FSANZ estimates that Australians aged two years and older eat an average of 2,150 mg of sodium per day from an average of 5,500 mg of salt (5.5 g). About 80 per cent of this would be from processed foods and 20 per cent from salt used at the table or in home cooking. This estimate of sodium intake from salt does not include the smaller amounts of sodium coming from naturally occurring sodium or sodium-containing food additives. Because this is an average, there will be a lot of Australians who eat more than this and more than the recommended maximum intakes. Read more about sodium and salt
Foods that contribute the most to Australians’ salt consumption are bread and bread rolls, meat, poultry and game products, including processed meat, and cereal products and cereal-based dishes such as biscuits and pizza.
How sodium intake is estimated
The best way to estimate sodium intake is to collect people’s urine over a 24-hour period and measure how much sodium it contains, since this will approximate sodium intake. However, this is not easy to do and is expensive, so has not been used in large scale national surveys in Australia. Measuring urinary sodium excretion does not tell you directly how much salt someone is eating because it measures all sources of sodium, not just sodium from salt. It also doesn’t provide information on which foods contribute most to sodium intake.
Sodium intake can also be estimated by recording how much food and drink a person has over a set period of time (usually one or two days). This information can be matched with data on how much sodium is in different foods, to estimate total sodium intake for that person. Although this approach works well for most nutrients, there are some particular difficulties for sodium because sodium intakes can be increased if salt is added at the table or during cooking. It is hard to accurately estimate how much of this ‘discretionary’ salt is added by each person. Also, different brands of the same food can vary widely in sodium content and this variability may not be captured in nutrient databases.
How salt intake is estimated
Once sodium intakes have been estimated, salt intake can then be calculated from this using the ratio of the molecular mass of sodium chloride (58.44 g/mol) to the atomic mass of sodium (23 g/mol). However it is important to first take into account the proportion of sodium that is not from sodium chloride. If you don’t do this, you will overestimate salt intake. For example, regular milk has a sodium content of 46 mg/100g (FSANZ, 2007) but does not contain any added sodium chloride. Biscuits can have substantial sodium contents (e.g. 290 mg/100g for plain sweet biscuit, see FSANZ, 2007) but a large amount of this will be from sodium bicarbonate or other leavening agents.
How FSANZ estimated Australians’ salt intake
FSANZ was interested in how best to deliver more iodine to Australians – salt with added iodine is an effective tool for increasing a population’s iodine intake. You can read about why we needed to do this in our proposal documents for Proposal P230 – Iodine Fortification.
Therefore we needed to estimate how much salt Australians eat each day.
We took food consumption data for 13,858 Australians, aged two years and above, from the 1995 Australian National Nutrition Survey. We matched this food consumption data with data for sodium levels in foods and calculated how much salt this was equivalent to, for every food that was recorded in the 1995 Survey, after adjusting for sodium that comes from non-salt sources. The sodium data we used were largely data from the early 2000s onwards, because we knew there had been some important changes since the time of the Nutrition Survey in the way the Australian food industry uses salt (e.g. see Williams et al, 2003). We then used our custom-built dietary modelling program, DIAMOND, to estimate salt intake for each person in the 1995 National Nutrition Survey. From this, we derived mean and high consumer salt intake estimates. You can read about what we did in more detail in our Salt Intake from Processed Food and Discretionary Use in Australia publication (pdf 204kb).
How do our estimates compare to other Australian research?
Some people have stated that they believe FSANZ’s estimates of salt intake are too low (e.g. see http://www.awash.org.au
Most other Australian research on sodium intakes has estimated total sodium intakes, not salt, although this may sometimes be referred to as salt intake or ‘salt equivalents’. FSANZ’s estimates are for sodium chloride and do not take account of the 10% or more of dietary sodium that comes from other sources. Therefore, it is to be expected that our estimates of sodium from salt will be lower than estimates of total sodium intake.
FSANZ estimated that across all Australians, including young children, mean salt intake is 5.5 grams per day. This amount of salt would deliver an average of 2,150 mg sodium per day. Therefore mean total sodium intakes in Australia, taking into account non-salt sources of sodium, would probably be around 2,500 mg/day. This is equivalent to about 110 mmol sodium per day, which is another way of expressing sodium intake. Because this is an average value, there will be many Australians whose sodium intake is higher than this.
Every technique for estimating nutrient intakes has its limitations. FSANZ acknowledges these limitations in its dietary intake assessment reports. In the case of FSANZ’s work on salt intakes, one of the major limitations is that we had to make assumptions about how much salt people add at the table or during cooking (assumed to be between 0.8 and 1.2 grams per day for those who reported often or usually adding salt). Another limitation is that we used sodium concentration data that are averages of concentrations across food groups (e.g. across all types of white breads), whereas some individuals may regularly eat foods that are higher (or lower) in sodium than the average value.
The 1995 National Nutrition Survey is the most recent national food consumption survey covering Australians from two years of age and above. In most cases it provides the most reliable data available for estimating nutrient intake from the wide range of foods that make up our diets. While consumption of some foods (e.g. sushi) might have changed since 1995, consumption of the main staples of our diet, such as bread and milk, is unlikely to have changed to such an extent that intake estimates would be invalidated.
FSANZ doesn’t believe that it is appropriate to assume salt intake in Australia is the same as that in other developed countries. In particular, the Australian National Heart Foundation, through its Tick program, has been a leader in encouraging reduced salt use in processed foods. Our salt intake estimates are based on Australian food consumption data and data for the composition of Australian foods.
Boorman, J., Cunningham, J. Mackerras, D. (2008). Salt intake from processed foods and discretionary salt use in Australia.
FSANZ (2007). NUTTAB 2006
Goodall, S., Gallego, G., Norman, R. (2008). Scenario Modelling of Potential Health Benefits Subsequent to the Introduction of the Proposed Standard for Nutrition, Health and Related Claim. Report developed for Food Standards Australia New Zealand. Centre for Health Economics Research & Evaluation.
James, WPT, Ralph A, Sanchez-Castillo CP (1987). The dominance of salt in manufactured foods in the sodium intake of affluent societies. Lancet ; i:426-9
NHMRC and Ministry of Health. 2006. Nutrient reference values for Australia and New Zealand. Commonwealth of Australia.
Williams P. McMahon A, Bousted R. (2003) . A case study of sodium reduction in breakfast cereals and the impact of the Pick the Tick food information program in Australia. Health Promotion International; 18:51-6.