Development of dietary supplement nutrient data

Preparation of TGA data

Development of nutrient profiles for dietary supplements consumed in the AHS

FSANZ used a recipe approach for developing nutrient profiles for dietary supplements consumed during the 2011-13 Australian Health Survey (AHS).

To undertake recipe calculations for dietary supplements FSANZ needs information on the:

• ingredients and their proportions
• nutrient content of each of the ingredients
• dosage type and amount. 

Where possible, FSANZ drew on the dietary supplement data provided by the Therapeutic Goods Administration (TGA) to obtain this information.

The AUSNUT 2011-13 dietary supplement database was developed following the key steps outlined below.

Preparation of data provided by TGA for use in recipe calculations

Development of a list of unique ingredients

FSANZ reviewed the data provided by TGA for close to 10,000 dietary supplements to produce a list of unique ingredients containing nutritionally relevant levels of the nutrients to be reported from the AHS.

Ingredients were removed from the initial list if:

• they did not contain any of the relevant nutrients 
• or the content of the relevant nutrients were so low that they would have little impact on population nutrient intake estimates.

Where nutrient data was not available for an ingredient (e.g. some plant/herbal ingredients), FSANZ reviewed the scientific literature to determine whether they were known to contain nutrients such as iodine (e.g. algae) or caffeine (e.g. guarana or green tea extract).

This process resulted in the initial list of unique ingredients being reduced from 1,500 to approximately 320 ingredients.

Development of a nutrient profile for each unique ingredient

A complete nutrient profile was developed for each of the 320 unique ingredients using a range of techniques. The most common techniques were the use of:

• molecular formula and formulation information
• reference books and scientific literature
• product labels and company websites.

Nutrient values were assigned to each ingredient on a per 100 mg basis. Where a nutrient was not present in an ingredient, a zero value was assigned.

Molecular formula and formulation information

Where the molecular formula of an ingredient was known, it was used to determine the proportion of each of the relevant nutrients in each ingredient.

For example, the molecular formula for the ingredient calcium ascorbate dihydrate was used to determine the calcium and vitamin C content as shown below:

Molecular formula = Ca(C6H7O6)2.2H2O
Molecular weight = 426.3 (Ca= 40.078, C=12.01, H=1.008 and O=16)
% of Ca = (40.078/426.3)*100 = 9.4%
% of Vitamin C = (350/426.3)* 100 = 82.16%

Formulation information, such as the proportion of a mineral within a chelate, was provided by the TGA in some cases.

Where vitamins (e.g. vitamins A and E) were expressed in International Units, these values were converted to the appropriate AHS unit for reporting according to NHMRC Nutrient Reference Values for Australia and New Zealand 2005 (NHMRC, 2006).

Some examples of the final nutrient profile for selected nutrients in a series of vitamin C containing ingredients are shown below.

For a complete list of unique dietary supplement ingredients and their nutrient profiles refer to:

• File

  Dietary supplement ingredient list (Excel 103KB)

Development of nutrient profiles for dietary supplements consumed

Defined dietary supplements

Where the respondent was able to identify the exact dietary supplement consumed, nutrient profiles were developed using the formulation data provided by TGA and the nutrient profiles of dietary supplement ingredients generated by FSANZ.

The nutrient profile of each dietary supplement consumed was calculated by multiplying the nutrient value for each ingredient by the ingredient quantity, then summing the total of each nutrient where it was present in more than one ingredient. For example, for a supplement containing four sources of vitamin C, its vitamin C and calcium content would be calculated as follows:

For herbal supplements, most nutrient profiles will be a series of zero values unless the herbal supplement contains ingredients with known nutrient values (e.g. Malpighia punicifolia and vitamin C). For homoeopathic dietary supplements, all values in the nutrient profile have been imputed as zero due to the highly diluted nature of these products.

Where the amount of an ingredient was missing, the gap was filled using values from sources such as company websites, products labels, similar products and other credible websites.

In contrast to the nutrient values for foods, nutrient values for dietary supplements are presented on a per dosage unit basis. Dosage units refer to the way in which dietary supplements are supplied and consumption is reported, and include defined dosage units (tablet, pill, capsule, etc.) and measurable dosage units (g or mL) for the supplements available in bulk (powder, granules or liquid).

 'Not further defined' dietary supplements

Two approaches were used to assign nutrient data to reports of dietary supplements where a respondent was unable to identify the exact dietary supplement consumed.

Nutrient data for these dietary supplements were derived using two approaches:

• Ensuring the nutrient data were representative of all dietary supplements consumed during the survey that had a similar description. For example, a nutrient profile for Dietary supplement, multivitamin and/or multi-mineral, not further defined drew on nutrient data for the most commonly consumed multi-vitamin and mineral supplements weighted according to consumption patterns observed in the AHS. This was the approach used for the majority of undefined supplements.
• On a few occasions, an undefined supplement was assigned the nutrient profile of the most frequently consumed product from the relevant category. For example, Folate, not further specified could be assigned the nutrient profile for the most frequently consumed brand of folate supplement reported during the NNPAS or NATSINPAS.