OBJECTIVE To examine the mortality experience of 4221 employees from 1973 to 1999 and the illness absence patterns for 2203 employees from 1990 to 1999 of a chemical and refinery facility in Louisiana. METHODS Mortality and illness absence data were extracted from the Shell Oil Company's health surveillance system (HSS). The standardised mortality ratio was used as a measure of mortality risk. Morbidity frequency and duration of absence were calculated by age, sex, and four health risk factors (cigarette smoking, high blood pressure, hypercholesterolaemia, and obesity). RESULTS Male employees experienced a significant deficit in mortality for all causes of death, all cancers, lung cancer, heart disease, and respiratory disease compared with the corresponding US population. Brain cancer was non-significantly increased, with six observed and five expected deaths; mortality from leukaemia was consistently lower than expected. The majority of employees had no illness absences of six days or longer during the 10 year study period. The loss of productivity (in terms of days of absence) was greater for employees with health risk factors. Ever smoking male employees had a 79% increase of heart disease and more than 50% higher rates of respiratory disease and musculoskeletal disorders compared to non-smokers. Smokers were absent 2.9 and 1.6 more days than non-smokers and ex-smokers, respectively. CONCLUSIONS Regardless of the comparison population, significantly fewer deaths were seen for all causes combined, all cancers, lung cancer, heart disease, or non-malignant respiratory disease. Illness absence rates and duration were higher among employees with health risk factors.