Food is made up of complex and multidimensional elements whose implementation requires significant measurement tools. The overall objective of this work is to assess food consumption within sedentary and pastoral households. This is a cross-sectional study. The methodology used is a food survey of 845 households. The result of this study is that the food groups most stocked by both urban and rural households are cereals (urban 95.7%, rural 79% and pastoral 94.4%) followed by spices/condiments (urban 74.3%, rural 67.5% and 74.4%). The least stocked food groups in urban areas are fruits and vegetables, while the availability of meats in stored speculation is almost non-existent in rural and pastoral areas. The average household food consumption score is 68.10±21.04; 43.29±13.27 and 44.96±13.77 respectively in urban, rural and pastoral areas. We also note a lack of diversity in the diet of rural and pastoral households. Indeed, the average household dietary diversity score is 6.84±1.91; 5.53±1.49 and 5.13±1.29 respectively in urban, rural and pastoral areas. Dietary diversity as well as the average consumption score are associated with household living conditions. In short, the assessment of consumption is an important element in characterizing the household diet.