STUDY DESIGN: This study is a re-evaluation of data previously published that had shown no excess number of workers at risk for leukemia in the Swedish population. Occupation in 1960 was used as occupation with no adjustment for changes since that time. Railway workers from this initial study were compared with the population at large to see if cancer rates differed for those exposed to higher magnetic fields in their jobs than for those who may have experienced lower fields. The study population included all men that were 20-64 years of age and employed in 1960. Two subgroups were examined because of a 50% decrease in the number of workers in railway occupations in the decade after 1960: 1961-1969 and 1970-1979. Such a significant change could impact the study outcome because presumably fewer workers would be exposed in the second decade of work. Job categories included engine drivers, conductors, railway workers, and linemen. The exposure assessment was obtained by using the measurement data collected in the previous study. Average daily means and medians of the magnetic field density were assigned to respective occupation codes. Cancer data were obtained from the Cancer Environment Registry.
CONCLUSIONS: Railway workers experienced higher relative risk estimates for exposure groups during the 1961- 1969 period of follow-up when compared to the 1970-1979 period. Statistically significant increases were observed for engine drivers for chronic lymphocytic leukemia (Odds Ratio[OR]=2.7, 95% Confidence Interval[CI]=1.0-7.4) and for breast cancer (OR=8.3, 95% CI=2.0-34.3). See Glossary. Pituitary gland tumors were increased among engine drivers and conductors. While results on chronic lymphocytic leukemia and breast cancer confirm previous findings, the observations for pituitary gland cancers, if real, are new. These new findings have important implications for potential pathways associated with EMF exposure and cancer.
LIMITATIONS: The precision of exposure estimates is limited, which may have resulted in misclassification. The exposure assessment was only an approximation because it was based on occupation recorded from the 1960 census data which may differ from occupation of longest exposure.