The influence of occupational exposures on total mortality and respiratory mortality and morbidity was examined, employing a population specific Job Exposure Matrix (JEM). Moreover, the relationship between time-related variables of exposure to dust and chronic nonspecific lung disease (CNSLD) incidence was analysed, using time since first exposure and duration of exposure. Occupational exposures in the Zutphen cohort were assessed by application of a JEM, arbitrarily considering jobs as exposed when at least 10% of men who had held the job of interest reported an exposure to one or more from a list of 27 chemical agents. None of the exposures was related to mortality due to CNSLD, although results were influenced by the healthy-worker effect and low mortality rates. Exposure to wood dust and a high probability of exposure to dust were associated with total mortality. Exposures to dust and solvents were statistically significantly related to CNSLD incidence. An exposure-response relationship was found for the probability of exposure to dust with CNSLD incidence. Time-related estimates of exposure to dust based on work history were negatively related to CNSLD incidence. The results suggest the presence of a healthy-worker effect, in a general population study, resulting in an underestimation of the relationship between occupational exposures and respiratory diseases based on the evidence published so far. The use of the full work history to determine exposure to dust leads to stronger relationships with CNSLD incidence, compared to conventional analyses using exposure at the start of follow-up.