Several studies have associated heart disease with job strain, defined as low job control and high job demands. We have studied incident heart disease (519 cases) and job strain among 3,575 males in NHANES1 survey who were currently employed at baseline in the early 1970s, and followed through 1987. Scores for job control and job demands were assigned to each subject based on current occupation at baseline. Controlling for conventional risk factors, we found no excess risk for those with the highest strain (lowest control and highest demands, rate ratio 1.08). Those with highest job control did have significantly decreased risk (rate ratio 0.71, 95% CI 0.54-0.93). In blue-collar workers (58% of subjects) there was a significant inverse trend in risk with increasing job demands. Control for level of physical activity did not change this finding. A combination of high control and demand was protective among blue-collar workers (odds ratio 0.69, 0.48-0.99). Our findings suggest that class-specific analyses are needed in studying job stress, and that "active" blue-collar workers with high control and high demand are protected against heart disease. The "job demand" variable may measure whether work is challenging rather than fast-paced. Our findings are limited by the use of assigned job scores based on job title.