Simon Chege Kimenju

Do supermarkets contribute to the obesity pandemic in developing countries

OBJECTIVE Many developing countries are undergoing a nutrition transition with rising rates of overweight and obesity. This nutrition transition coincides with a rapid expansion of supermarkets. The objective of the present research is to study whether supermarkets directly contribute to overweight and other changes in nutritional status. DESIGN This research builds on cross-sectional observational data. Household- and individual-level data were collected in Kenya using a quasi-experimental survey design. Instrumental variable regressions were employed to analyse the impact of supermarket purchase on nutritional status. Causal chain models were estimated to examine pathways through which supermarkets affect nutrition. SETTING Small towns in Central Province of Kenya with and without supermarkets. SUBJECTS A total of 615 adults and 216 children and adolescents. RESULTS Controlling for other factors, buying in a supermarket is associated with a significantly higher BMI (P=0·018) and a higher probability of overweight (P=0·057) among adults. This effect is not observed for children and adolescents. Instead, buying in a supermarket seems to reduce child undernutrition measured by height-for-age Z-score (P=0·017). Impacts of supermarkets depend on many factors including peoples initial nutritional status. For both adults and children, the nutrition effects occur through higher food energy consumption and changes in dietary composition. CONCLUSIONS Supermarkets and their food sales strategies contribute to changing food consumption habits and nutritional outcomes. Yet the types of outcomes differ by age cohort and initial nutritional status. Simple conclusions on whether supermarkets are good or bad for nutrition and public health are not justified.

Consumer Preferences for Maize Products in Urban Kenya

Background New maize varieties have been biofortified with provitamin A, mainly β-carotene, which renders the grain yellow or orange. Unfortunately, many African consumers prefer white maize. The maize consumption patterns in Africa are, however, not known. Objective To determine which maize products African consumers prefer to purchase and which maize preparations they prefer to eat. Methods A survey of 600 consumers was conducted in Nairobi, Kenya, at three types of maize outlets: posho mills (small hammer mills), kiosks, and supermarkets. Results Clients of posho mills had lower incomes and less education than those of kiosks and supermarkets. The preferred maize product of the posho-mill clients was artisanal maize meal; the preferred product of the others was industrial maize meal. Maize is the preferred staple for lunch and dinner, eaten as a stiff porridge (ugali), followed by boiled maize and beans (githeri), regardless of socioeconomic background. For breakfast, only half the consumers prefer maize, mostly as a soft porridge (uji). This proportion is higher in low-income groups. Consumers show a strong preference for white maize over yellow, mostly for its organoleptic characteristics, and show less interest in biofortified maize. Conclusions Maize is the major food staple in Nairobi, mostly eaten in a few distinct preparations. For biofortified yellow maize to be accepted, a strong public awareness campaign to inform consumers is needed, based on a sensory evaluation and the mass media, in particular on radio in the local language.

The nutrition transition and indicators of child malnutrition

This study examines how the nutrition transition, which involves a westernization of diets and increased consumption of calorie-dense, processed foods, affects child malnutrition in developing countries. It is often assumed that the nutrition transition affects child weight but not child growth, which could be one reason why child underweight decreases faster than child stunting. But these effects have hardly been analyzed empirically. Cross-country panel regressions show that the nutrition transition reduces child underweight, while no consistent effect on child overweight is found. Against common views, our results also suggest that the nutrition transition reduces child stunting. Further research is required to confirm these findings.