How accurate is a woman's diagnosis of threatened preterm delivery?

During a 1-year period all women entering hospital with signs or symptoms that could herald preterm delivery were studied. The primary complaints of these 178 women were contractions in 70, watery discharge in 59, bloody discharge in 33, and abdominal pain in 16. These complaints were considered as confirmed threats of preterm delivery if the woman was delivered or received tocolysis within 48 h (n = 63; 35%); they were labelled as an unconfirmed threat if the woman was allowed to return home without further action within 48 h (n = 45; 25%). Overall, 75% of the women were correct in their estimation that something was wrong in that they either gave birth or were admitted for greater than 48 h for diagnostic or therapeutic measures. Parity had only a small influence on women's predictive ability. It is concluded that efforts to increase women's awareness of the significance of various symptoms are not likely to contribute much to improvement in the early diagnosis of preterm labour. Greater effect can be anticipated from the development of methods to differentiate between contractions that will or will not lead to early delivery.