The ‘DIAMOND’ (DIetAry Modelling Of Nutritional Data) computer program was custom built by Food Standards Australia New Zealand (FSANZ) in the mid-1990s to calculate dietary exposure to food chemicals such as food additives, pesticides, contaminants, nutrients and other food ingredients.
FSANZ is currently undertaking a major redevelopment of this computer program, to provide an up-to-date secure, integrated data storage and management system called ‘Harvest’.
How is Harvest set up?
There are currently two main parts to the Harvest system: the data storage section and the dietary exposure assessment section. Planned future development of Harvest will incorporate capacity to store and analyse food composition data.
What data are stored in Harvest?
Harvest requires two main sets of data to estimate dietary exposure:
- the concentration of the food chemicals in specific foods
- data on the food consumption levels of those foods and information about the consumers such as their age, gender and body weight.
Other data sets used include:
- food classification or grouping systems, so similar foods can be grouped together for reporting purposes
- recipe datasets to break down mixed foods to their basic ingredients
- health-based guidance values which indicate a safe level of exposure or required level of intake.
Where do food chemical concentration data come from?
Food chemical concentration data come from a range of sources including:
- survey data for a wide range of chemicals, including data from total diet surveys and other government surveillance surveys
- trial data (for agricultural and veterinary chemicals)
- food composition tables (for nutrients)
- manufacturers' data for the level of use (for food additives and novel foods or ingredients)
- maximum permission levels from food standards (Australia New Zealand Food Standards Code or other relevant international standards) such as:
- maximum permitted level (MPL) for food additives
- maximum residue limits (MRL) for pesticide and veterinary drug residues
- maximum levels (MLs) for contaminants
- permitted levels of nutrient fortification.
What food consumption data are used?
Food consumption data used in Harvest are from national nutrition surveys from both Australia and New Zealand. Those currently used are:
- 1995 Australian National Nutrition Survey (NNS) for 2 years and above
- 1997 New Zealand National Nutrition Survey (NNS) for 15 years and above
- 2002 New Zealand Children’s Nutrition Survey (CNS) for 5 -14 year olds
- 2007 Australian Children’s Nutrition and Physical Activity Survey (CNS) for 2-16 year olds.
What types of dietary exposure assessments can Harvest run?
Harvest includes options for running assessments or ‘models’ using different food consumption data and food chemical data. These are:
- a chemical exposure model that estimates dietary exposure to pesticides, contaminants, food additives and food ingredients
- a nutrient intake model that can estimate nutrient intakes for either a single day or a second day adjusted intake
- a food intake model that derives summary food consumption data only for specified foods or food groups.
How does Harvest use the data to estimate dietary exposure?
Harvest uses every person’s food consumption data from the national nutrition survey selected and works out what foods they ate that may contain the food chemical being assessed. This includes calculating how much of each food they consumed from mixed foods by using specially developed recipes. For example, how much milk they drank as a glass of milk as well as how much milk they consumed from sauces, puddings, cakes and other foods that contain milk.
Then Harvest multiplies the amount of food by the concentration of the chemical in that food. It repeats this for each food that contains the chemical.
For each person it adds up the estimated dietary exposure from all foods.
Harvest then divides the dietary exposure by that person’s individual body weight (if required). It then ranks estimated dietary exposures for all people in the group being assessed and generates population statistics such as mean, median and 90th percentile.
We may report our results for ‘all respondents’, which means for everyone who took part in the national nutrition survey, or only for ‘consumers’, who are the subset of people who were exposed to the chemical in question, or for both respondents and consumers.
We may also subdivide respondents and consumers by age group or by gender, if this is relevant.
Average exposure or intake is reported as well as exposure or intake in the high consumer
, representing those who have a high exposure to, or intake of, a particular food chemical.
The dietary exposures are then compared to the relevant health-based guidance value (if applicable).
How Harvest is used in dietary exposure assessments
The following diagram summarises the process of conducting a dietary exposure or intake assessment using Harvest:
Dietary exposure and intake assessments at FSANZ
FSANZ Dietary Exposure Assessment Principles and Procedures
Food consumption data used in dietary exposure assessments