On the ability of birth defects monitoring to detect new teratogens.

Recent concerns have been raised about the ability of birth defects monitoring programs to detect increases in the incidence of birth defects following the introduction of new teratogens. The authors illustrate how most monitoring programs in the United States and Europe are limited in their ability to detect new teratogens because of a combination of parameters: the small population size, the low population frequency of exposure to the new teratogen, the weakness of many suspected teratogens (measured in terms of relative risk R), the low background rate, and the etiologic heterogeneity in the measured defects. In a system that monitors 25,000 births per year, it can be shown that although a new teratogen such as thalidomide (R = 175) can lead to a significant increase in the number of observed cases in 1-2 weeks of monitoring, even strong teratogens such as valproic acid and isotretinoin (R = 20-25) require more than 20 years of monitoring to show a significant increase in the number of cases because of low exposure frequency. Also, most mild to moderate teratogens (R = 2-5) can be totally missed. To improve the ability of birth defects monitoring programs to detect new teratogens, it is suggested that surveillance systems ought to examine subsegments of the population with maximal exposure potential, classify birth defects into more etiologically homogeneous groups, and expand the sample size of the monitored population.