Prospective assessment of pregnancy outcome: PATIENTS AND METHODS

Included in this study were all women who were prospectively counselled in the authors’ clinic about first trimester exposure to any BDZ between September 1986 and September 1991; each BDZ case was matched to a control temporally closest to the study case in our computerized database. Control cases sought counselling following exposure to nonterato-genic cheap drugs (agents known to be safe in human pregnancy). A team physician obtained information about maternal drug history, and medical, genetic, obstetric and occupational history in an in ter view at the hos pital. A trained in ter viewer as-cer tained preg nancy out come in a telephone in ter view with the mother, and the infant’s physician corroborated this by sending a written report to the clinic.

The primary outcome of interest was the rate of birth defects in the live births in each group. Secondary outcome measurements included rates of elective and spontaneous abortions, exposure to ethanol and tobacco, marital status, weight gain during pregnancy, use of forceps at delivery, presence ofmeconium, premature rupture ofthe membranes, gestational age, birth weight and attainment of developmental milestones. Data expressed as proportions were compared using contingency table analysis and unpaired continuous data using Student’s t test.