This study has four objectives. The major goal is to determine if rates of depressive symptoms and depressive disorders are elevated in women who miscarry. Second, we will examine whether these rates vary with characteristics of the woman (age, ethnicity, previous spontaneous abortions, psychiatric history, etc.), characteristics of the pregnancy (e.g., length of gestation), social support, and stressful life events. Third, we will explore whether the clinical picture within 6 weeks following miscarriage predicts depressive symptoms or disorders at 6 months. Fourth, we will study the source of discrepancies in the classification of subjects as cases arising from the use of symptom scale scores vs. a diagnostic assessment. 450 women miscarrying at Columbia Presbyterian Medical Center over a 19 month period will be recruited. These women will be interviewed with instruments measuring depressive symptoms, attitudes towards pregnancy and preparation for the child, social support, and life events at 1 week, 6 weeks, and 6 months. The 6 month contact also includes a diagnostic interview. A sample of subjects classified inconsistently as """"""""probable psychiatric cases"""""""" using symptom scales but diagnosed as """"""""noncases"""""""", or in the reverse manner, will be recruited for a clinical resolution interview. Symptoms and depressive disorders identified in the miscarrying women will be compared with those in two other groups: pregnant women and women not recently pregnant. The pregnant comparison group provides an estimate of the level of depressive symptoms and disorder expected in miscarried women had they remained pregnant; the nonpregnant comparison group, an estimate of the level expected among nonpregnant women in the community. Both comparison groups will be similar to the miscarriage group for risk factors for depression, other than miscarriage, or such factors will be controlled in the analysis. This study is the first prospective controlled epidemiologic study of the psychological sequelae of miscarriage. Information on the nature of this negative event is obtained independently of subject recall and prior to its psychological aftermath. The study explores an important but neglected area of loss and bereavement and expands the purview of previous studies of postpartum psychiatric disorders to include early reproductive loss.