A new approach to identifying food deserts using grocery store purchase data suggests that store proximity is not the driver of nutritionally deficient diets – it is financial and social inequality.

In a study led by the University of Adelaide, researchers found nutritional disadvantage is concentrated in low-income and minority communities, even in areas where stores are nearby.

“Food deserts refer to areas where residents are unable to access a nutritious diet, where barriers to obtaining healthy foods are thought to underpin dietary behaviour,” says Tayla Broadbridge, from the University’s School of Computer and Mathematical Sciences, who led the study.

“Previous attempts to identify food deserts have relied on assumptions about the relationships between store locations, sociodemographic factors, and access to healthy food.

“These methods typically classify areas as food deserts without any direct, quantitative link to food purchase data or dietary patterns.

“Treating food access as only a store distribution problem ignores critical factors such as cultural and economic landscapes that shape residents’ urban life and mobility, and, consequently, their shopping behaviour.

“Our study demonstrates that analysing the relationship between food purchasing patterns and sociodemographic factors can identify food deserts that aren’t picked up by other methods, as well as the factors that drive them.”

Tayla, who also holds a position in the School of Mathematics at the University of Nottingham, United Kingdom, completed her study by using food purchase records from London locations of the British multinational grocery chain, Tesco.

“We analysed supermarket-transaction data from 1.6 million London customers and identified neighbourhoods where residents’ purchases follow a nutrient-deficient pattern, highlighting areas that are characteristic of food deserts,” says Tayla.

“Our model identified several areas in London’s east, such as Newham, Barking, and Dagenham, and some areas of London’s northwest, such as Ealing and Brent, which adhered to high-sugar and high-carbohydrate purchases, suggesting residents in these areas are not accessing a nutritionally adequate diet.”

Though Tayla’s model identified that nutrient deficient purchasing is associated with household income and minority ethnic communities, the influence of these factors varies across the city.

“These spatial differences in associations may reflect a combination of socioeconomic conditions, local food retail environments, or varying food preference,” says Tayla, whose study was published in PLOS Complex Systems.

“The strong variation in these relationships across London’s neighbourhoods highlights the need for locally tailored, culturally sensitive strategies to improve access to healthy food.

“Further work with local experts is needed to better understand these context-specific relationships and to develop area-specific, context-sensitive interventions as part of local public health strategies.”

While this study focused on Tesco purchases in London, Tayla says it could be replicated in other contexts, including Australia.

“This study moves beyond the reliance on sociodemographic and environment characteristics alone when identifying food deserts,” Tayla says.

“Applied in Australia, our method could use loyalty card data from major grocery retailers such as Woolworths or Coles to show where nutritional disadvantage is most concentrated, and how it relates to local sociodemographic factors.

“This will help local councils and communities target support where it’s needed most.”