On June 28, 2007 the Seattle Times published an article about use of antidepressants by women who are pregnant. The article downplayed the risks to a developing fetus attributable to antidepressant use that have been well-documented in several previous studies.
The Times’ take-home message was that the U.S. Centers for Disease Control and Prevention believes study results are “generally reassuring” it is safe for pregnant women to continue taking drugs such as GlaxoSmithKline’s PAXIL. On the same day, June 28, 2007, The Times also published an article concerning exposure to secondhand smoke for pregnant women and the alleged risks that such exposure creates for a developing a fetus. That University of Washington study included data that purportedly demonstrates nicotine ingested through exposure to Environmental Tobacco Smoke (ETS) is strongly associated with later behavior problems in children and recommends further study on the subject.
The problems with the two news reports by the Seattle Times are manifold. Most glaringly apparent is the fact that many of the birth defects attributed to ETS by tobacco control advocates and in part by the University of Washington study are also attributed by other researchers to use of anti-depressants during pregnancy. Despite this confounding factor, the University of Washington study authors report no controls as to whether subjects were using prescribed antidepressants during their pregnancy. The University of Washington study authors also report no controls for use NRT products by study participants.
We observe the curious phenomenon of University of Washington study authors inventing analysis criteria and calculation variables concerning ETS to “prove” an association that does not plausibly exist, while ignoring well-documented and highly relevant confounding factors.