Identification and characterization of traditional diets from Burundi by multivariate analysis

Multivariate analyses were applied to food habits in Burundi in order to identify typical food combinations and to determine the extent to which these patterns are compatible with nutritional adequacy and diversity. Eight criteria were used in cluster analysis: food intake (g/pers/day), energy intake, protein, lipid, carbohydrates, calcium, iron and vitamin C intakes. Macronutrients expressed as a percentage of total energy intake were also taken into account in order to differentiate diets on a qualitative basis. Six clusters, each containing 5 to 12 meals, were identified as the most representative of Burundian diets in rural areas. These meal clusters were discriminated by their calcium, energy, protein and iron contents. In comparison with nutritional guidelines, most of them failed to meet the recommendations. The results obtained may be beneficial to nutrition educators and policy makers in targeting food programs and education efforts to improve diet adequacy.

Key words:

• cluster analysis
• food combination
• nutritional adequacy
• food diversity