OBJECTIVE: To examine positive and negative social support and other selected social context variables (age, education, marital status, gender, and exposure to other smokers inside and outside the home) as predictors of smoking cessation in non-hospitalised adults with diagnosed cardiovascular disease at follow up after one, six, and 12 months. DESIGN: Discriminant function analyses (DFA) and longitudinal "lag" analyses. SUBJECTS: 137 Non-hospitalised adults with diagnosed cardiovascular health problems. RESULTS: Examination of the concurrent DFAs revealed significant univariate F ratios for the predictor variables of gender and marital status at one year and low negative support at all three follow ups. Quitters reported significantly lower levels of negative support than non-quitters over the course of the year and tended to be male and married. Longitudinal "lag" analyses, however, revealed that higher positive social support at one month and higher negative support at six months were both predictive of smoking cessation at one year. At one year more men than women and more married than not married smokers were successful in quitting. No effects for age, education, or exposure to others smoking inside or outside the home were found on any of the concurrent DFAs or longitudinal analyses. CONCLUSION: A series of concurrent DFAs revealed that positive support was a significant predictor of quitting at one year and negative support was predictive of not quitting at all three follow ups. Longitudinal "lag" analyses showed that positive support at one month and negative support at six months both predicted quitting at one year. Being male and married were found to contribute to quitting on both sets of analyses. The effects for positive and negative support on the smoking behaviour of adults with cardiovascular disease tended to change over the course of a year. These findings suggest that positive and negative social support may have differential effects over time. As the smoker moves along the "quitting trajectory" it may be that more "nagging" or negative interactions are needed at some point to get smokers to quit, if positive support has not worked or is not working. Progression of disease also may have served as a stimulus for family members and friends to become more insistent and negative about the person's continued smoking. More research is needed to examine the quitting process to determine which and how social context variables contribute.