Gw. Olsen et al., MORTALITY FROM PANCREATIC AND LYMPHOPOIETIC CANCER AMONG WORKERS IN ETHYLENE AND PROPYLENE CHLOROHYDRIN PRODUCTION, Occupational and environmental medicine, 54(8), 1997, pp. 592-598
Objectives-A previous study reported a fivefold increase in mortality
from pancreatic cancer and a threefold increase in lymphopoietic and h
aematopoietic cancer among 278 men who were assigned to a now dismantl
ed Union Carbide chlorohydrin unit in the Kanawha Valley of West Virgi
nia. There were also significant trends with duration of employment. T
he purpose of this study was to determine whether a comparable increas
ed risk in mortality from pancreatic cancer and lymphopoietic and haem
atopoietic cancer occurred among male employees assigned to the Dow Ch
emical Company's ethylene and propylene chlorohydrin production proces
ses. Methods-The cohort consisted of 1361 male employees who worked at
the company's Freeport, Texas, Plaquemine, Louisiana or Midland, Mich
igan plants. Subjects were considered to have had a minimum of 30 days
of workplace experience in 1940-92, in the ethylene chlorohydrin and
propylene chlorohydrin process areas. These process areas were located
within the ethylene oxide and propylene oxide production plants. A to
tal of 300 deaths was observed to 31 December 1992. Results-The standa
rdised mortality ratio (SMR) for all malignant neoplasms was 94 (95% C
I 74 to 118). There was one pancreatic cancer death compared with 4.0
expected (SMR 25, 95% CI 1 to 140). There were 10 lymphopoietic and ha
ematopoietic cancer deaths compared with 7.7 expected (SMR 129, 95% CI
62 to 238). Additional analyses, which examined location, production
process, duration of employment, and a 25 year induction latency perio
d, were not significant. Conclusions-The results provide some assuranc
e that the Dow Chemical cohort, to date, has not experienced increased
risks of pancreatic cancer and lymphopoietic and haematopoietic cance
r as previously reported in a different cohort of chlorohydrin workers
. Possible reasons are discussed for the inconsistent findings between
the two cohorts.