href="#ulink_70448ef7-3795-5f3c-bc4b-adf27a844a01">5]. While more energy-dense diets are beneficial for people who suffer from hunger and insufficient energy intake, they foster overweight and obesity in others without necessarily contributing to more nutritious and sustainable diets.
The rapid spread and growth of supermarkets in LMICs, which is often referred to as the “supermarket revolution,” plays an important role in shaping food environments and consumers’ food choices and diets [6]. Supermarkets influence food prices, the types of foods offered, shopping atmosphere, and the way food procurement systems are organized [7]. While research on supermarkets and their associations with diets, nutrition, and health have mainly focused on high-income countries and on adults, evidence for African countries is scarce. Here, we review recent studies carried out in different LMICs. The main emphasis is on studies in Kenya, the only African country for which research on the nutrition effects of supermarkets is available. The studies in Kenya used panel data and statistical differencing techniques, thus also allowing robust causal inference about the effects of supermarket shopping on people’s diets, nutrition, and health. Furthermore, the studies in Kenya differentiated between the effects on adults and children, which is interesting as the impacts of supermarkets may not be uniform across age cohorts. After reviewing this research, we discuss possible policy implications and examples of interventions in the context of supermarkets in sub-Saharan Africa (SSA).
Background
Conversely to the growth of supermarkets in the USA and Europe (historically in the first two-thirds of the 20th century), the expansion of supermarkets in SSA is happening at a much faster pace and only started in the late 1990s [8]. Except for a few large supermarket stores in big cities, where fresh fruits and vegetables are available, the expansion of supermarkets in SSA is especially characterized by the increasing offer of processed, dry, and packed foods [6, 9].
While the transformation of agri-food systems can create opportunities for domestic agriculture and food industries, challenges and possibilities for consumers concerning food consumption, nutrition, and health exist [7, 10]. The way retail outlets and their businesses are organized changes food environments and consumers’ food choices. Pricing, advertising, positioning, and the availability of different products and brands directly shape food preferences and create desires. Due to higher quantities traded and more efficient supply chains, supermarkets can sell certain foods at lower prices. Furthermore, supermarkets offer a wider range of types and brands of foods than traditional retailers and a different shopping atmosphere. While over-the-counter purchases are common in traditional retailers, supermarkets are self-service outlets. Hence, shopping in supermarkets is more autonomous, encouraging spontaneous purchase behavior and thus also changing people’s food choices and diets [11]. Besides personal preferences, habitual and every-day shopping practices, consumers’ choices are also affected by changing lifestyles and society.
The implications of the supermarket revolution for nutrition and health in LMICs are not yet well understood. For adults, some studies reveal significant associations between supermarket purchases and changing diets and nutrition. Studies from Tunisia and Vietnam found improvements in dietary quality through supermarket shopping [12, 13]. Studies from Indonesia, Guatemala, and Kenya suggest an association with higher energy consumption and a shift towards more processed foods through supermarket shopping [9, 14, 15]. Furthermore, some studies show significant links between supermarket shopping, higher body mass index (BMI), and higher risk of overweight and obesity [16, 17].
While evidence of the associations between supermarkets and adult nutrition in LMICs is relatively limited, studies that have looked at the implications for child and adolescent nutrition are even scarcer. One study from Indonesia found an increased risk of child overweight through supermarket shopping, but only for children in high-income households [18]. Results from Kenya do not show significant effects of supermarket shopping on child overweight rates, but suggest a positive effect on child height [17]. Furthermore, a cross-country analysis, including data from 41 LMICs, revealed a significantly negative association between the country-level share of supermarket retailing and child stunting, without finding any significant association between supermarkets and child overweight [19].
This discussion shows that the effects of supermarket shopping on diets, nutrition, and health are not straightforward and differ between adults, children, and adolescents. Causal inference is also tricky because most of the studies discussed used cross-section observational data where endogeneity bias is difficult to control. In the following, we review additional evidence from Kenya, where recent studies have used panel data for more robust causal inferences.
Table 1. Effects of supermarket shopping on adults’ diets and nutritional status
Panel Data Evidence from Kenya
Kenya is one of the countries in SSA with the fastest supermarket growth rates during the last 20 years. Occurring mainly after 2002, Kenya became the country with the largest number of leading food retailers (a total of 10 in 2018) on the African continent [8]. While massive development is happening, Kenya currently suffers from a triple burden of malnutrition. While 26% of children aged under 5 years are stunted, more than 34% of female adults are overweight, and 27% of women suffer from anemia. Furthermore, about 6% of adults suffer from diabetes and over 26% from high blood pressure [20].
The following findings are based on panel data that were collected in medium-sized towns in central Kenya in 2012 and 2015. While some of the towns already had a supermarket at the time of data collection, others did not have a supermarket yet. This provides a quasi-experimental set-up, which allows comparing diets, nutrition, and health outcomes between people with different access to supermarkets and other types of retailers.
Supermarket Effects on Adults’ Diets and Nutrition
Table 1 summarizes findings by Demmler et al. [21] using the panel dataset on adults from Kenya (collected in 2012 and 2015) and panel regression models including fixed effects estimations (a form of statistical differencing), which allow for causal inference. The table depicts the effects of supermarket shopping on adults’ energy consumption from different food groups, the share of energy from highly processed foods, and the effects on nutritional outcomes. Supermarket shopping
