The effect of a biscuit with red palm oil as a source of beta-carotene was compared with the effect of a biscuit with beta-carotene from a synthetic source on the vitamin A status of primary school children in a randomised controlled trial. Children aged 5-11 years (n = 265) were randomly assigned to one of three groups: (1) placebo biscuit; (2) biscuit with synthetic beta-carotene as a vitamin A fortificant; and (3) biscuit with red palm oil as a source of beta-carotene. The two non-placebo biscuits were designed to provide 34% of the RDA for vitamin A per serving (4 x 15 g biscuits). The biscuits were distributed daily during the school week and compliance was closely monitored and recorded. Children were assessed at baseline and after 6 months of intervention. Mean serum retinol in all three groups increased significantly compared to baseline (P < 0.0001). The prevalence of low serum retinol levels (< 20 micrograms/dL) dropped from 50 to 24.4% in the control group, from 48.2 to 16.9% in the synthetic beta-carotene group, and from 50.6 to 22.8% in the red palm oil group. There was no significant treatment effect compared to the control group in either the synthetic beta-carotene or red palm oil group. The increase in the control group was probably due to a school feeding scheme (providing 33% of the RDA for vitamin A) introduced during the latter part of the study. Our results were thus confounded and the 'true' effect of the red palm oil biscuit on vitamin A status could not be established. The study has, however, shown that red palm oil can be incorporated in a biscuit and that the end product with regard to taste and appearance was well accepted by the school children. A follow-up study in a school where there is no school feeding is indicated.