The 1988 National Maternal and Infant Health Survey (NMIHS) was conducted by the National Center for Health Statistics to study factors related to poor pregnancy outcome, such as adequacy of prenatal care; inadequate and excessive weight gain during pregnancy; maternal smoking, drinking, and drug use; and pregnancy and delivery complications. The NMIHS is a nationally representative sample of 11,000 women who had live births, 4,000 who had late fetal deaths, and 6,000 who had infant deaths in 1988. Questionnaires were mailed to mothers based on information from certificates of live birth, reports of fetal death, and certificates of infant death. Information supplied by the mother, prenatal care providers, and hospitals of delivery was linked with the vital records to expand knowledge of maternal and infant health in the United States. The response rates in all three components of the NMIHS differed according to the mothers' characteristics. Mothers were more likely to respond if they were 20-39 years of age, were white, were married, had fewer than four children, entered prenatal care early, had more prenatal visits, had more years of education, or resided in the Midwest Region. The percent of respondents was lower for teenage mothers, mothers of races other than white, and mothers with four or more children, little prenatal care, or fewer years of education. Mothers whose infants weighed less than 2,500 grams were less likely to respond in the live-birth and infant-death components than mothers whose infants weighed 2,500 grams or more. The NMIHS will provide an invaluable tool for researchers and practitioners seeking solutions to perinatal and obstetric problems.