Background-The impact of QRS morphology and duration on the effectiveness of cardiac resynchronization therapy (CRT) has been usually assessed separately. The interaction between these 2 simple ECG parameters and their effect on CRT has not been systematically assessed in a large-scale clinical trial. Methods and Results-The Resynchronization-Defibrillation for Ambulatory Heart Failure Trial showed that implantable cardioverter defibrillator-CRT was associated with a significant reduction in the primary end point of all-cause mortality or heart failure hospitalization. For this substudy, we excluded patients in atrial fibrillation and those with a previous pacemaker. All baseline ECGs were reviewed by a panel of 3 experienced electrocardiographers. A total of 1483 patients were included in this study. Of these, 1175 had left bundle-branch block (LBBB) and 308 had non-LBBB. In patients with LBBB receiving implantable cardioverter defibrillator-CRT, there was a reduction in the primary outcome and in each individual component of the primary outcome. Furthermore, there was continuous relationship between QRS duration and extent of benefit. In patients with non-LBBB and QRS ≥160 ms, the hazard ratio for the primary outcome was 0.52 (0.29-0.96; P=0.033); in patients with QRS <160 ms, the hazard ratio was 1.38 (0.88-2.14; P=0.155). Conclusions-In patients with LBBB, there was a continuous relationship between broader QRS and greater benefit from implantable cardioverter defibrillator-CRT. However, our data do not support the use of implantable cardioverter defibrillator-CRT in patients with non-LBBB, especially when the QRS duration is <160 ms. There may be some delayed benefit when the QRS is ≥160 ms, but this needs further investigation. Clinical Trial Registration-URL: http://www.clinicaltrials.gov. Unique identifier: NCT00251251.