The NCI method can be used to estimate:
- the distribution of usual food or nutrient intake for a population or subpopulation;
- error-corrected estimates of regression slopes in a health outcome model where usual intake is considered a continuous exposure; and
- the effects of non-dietary covariates on usual food or nutrient consumption.
The particular statistical techniques used depend on the primary mode of dietary assessment (e.g., 24-hour recalls/food records vs. food frequency questionnaires/screeners), whether the dietary components of interest are consumed nearly every day or episodically, and on the number of dietary components to be modeled simultaneously.
SAS macros are currently available to facilitate modeling of:
- a single dietary component, whether consumed daily or episodically;
- ratios of two dietary components that are consumed nearly every day; or
- multiple dietary components, whether consumed daily or episodically.
As new papers are published documenting extensions to the NCI Method, additional accompanying macros will be posted for use by analysts.
Please send any comments and questions regarding the macros to firstname.lastname@example.org. We will consider feedback received from users when developing future versions of our software and other resources. We regret that we are unable to respond to every comment and question.