Background
To determine the dietary patterns, anthropometric status, prevalence and risk factors for anaemia among school children aged 5–11 years in a peri-urban area of Central Uganda.

Methods
A cross-sectional study involving both qualitative and quantitative methods was used. Anthropometric data were taken using standardised equipment, whereas haemoglobin was assessed using a haemoglobin meter. Food intake was assessed using a food frequency questionnaire. One hundred and twenty-two primary school children, aged 5–11 years, and their carers were recruited in the study.

Results
The proportion of anaemic children (haemoglobin <12 g/dl) was 37.7%; 36.9% of these had mild and 0.8% had moderate anaemia. The unadjusted odds ratio (OR) showed that children who never consumed fish had a nine-fold increased odds of being anaemic [OR = 9.0, 95% confidence interval (CI) = 1.6–50.7; P = 0.018], whereas those who had fewer meals (1–2 per day) had a 27.0% increased risk (OR = 3.7, 95% CI = 1.1–12.6; P = 0.021). The adjusted OR showed number of meals per day as the only independent predictor of anaemia (OR = 3.5, 95% CI = 1.1–11.8; P = 0.031). The proportion of malnutrition (<-2 SD) for wasting (body mass index for age) was 3.3%, being underweight was 5.8% and stunting was 6.6%. Children aged >8 years were associated with wasting (P = 0.041) and stunting (P = 0.034). One main dietary pattern was identified explaining approximately 20.4% of the variability of intake in the population. However, scores of this pattern were not significantly associated with child haemoglobin levels.

Conclusions
Anaemia but not macronutrient malnutrition in this cohort of school children is high. Patterns of the dietary intake observed did not explain nutritional status in this population.