Measuring and validating food insecurity in Embo, using the food insecurity scale and index

Abstract

Measurement of household food insecurity is needed to identify the magnitude of food insecurity and assess the impact of development interventions. However, there is no commonly agreed measure of household food insecurity. While researchers continually experiment with new measures, the resultant measures are often complex and include numerous variables that still do not distinguish clearly between the food secure and the food insecure. This study set out to prepare a quick and convenient tool to measure household food security, using common household demographic and socio-economic variables commonly collected through a variety of household surveys. This has minimised data collection costs and assisted national food security units to continually measure and monitor household food insecurity. Food insecurity levels were estimated using data from a baseline survey conducted in a community in KwaZulu-Natal, South Africa. Food security was estimated using a number of measures, including food quantity (adequacy), dietary diversity, dietary quality, coping strategies employed and the Coping Strategies Index. The study found that household food availability varied across the two seasons over which data were collected. Only the percentile of sample households with adequate food intakes (one third of the samples) consumed enough food during the lean period when agricultural production was low. Households with inadequate food intakes also had consumed insufficient energy and lower micronutrient intakes during the period when agricultural produce was more abundant. Energy, iron and protein consumption was positively related to the consumption of adequate food. Energy intake was a relatively good indicator of protein and micronutrient intakes during the leaner period. Consumption of foods from three food groups, namely cereals, legumes and vegetables and fruits were necessary for adequate food intake. Cereals were the most important foods, forming the base of most meals, while fats and animal sourced foods were not widely consumed. Diversifying consumption through fruits and vegetables contributed significantly and positively to improved household food intakes. Household dietary diversity and dietary quality improved during the period of plenty. The application of coping strategies was strongly related to household food intake and diversity. Engaging in more coping strategies and having resultant higher Coping Strategy Index scores was strongly associated with household food inadequacy intakes and low food diversity scores. As expected, sampled households employed more coping strategies during the lean season. The strong and significant relationships between the Coping Strategies Index scores, the number of coping strategies practised by households and the household food intake indices (the Household Food Intake Index and Nutrient Adequacy Ratios) show that food intake is a strong indicator for household food security. The Household Food Insecurity Index and the Household Food Insecurity Scale were developed using 13 potential household demographic and socio-economic variables to identify the food-insecure households. The results of these two new measures were correlated with the results of the common measures reported above and found to be useful determinants of food security. The study found that while the Household Food Insecurity Index explained the influence of demographic and socio-economic variables in household food insecurity, the Household Food Insecurity Scale is more convenient in application (easy data management and computation process), and it is strongly related to the Coping Strategies Index scores. Both the Household Food Insecurity Scale and the Household Food Insecurity Index were useful tools to measure household food security and differentiate between food security and food insecure households in Embo Community. More research is recommended to further test the usefulness of the proposed measures in various settings.