At an interdisciplinary workshop on home monitoring of uterine activity, participants reviewed the basis of this technology and its use in the identification of women at high risk of preterm delivery and in the prevention of preterm birth. Although the guard-ring devices in current use appear capable of detecting uterine activity, they do not clearly distinguish between Braxton Hicks contractions and the contractions of early labor. There was agreement that women destined for preterm delivery have more uterine activity on average than do other women of comparable gestational age but that there is substantial overlap between the two groups. Thus it is uncertain whether this difference can be used effectively for screening purposes. Conferees agreed that there is considerable evidence that twice-daily monitoring of very-high-risk women, in conjunction with daily nursing support and high-quality obstetric care, may prevent preterm births. Available evidence does not clearly distinguish the contribution of tocodynamometry from that of nursing support. A number of areas were identified in which further research is needed.