Bias against the null hypothesis: the reproductive hazards of cocaine.

To examine whether studies showing no adverse effects of cocaine in pregnancy have a different likelihood of being accepted for presentation by a large scientific meeting, all abstracts submitted to the Society of Pediatric Research between 1980 and 1989 were analysed. There were 58 abstracts on fetal outcome after gestational exposure to cocaine. Of the 9 negative abstracts (showing no adverse effect) only 1 (11%) was accepted, whereas 28 of the 49 positive abstracts were accepted (57%). This difference was significant. Negative studies tended to verify cocaine use more often and to have more cocaine and control cases. Of the 8 rejected negative studies and the 21 rejected positive studies, significantly more negative studies verified cocaine use, and predominantly reported cocaine use rather than use of other drugs. This bias against the null hypothesis may lead to distorted estimation of the teratogenic risk of cocaine and thus cause women to terminate their pregnancy unjustifiably.