[Changes in vitamin A intake following the social marketing of red palm oil among children and women in Burkina Faso].

This paper focuses on changes in vitamin A (VA) intakes as part of the evaluation of a pilot project on social marketing of red palm oil (RPO) as a VA supplement for mothers and children in central-north Burkina Faso. The objectives of the 30-month project are to demonstrate the feasibility and effectiveness of introducing RPO in non-consuming areas. RPO is collected from women in the South-West region and it is sold in project sites by village volunteers. RPO is promoted by community workers trained in persuasive communication and social marketing. The target population is free to buy and consume RPO. Evaluation design includes data collected at onset, then 12 and 24 months later, from the same sample of 210 mothers and their children randomly selected in seven project sites. Children were 1 to 3 years old at onset. Blood samples were collected at baseline from mothers and children for serum retinol determination by HPLC. VA intakes are estimated by a semi-quantitative food frequency questionnaire, using the conventional beta-carotene to retinol conversion factors and the newly revised lower factors. VA deficiency is a major public health problem in the area: 64% of mothers and 85% of children had serum retinol concentrations < 0,70 mumol/l at baseline. VA came mainly from plant foods, particularly fruits and dark green vegetables which provided more than 90% of the dietary VA at onset of the project. Mean vitamin A intakes are low. We found 138 106 mug ER for the children and 302 +/- 235 microg ER for the mothers with conventional factors and 64 +/- 58 microg ER and 133 +/- 162 microg ER, respectively, with the revised factors. One year later, one third of respondents had consumed RPO in the previous week, and it supplied around 56% of the VA intake of children and 67% of mothers (36% and 46% respectively for the whole group). VA intakes were significantly increased at 510 +/- 493 microg ER and 801 +/- 913 microg ER for the children and their mothers respectively (347 +/- 443 microg ER and 568 +/- 803 microg ER respectively, with the revised factors). Analyzing serum retinol and dietary data collected at baseline, it was found that VA intakes < 62,5% of safe level of intake were highly sensitive to low serum retinol (< 0,70 micromol/l) and using revised conversion factors to assess total VA intake slightly enhanced sensitivity. The proportion of mothers and children at risk of inadequate VA intake changed from nearly 100% at baseline to 60% one year later. The results show that promoting RPO (and other VA rich foods) was effective in improving VA intakes. This improvement will hopefully be sustained and even further enhanced during the remaining 12 months of the project, after which repeated measurement of serum retinol and VA intakes will allow the actual impact of the project to be truly assessed.