Gustavsson et al. (2001) conducted a case-control study of myocardial infarction among men and women (45–70 years old) in Stockholm, Sweden. Strict diagnostic criteria and a population-based design were used to identify the first, nonfatal myocardial infarction. A detailed occupational history was obtained with a questionnaire, which was followed by a sophisticated job-exposure matrix to provide quantitative exposure intensity of motor-vehicle exhaust as assessed with carbon monoxide (CO) and other combustion products as assessed with respirable particles. Exposures were expressed in terms of highest intensity during at least 1 year of work and cumulative exposure. Exposures were presumably higher in earlier decades, and time in occupation contributed to the cumulative-exposure estimate. Information on potential confounders (age, sex, smoking, hypertension, obesity, and diabetes), was obtained and incorporated into the analysis. It is not clear whether SES was used in the analysis, considering that referents were somewhat more likely to be of higher SES (25% of referents vs 19% of cases) and cases were substantially more likely to be manual workers (34% of cases vs 25% of referents). Fairly consistent effects were observed for combustion products, and there was evidence of a dose-response relationship. Myocardial infarction was, after adjustment, increased in the combustion-product group with high intensity during at least 1 year of occupational exposure (RR 2.11, 95% CI 1.23–3.60) and with intermediate intensity (RR 1.42, 95% CI 1.05–1.92). The analysis of the exposure-response trend generated an RR of 1.24, 95% CI 1.07–1.51. An exposure-response trend was also found for the other exposure category, cumulative exposure (in milligrams of respirable particles per cubic meter per year). The group with high cumulative exposure to combustion products had RR 1.35, 95% CI 1.02–1.79, and the group with intermediate cumulative exposure had RR 1.22, 95% CI 0.91–1.64. For the motor-vehicle exhaust group (which included many of the same subjects as the combustion-products group), myocardial infarction was not increased with high exposure (after adjustment), but was increased with intermediate exposure (RR 1.32, 95% CI 1.01–1.73). The trend analysis for the motor-vehicle exhaust group was not noteworthy. Examples of high combustion-product exposure referred to exposures at the threshold limit value. Examples of occupations with high combustion-product exposure were ship engine room crew, firefighters, engineers and technicians in energy production, chimney sweeps, and blacksmiths.

Mortality was studied in a subset of the large ACS cohort to assess occupational exposure to diesel exhaust (Boffetta et al. 1988). The study, initiated in 1984, enrolled 1.2 million people. The focus was on men (40–79 years old) and used a crude job-exposure matrix (based on 2-digit standard industrial classification codes) to determine exposure to diesel exhaust. Men with and without exposure to diesel exhaust were compared. Exposure was not associated with an increased risk of ischemic heart disease (RR 0.98) on the basis of 398 deaths, but there was an increased risk of cerebrovascular disease (codes 430–438: RR 1.61, p<0.05) on the basis of 62 deaths, and of “arteriosclerosis” a minor ICD category (code 440: RR 3.12, p<0.05) on the basis of 10 deaths. However, mortality attributed on death certificates to arteriosclerosis is nonspecific and thus difficult to interpret.

Alfredsson et al. (1993) performed an incidence and mortality study of myocardial infarction among bus drivers in Sweden over a 15-year period. It relied on two population-based registries—one of first acute myocardial-infarction hospitalizations and another of all deaths. Registries were linked to census records of occupation. Comparing male bus drivers with people in other occupations, the study found a higher incidence of myocardial infarction in bus drivers (RR 1.4, 95% CI 1.1–1.8, adjusted for age, calendar year, county, and SES); the effect appeared