4363.0.55.001 - Australian Health Survey: Users' Guide, 2011-13  
Latest ISSUE Released at 11:30 AM (CANBERRA TIME) 06/03/2015   
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product  
Contents >> Nutrition >> Usual Nutrient Intakes >> Reporting Against Nutrient Reference Values

REPORTING AGAINST NUTRIENT REFERENCE VALUES

The main purpose of calculating group usual nutrient intakes is to give a more accurate assessment of the proportion of people in the population who are not meeting their long-term nutrient requirements, or usually exceeding the upper safe level of intake.1 The proportion of the group outside advisable limits on the balance of energy from different macronutrients, is also of interest. To calculate these figures, usual nutrient intakes are compared with Nutrient Reference Values (NRVs), specifically Estimated Average Requirements (EARs), Upper Levels of Intake (ULs), and Acceptable Macronutrient Distribution Ranges (AMDRs).2

THE USE OF NRVS IN ESTIMATING THE PREVALENCE OF INADEQUATE INTAKES

Individuals within any age and sex group will vary in their actual personal requirements for a nutrient due to, for example, differences in their age, body weight, physiology, and health. If it were possible to measure the actual requirements of every individual in a group, and each individual’s usual intake, then the prevalence of inadequacy would simply be the proportion of those individuals whose measured usual intake was below their measured personal requirement. However, neither measurement of individuals’ personal requirements for nutrients, nor measurement of their usual nutrient intakes, is feasible for reasonably-sized samples. Alternative approaches, which work on a group rather than individual basis, are therefore used.1

To estimate the prevalence of inadequate nutrient intake on a group basis, three steps are performed:

    1. Estimate the group’s usual nutrient requirements
    2. Estimate the group’s usual nutrient intake
    3. Calculate the prevalence of inadequate intakes by comparing the group’s usual intakes and usual requirements 1

The first step of estimating group usual nutrient intake requirements has been performed in setting the NRVs. The NRVs are a set of recommendations made by the Australian National Health and Medical Research Council and the New Zealand Ministry of Health for nutritional intake, based on currently available scientific knowledge.2 Where comparisons with guideline values (Nutrient Reference Values or NRVs) have been made, any results outside of these guideline values need to be considered along with how the guideline values were established in order to appropriately interpret the quality of the resulting estimates. More information on the methods used to derive the NRVs for each nutrient is available at Nutrient Reference Values for Australia and New Zealand. The most recent NRVs available at the time of publishing the usual nutrient intakes were the 2006 NRVs.

The second step is to derive a group usual nutrient intake distribution. In the Australian Health Survey: Usual Nutrient Intakes, 2011-12 the NCI method has been used to do this, based on two days of estimated nutrient intakes (24 hour dietary recalls) of individuals from the 2011-12 NNPAS, representing each of the NRV age and sex groups. The output of the method is a simulated distribution of usual nutrient intakes for each of these age and sex groups in the population.

To perform the third step of comparing group usual nutrient intakes with group usual nutrient requirements, two methods have been used in this publication: the EAR cut-point method (most nutrients) and Beaton’s full probability method (iron). Summary information on these two methods is outlined below.

EAR CUT-POINT METHOD

The EAR cut-point method can be used to estimate the proportion of the population with inadequate intakes when:
    • intakes and requirements are not correlated
    • the distribution of requirements is symmetrical
    • the distribution of intakes is more variable than the distribution of requirements. 1
Where these conditions are met, the proportion of the group’s usual intake distribution falling below the EAR is taken to be equivalent to the prevalence of inadequate intakes (that is, the proportion of the group not meeting their nutrient requirements).3

The bias due to applying the EAR cut-point method is expected to be low when the mean intake and EAR are similar. As the difference between mean intakes and the EAR increases, the size of the relative bias from this method increases. When the true prevalence of inadequacy in the population is around 1 to 2%, the EAR cut-point method may result in an estimate of prevalence of around 3%.4 This bias will not be included in the estimates of sampling error in the data cubes (see Data Quality).

More information on the EAR cut-point method, including an illustration of the workings of the method, is available at Dietary Reference Intakes: Applications in Dietary Assessment.1

At the time of publishing the usual nutrient intakes, EARs were available for:
  • Macronutrients:
    • Protein
  • Vitamins:
    • Vitamin A (retinol equivalents)
    • Thiamin (B1)
    • Riboflavin (B2)
    • Niacin equivalents (B3)
    • Vitamin B6
    • Vitamin B12
    • Dietary folate equivalents
    • Vitamin C
  • Minerals:
    • Calcium
    • Iodine
    • Iron
    • Magnesium
    • Phosphorus
    • Selenium
    • Zinc
The conditions for use of the EAR cut-point method are assumed to be satisfactorily met, and the EAR cut-point method has been used for all of these nutrients, except iron.1 Beaton’s full probability method has been used for iron (see next section).

BEATON'S FULL PROBABILITY METHOD

The full probability approach must be used in place of the simpler EAR cut-point method when the requirements distribution is skewed. This is known to be the case for the iron requirements distributions of the following groups: children aged 1 to 3 years, children aged 4 to 8 years, and menstruating females aged over 14 years.5 For consistency, the full probability method has been used for all age and sex groups for iron.

Unlike the EAR cut-point method, detailed information on the shape of the requirements distribution is required for the full probability method.1 Information on the iron requirements distribution used, and how the method was implemented in this publication to estimate the prevalence of inadequate iron intakes, are in Beaton’s Full Probability Method for Iron.

