Commentary
A comparison of dietary estimates from the National Aboriginal and Torres Strait Islander Health Survey to food and beverage purchase data

https://doi.org/10.1111/1753-6405.12718Get rights and content
Under a Creative Commons license
open access

Abstract

Objective

We compared self‐reported dietary intake from the very remote sample of the National Aboriginal and Torres Strait Islander Nutrition and Physical Activity Survey (VR‐NATSINPAS; n=1,363) to one year of food and beverage purchases from 20 very remote Indigenous Australian communities (servicing ∼8,500 individuals).

Methods

Differences in food (% energy from food groups) and nutrients were analysed using t‐test with unequal variance.

Results

Per‐capita energy estimates were not significantly different between the surveys (899 MJ/person/day [95% confidence interval −152,1950] p=0.094). Self‐reported intakes of sugar, cereal products/dishes, beverages, fats/oils, milk products/dishes and confectionery were significantly lower than that purchased, while intakes of meat, vegetables, cereal‐based dishes, fish, fruit and eggs were significantly higher (p<0.05).

Conclusion

Differences between methods are consistent with differential reporting bias seen in self‐reported dietary data.

Implications for public health

The NATSINPAS provides valuable, much‐needed information about dietary intake; however, self‐reported data is prone to energy under‐reporting and reporting bias. Purchase data can be used to track population‐level food and nutrient availability in this population longitudinally; however, further evidence is needed on approaches to estimate wastage and foods sourced outside the store. There is potential for these data to complement each other to inform nutrition policies and programs in this population.

Keywords

store sales data
population nutrition
dietary assessment
Indigenous Australians
24‐hour recall

Cited by (0)

The authors have stated they have no conflict of interest.