Records, recall loss, and recall bias in pregnancy: a comparison of interview and medical records data of pregnant and postnatal women.

To determine the agreement between interview-based and medical records data concerning illnesses and medications early in pregnancy, we compared the interviews of 202 women with the ongoing records collected during their pregnancies. Substantial underrecording of most transient illnesses was found. Fewer short-term illnesses were reported by postpartum women than sill-pregnant women, suggesting the potential for recall bias or loss.