Contribution of major food companies and their products to household dietary sodium purchases in Australia

Abstract

Background: The Australian federal government will soon release voluntary sodium reduction targets for 30 packaged food categories through the Healthy Food Partnership. Previous assessments of voluntary targets show variable industry engagement, and little is known about the extent that major food companies and their products contribute to dietary sodium purchases among Australian households. Methods: The aim of this cross-sectional study was to identify the relative contribution that food companies and their products made to Australian household sodium purchases in 2018, and to examine differences in sodium purchases by household income level. We used 1 year of grocery purchase data from a nationally representative consumer panel of Australian households who reported their grocery purchases (the Nielsen Homescan panel), combined with database that contains product-specific sodium content for packaged foods and beverages (FoodSwitch). The top food companies and food categories were ranked according to their contribution to household sodium purchases. Differences in per capita sodium purchases by income levels were assessed by 1factor ANOVA. All analyses were modelled to the Australian population in 2018 using sample weights. Results: Sodium data were available from 7188 households who purchased 26,728 unique products and purchased just under 7.5 million food product units. Out of 1329 food companies, the top 10 accounted for 35% of unique products and contributed to 58% of all sodium purchased from packaged foods and beverages. The top three companies were grocery food retailers each contributing 12-15% of sodium purchases from sales of their private label products, particularly processed meat, cheese and bread. Out of the 67 food categories, the top 10 accounted for 73% of sodium purchased, particularly driven by purchases of processed meat (14%), bread (12%) and sauces (11%). Low-income Australian households purchased significantly more sodium from packaged products than highincome households per capita (452 mg/d, 95%CI: 363-540 mg/d, P < 0.001). Conclusions: A small number of food companies and food categories account for most of the dietary sodium purchased by Australian households. Prioritizing government engagement with these groups could deliver a large reduction in population sodium intake.

The estimated health impact of sodium reduction through food reformulation in Australia: A modeling study

“…The datasets used are described in detail in Table 1 , and the detailed modeling steps are in Text A in S1 File . In brief, the modeling approach involved 4 stages ( Fig 1 ) , (1) estimating pre-reformulation sodium intakes from each packaged food group with a sodium target using nationally representative 24-hour dietary recall data of the Australian population (National Nutrition and Physical Activity Survey (NNPAS) and Food Standards Australia New Zealand Food Composition Database) for each age–sex group [ 15 ], adjusted for underreporting using sex-specific 24-hour urinary sodium excretion estimates of nondiscretionary sodium intake (Table A in S1 File ) [ 10 ]; (2) using a nationally representative household grocery shopping panel data (NielsenIQ Homescan [ 16 ]) and a brand-specific food composition database (FoodSwitch) [ 17 ] to estimate the pre- and post-reformulation sales-weighted average sodium content, and the expected percentage reduction in sales-weighted average sodium content for each targeted food group; (3) estimating the reduction in sodium intake from each individual targeted food group or company based on steps (1) and (2), i.e., multiply prereformulation sodium intake in each food group by the estimated percentage reduction in sales-weighted average sodium content following reformulation (reductions are then summed across food groups to estimate overall reductions in sodium intake); and (4) estimating the effect of reduced sodium intake on CVD, CKD, and stomach cancer outcomes via comparative risk assessment analysis for each age–sex group. Table B in S1 File outlines the model assumptions and restrictions.…”

“…Based on data usage agreements with NielsenIQ, for the purposes of this analysis, food companies were deidentified; however, we classified whether a company was a “retailer” or a “manufacturer.” Manufacturers were classified as food companies that manufacture and distribute items (also known as “branded products”) for general trade. Retailers were defined as supermarkets that sell their own “private-label” products in their own stores [ 16 ], and the estimates for retailers were based on reformulation of their “private-label” products.…”

Estimating the potential impact of Australia’s reformulation programme on households’ sodium purchases

“…Using the categorisation system developed by the Global Food Monitoring Group, foods and beverages in FoodSwitch were classified into a hierarchical category tree to allow for comparison of nutritionally similar foods. [27][28][29] *Food categories and subcategories listed are those that are targeted as part of the Australian sodium reformulation programme. †Results are sales-weighted and projected to the Australian population using sample weights provided by Nielsen.…”

Deconstructing the Supermarket: Systematic Ingredient Disaggregation and the Association between Ingredient Usage and Product Health Indicators for 24,229 Australian Foods and Beverages

“…The 2019 FoodSwitch Annual Database, the dataset used in this study, was generated from in-store surveys carried out in Australia at five large supermarkets (IGA, ALDI, Woolworths, Coles and Harris Farm) between August and November 2019 [18]. The dataset represents the majority of supermarket food and beverage products purchased by Australian households [19,20].…”

“…The median (interquartile range (IQR)) number of ingredients across all products was 8 (3-14) (Table 3). 'Convenience foods' had the greatest median number of ingredients per product (20 (14-27)), followed by 'Foods for specific dietary use' (12 (7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)) and 'Bread and bakery products' (13 (9-19)) (Table 3).…”