Objective: To evaluate the long-term effect on micronutrient status of a b-carotene-, iron-and iodine-fortified biscuit given to primary school children as school feeding. Design: Children receiving the fortified biscuit were followed in a longitudinal study for 2.5 years n 108; in addition, cross-sectional data from three subsequent surveys conducted in the same school are reported. Setting: A rural community in KwaZulu-Natal, South Africa. Subjects: Children aged 6±11 years attending the primary school where the biscuit was distributed.Results: There was a significant improvement in serum retinol, serum ferritin, haemoglobin, transferrin saturation and urinary iodine during the first 12 months of the biscuit intervention. However, when the school reopened after the summer holidays, all variables, except urinary iodine, returned to pre-intervention levels. Serum retinol increased again during the next 9 months, but was significantly lower in a subsequent cross-sectional survey carried out directly after the summer holidays; this pattern was repeated in two further cross-sectional surveys. Haemoglobin gradually deteriorated at each subsequent assessment, as did serum ferritin (apart from a slight increase at the 42-month assessment at the end of the school year). Conclusions: This study has shown that fortification of a biscuit with b-carotene at a level of 50% of the Recommended Dietary Allowance (RDA) was enough to maintain serum retinol concentrations from day to day, but not enough to sustain levels during the long school holiday break. Other long-term solutions, such as local food production programmes combined with nutrition education, should also be examined. The choice of the iron compound used as fortificant in the biscuit needs further investigation.

Objective: There is a paucity of data on the micronutrient status of low-income, lactating South African women and their infants under 6 months of age. The aim of this study was to elucidate the level of anaemia and vitamin A deficiency (VAD) in peri-urban breast-feeding women and their young infants. Design: Cross-sectional study including anthropometric, biochemical and infant feeding data. Setting: Peri-urban settlement in Cape Town, South Africa. Subjects: Breast-feeding women (n ¼ 113) and their infants (aged 1-6 months) attending a peri-urban clinic. Results: Mean (standard deviation (SD)) haemoglobin (Hb) of the lactating mothers was 12.4 (1.3) g dl 21, with 32% found to be anaemic (Hb , 12 g dl 21 ). Maternal serum retinol was 49.8 (SD 13.3) mg dl 21 , with 4.5% VAD. Using breast milk, mean (SD) retinol concentration was found to be 70.6 (24.6) mg dl 21 and 15.7 (8.3) mg/g milk fat, with 13% below the cut-off level of , 8 mg/g fat. There was no correlation found between breast milk retinol and infant serum retinol. Z-scores (SD) of height-for-age, weight-for-age and weight-for-height were 2 0.69 (0.81), 0.89 (1.01) and 1.78 (0.83), respectively. Mean (SD) infant Hb was 10.9 (1.1) g dl 21, with the prevalence of anaemia being 50%, 33% and 12% using Hb cut-offs below 11 g dl 21 , 10.5 g dl 21 and 9.5 g dl 21 , respectively. Mean (SD) infant serum retinol was 26.9 (7.2) mg dl 21 , with 10% being VAD. None of the infants was exclusively breast-fed, 22% were predominantly breast-fed and 78% received complementary (mixed) breast-feeding. Thirty-two per cent of infants received weaning foods at an exceptionally young age (# 1 month old). Conclusion: A high rate of anaemia is present in lactating women residing in resourcepoor settings. Moreover, their seemingly healthy infants under 6 months of age are at an elevated risk of developing early-onset anaemia and at lower risk of VAD.

Objective: To assess primary health care (PHC) facility infrastructure and services, and the nutritional status of 0 to 71-month-old children and their caregivers attending PHC facilities in the Eastern Cape (EC) and KwaZulu-Natal (KZN) provinces in South Africa.Design: Cross-sectional survey. Setting: Rural districts in the EC (OR Tambo and Alfred Nzo) and KZN (Umkhanyakude and Zululand).Subjects: PHC facilities and nurses (EC: n = 20; KZN: n = 20), and 0 to 71-month-old children and their caregivers (EC: n = 994; KZN: n = 992).Methods: Structured interviewer-administered questionnaires and anthropometric survey.