« April 2007 | Main | July 2008 »
May 21, 2007
When is a "study" a study?
A report on phthalates and their possible health effects recently emerged from the University of Maryland. It was requested by the state legislature to support his bill calling for banning phthalates from vinyl toys. The report was written by two university faculty affiliated with pressure groups that are actively supporting anti-phthalates legislation. Not surprisingly, the picture that emerged of phthalates was highly negative. Media coverage played it as a "study." One news report referred to "new research." Worse, some politicians pushing for phthalate bans have called it a study that counters the risk assessments done in Europe and America — assessments that objectively reviewed many hundreds of studies. These risk assessments have generally concluded that phthalates can continue to be safely used in their current applications in consumer products.
The moral of this blog is "check it out." Be careful, journalists and Internet junkies, about exactly what it is you are reading or quoting. There are reports; there are comprehensive literature reviews, or risk assessments; there is real scientific research. They are not the same. Only real research should be called a "study." The Maryland document is not a "study." It contains no new research. It is a report, or review.
And just what do reviews review? They review the scientific studies. And not all reviews are equal, by any means. The Maryland report was completed in a few months and has 103 references. That may sound like a lot, but it isn't. The European Union's risk assessment of just one phthalate, DEHP, took many years and contains 971 citations of scientific research. In other words, the EU scientists looked at just about every well-conducted scientific study, but the Maryland authors took an oddly narrow, limited approach. And the Maryland authors didn't just include citations to scientific studies as part of their review — they also included citations to a number of activist Web sites. None of their risk assessments which did review all the solid science are cited.
By the way, the chairman of our Toxicology Research Task Group has counted 1,300 scientific studies on phthalates in his collection. He tells me that "a 'study' has an objective (hypothesis), outline of approach, methods, rationale, test system, and results. It is an investigation using a test system (which could be human if we are talking about an epidemiology study)." In grade school we called this the scientific method. So when we say that phthalates are among the most extensively studied families of chemicals in commerce today, that is a very studied statement.
Posted by Marian at 9:14 PM | Comments (0)