OVERVIEW OF THE NCI METHOD

The NCI method is able to estimate group usual intakes for nutrients or foods. Some summary information and important features of the NCI method are outlined below. Further details on the method are available from the NCI method website.

GROUP vs INDIVIDUAL USUAL INTAKES

The NCI method uses the pooled intake records for a group to estimate the group’s usual intake distribution. It uses the available day two records for the group to estimate and remove within-person (or day-to-day) variation from the group’s day one reported intakes, thus drawing in the tails of the group intake distribution. As all NCI method calculations in this publication are done drawing on the strength of pooled group intake data, usual intakes for individuals within the group are not produced.¹

MEAN USUAL NUTRIENT INTAKES vs DAY ONE MEAN NUTRIENT INTAKES

Usual nutrient intake calculations aim to remove within-person variation from group intakes, thus drawing in the tails of the intake distribution. The group’s mean usual intakes, in general, will be similar to the mean day one intakes. Some differences may occur because of the weekend/weekday weighting adjustment, exclusion of pregnant and breastfeeding women from the input data, or bias (e.g. from the model simulation/transformation process).² Mean day one intakes are presented in Australian Health Survey: Nutrition First Results - Food and Nutrients, 2011-12 (cat. no. 4364.0.55.007).

NCI METHOD MODEL TYPES

Different versions of the NCI method are available for use, each using a slightly different underlying model. For nutrients or foods that are consumed nearly every day by nearly everyone, a simpler form of the model (“one-part” or “amount-only”) is used. The amount only model has been used for most nutrients in the usual nutrient intakes publication. For nutrients or foods that are less commonly consumed (i.e., episodically), a more complex form of the model (“two-part”) is used. The two-part model was used for folic acid, alcohol, percentage energy from alcohol, and caffeine intakes.¹ Note that there are two forms of the two-part model. For more information on NCI method model types see Model type in Model implementation: data used and model specification.

COVARIATES IN THE NCI METHOD

Covariates are characteristics which influence people’s nutrient intakes, such as their age and sex. The NCI method can use covariates to output results for sub-groups, when the model is run on the pooled intakes of a larger group.³ Covariates can also be used to adjust for the influence that aspects of the dietary collection can have on the reported intakes, such as the day of the week of the dietary recall and whether it was the first or second dietary recall for that respondent. Additionally, it is possible to include covariates purely as internal model parameters to aid in the model fitting process, including estimation of within- and between-person variation.^4,5

OVERVIEW OF THE NCI METHOD MODELLING

The NCI method modelling process, as implemented by the two NCI method SAS macros (mixtran and distrib), can be summarised as:⁶

More information on general features of this process is below. All model types that estimate group usual intake distributions use both the mixtran and distrib macros to perform the five basic steps above.

A model is fit to the input data set (steps one and two) in the first of the two NCI method SAS macros (the mixtran macro).⁶ The main purpose of the model is to estimate how much of the variation in the reported intakes is within-person (or day-to-day) variation, so that this can then be omitted from the usual intake distribution simulated in the second macro. The rest of the variation in the reported intakes (between-person variation) is described using covariates (if covariates are used), and a residual-between person variation term. In fitting the model, a Box-Cox transformation, within-person variation, covariates (if used), and residual between-person variation terms are all jointly estimated (maximum likelihood method).^2,8 Some information on each of these model terms is below.

