Measuring dietary diversity in rural Burkina Faso: comparison of a 1-day and a 3-day dietary recall

Objectives: To compare dietary diversity scores measured over a 1-day and a 3-day period, and to assess their relationships with socioeconomic characteristics and the nutritional status of rural African women. Design: A qualitative dietary recall allowed calculation of a dietary diversity score (DDS; number of food groups consumed out of a total of nine). Body mass index (BMI) and body fat percentage (BFP) were used to assess the nutritional status of women. Setting and subjects: A representative sample of 550 mothers in northeast Burkina Faso. Results: The DDS increased from 3.5 to 4.4 when calculated from a 1-day or a 3-day recall (P , 0.0001), although for the latter the DDS was affected by memory bias. The DDS calculated from a 1-day recall was higher when a market day occurred during the recall period. Both scores were linked to the sociodemographic and economic characteristics of the women. Women in the lowest DDS tertile calculated from the 1-day recall had a mean BMI of 20.5 kg m 22 and 17.7% of them were underweight, versus 21.6 kg m 22 and 3.5% for those in the highest tertile (P ¼ 0.0003 and 0.0007, respectively). The DDS calculated from the 1-day recall was also linked to mean BFP; all these links remained significant after adjustment for confounders. For the 3-day period, no such relationships were found to be significant after adjustment. Conclusion: The DDS calculated from a 1-day dietary recall was sufficient to predict the women's nutritional status. In such a context attention should be paid to market days.