Study type: Epidemiological study (observational study)

Incidence of selected cancers in Swedish railway workers, 1961-79. epidem.

Published in: Cancer Causes Control 1994; 5 (2): 189-194

Aim of study (acc. to author)

The incidence of leukemia, lymphoma, tumors of the brain, breast, and the pituitary gland was investigated among railway workers in Sweden.

Further details

The present study is a re-analysis of the study of Floderus et al (1993) in which the 1961-1979 incidence data showed no increase in risk for leukemia and brain tumors for railway workers. In the present study, follow-up was divided into two 10-year periods 1961-1969 and 1970-1979 because the structural change in the railway industry could mean that a large proportion of subjects in the exposure groups was exposed to high levels of EMF only during the very first period, and that this could dilute a potential effect.

Endpoint/type of risk estimation

Type of risk estimation:


Exposure groups

Group Description
Group 1 engine drivers
Group 2 conductors
Group 3 engine drivers and conductors
Group 4 railway workers (station masters, train dispatcher, railroad assistants and linemen)
Group 5 workers in the railway industry



17,150,940 person-years in 1961-1969 and 19,056,600 person-years in 1970-1979

Conclusion (acc. to author)

Overall, there seem to be elevated estimates for the exposure groups during the first decade, which are not seen during the second period of follow-up.
For the first decade 1961-1969, engine drivers and conductors combined had an relative risk (RR) of 1.9 (CI 0.9-4.0) for chronic lymphocytic leukemia, of 1.4 (CI 0.4-4.3) for acute myeloid leukemia, and 1.0 (CI 0.5-1.9) for lymphoma. For all brain tumors, the RR was 1.2 (CI 0.8-1.9), with a higher risk estimate for those below age 30 (RR 12.2, CI 2.8-52.5). Three cases of breast cancer and nine cases of tumors of the pituitary gland occurred among engine drivers and conductors (RR 4.9, CI 1.6-11.8; RR 3.2, CI 1.6-6.2, respectively).
The authors conclude that the results give some support to the hypothesis of an association between EMF and certain types of cancers in railway workers.

Limitations (acc. to author)

Occupation was only defined at baseline in 1960, therefore a potential occupational change is not incorporated.

Study funded by

Related articles