This review addresses the question of whether a short birth interval is associated with adverse nutritional outcomes for the mother or the child. Indices of anthropometric status (maternal weight or body mass index; child growth) and micronutrient status (e.g. iron or vitamin A) were included as outcomes. A computerized search of all relevant papers published since 1966 was completed, and the 'snowball' method was used to identify additional relevant published or unpublished papers. In total, 57 papers were found to contain data regarding the relationship between birth spacing and nutritional outcomes (35 for child nutrition, 11 for maternal anthropometric status, and 11 for maternal anaemia or micronutrient status). Of these, 23 papers were excluded from further consideration because they did not include any multivariate analysis, leaving 34 papers that met the criteria for the review (22 for child nutrition, eight for maternal anthropometric status, and four for maternal anaemia or micronutrient status). The studies on child nutrition outcomes indicate that a longer birth interval is associated with a lower risk of malnutrition in some populations, but not all. In those countries in which the relationship was significant, the reduction in stunting associated with a previous birth interval >or=36 months ranged from approximately 10% to 50%. Some of this reduction may be due to residual confounding, i.e. to factors not included in the analysis (such as breastfeeding and maternal height). The studies on maternal anthropometric outcomes yielded mixed results. Because the nutritional burden on the mother between pregnancies depends on the extent of breastfeeding, the interpregnancy interval is not the best measure of whether the mother has had a chance to recover from the pregnancy, in terms of repleting her nutritional status. Therefore, some studies examined the 'recuperative interval' (duration of the non-pregnant, non-lactating interval) instead. Taken as a whole, the studies do not provide clear evidence of a link between interpregnancy or recuperative interval and maternal anthropometric status. This may be due, in part, to changes in the hormonal regulation of nutrient partitioning between the mother and the fetus when a mother is malnourished. Only four papers were identified that related to micronutrient status, three of which examined maternal anaemia. One study showed an increased risk for maternal anaemia when the interpregnancy interval was <6 months, but the analysis did not control for iron supplementation during pregnancy. The other two studies did not show a significant association between interpregnancy interval and maternal anaemia. One study of micronutrient status indicated no significant relationship between interpregnancy interval and maternal serum zinc, copper, magnesium, ferritin, folate or thyroid-stimulating hormone. Important methodological limitations were apparent in most of the studies. Thus, further research with more comprehensive control of potentially confounding variables is needed.