Case et al22 conducted a prospective, cohort, multicenter Canadian-US study that followed up 1234 early post-MI patients for 1 to 4 years (mean, 2.1 years). Two psychosocial variables, living alone and disrupted marriage, were part of the risk model. Living alone independently predicted mortality, with an odds ratio of 1.5 (95% confidence interval, 1.0-2.9), but disrupted marriage was not a risk factor. The excess mortality associated with living alone showed a graded response, with incidence of cardiac death at 12.4%, 6.6%, and 4.4% for those living alone, those living with 1 person, and those living with more than 2 persons, respectively. The usual risk predictors of morbidity and mortality post-MI (function class, ejection fraction <40%, lower level of education, β-blocker nonuse, premature ventricular contractions, and a history of MI) remained essentially unchanged when living alone was factored into the model or left out, identifying it as an independent hazard of 1.58. This risk hazard is similar to 4 of the 6 risk predictors previously listed. This study did not identify the features of living alone that may be responsible for the excess hazard. It also showed that living with 1 vs 2 people does not provide an incremental mortality reduction. Marriage disruption was not a risk as long as cohabitation continued.