Five years ago Rohm & Haas, a chemical company in Montgomery County, PA, alerted employees about a mysterious series of brain cancers. Now Rohm & Haas says its latest study has turned up no problems. But an epidemiology expert witness from Columbia University has concluded that even 12 brain cancer deaths at the Spring House research center “significantly exceeds the expected number.” The expected number is 3.45 cases per 100,000 people, expert witness Richard Neugebauer wrote in his analysis. Philly.com also reports:
He estimated that brain cancer deaths among Spring House workers were at least three times more than expected, and possibly eight times more, depending on the size of the work force.
“Their own panel is saying ‘We don’t think you got all the deaths; something doesn’t look right,’ ” Freiwald said, referring to the critique by the three outside experts. “Epidemiology is a science where it’s very easy to turn it into a shell game.”
Proving the existence of a “cancer cluster” is notoriously difficult. Even when the number of cases is high enough to be what scientists consider “statistically significant” – and it rarely is – linking it to a chemical, building material, or other exposure is like nailing jello.