THE USE OF NRVS IN DETERMINING THE PROPORTION AT RISK OF ADVERSE EFFECTS FROM EXCESS INTAKE

The proportion of the population above the UL reflects the proportion at risk of adverse effects from excess intake. Although the UL is used as a cut-off value, not all intakes above the UL are necessarily excessive or unsafe. Intakes that are much greater than the UL will carry a higher risk of adverse effects than intakes just above the UL.2

The proportion of the population above the ULs for niacin (set for both vitamin B3 as nicotinic acid and as nicotinamide), and vitamin B6 (set for pyridoxine) have not been published. This is because, based on the food composition data, it was not possible to report on the specific chemical forms of these nutrients to which these ULs apply. The UL for magnesium is set for supplemental magnesium only, and so the intakes from foods were not compared with this value.2 The published usual intake distributions for niacin (vitamin B3), vitamin B6 and magnesium should not be compared with these ULs.

Proportions above the ULs for the following nutrients were therefore reported:
  • Macronutrients:
    • Long chain omega 3 fatty acids
  • Vitamins:
    • Preformed vitamin A (retinol)
    • Folic acid
    • Vitamin E
  • Minerals:
    • Calcium
    • Iodine
    • Iron
    • Phosphorus
    • Selenium
    • Zinc
    • Sodium

ACCEPTABLE MACRONUTRIENT DISTRIBUTION RANGES

Acceptable macronutrient distribution ranges (AMDR) are another set of NRVs. AMDRs form part of the recommendations for optimising the balance of intake of each of the macronutrients to lower chronic disease risk while allowing for adequate micronutrient intakes. Usual intakes outside of the AMDR indicate imbalance in the contribution of energy from each of the macronutrients, and an increased risk of chronic disease over time.6 The proportion of the population with usual intake outside (both above and below) each of these ranges has been presented. For more information on the methods used to report against AMDRs see the modelling ratios section of Model implementation.

OTHER NRVS

Adequate Intakes (AI)
Where insufficient evidence exists to set an EAR, in some cases an AI has been set. The basis for each AI varies, but in general is the average daily nutrient intake level of a group of apparently healthy people.2 Usual nutrient intake distributions for nutrients with an AI are available in the data cubes. However, no comparisons have been made with AIs in this publication. Any comparisons of the published distributions with AIs should be made with caution, and interpreted considering the basis on which the AI has been set.

Recommended Dietary Intakes (RDI)
RDIs should not be used to estimate the prevalence of inadequate intakes at the group level. They are appropriate for use for individuals. The appropriate NRV for assessing nutrient intakes of groups is the EAR. RDIs are the average dietary intake level that is sufficient to meet the nutrient requirements of nearly all (97-98%) healthy individuals in a particular age and sex group. As RDIs are set much higher than EARs (EAR + 2 standard deviations), their use as a cut-point overestimates the prevalence of inadequacy.1,2

Suggested Dietary Targets (SDT)
SDTs have been set for some nutrients where there is evidence that higher intakes (e.g. above the RDI or AI) could have benefits in reducing chronic disease risk. Based on data available at the time, in most cases in the 2006 NRVs the SDT was set at the 90th percentile of intake in the Australian and New Zealand population,.2,7 No comparisons have been made with SDTs in this publication. Any comparisons should be made considering the basis for the SDT used, and any developments in the evidence since it was set.

PREGNANCY AND LACTATION

Different NRVs are set for women who are pregnant or breastfeeding, because they have different nutrient requirements.2 No results are presented against these NRVs, as the sample of pregnant or breastfeeding women in the 2011-12 NNPAS was not large enough to estimate their usual nutrient intakes.

As women who are pregnant or breastfeeding may eat differently, intake data from all females who reported that they were pregnant or breastfeeding (n=228) was removed from the input data set. The day one sample was therefore 11,925, with 64% of these records having a second day of intake (n=7,585).

REFERENCE BODY WEIGHTS

The 2006 NRVs used in the usual nutrient intakes publication are underpinned by a set of reference body weights for each age and sex group.2 The NRVs in the usual nutrient intakes publication have been used as published for all nutrients. They are based on the NRV reference body weights, and not the body weights measured in the 2011-12 NNPAS sample.

ENDNOTES

1 Food and Nutrition Board: Institute of Medicine, 2000, Dietary Reference Intakes: Applications in dietary assessment, National Academy Press, Washington, D.C., p.p.73-105.
2 National Health and Medical Research Council and New Zealand Ministry of Health, 2006, Nutrient Reference Values for Australia and New Zealand, <https://www.nrv.gov.au/>, last accessed 4/2/2015.
3 The proportion of a group’s usual intake distribution falling below the EAR is taken to be equivalent to the prevalence of inadequate intakes. This is because although some individuals with intakes below the EAR will actually also have low personal requirements and be meeting their personal nutritional needs, and some individuals with intakes above the EAR will actually have high personal requirements and not be meeting their nutritional needs, the size of these two groups should be approximately equal (given certain conditions, as outlined).
4 The bias from applying the EAR cut-point method is highest when the cut-point is furthest from the mean usual nutrient intake, that is, on either tail of the usual nutrient intakes distribution.
5 Institute of Medicine, 2001, Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. National Academy Press, Washington, D.C., pp. 697-703 <http://www.iom.edu/Reports/2001/Dietary-Reference-Intakes-for-Vitamin-A-Vitamin-K-Arsenic-Boron-Chromium-Copper-Iodine-Iron-Manganese-Molybdenum-Nickel-Silicon-Vanadium-and-Zinc.aspx>
6 National Health and Medical Research Council and New Zealand Ministry of Health, 2006, Nutrient Reference Values for Australia and New Zealand, <https://www.nrv.gov.au/chronic-disease/macronutrient-balance>, last accessed 4/2/2015.
7 Data available at the time of setting the NRVs included the 1995 National Nutrition Survey.


This section contains the following subsection :
          Beaton’s Full Probability Method for Iron

Previous PageNext Page