• Box-Cox transformation: usual intake calculations in the NCI method require a normal or near normal distribution. However, nutrient intake data are usually positively skewed, so the NCI method models use a Box-Cox function to transform the input data to normality or near normality. The strength of the Box-Cox transformation is chosen (i.e. the lambda⁷ value in the function is selected) as part of the overall model fitting process.
• Covariates: In the model fitting process, the relationship of covariates to the reported intakes is estimated (i.e. their linear regression coefficients).
• Residual between-person variation: also referred to as ‘person-specific effects’, this is the between-person variation which is not explained by covariates. It is modelled as a normal distribution (after transformation) with a mean of zero.
• Within-person variation is also modelled as a normal distribution (after transformation) with a mean of zero.^2,8

Note that the two-part model is more complex. For episodically consumed nutrients, in the two-part model the NCI method also estimates the probability of consuming the nutrient on a given day, using logistic regression with an additional person-specific random effect, and covariates (if used).²

Information on the model fitted to the data is output from the ‘mixtran’ macro and passed to the second NCI method SAS macro (the ‘distrib’ macro).¹²

The distrib macro runs steps three, four, and five above, starting with a Monte Carlo simulation component.⁶ A number of pseudo persons (n=100) are generated to represent each individual in the ingoing data set, using the model estimated in the mixtran macro. Each of the 100 pseudo persons for an individual has the covariate information corresponding to that individual (such as the same age and gender), but different simulated person-specific effects. ^2,9 Within-person or day-to-day variation is not included in the simulated intakes, as it does not contribute to usual intakes. The simulated pseudo-person intakes are back-transformed to the original scale (or units) to give a simulated population usual intake distribution. Population mean and percentiles of intake, along with proportions above and below cut-points (NRVs), are derived empirically from this distribution. Sample weights are taken into account to ensure the results represent the population.²

NCI METHOD MACROS

The NCI method website provides SAS macros that can be downloaded and used for usual intake calculations. The macros used by ABS, as programmed by NCI, were more recent versions of mixtran and distrib (version 2) provided by the NCI. This version uses a different back-transformation (nine-point approximation) which is better able to handle very skewed nutrient intakes.²

ENDNOTES

1 National Cancer Institute, 2013, Usual dietary intakes: details of the method, <http://appliedresearch.cancer.gov/diet/usualintakes/details.html>, last accessed 16/02/2015.
2 Tooze, JA et al. 2010, ‘A mixed-effects model approach for estimating the distribution of usual intake of nutrients: The NCI method’, Statistics in Medicine, vol. 140, pp.111-116, <http://jn.nutrition.org>, last accessed 09/02/2015.
3 Dodd, KW et al. 2006, ‘Statistical methods for estimating usual intake of nutrients and foods : a review of the theory’, Journal of the American Dietetic Association, vol. 106, pp. 1640-1650, <http://www.andjrnl.org/>, last accessed 09/02/2015.
4 Tooze, JA et al. 2006, ‘A new statistical method for estimating the usual intake of episodically consumed foods with application to their distribution’, Journal of the American Dietetic Association, vol. 106, pp. 1575-1587.
5 Kipnis, V et al. 2009, ‘Modeling data with excess zeros and measurement error: application to evaluating relationships between episodically consumed foods and health outcomes’, Biometrics, vol. 65, no. 4, pp. 1003-1010.
6 National Cancer Institute, 2013, Usual dietary intakes: SAS macros for analysis of a single dietary component, <http://appliedresearch.cancer.gov/diet/usualintakes/macros_single.html>, last accessed 16/02/2015.
7 The Box Cox transformation is a function that transforms data to a near-normal distribution using a variable called a lambda (). This variable affects the strength of the transformation, so that the transformation can be adjusted to suit the characteristics of the input data set. In the NCI method, the lambda is selected such that the within-person errors are normally distributed around mean zero on the transformed scale. For more information see endnote 2. The Box-Cox function is , where r is the input nutrient intake. Note that a minimum lambda bound of 0.01 has been set when using this function. Therefore, although in a Box-Cox transform the limiting case for =0 is formally defined as the natural logarithm, it has not been employed in the usual nutrient intakes publication.
8 Zhang, S et al. 2011. ‘Fitting a bivariate measurement error model for episodically consumed dietary components’, The International Journal of Biostatistics, vol. 7, no. 1, <http://www.bepress.com/ijb/vol7/iss1/1>, last accessed 01/03/2014.
9 A set seed was used in simulations in the distrib macro.

Previous Page

Next Page