The current study estimated the contribution of job characteristics to socioeconomic inequalities in incidence of myocardial infarction (MI) during a 12-year follow-up period. Data were from the working population (aged 25-64 years) in the Netherlands longitudinal GLOBE study (N = 5757). Self-reported information was available from baseline measurement (in 199 1) for education, occupation, job demand, job control, fear of becoming unemployed, adverse physical working conditions, and smoking and alcohol use. Information on hospital admissions for MI among study participants was available until 2003, and was linked to baseline data via record linkage. Cox regression analyses were performed to estimate the hazard of MI in different socioeconomic groups before and after adjustment for job characteristics and health-related behaviours. Lower educated and manual workers had a higher risk of MI during follow-up, after adjusting for age, sex and marital status than higher educated and non-manual workers, respectively. After adjustment for occupation, the lowest educated still had an elevated risk of MI. After adjustment for education, no significant association of occupation with MI was observed. Job control and adverse physical working conditions were not significantly associated with MI after adjustment for socioeconomic position. These results suggest that the reduction of the socioeconomic position-MI association after adjustment for the two specific job characteristics reflect the effect of other unobserved factors closely related to both socioeconomic position and job characteristics. The results of this study point toward education as being the stronger predictor of hospital admitted MI, compared to occupational position and job characteristics, in the Dutch working population. (C) 2008 Elsevier Ltd. All rights reserved.