Micronutrient malnutrition is a challenge for women of reproductive age, who are particularly vulnerable due to greater micronutrient needs. The minimum dietary diversity for women (MDD‐W) indicator is a micronutrient adequacy's proxy for those women, but little is known about its relation to other dimensions. We assessed MDD‐W and its association with other socioeconomic, food security and purchasing practices in urban Burkina Faso. We conducted multi‐stage cluster sampling in two main cities of Burkina Faso, stratified by type of district, and interviewed 12 754 women in the 2009‐2011 period. We obtained food consumption data through unquantified 24 hour recalls and computed MDD‐W as consuming at least five out of ten predefined food groups. We constructed multivariable regression models with sociodemographic and food security covariates. MDD‐W in urban Burkina Faso was 31%, higher in Ouagadougou (33%) than in Bobo‐Dioulasso (29%), and lower in unstructured districts. The most frequently consumed food groups were ‘all starchy', ‘vitamin A rich dark green leafy vegetables' and ‘other vegetables'. Household's expenses were associated with higher likelihood of MDD‐W, while the association with household food security indicators varied by year and type of district. Purchasing foods in markets and choosing the place of purchase based on large choice rather than proximity showed a positive association with the MDD‐W. Only one in three women in urban Burkina Faso reached the minimum dietary diversity, and although socioeconomic and food security variables had the greatest effect on MDD‐W, purchasing practices, like going to the market, also showed a positive effect.

This paper studies the determinants of poverty dynamics in several rural areas from Madagascar. A particular attention is devoted to testing if rural poverty persistence in Madagascar could be explained by a vicious circle leading to a poverty trap. Annual poverty transitions retrieved from an original household panel data survey covering the 1996-2006 periods show that differences in household and environment characteristics are clearly associated with di¤erences in poverty transitions probabilities. Poverty-vulnerable households have higher dependency ratio and are less educated. They also show different income-generating activities pattern. They cultivate less market-oriented crops and have more vulnerable activities such as agricultural wage work. A Markovian poverty transition model is used to evaluate the role of past poverty in this situation. Past poverty is allowed to have both an intercept and a slope effect on poverty transitions probabilities. Our results show that a substantial share of the di¤erences in household poverty transitions probabilities is attributable to past poverty status. These results encourage the development of social protection to prevent households to fall into a poverty trap as well as speciffic measures to encourage paths out of poverty traps.

No abstract

No abstract