OBJECTIVE To examine the relationship between smoking, heavy drinking and depression among U.S. middle-aged and older adults. METHODS Individual-level data came from 1992-2012 waves of the Health and Retirement Study. Smoking was ascertained from self-reported cigarette smoking status at the time of interview. Heavy drinking was defined as one or more drinks per day on average or four or more drinks on any occasion in the past three months for women, and two or more drinks per day on average or four or more drinks on any occasion in the past three months for men. Depression was defined as scoring three and above on the eight-item Center for Epidemiologic Studies Depression Scale. Cox proportional hazards regressions were performed to examine the relationship between smoking, heavy drinking and depression. RESULTS Compared to non-smokers, smokers free from depression and heavy drinking at baseline were 20% (95% confidence interval: 12-28%) and 34% (20-50%) more likely to develop depression and engage in heavy drinking during follow-up period, respectively. Compared to non-depressed participants, participants with depression who were nonsmokers and non-heavy drinkers at baseline were 41% (14-74%) and 18% (6-31%) more likely to smoke and engage in heavy drinking during follow-up, respectively. Compared to non-heavy drinkers, heavy drinkers who were nonsmokers at baseline were 60% (26-104%) more likely to smoke during follow-up. CONCLUSIONS Health promotion programs in midlife and older age should be mindful of the associations between smoking, heavy drinking and depression in order to improve intervention effectiveness.