[The ice storm: an opportunity to study the effects of prenatal stress on the baby and the mother.].

Over the last several years, studies have shown that stressful experiences during the pregnancy can predict levels of neurological development, as well as cognitive and psychological functioning, during childhood and adulthood. For example, Mednick (1997) has been studying the effects of a major earthquake in China on the psychological and intellectual development of the unborn child. Twenty-three years after the quake, significant differences have been found between the earthquake group and a control group born one year later in terms of intellectual functioning, depression, and the size of certain brain regions. Less severe events, such as a divorce or job loss during the pregnancy, may also increase the risk of obstetric complications and may have an effect on the baby's neurological well being, weight and head circumference at birth. Death of the baby's father during the pregnancy and natural disasters have both been associated with increased rates of depression, schizophrenia and criminality in adulthood. Several of these same effects have been found in studies of prenatal stress in non-human primates. Many of these studies suggest that the second trimester of pregnancy is a particularly critical period during which stressful events may compromise development of the fetus. Methodological constraints limit research on prenatal maternal stress. Animal studies are able to control for pre- and postnatal environments. However, animal studies have limited generalizability to humans for whom numerous risk and protective factors are in operation. Studies of human pregnancies cannot randomly assign subjects to stress conditions. Maternal personality and temperament may be associated with characteristics of a woman's child not only through genetic transmission of personality, but possibly also through differential exposure to difficult life conditions which may, in part, be self-imposed. In addition, studies of prenatal life events in humans have severely restricted variance; very large samples of women must be screened to identify even small numbers of women who have experienced major life events during the pregnancy. Finally, follow-back studies which show an association between prenatal events and later rates of mental illness, do not include timely evaluations of actual rates of exposure to the event, nor the pregnant woman's subjective or biological reactions to the event. In this paper, we present a review of the literature on prenatal maternal stress followed by a discussion of how the January 1998 Québec ice storm could be used to study the mechanisms by which prenatal stress may influence mental health outcomes in the unborn child.