Cancer in the Semiconductor Industry

Abstract

THE POSSIBLE ASSOCIATION between chemical exposures and brain cancer is receiving renewed interest in light of the recently published results of the United Kingdom’s Health and Safety Executive (HSE) investigation of cancer among current and former workers of the National Semiconductor, Ltd., Facility (NSUK) in Creenock, Scotland.’ The HSE investigators found not only a higher than expected incidence of breast, lung, and stomach cancers among female workers, but approximately 4 times as many brain cancer deaths in males as expected, on the basis of comparisons with age and sex-specific mortality rates for Scotland. Although recognizing the need for additional, broader cancer studies among workers in the semiconductor industry, the HSE investigators took a questionable stance on the brain cancer findings: “In view of the fact that brain cancer was not of specific interest at the outset of the investigation and the short latency for 3 of the 4 cases, it i s most probably not work-related,” the authors wrote.* The statement is worrisome, suggesting that if the HSE does not initially suspect a particular type of cancer as being work related, then indications to the contrary may be dismissed. It is especially worrisome when the cancer in question has been associated with exposure to toxic chemicals since the mid-I 970s3 and when several of the associated chemicals and agents-including the organic solvents trichloroethane and trichloroethylene, and both ionizing and nonionizing radiation-are found in Table 1 of the HSE report, in which the known or suspected carcinogens are listed that had been used or had been present at the NSUK Creenock plant since operations began in 1970.4 In fact, a review of the epidemiological literature over the past 2 decades-which, as the HSE investigators note, is generally based on electronics manufacturing industries-gives reason for one to suspect that the increased risk of brain cancer among NSUK workers is work related. The HSE investigators argue that because exposures in electronics assembly work are not identical to those in semiconductor manufacturing, “It would be unwise to draw any conclusions about the semiconductor industry from more broadly based studies.” However, as was noted earlier, several of the suspected carcinogens associated with brain cancer in these broader studies are as common, if not more so, in the semiconductor industry. At the very least, the studies should caution against concluding that the excess of brain cancer at NSUK is not work related. In 1983, the first (and only) evaluation of the general cancer incidence pattern in the electronics industry-as opposed to a study of a particular cancer or of a subpopulation, such as workers of a particular company or gender-was conducted in Sweden by linking the Swedish Cancer Registry with Swedish census records for the period 1961-1973.5 Of more than 75,000 subjects, the investigators found an increased risk of cancer of 1.1 5 times for men and 1.08 for women who were employed in the electronics manufacturing industry. That may not sound profound, but keep in mind the size of the study. More than 75,000 subjects were included and were taken from all sectors of the industry-from manufacturing to administrative to sales. Listen to the authors’ own cautionary statement: “An estimated slight excess risk [of cancer], referring to the electronics industry as a whole, could reflect some hazardous practice of a more severe type in some sectors of the industry. Since the registry does not include any specific exposure data, risk estimates should be taken as starting points for further inquiry, focusing on particular features of the work environment.”6 Two years later, Milham7 looked for links between cancer and exposure to electromagnetic fields (EMFs) by linking the death records with occupational codes of males in Washington State. He found an increased risk of death from lung, pancreas, kidney, and brain cancer “usually greatest in those occupations which have [chemical] inhalation exposures in addition to EMF [electromagnetic field] exposure^."^ In that same year, Lin et aL8 conducted a study and looked specifically at brain tumors among electrical workers. They found an excess death rate from brain cancers in their data set. “It is not known,” the authors concluded, ”whether the increased risk of brain tumors observed among electrical workers is due to the magnetic or electric fields themselves or possibly to a common chemical exposure, for example polychlorinated bi p hen y Is, organic solvents, or metal fumes.”8