Many studies have examined effects of prenatal stress on pregnancy and fetal development, especially on prematurity and birthweight, and more recently long-term effects on child behavioral and emotional development. These studies are reviewed and their limitations are discussed with regard to definitions (including the concepts of stress and anxiety), stress measurements, samples, and control for confounds such as depression. It appears necessary to assess individual stress reactivity prospectively and separately at each trimester of pregnancy, to discriminate chronic from acute stress, and to take into consideration moderator variables such as past life events, sociocultural factors, predictability, social support and coping strategies. Furthermore, it might be useful to examine simultaneously, during but also after pregnancy, stress, anxiety and depression in order to understand better their relationships and to evaluate their specific effects on pregnancy and child development. Finally, further research could benefit from an integrated psychological and biological approach studying together subjective perceived stress and objective physiological stress responses in pregnant women, and their effects on fetal and child development as well as on mother-infant interactions.