Major depression is a prevalent mental disorder in the working population. Improving the work environment may reduce the risk of major depression. The authors examined data from the longitudinal cohort of the Canadian National Population Health Survey from 1994-1995 to 2004-2005. Survey participants were classified into 4 groups by changes in job strain status from 1994-1995 to 2000-2001 (no change in low job strain, no change in high job strain, changing from high to low job strain, and changing from low to high job strain). The incidence proportion of major depressive episodes in each of the 4 groups was 4.0%, 8.0%, 4.4%, and 6.9%, respectively. Participants who reported a change from high to low job strain had a risk of major depression similar to those exposed to persistently low job strain. Among those exposed to persistent high job strain, only participants who reported good or excellent health at baseline had a higher risk of major depression, but those who reported fair or poor health did not. Reducing job strain may have positive impacts on the risk of depression. Self-rated health is a strong predictor of depression and plays an important role in the relation between job strain and depression.