There is a general perception that medicinal drugs are not safe in pregnancy despite the fact that fewer than 30 drugs have been shown to cause major malformations in humans. A large number of women need medications in pregnancy to treat pregnancy-induced conditions, acute illnesses, and chronic diseases. The objectives of this study were the following: (1) to characterize the perception of teratogenic risk by pregnant women and their partners and by health professionals and (2) to examine the most reassuring way to present data on a drug for nausea and vomiting of pregnancy that has been proven to be safe to the fetus. A convenience sample of pregnant Canadian women and their partners, pharmacists, nurses, physicians, and hospital workers were asked to choose the "safest" among four drugs by statements describing their safety. Although the text of all four was similar, the title and narrative were modified to be more or less "reassuring" by the use of more or less terms such as malformations and abnormalities. Health professionals rated the teratogenic risk significantly lower than the parents, but even they rated the drugs as not safe, despite a scientifically reassuring text. Sixty percent of the 240 participants, regardless of their perception of teratogenic risk, believed the four drugs were of similar risks. However, in the other 40%, the less "reassuring" text led to higher teratogenic perception, and the more reassuring options tended to decrease the false perception of teratogenic risk. It was concluded that in general, four different versions of reassuring text describing a scientifically proven safe drug in pregnancy did not lead expecting parents to believe they were safe. Among those who did not rank the four drugs as having equal safety/risk, the less "reassuring" text led to a higher perception of teratogenic risk. Even health professionals reading the labels describing safe drugs rated them as unsafe. Presently, the perception of teratogenic risk is strong even for safe drugs and is difficult to change even with evidence-based facts.