to-eat foods for infants; fruit drink, vegetable juice and ready-to-eat fruit-based foods | 100 mg |
That is three numbers across five classes, because the 100 mg figure covers three of them. In full, the middle class of that third row reads “ready-to-eat foods for infants (including cereal-based foods for infants other than rusks and biscuits)”. That parenthetical is what pulls an ordinary infant cereal into the 100 mg group, and it does the job by pointing back at the two classes above it.
Those figures haven’t moved. The sodium paragraph was not touched by any of the later amendments recorded at the foot of the compilation, including the one that added native bee honey to a different paragraph of the same section.
3.5 times the sodium, gram for gram
In percentages, 100 mg in 100 g is 0.1% sodium by weight, and 350 mg is 0.35%. At their respective ceilings a rusk may lawfully carry 3.5 times the sodium of the classes held to 100 mg, gram for gram, and a biscuit three times. Same standard, same age bracket, and the only thing between them is a category name.
A 120 g pouch at the 100 mg ceiling carries 120 mg of sodium. Against the NHMRC Adequate Intake for a 7 to 12 month old, 170 mg a day, that is about 71% of it (more on that number below), and 120 mg is also exactly the Adequate Intake for a baby under six months. A rusk is a different case. Take a tenth of 100 g: 10 g of rusk at its ceiling carries 35 mg of sodium, about a fifth of the same 170 mg. No rusk was weighed for this post, so that 35 mg is arithmetic on the ceiling and not a measured product.
The ceilings reach less than the aisle does. They apply to food for infants as the Code defines it: food intended or represented for use as a source of nourishment for infants. The Code defines an infant as “a person under the age of 12 months”. A pack represented for use from 12 months is not food for infants, so the sodium ceilings in 2.9.2 do not reach it. The definition of food for infants also excludes, among others, infant formula products, which have their own standard, and unprocessed fruit and vegetables, so no sodium ceiling in 2.9.2 applies to a banana.
Salt is banned outright in some of them
Fruit drink, vegetable juice and ready-to-eat fruit-based foods get a second rule, in the same Standard: no added salt at all. That is a prohibition, which is different from a cap.
“Salt” has a definition, in Standard 1.1.2: a food that is the crystalline product consisting predominantly of sodium chloride, obtained from the sea, underground rock salt deposits or natural brine. So the ban bites on salt as an ingredient. Sodium that arrives some other way, through an additive or naturally in the ingredients, isn’t “added salt”, and it still counts toward the 100 mg ceiling.
Separately, the NHMRC’s Infant Feeding Guidelines summary advises (February 2013): “Do not add salt to foods for infants. This is an important safety issue as infant kidneys are immature and unable to excrete excess salt.”
“Rusk” and “biscuit” in the definitions standard
Search the Code’s definitions standard and “rusk” turns up zero times. So does “biscuit”. The two words appear in Standard 2.9.2 only inside the sodium paragraph itself.
So the step from 100 mg to 300 or 350 turns on a category name that Standard 1.1.2 doesn’t define. Is a baked finger of dried toast a rusk? Is a teething cracker a biscuit? Neither Standard 1.1.2 nor 2.9.2 says, and no pack-front claim settles it, because the pack’s own name for the product isn’t a legal definition either.
A definition in Standard 1.1.2, or FSANZ naming the categories, would settle it. Peeka searched for an FSANZ proposal reviewing sodium in Standard 2.9.2 and found none, so there is nothing to wait for and no date to cite.
Lining a pack up against the law
For solid and semi-solid foods, the Code does give you one thing for free: the same units. A pack’s nutrition information panel has to show sodium, per serving and per unit quantity, “expressed in milligrams or both milligrams and millimoles”. The same standard defines unit quantity as 100 grams for a solid or semi-solid food, and 100 millilitres for a liquid.
Those words are on the pack because a third instrument prescribes them. Food for infants gets its own panel layout, set out at S12-6 of Schedule 12, and the format there names a Sodium row and a Quantity per 100 g (or 100 mL) column. So for a solid or semi-solid food, that row, in that column, is in the same unit as the ceilings in 2.9.2. No conversion. Read the number across and compare it with 350, 300 or 100. Fruit drink and vegetable juice are liquids: their panel figure is per 100 mL while their ceiling is per 100 g, and the two do not line up without the product’s density.
Infant food is excused from some panel rows. Section 2.9.2-11 lets it skip some of them, saturated fat and potassium for two. Sodium is not one of them, so the sodium row stays required.
The comparison has two limits. You can’t tell from the panel which of the three classes the maker put the product in, and a number under the ceiling tells you the product is under that ceiling. It doesn’t tell you the ceiling was the right one for the food in your hand, and you can’t work that out from the panel.
A ceiling and a daily figure are different numbers
The NHMRC’s Nutrient Reference Values give an Adequate Intake for sodium of 120 mg a day from birth to six months and 170 mg a day from seven to twelve months. An Adequate Intake is an estimate of what is enough. It isn’t a cap, and going over it breaches nothing.
The birth-to-six-month figure was calculated from average breast milk intake (0.78 L a day) at an average sodium concentration of 160 mg/L. The seven-to-twelve-month figure was extrapolated from it, “from a consideration of metabolic body weights and relative energy requirements”. The infant values were not reviewed in the 2017 update either, so they stand as set in 2006.
The ceiling is firm as law: a written rule with a fixed figure, unchanged since the Standard was published in 2015. Standard 2.9.2 states no health derivation for 350, 300 or 100 mg/100 g. The 170 mg is firm about adequacy: it is NHMRC’s estimate of what an infant’s sodium intake looks like when the sodium comes from breast milk, scaled up for body size and energy needs at seven to twelve months. It is derived rather than measured in that age group and last reviewed in 2006, and nothing about the derivation is hidden; NHMRC sets it out on the page. A fifth of 170 is a sense of scale, not a threshold you have crossed or avoided.
If you go looking on the NHMRC page for the number that marks too much, the entry is against the Upper Level of Intake for infants from birth to twelve months. NHMRC’s stated reason for having no figure is a gap in the data: the 2006 Upper Level “could not be established because of insufficient data documenting the adverse effects of chronic over-consumption of sodium in this age group.” The entry reads, in full: “Not possible to establish. Source of intake should be through breast milk, formula and food only.”
General information, not medical advice. Chat with your GP or child health nurse.