Non-ST-segment elevation acute coronary syndromes include unstable angina and non-ST-segment elevation myocardial infarction. Most randomized controlled trials of routine versus selective invasive coronary angiography have high rates of crossover from control to intervention groups and do not include subgroup analysis for unstable angina. Consequently, no clear, specific recommendations exist regarding the use of angiography in unstable angina.To assess the effect of angiography on mortality in unstable angina, incorporating the results of additional cardiac procedures and events.Longitudinal study using hospital discharge data, discrete-time survival analysis with propensity score adjustment, and sensitivity analysis.Victoria, Australia, 2001 to 2011.All residents, all ages.Routine invasive coronary angiography.12-month all-cause mortality.Emergently admitted patients with unstable angina (n = 33 901) who did or did not receive angiography during their first hospitalization were balanced on 44 covariates of propensity score. Routine angiography was associated with a 52% decrease in 12-month mortality (hazard ratio, 0.48 [95% CI, 0.38 to 0.61]); revascularization offered no additional statistical mortality benefit compared with diagnostic angiography alone. The predicted cumulative probability of death at 12 months was 0.024 (CI, 0.021 to 0.027) for patients receiving angiography within 2 months of their index unstable angina versus 0.097 (CI, 0.090 to 0.105) for those not receiving it. Sensitivity analysis demonstrated that to negate the observed effect size, an unmeasured confounder must independently decrease mortality by 90% and have a prevalence gap of 15% or greater between the angiographic groups.Nonrandom allocation of angiography.Patients with unstable angina benefit from an invasive management pathway initiated by invasive coronary angiography during their hospitalization and up to 2 months after discharge.National Health and Medical Research Council, Australia and BUPA Health Foundation.