This study tested the hypothesis that prolonged QRS duration independently predicts long-term mortality in patients who underwent risk stratification and treatment for ventricular arrhythmias. Patients who underwent risk stratification by electrophysiologic study were identified. Electrophysiologic study results were defined as positive if sustained monomorphic ventricular tachycardia was induced. Mortality was the primary end point. Of 915 patients studied, mean left ventricular (LV) ejection fraction (EF) was 35.3 +/- 15.7%, 608 (66.4%) had coronary artery disease, 233 (25.5%) had positive electrophysiologic study findings, 298 (32.6%) received implantable cardioverter-defibrillators, and 174 (19%) died (mean follow-up 35.0 +/- 15.0 months). Cox regression analysis identified older age, coronary artery disease, digoxin use, absence of beta blockers, lower LVEF, and prolonged QRS duration to be independent predictors of mortality. QRS duration > or =130 ms, present in 33.6% of patients, was associated with a twofold increase in mortality (hazard ratio 2.1, 95% confidence interval 1.5 to 2.8; p <0.0001). For every 10 ms increase in QRS duration, mortality rate increased 10%. In a subgroup of patients with coronary artery disease and LVEF < or =30%, prolonged QRS duration remained an independent predictor of mortality (hazard ratio 2.6, 95% confidence interval 1.6 to 4.2; p <0.0001). Thus, prolonged QRS duration is a strong independent marker of long-term mortality in patients who undergo risk stratification for ventricular arrhythmias. Whether QRS duration represents only a marker for mortality or if modification of this factor using resynchronization therapies will impact mortality merits further study.