Simple, cost-effective and convenient instruments like food group-based scores are proposed to assess micronutrient adequacy of children in developing countries. We assessed the predictive ability and seasonal stability of a dietary diversity score (DDS) to indicate dietary quality of infants.

A 24 h dietary recall assessment was carried out on a sample of 320 and 312 breast-fed infants aged 6–12 months during harvest (HS) and pre-harvest (PHS) seasons, respectively, in Ethiopia. DDS was calculated based on seven food groups, while mean micronutrient density adequacy (MMDA) was calculated for eight micronutrients. Multiple linear regression models were used to assess the relationship between DDS and MMDA, and differences in nutrient intake between the two seasons. A receiver-operating characteristic curve analysis was performed to derive DDS cut-offs that maximized sensitivity and specificity of assessing dietary quality.

The study was conducted in the catchment of the Gilgel Gibe Field Research Centre of Jimma University, south-west Ethiopia.

The mean (sd) DDS for HS and PHS was 2·1 (0·94) and 2·3 (1·1), respectively. The DDS was associated with MMDA (β=0·045, P<0·0001 in HS; β=0·044, P<0001 in PHS). A DDS of ≤2 food groups best predicted ‘low’ MMDA (<50 %) with 84 % and 92 % sensitivity, 36 % and 43 % specificity, and 47 % and 51 % correct classification for the HS and PHS, respectively.

DDS is predictive of dietary quality of breast-fed infants. The study supports the use of DDS to indicate inadequate intakes of micronutrients by breast-fed infants in different seasons.