Despite the many studies on the socioeconomic gradient in health, our understanding of the underlying mechanisms remains incomplete This study used path analysis to explore the association between childhood socioeconomic disadvantage and adult health and the extent to which this association is mediated by health behaviours. The study used information from two cohorts, a central aim was to explore the extent to which the associations examined in the study changed over time. A sample of 825 people aged 15 to 20 in 1968 and 1981 was derived from the Swedish Level of Living Survey (LNU). In our examination of individual-level life-course data, we found evidence that disadvantage was transmitted from parents to children and from childhood to adulthood, at least in the later-born cohort. The results of our cohort analyses suggest that childhood socioeconomic disadvantage was not related to health or health behaviours in the same way in the two cohorts. Specifically, the social patterning of both health and health behaviours was stronger in the later-born cohort. The health behaviours included in the analyses were thus not equally important mediators of the association between childhood socioeconomic disadvantage and midlife health in the cohorts. Because health behaviours in adolescence did not directly affect midlife health in either of the two cohorts, the results also suggest that it is the accumulated effects of health behaviours that have the greatest influence on midlife health.