The benefits of leisure time physical activity (LTPA) in cardiovascular prevention are well established. While cardiac rehabilitation programmes have been demonstrated as improving myocardial infarction (MI) prognosis, the strength of the association between LTPA and post-MI survival has yet to be quantified. We evaluated long-term survival after MI of inactive, irregularly active, and regularly active patients and examined trajectories of LTPA and their relationship to mortality risk. Consecutive patients aged ≤65 years (n = 1,521), discharged from 8 hospitals in central Israel after first MI in 1992-1993, were followed through 2005. Extensive clinical and sociodemographic data, including self-reported LTPA habits, were obtained at baseline and at 4 subsequent interviews. Pre-MI inactive patients (54%) had lower socioeconomic status, higher prevalence of risk factors and comorbidities and more severe MI. The point prevalence rate of regular LTPA at all follow-up interviews was approximately 40% and 18% were regularly active throughout the entire follow-up. Over a median follow-up of 13.2 years, 427 deaths occurred. After multivariable adjustment, no association was observed between pre-MI LTPA and death. However, with LTPA categories modelled as time-dependent variables, providing an estimation of cumulative assessment and accounting for changes in LTPA post-MI, a strong inverse graded association was revealed (multivariable-adjusted hazard ratios, 0.56 [95% CI: 0.42-0.74] for regular and 0.71 [95% CI: 0.54-0.95] for irregular activity vs. none). Similar estimates were obtained among pre-MI sedentary patients. In summary, after MI, regularly active patients had about half the risk of dying compared with inactive patients, irrespective of pre-MI habits.