The heterogeneous association between education and the adoption of safe food handling practices in Ethiopia.

Abstract

Background: Foodborne disease is a great concern to low- and middle-income countries. To prevent illness and death, intervention strategies need to be implemented across the food safety system and should include promoting the adoption of safe food handling practices. The positive association between education and health has been well-established, and one possible mechanism is that education may improve health by encouraging individuals to adopt more appropriate protective practices. Decisions regarding adoption of these practices may also be influenced by the food safety risks individuals face, the trade-offs they make to maximize utility, or behavior biases which may be correlated with education. This study aims to estimate the heterogeneous association between education and the adoption of safe food handling practices among people facing different levels of food safety risk.

Methods: Models were constructed based on the Grossman health model and risk as well as behavior bias theories. Multivariate logistic regression models were estimated to explore the heterogeneous associations using data from a community survey conducted in Ethiopia. Agricultural household status and livestock presence were used as proxies to represent varying risk levels. Average marginal effects were estimated to provide a more accessible interpretation of the results.

Results: Results showed that the association between education and certain safe food handling practices was positive among individuals in households assumed to face higher food safety risks, while the association was less pronounced (or even negative) for those facing lower levels of risk. We observed that secondary education attainment was associated with a 20 percentage points increase (p < 0.01) in the probability of washing hands compared to the reference group (illiterate) in agricultural households. However, for non-agricultural households, secondary education was associated with a 10 percentage points decrease (p < 0.05) in probability. Similar patterns were found for washing surface.

Conclusions: Education is associated with increased adoption of safe food handling practices among individuals facing higher food safety risks. This has important implications for developing targeted policies focused on individuals most susceptible to foodborne diseases. Future policies aimed at increasing the adoption of safe food handling practices should also integrate individuals' decision-making processes and behavior biases in the context of varying risk levels.