The significance of uncertain gestation for obstetric outcome.

An analysis of 11 454 singleton births in a 'total obstetric population' (Aberdeen City District) from 1976 to 1980 showed that women whose gestation was assessed as approximate or uncertain had less favourable characteristics than the certain group. Uncertainty was significantly related to adverse pregnancy outcomes such as perinatal mortality, low birthweight, and spontaneous preterm delivery, and the association with the adverse outcomes was independent of unfavourable maternal characteristics. Moreover, uncertainty 'explained' adverse outcomes better than the traditional sociodemographic variables used in the analysis. It is argued that research findings based only on women of certain gestation cannot be generalized to total populations including women of uncertain gestation and will be subject to bias.