Mortality and cancer incidence in workers in two Australian prebake aluminium smelters

Malcolm Ross Sim ^1*, Anthony Del Monaco ¹, Jan Lucas Hoving ², Ewan MacFarlane ¹, Dean McKenzie ¹, Geza Benke ¹, Nicholas de Klerk ³ and Lin Fritschi ⁴

¹ Monash University, Australia
² University of Amsterdam, Netherlands
³ Institute for Child Health Research, Australia
⁴ Western Australia Institute for Medical Research, Australia

^* To whom correspondence should be addressed. E-mail: malcolm.sim{at}med.monash.edu.au.

Accepted 30 January 2009

Abstract

Background: Exposures in the aluminium production industry have been declining and this study aimed to investigate cancer and mortality in two Australian pre-bake aluminium smelters.

Methods: The cohort included 4396 males who had worked in a pre-bake smelter for at least three months. They were matched against the National Death Index and the National Cancer Statistics Clearing House to obtain cause of death and type of cancer. Standardised Mortality Ratios (SMR) and Standardised Incidence Ratios (SIR) were calculated for the whole cohort, for production or maintenance work categories and for duration of employment categories.

Results: The SMRs for all causes, circulatory, respiratory and injury deaths were at or below expected. Mesothelioma was the only significantly increased cause of death (SMR 3.52, 95% CI 1.46, 8.46). Death from prostate cancer in production workers was elevated (SMR 2.39, 95% CI 1.29, 4.44) and in those who had worked for more than 20 years in production or maintenance jobs (SMR 3.67, 95% CI 1.53, 8.81). There were statistically significant excesses of incident stomach cancer, mesothelioma, and kidney cancer cases, while the SIR for melanoma was significantly reduced. There was no significant trend for duration of exposure for any type of incident cancer.

Conclusions: This study found no overall excess of mortality or cancer, but incident mesothelioma and kidney cancer risks were elevated. The lack of excess risk for lung or bladder cancer or deaths from respiratory disease may be related to the different level and pattern of exposure between Søderberg and pre-bake smelters.