BACKGROUND: Previous literature has explored the relationship between television viewing and cardiovascular disease (CVD) in adults; however, there remains a paucity of longitudinal data describing how young adult television viewing relates to premature CVD events. OBJECTIVE: To ascertain the relationship between level and annualized changes in television viewing from young adulthood to middle age and the incidence of premature CVD events before age 60. DESIGN: The Coronary Artery Risk Development in Young Adults (CARDIA) study, a prospective community-based cohort with over 30 years of follow-up (1985-present). PARTICIPANTS: Black and White men and women who were 18-30 years old at baseline (1985-1986). MAIN MEASURES: Independent variables: Individualized television viewing trajectories were developed using linear mixed models. DEPENDENT VARIABLES: Fatal and nonfatal coronary heart disease (CHD), heart failure, and stroke outcomes were analyzed separately and as a combined CVD event outcome. KEY RESULTS: Among 4318 included participants, every 1-h increase in daily hours of television viewing at age 23 was associated with higher odds of incident CHD (adjusted odds ratio [AOR] 1.26, 95% confidence interval [CI] 1.06-1.49) and incident CVD events (AOR 1.16, 95% CI 1.03-1.32). Each additional hour of daily television viewing annually was associated with higher annual odds of CHD incidence (AOR 1.55, 95% CI 1.06-2.25), stroke incidence (AOR 1.58, 95% CI 1.02-2.46), and CVD incidence (AOR 1.32, 95% CI 1.03-1.69). Race and sex modified the association between television viewing level at age 23 and CHD, heart failure, and stroke, with White men most consistently having significant associations. CONCLUSIONS: In this prospective cohort study, greater television viewing in young adulthood and annual increases in television viewing across midlife were associated with incident premature CVD events, particularly CHD. Young adulthood as well as behaviors across midlife may be important periods to promote healthy television viewing behavior patterns.