Delzell/Herrick study on IBM microprocessor workers

There had been a controversy with Elsevier's journal Clinics in Occupational and Environmental Medicine not publishing an article by Richard Clapp on cancer rates among IBM microprocessor workers.

Now a study ordered by IBM reports opposite conclusions to those of Boston University's Clapp. Underlying the different results is a question of the appropriate baseline for comparison, an issue which also is found in the Vioxx discussion.


from The Scientist (Nov. 9, 2004):

An IBM-commissioned study has found that employees at the company's microprocessor manufacturing plants had a lower rate of cancer than would be expected, contradicting results of a yet-unpublished study that found higher rates of cancer among workers and is at the root of a controversy involving the journal that was to publish it.

"Overall mortality among IBMers at the study locations was 35% lower than the comparison populations," Martin Sepulveda, IBM's vice president for occupational medicine, wrote in a company-wide employee memo. "Cancer among IBM employees was 16% lower."

The findings fly in the face those of another study, by Richard Clapp, at Boston University, who used IBM's employee dataset and concluded that there were higher mortality and cancer rates at plant facilities. The judge forbade Clapp from testifying during a recently resolved court case in California between IBM and former employees diagnosed with cancer. Because of the case, no one has seen the study.