OBJECTIVETo compare the effects of dust exposure and smoking on mortality.

METHODSBased on the Guangzhou Occupational Health Surveillance Record System established in 1989-1992, 80 987 factory workers, aged > or = 30, occupationally exposed to dust and non-exposure to dust, were included in a prospective cohort study.

RESULTS(1) The mean age of the cohort was 43.5. Most subjects were workers, with secondary education, and married. Out of them 16.3% having ever exposed to dust, 43.7% ever smoked and 33.5% drank alcohol. (2) The cohort were traced for 8 years on average, but 35 were lost for follow-up. Malignant neoplasm was most commonly seen among 1539 deaths. (3) The adjusted relative risk (RRs) of death of all causes, malignant neoplasm and cerebro-cardiovascular diseases for dust exposed workers were close to those for smokers. However, the adjusted RRs of death of nasopharynx cancer and respiratory system diseases for dust exposed workers were higher than those for smokers. The adjusted RRs of death of lung cancer and stomach cancer for smokers were 2.2 times and 1.5 times of that for dust exposed workers respectively. Dust exposure combined with smoking caused significantly higher death RR. (4) In males, the adjusted RRs of death of all causes, malignant neoplasm and respiratory system diseases for silica dust exposed workers and the adjusted RR of death of cerebro-cardiovascular diseases for wood dust exposed workers were higher those in smokers. (5) The population attributable risk percentage (PARP) of all causes of death for smokers was 2.5 times of that for dust exposed workers. (6) In males, the risks of death for all causes, malignant neoplasm, lung cancer and stomach cancer increased with the amount of smoking per day and the duration of smoking. The risks of death caused by coronary heart disease and respiratory system diseases were increasing with the amount of smoking per day and the duration of smoking respectively.

CONCLUSIONSThe adjusted RRs for dust exposed workers were close to those for smokers with dust exposure and smoking having synergistic effects. Some of the adjusted RRs for smokers were higher than those for dust exposed workers PARP for smokers was higher than that for dust exposed workers. Our findings suggested that there was a significant dose-effect between smoking and risk of deaths.