Vitamin A deficiency continues to be a public health problem in developing countries. Although overt clinical signs of deficiency are disappearing as industrialization progresses, a marginal vitamin A status is still prevalent;Because plasma retinol concentrations are homeostatically controlled throughout a wide range of liver reserves of vitamin A, serum retinol concentrations alone are not a good indicator of vitamin A status. Several methods have been developed, including the modified relative dose response (MRDR) assay, which diagnose the vitamin A status of both communities and individuals. The current work refines the use of the MRDR in assessing the vitamin A status of children and adults at the individual and community levels;The reproducibility of the MRDR in vitamin A sufficient individuals was assessed in 6 adults and 1 child at four different times during a 7 month period. The response remained normal at all times studied. In the same subjects, another indicator of marginal vitamin A status, conjunctival impression cytology (CIC), also gave normal impressions of goblet and epithelial cells;Next, the MRDR, relative dose response (RDR) and CIC assays were applied to two groups of children from different Indonesian villages. The MRDR and CIC tests identified the same group as being at an increased risk of vitamin A deficiency. The RDR gave results similar to the MRDR in the second group of children when a dose of 3.5 [mu]moles of retinyl acetate was used but not in the first group when only 1.57 [mu]moles was administered. The period between repeated tests was also longer in the second group of children;In addition, the MRDR assay was applied to three groups of Indonesian women: (1) lactating, (2) pregnant and (3) non-pregnant, non-lactating women. Only 5 to 7% of the variability in the 3,4-didehydroretinol to retinol (DR/R) ratio could be explained by body weight when a standard 8.8 [mu]mole dose of 3,4-didehydroretinyl acetate was administered. Seventy percent of the lactating women were in a marginal vitamin A status as judged by the MRDR test. At all times studied (3 to 6 h after dosing), the serum DR/R ratio was three-fold higher in the lactating women than in the non-pregnant, non-lactating women. Seventeen percent of the Indonesian pregnant women were in a marginal vitamin A status by use of the same indicator. Clearly, both pregnant and lactating Indonesian women in West Java are at risk of vitamin A deficiency. Appropriate intervention strategies should consequently be defined to alleviate the vitamin A deficiency problem.